[ { "id": "1-hive", "title": "1Hive", "domain": "governance", "domain_name": "Governance", "definition": "A decentralised community and DAO that issues the Honey community currency and develops governance tooling such as the Gardens framework for conviction voting.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:1-hive", "labels": [ "1Hive" ], "is_subclass_of": [ "Decentralized Autonomous Organization" ], "wikilinks": [ "Decentralized Autonomous Organization", "Governance Token", "Quadratic Funding", "Liquid Democracy" ] }, { "id": "1-inch", "title": "1inch", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised exchange aggregator that routes trades across multiple liquidity sources to find favourable execution prices on Ethereum and other chains.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:1-inch", "labels": [ "1inch" ], "is_subclass_of": [ "Decentralized Exchange" ], "wikilinks": [ "Automated Market Maker", "Smart Contract", "Ethereum", "Liquidity Pool", "Uniswap", "Decentralized Exchange" ] }, { "id": "2-d-li-dar", "title": "2D LiDAR", "domain": "robotics", "domain_name": "Robotics", "definition": "2D LiDAR is a laser scanning sensor that emits a rotating beam in a single horizontal or vertical plane, producing a planar point cloud used for obstacle detection, proximity sensing, and 2D mapping. It is widely deployed on mobile robots and autonomous guided vehicles where full 3D sensing is unnecessary or cost-prohibitive.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:2-d-li-dar", "labels": [ "2D LiDAR" ], "is_subclass_of": [ "Perception and Sensing", "Lidar" ], "wikilinks": [ "Lidar", "Robotics" ] }, { "id": "360-video", "title": "360 Video", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A spherical video format captured with omnidirectional camera rigs that records the full 360-degree field of view around a single vantage point, allowing viewers to look in any direction during playback on headsets, browsers, or mobile devices. Because the footage is fixed to the capture position, it offers rotational (three-degrees-of-freedom) immersion but no positional movement, distinguishing it from volumetric formats that reconstruct scene geometry.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:360-video", "labels": [ "360 Video", "360-Degree Video" ], "is_subclass_of": [ "Immersive Media" ], "wikilinks": [ "Immersive Media", "VR Experiences", "Volumetric Video", "Virtual Reality" ] }, { "id": "3-d-animation", "title": "3D Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Animation is the discipline of creating motion in three-dimensional digital environments using techniques such as keyframe animation, motion capture, and procedural simulation. It underpins character movement, environmental dynamics, and cinematic sequences in spatial computing platforms, game engines, and metaverse applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:3-d-animation", "labels": [ "3D Animation", "3DAnimation" ], "is_subclass_of": [ "Animation Technique" ], "wikilinks": [ "ISO/IEC JTC 1/SC 24" ] }, { "id": "3d-asset-creation", "title": "3D Asset Creation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The production of three-dimensional digital assets\u2014models, characters, environments, props, and their materials, textures, rigs, and animations\u2014ready for use in games, film, extended reality, simulation, and digital-twin applications. 3D asset creation is the spatial specialisation of general asset creation, spanning manual modelling and sculpting, photogrammetry and scanning, procedural generation, and increasingly generative AI pipelines that produce textured meshes from text or image prompts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:3d-asset-creation", "labels": [ "3D Asset Creation" ], "is_subclass_of": [ "Asset Creation" ], "wikilinks": [ "Asset Creation", "3D Modelling", "Creative Tools" ] }, { "id": "3-d-asset-standard", "title": "3D Asset Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D Asset Standard is a formal specification defining the file formats, metadata schemas, coordinate systems, material models, animation encodings, and interoperability protocols required for consistent creation, exchange, and real-time rendering of three-dimensional digital content across diverse software platforms and runtime environments. Such standards normalise vertex data structures, physically based rendering (PBR) material pipelines, level-of-detail hierarchies, skeletal animation rigs, and compression algorithms so that assets created in one authoring tool function correctly in another without manual conversion or data loss. Prominent examples include the Khronos Group glTF 2.0 specification for web and mobile 3D delivery, Pixar's Universal Scene Description (USD/USDZ) for complex scene composition, and the Metaverse Standards Forum interoperability profiles. These specifications underpin spatial computing, extended reality, game engines, digital-twin platforms, and emerging metaverse ecosystems by ensuring deterministic visual fidelity and semantic asset identity across the entire content pipeline.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:3-d-asset-standard", "labels": [ "3D Asset Standard", "3DAssetStandard" ], "is_subclass_of": [ "Interoperability Standard" ], "wikilinks": [ "ISO/IEC 14496-16" ] }, { "id": "3-d-asset", "title": "3D Asset", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D Asset is a discrete, self-contained digital artefact encoding three-dimensional geometry, surface materials, skeletal animation rigs, and associated metadata in a machine-readable interchange format such as glTF 2.0, USD, FBX, or OBJ. Such assets are produced by digital content creation tools, stored in asset management systems with versioning and provenance records, and consumed at runtime by rendering engines, game engines, and spatial computing platforms. Production-quality 3D assets typically incorporate multiple levels of detail (LOD), physically-based rendering (PBR) material maps (albedo, metallic-roughness, normal, occlusion, emissive), collision meshes for physics simulation, and compressed texture atlases optimised for a target hardware envelope. Cross-platform delivery relies on open standards maintained by the Khronos Group and Pixar/Academy Software Foundation, enabling asset reuse across metaverse environments, digital twins, extended reality (XR) applications, and game titles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:3-d-asset", "labels": [ "3D Asset", "3D Asset Dataset", "3DAsset" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [ "W3C" ] }, { "id": "3-d-content-creation", "title": "3D Content Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Content Creation encompasses the end-to-end process of designing, modelling, texturing, rigging, and rendering three-dimensional digital assets and environments for use in games, film, XR experiences, and digital twins. It spans artistic and technical disciplines ranging from polygon modelling and UV unwrapping to physically based material authoring and real-time rendering optimisation. The discipline bridges human creative intent and the technical pipelines that deliver interactive or cinematic experiences.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:3-d-content-creation", "labels": [ "3D Content Creation" ], "is_subclass_of": [ "Content Creation" ], "wikilinks": [] }, { "id": "3-d-content-generation", "title": "3D Content Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The computational process of producing three-dimensional digital content through algorithmic and AI-driven methods, encompassing procedural generation, neural rendering, NeRF-based scene synthesis, and GAN-based 3D model creation. Applications span game development, virtual reality, digital twins, and automated CAD, enabling scalable production of photorealistic 3D assets from text or image inputs.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:3-d-content-generation", "labels": [ "3D Content Generation", "3D Generation", "Generative 3D" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Computer Graphics", "Autonomous Robot", "Digital Asset", "Generative AI", "Neural Rendering", "Procedural Generation" ] }, { "id": "3-d-content-pipeline", "title": "3D Content Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The comprehensive workflows, tools, and methodologies for creating, processing, optimising, and delivering three-dimensional assets for metaverse and spatial computing applications. A 3D content pipeline spans concept art, geometry modelling, UV mapping, PBR texturing, rigging, LOD generation, and platform-specific optimisation, integrating DCC tools, game engines, version control, and CI/CD systems to support distributed teams and continuous asset delivery.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:3-d-content-pipeline", "labels": [ "3D Content Pipeline", "3DContentPipeline" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "3-d-design", "title": "3D Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "3D Design is the creative and technical discipline of conceiving, modelling, and refining three-dimensional digital objects and environments for interactive, immersive, or real-time applications. It encompasses spatial composition, form language, and visual hierarchy adapted for game engines, virtual reality, and metaverse platforms, requiring performance-conscious workflows such as level-of-detail strategies, UV unwrapping, and real-time shader design.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:3-d-design", "labels": [ "3D Design", "3DDesign" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "3-d-development", "title": "3D Development", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Development encompasses the modologies, tools, and workflows for creating, iterating, and deploying three-dimensional digital assets and environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:3-d-development", "labels": [ "3D Development" ], "is_subclass_of": [ "Platform and Environment", "Software Development" ], "wikilinks": [ "Game Development", "Metaverse Creation", "Software Development", "Virtual Environment Design", "XR Applications", "3D Graphics Standard", "3D Modeling", "3D Rendering Engine", "Computer Vision", "Game Engine" ] }, { "id": "3-d-engine", "title": "3D Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D engine is software that handles the representation, rendering, physics simulation, and animation of three-dimensional scenes, providing an abstraction layer over GPU hardware and platform-specific graphics APIs so developers can build interactive applications and simulations without writing low-level graphics code.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:3-d-engine", "labels": [ "3D Engine" ], "is_subclass_of": [ "Game Engine" ], "wikilinks": [ "Computational Geometry", "3D Engine", "Physics Simulation Engine", "Game Engine", "https://en.wikipedia.org/wiki/Game_engine", "https://docs.unity3d.com/Manual/index.html" ] }, { "id": "3-d-file-format", "title": "3D File Format", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D File Formats are standardised container structures for encoding three-dimensional geometric data, textures, materials, animations, and metadata, enabling asset portability across tools, rendering engines, and platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:3-d-file-format", "labels": [ "3D File Format", "C3D Format", "STEP File Format" ], "is_subclass_of": [ "Data Format Standard" ], "wikilinks": [ "3D Rendering", "Animation Keyframes", "Asset Pipeline", "Cross-Platform Asset Exchange", "glTF", "Material Definition", "Mesh Data", "3D Model", "Data Format Standard", "Digital Asset Management", "Interoperability" ] }, { "id": "3-d-gaussian-splatting", "title": "3D Gaussian Splatting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A neural rendering technique that represents 3D scenes as collections of millions of 3D Gaussian primitives with learnable positions, colours, opacities, and covariances, enabling photorealistic real-time rendering at 100+ frames per second through GPU-accelerated rasterisation, revolutionising telepresence and immersive collaboration with unprecedented visual fidelity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:3-d-gaussian-splatting", "labels": [ "3D Gaussian Splatting", "TELE-051-3d-gaussian-splatting" ], "is_subclass_of": [ "Neural Rendering", "TELE-050-neural-rendering-telepresence" ], "wikilinks": [ "DifferentiableRendering", "PhotorealisticTelepresence", "TELE-020-virtual-reality-telepresence", "TELE-050-neural-rendering-telepresence", "TELE-052-neural-radiance-fields", "TELE-053-volumetric-video-conferencing", "TELE-060-instant-ngp" ] }, { "id": "3-d-generation", "title": "3D Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "3D Generation refers to the automated or semi-automated creation of three-dimensional geometry, appearance, and scene representations using machine learning models, procedural algorithms, or hybrid approaches. Techniques include text-to-3D, image-to-3D, and scene-level generation via neural radiance fields, Gaussian splatting, diffusion models, and large multi-modal networks. The field aims to dramatically reduce the time and skill threshold required to produce photorealistic or stylised 3D content for gaming, film, spatial computing, e-commerce, digital twins, and robotics simulation.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:3-d-generation", "labels": [ "3D Generation", "3D Model Generation" ], "is_subclass_of": [ "3D Content Generation", "Generative Model", "Computer Vision", "Deep Learning" ], "wikilinks": [ "Diffusion Model", "NeRF", "Gaussian Splatting", "Text-to-3D", "Neural 3D Generation", "3D Reconstruction", "3D Content Creation", "Digital Twin Generation", "Game Asset Generation", "Score Distillation Sampling", "Generative Model", "Computer Vision", "Multimodal AI", "Image Generation", "Deep Learning", "Generative Adversarial Network", "Variational Autoencoder", "Transformer Architecture", "Implicit Neural Representation", "Point Cloud" ] }, { "id": "3-d-geometry", "title": "3D Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D geometry is the mathematical description of shapes, positions, and transformations of objects in three-dimensional space, using representations such as points, vectors, meshes, and coordinate transforms. It provides the underlying spatial data model that computer graphics rendering pipelines and photorealistic rendering engines operate on, from vertex positions through to camera projection matrices. Practical 3D geometry combines linear algebra with computational geometry techniques to represent and manipulate scenes efficiently.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:3-d-geometry", "labels": [ "3D Geometry" ], "is_subclass_of": [ "Computational Geometry" ], "wikilinks": [] }, { "id": "3-d-graphics-standard", "title": "3D Graphics Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Graphics Standards are technical specifications and conventions governing the representation, rendering, and interchange of three-dimensional visual data, including APIs like OpenGL, DirectX, and Vulkan, shader languages, and coordinate system conventions.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:3-d-graphics-standard", "labels": [ "3D Graphics Standard" ], "is_subclass_of": [ "Display and Rendering", "Content and Assets", "Rendering Standard", "Technical Specification" ], "wikilinks": [ "Cross-Platform Compatibility", "DirectX", "Driver Implementation", "GPU Architecture", "GPU Support", "OpenGL", "Rendering Standard", "Technical Specification", "Vulkan", "3D Rendering Engine", "Computer Vision", "Hardware Acceleration", "Real-time Rendering", "Rendering Pipeline", "Shader Language" ] }, { "id": "3-d-li-dar", "title": "3D LiDAR", "domain": "robotics", "domain_name": "Robotics", "definition": "3D LiDAR is a active ranging sensor that emits pulsed laser light across multiple vertical channels to capture dense three-dimensional point clouds of the surrounding environment. Spinning or solid-state variants measure range and intensity for hundreds of thousands of points per second, enabling robots and autonomous vehicles to perform obstacle detection, SLAM, and high-fidelity scene reconstruction in real time.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:3-d-li-dar", "labels": [ "3D LiDAR" ], "is_subclass_of": [ "Perception and Sensing", "Lidar" ], "wikilinks": [ "Lidar", "Robotics" ] }, { "id": "3d-mapping", "title": "3D Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D mapping is the process of building a spatially coherent three-dimensional representation of an environment from sensor observations, such as camera images, LiDAR scans, or depth data, typically as a robot or capture device moves through the space. It relies on techniques including structure-from-motion and triangulation to recover the geometry of the scene and the trajectory of the sensor simultaneously. 3D mapping outputs, including point clouds, meshes, or occupancy grids, support downstream applications in robotics navigation, spatial computing, and photogrammetry-based 3D reconstruction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:3d-mapping", "labels": [ "3D Mapping" ], "is_subclass_of": [ "3D Reconstruction" ], "wikilinks": [] }, { "id": "3-d-model", "title": "3D Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D Model is a digital representation of a three-dimensional object or environment constructed from vertices, edges, faces, and materials, encoded in standard formats such as glTF, OBJ, FBX, or USD, and used as a foundational asset in spatial computing, simulation, gaming, and metaverse applications.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:3-d-model", "labels": [ "3D Model" ], "is_subclass_of": [ "Content and Assets", "Digital Asset" ], "wikilinks": [ "3D Rendering", "Interactive Visualization", "Material Definition", "Mesh Data", "Virtual Environment Design", "3D Development", "3D File Format", "3D Modeling", "Animation Retargeting", "Computer Vision", "Digital Asset", "Metaverse", "Texture Mapping" ] }, { "id": "3-d-modeling-software", "title": "3D Modeling Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Modeling Software is a category of applications that provide tools for constructing, editing, and organising three-dimensional geometric representations of objects and scenes, typically via polygon mesh editing, subdivision surfaces, NURBS curves, sculpting, or parametric operations. These applications underpin virtually all digital content creation workflows for games, film, architecture, product design, and XR environments. They commonly integrate texturing, rigging, animation, and basic rendering capabilities alongside the core modelling toolkit.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:3-d-modeling-software", "labels": [ "3D Modeling Software" ], "is_subclass_of": [ "3D Modeling" ], "wikilinks": [] }, { "id": "3d-modelling", "title": "3D Modelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D modelling is the creation of a mathematical representation of a three-dimensional surface or volume using specialised software. Models are commonly built from polygon meshes, parametric surfaces or volumetric data and define the geometry that is later textured, rigged, animated and rendered. It is the foundational discipline for games, film, visualisation, product design and spatial computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:3d-modelling", "labels": [ "3D Modelling", "3D Modeling", "3D Modeling API", "3D Modeling Tools", "Automated 3D Modeling", "Manual 3D Modeling" ], "is_subclass_of": [ "Computer Graphics", "3D Development" ], "wikilinks": [] }, { "id": "3d-parallelism", "title": "3D Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "3D parallelism is a distributed training strategy that combines data parallelism, tensor parallelism and pipeline parallelism along three independent axes to train models too large for a single accelerator or single parallelism scheme alone. Each axis partitions a different dimension of the problem: data parallelism splits the batch, tensor parallelism splits individual layers across devices, and pipeline parallelism splits the layer stack across stages. Frameworks such as Megatron-LM and DeepSpeed implement 3D parallelism to scale training to thousands of GPUs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:3d-parallelism", "labels": [ "3D Parallelism" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "3-d-perception", "title": "3D Perception", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Perception is the computational capability to interpret sensor data \u2014 from cameras, LiDAR, radar, and depth sensors \u2014 and derive accurate three-dimensional understanding of the surrounding environment, including object detection, pose estimation, scene structure, and semantic labelling. It forms a foundational layer for autonomous systems, robotic manipulation, augmented reality registration, and spatial computing. The discipline draws on computer vision, geometry, and deep learning to transform raw observations into actionable 3D representations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:3-d-perception", "labels": [ "3D Perception", "3D Sound Perception" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "3-d-reconstruction", "title": "3D Reconstruction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Reconstruction is the computational process of recovering three-dimensional geometric and structural information from multiple 2D images or sensor data (such as LiDAR or depth cameras) using techniques including Computer Vision, photogrammetry, and Structure-from-Motion (SfM), enabling digital capture of real-world objects and environments for Digital Twin creation, immersive environment mapping, and spatial analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:3-d-reconstruction", "labels": [ "3D Reconstruction", "3D Capture", "3D Reconstruction Pipeline", "Incremental Reconstruction", "Real-Time 3D Reconstruction", "Stereo Reconstruction" ], "is_subclass_of": [ "Content and Assets", "Computer Vision" ], "wikilinks": [ "Camera Calibration", "Environmental Mapping", "Feature Matching", "Image Processing", "Point Cloud Generation", "Real-world Digitisation", "Structure-from-Motion", "Computer Vision", "Digital Twin", "Photogrammetry", "Point Cloud" ] }, { "id": "3-d-rendering-engine", "title": "3D Rendering Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D rendering engine is software that converts three-dimensional geometric data into two-dimensional images through processes including geometry processing, lighting calculation, texture mapping, and rasterisation or ray tracing. In real-time contexts it targets interactive frame rates (90 Hz+) and must minimise motion-to-photon latency for XR presence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:3-d-rendering-engine", "labels": [ "3D Rendering Engine" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Computer Graphics", "Real-time Visualisation", "XR Applications", "Immersive Experiences", "Metaverse" ] }, { "id": "3-d-rendering", "title": "3D Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D rendering is the computational process of generating a two-dimensional image or animation from a three-dimensional scene description by simulating the interaction of light with surfaces, materials, and geometry. It encompasses techniques ranging from real-time rasterization used in interactive applications to physically-based ray tracing and path tracing used for photorealistic offline production. The pipeline converts geometric primitives, shader programs, texture maps, and lighting data into final pixel colours via a GPU compute pipeline or software renderer. Contemporary rendering also incorporates neural methods such as Neural Radiance Fields and 3D Gaussian Splatting, as well as AI-driven upscaling and denoising, blurring the boundary between classical computer graphics and machine learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:3-d-rendering", "labels": [ "3D Rendering", "Web-Based 3D Rendering" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "3-d-scanning", "title": "3D Scanning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Scanning is the process of capturing the three-dimensional shape, and optionally the colour and texture, of real-world objects, people, or environments using hardware such as structured-light scanners, time-of-flight LiDAR, photogrammetry rigs, or depth cameras, producing digital point clouds or meshes that represent the physical source. The resulting data feeds into digital preservation, reverse engineering, visual-effects production, quality inspection, and spatial computing pipelines. Accuracy, resolution, and scan volume are the primary quality axes that distinguish scanning technologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:3-d-scanning", "labels": [ "3D Scanning" ], "is_subclass_of": [ "3D Reconstruction" ], "wikilinks": [] }, { "id": "3-d-scene-exchange-protocol-sxp", "title": "3D Scene Exchange Protocol (SXP)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Open standards framework enabling interoperable transfer of complete 3D scene graphs including geometric meshes (polygon topology vertex positions normals texture coordinates with indexed triangle lists optimized for GPU rendering), physically-based rendering (PBR) material systems (metallic-roug...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:3-d-scene-exchange-protocol-sxp", "labels": [ "3D Scene Exchange Protocol (SXP)" ], "is_subclass_of": [ "Standards and Interoperability", "Data Format", "Interchange Protocol", "3D Graphics Standard", "Metaverse Infrastructure", "Open Standard", "Scene Graph Format" ], "wikilinks": [ "3D Asset Interchange", "3D Modeling Software", "Academy Software Foundation OpenUSD", "Adobe Substance Painter", "Alembic Framework", "Animation Rig", "Animation Tools", "AR/VR Experiences", "Asset Portability", "Autodesk FBX Documentation", "Babylon.js Framework", "Blender glTF Exporter", "Camera Definition", "Collaborative Workflows", "COLLADA Specification", "Cross-Platform Compatibility", "Draco Compression", "Draco Mesh Compression", "E-Commerce 3D", "Eurographics Interchange Standards" ] }, { "id": "3-d-scene-reconstruction", "title": "3D Scene Reconstruction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "3D Scene Reconstruction is the computational process of recovering a complete, coherent three-dimensional model of an environment or scene from sensor observations such as images, depth maps, or LiDAR returns, integrating multiple viewpoints and modalities into a unified representation. Methods include volumetric fusion, truncated signed distance function (TSDF) integration, neural implicit representations, and Gaussian splatting, each offering different trade-offs between completeness, accuracy, and real-time capability. The output is used in robotics, autonomous driving, spatial computing, and cultural heritage documentation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:3-d-scene-reconstruction", "labels": [ "3D Scene Reconstruction", "Scene Reconstruction" ], "is_subclass_of": [ "3D Reconstruction" ], "wikilinks": [] }, { "id": "3-d-scene-representation", "title": "3D Scene Representation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D scene representation is a data structure that encodes the geometry, appearance and spatial layout of a three-dimensional environment, such as point clouds, meshes, voxel grids or neural implicit fields. It underpins 3D content generation and reconstruction pipelines, providing the intermediate format from which renderable or editable scenes are produced. Point clouds are one common instance, forming a sparse, unstructured representation that is often converted into denser structures for downstream use.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:3-d-scene-representation", "labels": [ "3D Scene Representation" ], "is_subclass_of": [ "Scene Representation" ], "wikilinks": [] }, { "id": "3-d-user-interface", "title": "3D User Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D User Interface (3DUI) is an interactive control system within three-dimensional virtual environments enabling users to manipulate objects, navigate spaces, and access functionality through spatial gestures, hand tracking, gaze-based selection, and voice commands. 3DUI design balances accessibility standards with intuitive spatial affordances appropriate for VR, AR, and metaverse interactions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:3-d-user-interface", "labels": [ "3D User Interface" ], "is_subclass_of": [ "Interaction Technology", "User Interface", "Interaction Design" ], "wikilinks": [ "Accessible Immersion", "Gaze Tracking", "Gesture Recognition", "Interaction Design", "Intuitive Interaction", "Sensor Input", "User Navigation", "Voice Input", "Accessibility Standard", "Hand Tracking", "Haptic Feedback", "User Interface" ] }, { "id": "3-d-and-4-d", "title": "3D and 4D", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The domain of three-dimensional and four-dimensional content creation, covering tools, techniques, and AI-powered pipelines for generating, editing, and rendering 3D assets and temporally dynamic (4D) representations. Encompasses text-to-3D systems (NVIDIA Edify, Luma Genie, Point-E), neural radiance fields, Gaussian splatting, and diffusion-based mesh generation, as well as 6D pose estimation, scene-scale generation, and integration with AR/VR workflows.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:3-d-and-4-d", "labels": [ "3D and 4D" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "automation", "community", "decision making", "design thinking", "documentation", "innovation", "modeling", "neural networks", "open source", "optimization", "organisation", "performance", "research", "scalability", "skills development", "user experience", "visualization", "artificial intelligence", "bias", "collaboration" ] }, { "id": "3-gpp", "title": "3GPP", "domain": "standards", "domain_name": "Standards", "definition": "The 3rd Generation Partnership Project (3GPP) is a global consortium of regional telecommunications standards bodies that develops and maintains versioned technical specifications governing mobile communication systems, from UMTS (3G) through LTE (4G) to NR (5G) and the emerging 6G framework. Operating through working groups (RAN, SA, CT), it produces Releases that introduce new radio access technologies, core network architectures, and service-layer capabilities. 3GPP specifications are the foundational technical substrate for the global cellular infrastructure serving billions of devices.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:3-gpp", "labels": [ "3GPP", "3GPP 5G Standards" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "51-percent-attack", "title": "51 Percent Attack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A 51% Attack is a consensus-layer attack on a Proof-of-Work blockchain in which a single entity or coalition controls more than half of the network's hash rate, enabling double-spending, transaction censorship, and chain reorganisation. The attack exploits the longest-chain rule: the attacker mines a private fork containing fraudulent transactions and, once it exceeds the honest chain in cumulative work, broadcasts it to override confirmed history.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:51-attack", "labels": [ "51 Percent Attack", "51% Attack", "51 Percent Attack Resistance" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain" ], "wikilinks": [ "AI Energy Optimisation", "altcoins", "ASIC", "Bitcoin Gold", "blockchain reorganization", "Blockchain Reorganization", "blockchain security", "Coinbase", "Consensus", "consensus", "Consensus Attack", "consensus mechanisms", "cryptocurrencies", "cryptographic", "DDoS attack", "Decred", "Ethereum Classic", "hash rate", "Hash Rate", "IEEE 2418.1" ] }, { "id": "5-g-connectivity", "title": "5G Connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "5G Connectivity is the set of radio access and core network capabilities delivered by a 5G NR network that provides devices and applications with wireless communication services characterised by high throughput, ultra-low latency, and high device density. It encompasses the radio link between user equipment and the gNB base station, quality-of-service mechanisms, network slicing for application-specific service guarantees, and the APIs through which applications access network capabilities. 5G Connectivity is a critical enabler for cloud XR, autonomous systems, and real-time industrial applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:5-g-connectivity", "labels": [ "5G Connectivity" ], "is_subclass_of": [ "5G" ], "wikilinks": [] }, { "id": "5-g-network", "title": "5G Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A 5G Network is the end-to-end fifth-generation mobile telecommunications system comprising a 5G New Radio (NR) radio access network built around gNB base stations, a cloud-native 5G Core implementing a service-based architecture with independently deployable network functions (AMF, SMF, UPF, PCF, NRF, AUSF), and the transport infrastructure interconnecting them. The architecture fundamentally separates the control plane from the user plane, enables network slicing to provision multiple virtualised logical networks on shared physical resources, and integrates multi-access edge computing nodes for deterministic low-latency application hosting. Defined across 3GPP Release 15 (frozen 2018) through Release 18 (5G-Advanced), 5G networks support three canonical service classes: enhanced Mobile Broadband (eMBB), Ultra-Reliable Low-Latency Communications (URLLC), and massive Machine-Type Communications (mMTC), making them the foundational connectivity layer for industrial IoT, autonomous systems, extended reality, and private campus networks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:5-g-network", "labels": [ "5G Network", "5G Low-Latency Networks", "5G Networks" ], "is_subclass_of": [ "Telecommunications Infrastructure" ], "wikilinks": [] }, { "id": "5-g", "title": "5G", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "5G (Fifth Generation) is the latest generation of mobile cellular network technology, standardised by 3GPP from Release 15 onwards, characterised by peak data rates exceeding 20 Gbps, radio latencies below 1 ms, and support for up to one million connected devices per square kilometre. It introduces three primary use-case families: enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communications (mMTC). 5G underpins advances in spatial computing, autonomous vehicles, industrial automation, and IoT at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:5-g", "labels": [ "5G", "5G NR", "5G New Radio" ], "is_subclass_of": [ "Telecommunications" ], "wikilinks": [] }, { "id": "6-do-f-tracking", "title": "6DoF Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "6DoF (Six Degrees of Freedom) Tracking is the measurement and continuous estimation of an object's complete rigid-body pose in three-dimensional space, encompassing three translational components (x, y, z position) and three rotational components (pitch, yaw, roll orientation). It is the foundational capability for XR headsets, controllers, and spatial computing devices to understand and respond to the user's physical motion with sub-millimetre accuracy and minimal latency. 6DoF tracking is achieved through sensor fusion of inertial measurement units, camera-based visual odometry, and optionally external beacons or reference markers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:6-do-f-tracking", "labels": [ "6DoF Tracking", "6-DoF Tracking" ], "is_subclass_of": [ "Simultaneous Localisation and Mapping" ], "wikilinks": [] }, { "id": "6-g-network-slice", "title": "6G Network Slice", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Virtual partition of 6G infrastructure guaranteeing specified quality-of-service levels for immersive workloads through isolated resource allocation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:6-g-network-slice", "labels": [ "6G Network Slice" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "3GPP Release 21", "6G Network Infrastructure", "Dynamic Resource Allocation", "ETSI ENI 008", "Low Latency Service", "Network Slicing Orchestrator", "QoS Policy", "Resource Allocation Unit", "SDN Controller", "Service Level Agreement", "Traffic Classifier", "Workload Isolation", "Guaranteed Bandwidth", "InfrastructureDomain", "NetworkLayer" ] }, { "id": "a-star-algorithm", "title": "A Star Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The A* algorithm is an informed best-first graph-search method that finds a least-cost path from a start node to a goal node by evaluating each candidate node n with the function f(n) = g(n) + h(n), where g(n) is the exact cost accumulated from the start node to n and h(n) is an admissible heuristic estimating the remaining cost from n to the goal. By always expanding the open-list node with the lowest f-value, A* is guaranteed to find the optimal path when h is admissible (never overestimates) and is also consistent (satisfies the triangle inequality h(n) \u2264 c(n,n') + h(n') for every edge (n,n')). A* generalises Dijkstra's algorithm by adding goal-directed heuristic guidance, and subsumes greedy best-first search as the special case g(n)=0. First described by Peter Hart, Nils Nilsson, and Bertram Raphael at the Stanford Research Institute in 1968, A* remains the most widely deployed optimal graph-search algorithm in artificial intelligence, robotics, game development, and logistics.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:a-star-algorithm", "labels": [ "A Star Algorithm", "A* Algorithm", "A-Star Algorithm" ], "is_subclass_of": [ "Search Algorithm", "Informed Search", "Graph Search", "Pathfinding" ], "wikilinks": [ "Search Algorithm", "Pathfinding", "Dijkstra Algorithm", "Cost Function", "Graph Search", "Informed Search", "Heuristic Methods", "Priority Queue", "Optimisation", "Robotics", "Game AI", "Autonomous Navigation", "Path Planning", "State Space", "Graph Theory", "Weighted Graph", "Shortest Path", "Dynamic Programming", "IDA*", "Jump Point Search" ] }, { "id": "a-b-testing", "title": "A/B Testing", "domain": "data", "domain_name": "Data", "definition": "A/B testing is a controlled experimentation method that compares two or more variants by randomly assigning subjects to each and measuring a defined outcome metric. By holding all factors constant except the variant under test, it isolates causal effects and supports data-driven decisions with statistical rigour. It is widely used to optimise digital products, content and user experiences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:a-b-testing", "labels": [ "A/B Testing", "A B Testing" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "a2-a-protocol", "title": "A2A Protocol", "domain": "ai", "domain_name": "Ai", "definition": "The Agent-to-Agent (A2A) Protocol is an open communication specification, initially proposed by Google in April 2025 and transferred to the Linux Foundation in June 2025, that defines how autonomous AI agents discover one another, negotiate capabilities, delegate tasks, and exchange results across heterogeneous agent frameworks and cloud environments. It uses HTTP/HTTPS transport with JSON-RPC 2.0 structured messages and Server-Sent Events for streaming, an Agent Card system at a well-known URI for capability advertisement, and enterprise-grade authentication through OAuth 2.0, mTLS, and JWT, enabling cross-vendor agent interoperability without requiring shared infrastructure. A2A complements tool-access protocols such as the Model Context Protocol by standardising the agent-to-agent interaction layer, and reached v1.0 production readiness in 2026 with adoption by over 150 organisations including Microsoft, AWS, Salesforce, SAP, and ServiceNow.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:a2-a-protocol", "labels": [ "A2A Protocol" ], "is_subclass_of": [ "Agent Communication Protocol", "Agent-to-Agent Protocol" ], "wikilinks": [ "Agent-to-Agent Protocol", "Model Context Protocol", "Multi-Agent System", "Multi-Agent Orchestration", "Agentic Workflow", "Inter-Agent Communication", "Agent Frameworks", "Agentic AI", "Function Calling", "Agent Communication Protocol", "Service Discovery", "JSON-LD", "HTTP Protocol", "JSON-RPC 2.0", "Server-Sent Events", "OAuth 2.0", "Mutual TLS", "JSON Schema", "Decentralised Identifier", "Autonomous Agent" ] }, { "id": "aaai", "title": "AAAI", "domain": "ai", "domain_name": "Ai", "definition": "AAAI (the Association for the Advancement of Artificial Intelligence), founded in 1979 under the original name 'American Association for Artificial Intelligence' and renamed in 2007, is the principal non-profit scientific society devoted to advancing research in and responsible use of artificial intelligence. It is best known for its flagship annual peer-reviewed conference \u2014 one of the most selective and broadly scoped venues in AI \u2014 alongside the AI, Ethics and Society (AIES) conference, symposia, workshops, and the journal AI Magazine. AAAI also engages in education, public communication, and policy discussion concerning AI; administers the AAAI Fellows Program recognising sustained contributions to the discipline; and maintains historical continuity as the first professional society dedicated exclusively to AI, founded by leaders including Allen Newell, Edward Feigenbaum, Marvin Minsky, and John McCarthy. AAAI 2026 \u2014 the 40th annual conference \u2014 received 23,680 submissions and accepted 4,167 papers at a 17.6% acceptance rate, making it the most competitive edition in the conference's history.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:aaai", "labels": [ "AAAI" ], "is_subclass_of": [ "Academic Conference", "AI Research Area" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning", "Academic Conference", "AAAI Conference", "Peer Review", "NeurIPS", "ICML", "ICLR", "CVPR", "IJCAI", "Research Publication", "Scientific Conference", "Research Institution", "Deep Learning", "Natural Language Processing", "Computer Vision", "Knowledge Representation", "Planning and Scheduling", "Reasoning", "Symbolic AI" ] }, { "id": "abi-encoding", "title": "ABI Encoding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ABI encoding is the Application Binary Interface encoding scheme used by the Ethereum Virtual Machine to serialise function calls and their arguments into the byte layout that smart contracts expect on-chain. It defines fixed rules for encoding primitive types, dynamic types such as strings and arrays, and nested structures, so that a compiled contract can decode calldata deterministically regardless of the source language or tool that produced it. Development toolchains such as Foundry and compilers such as Vyper generate ABI-encoded calldata automatically from a contract's interface definition.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:abi-encoding", "labels": [ "ABI Encoding" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "acid-properties", "title": "ACID Properties", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "ACID Properties are the four guarantees - Atomicity, Consistency, Isolation and Durability - that define a reliable database transaction. Atomicity ensures a transaction is all-or-nothing; Consistency preserves invariants; Isolation hides concurrent intermediate states; and Durability persists committed results across failures. Together they let applications reason about correctness despite concurrency and crashes, distinguishing strongly consistent transactional systems from eventually consistent alternatives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:acid-properties", "labels": [ "ACID Properties" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "acinq", "title": "ACINQ", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ACINQ is a Paris-based Bitcoin technology company specialising in the development and operation of Lightning Network infrastructure, most notably the Eclair Lightning Network node implementation written in Scala and the Phoenix self-custodial mobile Bitcoin wallet. The company operates one of the largest publicly reachable Lightning Network routing nodes, providing significant liquidity and routing capacity to the network. ACINQ contributes to the BOLT (Basis of Lightning Technology) protocol specification process and has pioneered features such as trampoline routing to reduce the on-device computation required by mobile Lightning clients.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:acinq", "labels": [ "ACINQ" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Bitcoin", "Payment Channel", "Lightning Network" ] }, { "id": "acl", "title": "ACL", "domain": "security", "domain_name": "Security", "definition": "An Access Control List (ACL) is a data structure attached to a resource \u2014 such as a file, directory, or network interface \u2014 that enumerates which subjects (users, groups, or processes) are permitted to perform which operations on that resource. Each entry in the list is called an Access Control Entry (ACE) and specifies a principal, a set of permissions (read, write, execute, delete), and whether those permissions are granted or denied. ACLs originated in file-system security (POSIX, NTFS) and were subsequently extended to networking, where routers and firewalls use IP-level ACLs to filter packets by source address, destination port, and protocol. In distributed and cloud environments ACLs underpin fine-grained authorisation that complements role-based and attribute-based access-control models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:acl", "labels": [ "ACL" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "acm", "title": "ACM", "domain": "standards", "domain_name": "Standards", "definition": "The Association for Computing Machinery (ACM) is the world's largest and oldest scientific and educational computing society, founded in 1947, that advances the computing profession through peer-reviewed publication, professional development, and conference organisation. ACM maintains the ACM Digital Library, one of the most comprehensive archives of computing literature, and publishes flagship journals including Communications of the ACM and ACM Transactions series. It confers the ACM Turing Award \u2014 widely regarded as computing's highest honour \u2014 and promulgates ethical codes of professional conduct. ACM also co-sponsors and organises foundational conferences such as SIGGRAPH, STOC, CCS, and CHI through its network of Special Interest Groups.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:acm", "labels": [ "ACM" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "Standards Body", "Software Engineering" ] }, { "id": "adas", "title": "ADAS", "domain": "robotics", "domain_name": "Robotics", "definition": "Advanced Driver Assistance Systems (ADAS) are electronic systems that assist vehicle operators with driving and parking functions through automated technologies including adaptive cruise control, lane keeping assist, automatic emergency braking, blind spot detection, and parking assistance. ADAS operates at SAE Level 1\u20132 automation, augmenting rather than replacing the driver, and relies on sensor fusion across cameras, radar, and ultrasonic systems to perceive the vehicle's environment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:adas", "labels": [ "ADAS" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "ISO 26262", "SAE J3016", "Autonomous Robot", "Autonomous Vehicle", "MetaverseDomain", "Perception System", "Sensor Fusion" ] }, { "id": "ade20-k", "title": "ADE20K", "domain": "ai", "domain_name": "Ai", "definition": "ADE20K is a large-scale image segmentation dataset containing over 27,000 images annotated with 150 semantic categories covering both indoor and outdoor scenes, released by MIT CSAIL in 2017. It provides pixel-level semantic, instance, and part-level annotations enabling training and benchmarking of scene parsing and semantic segmentation models. ADE20K serves as the foundational benchmark for the ImageNet Scene Parsing Challenge and has driven significant advances in dense prediction architectures, with state-of-the-art models achieving 62.8 mIoU by 2022 (BEiT-3).", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:ade20-k", "labels": [ "ADE20K" ], "is_subclass_of": [ "Benchmarks", "Benchmark Dataset" ], "wikilinks": [ "Semantic Segmentation", "Panoptic Segmentation", "Instance Segmentation", "Data Annotation", "Computer Vision", "COCO Dataset", "Image Segmentation", "Benchmarks", "Benchmark Dataset", "Object Detection", "Deep Learning", "Convolutional Neural Network", "Transformer Architecture", "Scene Understanding", "Foundation Models", "Image Classification", "Data Curation", "Transfer Learning", "Autonomous Vehicle Navigation", "Robotics Perception" ] }, { "id": "adr-008", "title": "ADR-008", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An architecture decision record numbered 008, documenting a specific architectural choice within a project together with its context, options considered, and consequences. ADRs form a chronological log of design decisions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:adr-008", "labels": [ "ADR-008" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [ "Software Architecture", "Software Engineering" ] }, { "id": "adr-012", "title": "ADR-012", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An architecture decision record numbered 012, capturing a particular architectural decision within a project alongside its context and consequences. It is part of a numbered sequence of design records.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:adr-012", "labels": [ "ADR-012" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [ "Software Architecture", "Software Engineering" ] }, { "id": "aes-encryption", "title": "AES Encryption", "domain": "security", "domain_name": "Security", "definition": "AES encryption is the Advanced Encryption Standard, a symmetric block cipher adopted as a US federal standard that operates on fixed-size blocks using a shared secret key for both encryption and decryption. It is the most widely deployed symmetric cipher, used by hardware security features such as Intel SGX for memory encryption and by low-power wireless protocols such as Zigbee for link-layer security. It supports 128, 192 and 256-bit key lengths, with 128-bit remaining secure against all known practical attacks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:aes-encryption", "labels": [ "AES Encryption" ], "is_subclass_of": [ "Symmetric Encryption" ], "wikilinks": [] }, { "id": "aes-gcm", "title": "AES-GCM", "domain": "security", "domain_name": "Security", "definition": "AES-GCM (Advanced Encryption Standard \u2013 Galois/Counter Mode) is an authenticated encryption with associated data (AEAD) cipher mode that combines the AES block cipher operating in Counter Mode with the Galois Message Authentication Code. It provides both confidentiality and data integrity in a single pass, making it the dominant symmetric encryption scheme in modern secure communications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:aes-gcm", "labels": [ "AES-GCM" ], "is_subclass_of": [ "Symmetric Encryption" ], "wikilinks": [] }, { "id": "agi-timelines", "title": "AGI Timelines", "domain": "ai", "domain_name": "Ai", "definition": "AGI timelines are structured probabilistic forecasts estimating when artificial general intelligence \u2014 AI matching or exceeding human cognitive ability across most economically valuable tasks \u2014 might be achieved. Such forecasts aggregate expert surveys (AI Impacts ESPAI 2022/2023), compute-scaling extrapolations (biological anchors model, Cotra 2020/2022), benchmark-progress curves (Metaculus, Manifold Markets), and Bayesian models of research milestones to produce probability distributions over future dates. As of mid-2026, median estimates cluster in the 2030\u20132047 range depending on methodology. Timelines are inherently uncertain and contested, and they materially influence AI safety research prioritisation, alignment investment, compute governance thresholds, and public policy.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:agi-timelines", "labels": [ "AGI Timelines" ], "is_subclass_of": [ "Time-Series Forecasting", "Capability Forecasting" ], "wikilinks": [ "Artificial General Intelligence", "Capability Forecasting", "AI Safety", "AI Alignment", "AI Safety Research", "ARC-AGI", "Scaling Laws", "Performance Benchmarks", "Gartner Prediction", "Time-Series Forecasting", "Compute Governance", "Compute Resources", "Large Language Models", "Deep Learning", "Neural Networks", "Reinforcement Learning", "Model Evaluation", "Evaluation Benchmarks and Leaderboards", "AI Governance", "AI Safety Institute" ] }, { "id": "agi", "title": "AGI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial General Intelligence refers to the hypothetical capability of an AI system to perform any intellectual task that a human can do, characterized by flexibility, adaptability, and generalization across diverse domains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agi", "labels": [ "AGI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-acceleration-gap", "title": "AI Acceleration Gap", "domain": "economics", "domain_name": "Economics", "definition": "The compounding divergence between the rapid advancement of frontier AI capabilities and the slower adoption and integration of these technologies by the median enterprise or individual.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-acceleration-gap", "labels": [ "AI Acceleration Gap" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "ai-accelerator", "title": "AI Accelerator", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An AI accelerator is a class of specialised hardware designed to speed up machine learning workloads, particularly the dense linear algebra of neural network training and inference. Common forms include GPUs, tensor processing units, neural processing units, and custom application-specific integrated circuits that exploit massive parallelism, reduced-precision arithmetic, and high-bandwidth memory. By offloading matrix and tensor operations from general-purpose CPUs, AI accelerators deliver order-of-magnitude gains in throughput and energy efficiency for deep learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ai-accelerator", "labels": [ "AI Accelerator" ], "is_subclass_of": [ "Hardware Accelerator" ], "wikilinks": [] }, { "id": "ai-accountability", "title": "AI Accountability", "domain": "ai", "domain_name": "Ai", "definition": "AI Accountability is the set of mechanisms, obligations, and institutional structures that ensure developers, deployers, and operators of AI systems can be held responsible for the outcomes those systems produce. It encompasses technical auditability, legal liability, organisational governance, and redress pathways for harms caused by algorithmic decisions. Accountability frameworks bind technical transparency measures to enforceable consequences, distinguishing it from voluntary explainability efforts. Effective AI accountability requires clear assignment of responsibility across the AI value chain\u2014from data collection through deployment and monitoring.", "entityType": "Class", "qualityScore": 0.87, "maturity": "emerging", "iri": "urn:ngm:class:ai-accountability", "labels": [ "AI Accountability" ], "is_subclass_of": [ "Accountability", "AI Governance", "Responsible AI" ], "wikilinks": [ "Responsible AI", "Explainability", "Artificial Intelligence", "AI Governance", "AI Ethics", "Audit Trail", "Human Oversight", "Regulatory Compliance", "EU AI Act Regulatory Instrument", "Algorithmic Accountability", "AI Trustworthiness", "ISO/IEC JTC 1/SC 42", "NIST AI Risk Management Framework", "Model Card", "Algorithmic Auditing", "Transparency", "Data Governance", "AI Impact Assessment", "Bias Mitigation", "Fairness" ] }, { "id": "ai-adoption-barriers", "title": "AI Adoption Barriers", "domain": "economics", "domain_name": "Economics", "definition": "The economic, psychological, and social factors that hinder the widespread integration and utilization of artificial intelligence in society.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-adoption-barriers", "labels": [ "AI Adoption Barriers" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "ai-adoption-platform", "title": "AI Adoption Platform", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An integrated software suite that supports the strategic planning, execution, and tracking of AI implementation within an organization.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-adoption-platform", "labels": [ "AI Adoption Platform" ], "is_subclass_of": [ "Enterprise Ai" ], "wikilinks": [] }, { "id": "ai-adoption-rates", "title": "AI Adoption Rates", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantitative metrics tracking the penetration, frequency, and demographic distribution of artificial intelligence tool usage across consumer and enterprise populations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-adoption-rates", "labels": [ "AI Adoption Rates" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-adoption", "title": "AI Adoption", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Adoption is the multi-dimensional socio-technical process by which organisations, sectors, national economies and individual workers progressively integrate artificial-intelligence systems \u2014 classical machine learning, deep learning, foundation models, generative AI and agentic AI \u2014 into produ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-adoption", "labels": [ "AI Adoption" ], "is_subclass_of": [ "AI Governance and Ethics", "Technology Adoption", "Digital Transformation", "Innovation Diffusion", "Organisational Change", "Socio-Technical Process" ], "wikilinks": [ "Acemoglu Restrepo 2024 Simple Macroeconomics of AI NBER", "Agentic AI", "AI Investment Cycle", "AI Productivity Paradox", "AI Regulation", "AI Rejection", "AI Strategy", "AI Talent", "AI Talent Capacity", "AI Talent War", "AI Use Case Portfolio", "AI Vendor Stack", "BCG Where Is the Value in AI 2024", "Brynjolfsson Li Raymond 2023 Generative AI at Work NBER 31161", "Brynjolfsson McAfee 2014 Second Machine Age", "Brynjolfsson Rock Syverson 2021 Productivity J-Curve", "Build-Internally-Only Posture", "Change Management", "Change Management Programme", "Copilot Systems" ] }, { "id": "ai-advertising", "title": "AI Advertising", "domain": "economics", "domain_name": "Economics", "definition": "The integration of advertising mechanisms, such as sponsored content or product recommendations, directly into the interfaces and decision-making processes of artificial intelligence agents and chatbots.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-advertising", "labels": [ "AI Advertising" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "ai-agent-acquisition", "title": "AI Agent Acquisition", "domain": "economics", "domain_name": "Economics", "definition": "The strategic purchase of AI agent startups by larger technology companies to gain capabilities, talent, or market position.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-acquisition", "labels": [ "AI Agent Acquisition" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-agent-capabilities", "title": "AI Agent Capabilities", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The specific functional attributes, such as tool use, multi-session support, and autonomous decision-making, that define the operational scope of an AI agent.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-capabilities", "labels": [ "AI Agent Capabilities" ], "is_subclass_of": [ "Tool Use" ], "wikilinks": [] }, { "id": "ai-agent-coordination", "title": "AI Agent Coordination", "domain": "ai", "domain_name": "Ai", "definition": "AI Agent Coordination is the set of mechanisms by which multiple autonomous AI agents align their actions, share state, allocate tasks, and resolve conflicts to achieve goals that exceed any single agent's capability. It covers communication protocols, role assignment, consensus, and negotiation, and may be centralized through an orchestrator or fully decentralized. Effective coordination is essential for multi-agent systems operating over shared resources or distributed ledgers.", "entityType": "Class", "qualityScore": 0.85, "maturity": "emerging", "iri": "urn:ngm:class:ai-agent-coordination", "labels": [ "AI Agent Coordination" ], "is_subclass_of": [ "AI Agent System", "Multi-Agent System", "Decentralised Coordination" ], "wikilinks": [ "Decentralised Coordination", "RGB and Client-Side Validation", "AI Agent System", "Multi-Agent System", "Agent Orchestrator", "Task Allocation", "Consensus Mechanism", "Communication Protocol", "Contract Net Protocol", "Blackboard System", "Message Passing", "Role Assignment", "Model Context Protocol", "Agent-to-Agent Protocol", "Swarm Intelligence", "Reinforcement Learning", "Large Language Model", "Autonomous AI Agents", "Planning Algorithm", "Negotiation Protocol" ] }, { "id": "ai-agent-harness", "title": "AI Agent Harness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The orchestration layer, tooling, and execution environment that wraps a foundational language model to enable autonomous task completion and reasoning.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-harness", "labels": [ "AI Agent Harness" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-agent-misalignment", "title": "AI Agent Misalignment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The phenomenon where AI agents exhibit behaviors or strategies that deviate from intended goals, such as unsanctioned social engineering.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-misalignment", "labels": [ "AI Agent Misalignment" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "ai-agent-payments", "title": "AI Agent Payments", "domain": "finance", "domain_name": "Finance", "definition": "AI Agent Payments refers to the infrastructure, protocols, and economic mechanisms that enable autonomous AI agents to initiate, authorise, and settle financial transactions on behalf of users or autonomously as principals, spanning micropayment rails for API service consumption, programmatic access to payment networks, and identity-bound wallet abstractions that constrain agent spending to authorised scopes and budgets.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ai-agent-payments", "labels": [ "AI Agent Payments", "AI Economy Payments" ], "is_subclass_of": [ "Agentic Finance" ], "wikilinks": [] }, { "id": "ai-agent-risk-disclosure", "title": "AI Agent Risk Disclosure", "domain": "governance", "domain_name": "Governance", "definition": "The practice of companies identifying AI agents as a material risk to their business model in regulatory filings such as SEC reports.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-risk-disclosure", "labels": [ "AI Agent Risk Disclosure" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-agent-system", "title": "AI Agent System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An autonomous software entity that perceives its environment through Sensor Input|sensors, makes decisions using AI Techniques, and takes actions to achieve specific goals, capable of Machine Learning|learning from experience and adapting Adaptive Behavior|behaviour over time.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:ai-agent-system", "labels": [ "AI Agent System" ], "is_subclass_of": [ "AI Application", "Artificial Intelligence Core" ], "wikilinks": [ "A-SWE", "Action Executor", "Action Space", "Adaptive Behavior", "Adaptive Behavior", "Address Clustering", "Agent Autonomy", "Agentic System", "AI Model", "AI Techniques", "Alexa", "Algorithmic Trading", "AlphaGo", "Anthropic", "Anthropic Computer Use", "Anthropic Computer Use Documentation", "API Access", "API Integration", "API Integration", "Arbitrage" ] }, { "id": "ai-agent-tool-selection", "title": "AI Agent Tool Selection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process by which AI agents autonomously evaluate, choose, and invoke specific external tools or APIs to accomplish tasks, often based on dynamic criteria rather than fixed user preferences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-agent-tool-selection", "labels": [ "AI Agent Tool Selection" ], "is_subclass_of": [ "AI Agent" ], "wikilinks": [] }, { "id": "ai-agent", "title": "AI Agent", "domain": "ai", "domain_name": "Ai", "definition": "A software system that perceives its environment, makes decisions and takes actions to achieve goals, often using a language model together with tools and memory.", "entityType": "Class", "qualityScore": 0.89, "maturity": "growing", "iri": "urn:ngm:class:ai-agent", "labels": [ "AI Agent", "Ai Agent" ], "is_subclass_of": [ "Autonomous Agent", "AI Application", "Goal-Directed System" ], "wikilinks": [ "Tool Use", "Large Language Models", "Automated Planning", "Agentic AI", "Autonomous Agent", "Multi-Agent Systems", "Reinforcement Learning", "Retrieval-Augmented Generation", "Function Calling", "Chain of Thought", "Vector Database", "ReAct Pattern", "Foundation Model", "Memory Management", "Reasoning Engine", "Task Automation", "Workflow Orchestration", "AI Safety", "Chatbot", "Robotic Process Automation" ] }, { "id": "ai-agents", "title": "ai agents", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Agents are software entities that combine a reasoning core (typically a large language model) with tool-use capabilities, memory, and a perception-action loop to autonomously pursue user-specified goals across multiple steps. They differ from single-shot inference systems by operating in iterative observe-think-act cycles, invoking external APIs, executing code, browsing the web, or delegating sub-tasks to specialised agents. The architecture integrates classical notions of rational agency with modern deep learning, spanning planning, grounding, and self-correction mechanisms. Safety, controllability, and alignment are first-class concerns because agents can initiate irreversible real-world side-effects.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:ai-agents", "labels": [ "AI Agents" ], "is_subclass_of": [ "AI Application", "Autonomous Agent", "Goal-Directed System" ], "wikilinks": [ "Large Language Models", "Tool Use", "Memory", "Reasoning", "Planning and Scheduling", "Perception", "Action Execution", "Multi-Agent Systems", "Autonomous Decision Making", "Workflow Automation", "Cognitive Architecture", "Vector Databases", "Function Calling", "ReAct Framework", "Plan and Execute", "Prompt Engineering", "Retrieval-Augmented Generation", "Sandboxing", "Chatbots", "Robotic Process Automation" ] }, { "id": "ai-alignment", "title": "AI Alignment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of making AI systems' behaviour and goals consistent with human values, preferences, and intentions. AI alignment encompasses technical methods and research aimed at ensuring AI systems act in accordance with human interests, even as they become more capable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ai-alignment", "labels": [ "AI Alignment", "Model Alignment" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "AI Agent System", "MetaverseDomain", "natural language processing" ] }, { "id": "ai-anxiety", "title": "AI Anxiety", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The psychological and social unease experienced by individuals and communities in response to the rapid advancement and potential disruptive impact of artificial intelligence.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-anxiety", "labels": [ "AI Anxiety" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "ai-app-store", "title": "AI App Store", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A centralized digital marketplace platform that enables the discovery, installation, and management of third-party applications and integrations for AI assistants and agents.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-app-store", "labels": [ "AI App Store" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "ai-applications", "title": "AI Applications", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Domain-specific deployments of artificial intelligence technologies delivering measurable value across industries including healthcare diagnostics, autonomous vehicles, industrial automation, financial services, and personal assistants. AI Applications translate research-level techniques\u2014machine learning models, natural language processing, computer vision\u2014into production systems operating within real-world constraints of safety, reliability, and regulatory compliance.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:ai-applications", "labels": [ "AI Applications" ], "is_subclass_of": [ "AI Application", "Artificial Intelligence", "Machine Learning" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Digital Asset", "Smart Contract", "Machine Learning", "Natural Language Processing", "Computer Vision", "Deep Learning", "Generative AI", "Reinforcement Learning", "Autonomous Vehicle", "Autonomous System", "AI Governance", "AI Risk Management", "Large Language Models", "Transformer", "Recommendation System", "Industrial Robotics", "Autonomous Navigation", "Image Recognition" ] }, { "id": "ai-art-generation", "title": "AI Art Generation", "domain": "ai", "domain_name": "Ai", "definition": "AI Art Generation is the production of visual artwork by generative models, most commonly text-to-image diffusion systems that synthesize images from natural-language prompts. Outputs can be steered through prompting, fine-tuning, and lightweight adapters such as LoRA or DreamBooth that teach a model new subjects or styles. It has reshaped creative workflows in illustration, concept art, and design while raising questions about training-data provenance and authorship.", "entityType": "Class", "qualityScore": 0.89, "maturity": "established", "iri": "urn:ngm:class:ai-art-generation", "labels": [ "AI Art Generation", "AI Art", "AI Generation Systems", "Local AI Art Creation" ], "is_subclass_of": [ "Generative AI", "Creative AI", "Image Generation" ], "wikilinks": [ "Generative AI", "Diffusion Models", "Generative Adversarial Networks", "Image Generation", "Deep Learning", "Neural Network", "Transformer", "Large Language Models", "Training Data", "Machine Learning", "Image-to-Image", "Creative AI", "Kohya, DreamBooth and Similar", "Generative Model", "VAE", "GPU Compute", "Stability AI", "Synthetic Media", "Content Creation", "Creative Industries" ] }, { "id": "ai-audit", "title": "AI Audit", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A systematic, independent examination and evaluation of an artificial intelligence system's design, development processes, deployment procedures, operational performance, documentation, governance arrangements, and compliance with applicable requirements, conducted by qualified assessors to verify conformity with specified standards, regulations, ethical principles, or organisational policies, and to identify deficiencies, risks, or opportunities for improvement, producing documented findings and recommendations that support accountability, transparency, and continuous enhancement of AI system trustworthiness.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:ai-audit", "labels": [ "AI Audit" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "access to systems and personnel", "AI documentation", "Conformity Assessment", "Internal Audit", "ISO 19011", "ISO/IEC 23894:2023", "ISO/IEC 42001:2023", "ISO/IEC 5338:2023", "NIST AI Risk Management Framework", "Third-Party Certification", "Accountability", "AI Agent System", "AI Governance", "AI Impact Assessment", "AI Monitoring", "EU AI Act", "MetaverseDomain", "Quality Assurance", "Risk Management", "Transparency" ] }, { "id": "ai-avatars", "title": "AI Avatars", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"Virtual representations of human users generated and animated by artificial intelligence systems, employing machine learning models to synthesise photorealistic appearance, facial expressions, body movements, and speech from sensor inputs or user intentions, enabling realistic telepresence witho...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-avatars", "labels": [ "AI Avatars", "TELE-100-ai-avatars" ], "is_subclass_of": [ "Telepresence", "ArtificialIntelligence" ], "wikilinks": [ "GDPR (General Data Protection Regulation)", "PhotorealisticTelepresence", "SpeechSynthesis", "TELE-020-virtual-reality-telepresence", "TELE-051-3d-gaussian-splatting", "TELE-052-neural-radiance-fields", "TELE-060-instant-ngp", "TELE-102-codec-avatars", "TELE-103-metahuman-creator", "TELE-104-readyplayerme", "TELE-114-lip-sync-technology", "TELE-116-facial-expression-analysis", "UK Online Safety Act (2024)", "ArtificialIntelligence", "TELE-001-telepresence" ] }, { "id": "ai-benchmark-epistemological-critique", "title": "AI Benchmark Epistemological Critique", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A critical epistemological lens applied to AI benchmark claims, model capability assessments, and statistical presentations that may mislead through selective metrics, dataset contamination, cherry-picked results, or hallucination. The page collects resources and reasoning for evaluating AI performance claims with rigour, highlighting how large language models can generate plausible but false outputs that resemble statistical truth.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-benchmark-epistemological-critique", "labels": [ "AI Benchmark Epistemological Critique", "Lies Damn Lies and Statistics" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Large Language Models" ] }, { "id": "ai-benchmarking", "title": "AI Benchmarking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic process of evaluating and comparing the performance, quality, and capabilities of AI models and systems against standardized or internal metrics.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-benchmarking", "labels": [ "AI Benchmarking" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-benchmarks", "title": "AI Benchmarks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Standardized datasets and evaluation metrics used to quantitatively assess the performance, capabilities, and limitations of artificial intelligence models across specific tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-benchmarks", "labels": [ "AI Benchmarks" ], "is_subclass_of": [ "Benchmarks" ], "wikilinks": [] }, { "id": "ai-board", "title": "AI Board", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "EU Advisory body under Articles 65\u201367 of the AI Act, composed of one senior representative per Member State and supported by the AI Office as secretariat. Its mandate spans issuing harmonisation guidelines, resolving cross-border jurisdictional conflicts, coordinating joint enforcement actions, and advising the AI Office on GPAI systemic-risk classification and codes-of-practice adequacy.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-board", "labels": [ "AI Board" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU AI Act Articles 65-67", "AI Agent System", "MetaverseDomain" ] }, { "id": "ai-bubble-dynamics", "title": "AI Bubble Dynamics", "domain": "economics", "domain_name": "Economics", "definition": "The economic and market conditions characterized by rapid valuation increases in AI-related assets, driven by speculative investment and high growth expectations, which carry the risk of a subsequent market correction or crash.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-bubble-dynamics", "labels": [ "AI Bubble Dynamics" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-bubble-narrative", "title": "AI Bubble Narrative", "domain": "economics", "domain_name": "Economics", "definition": "A recurring economic and cultural discourse arguing that the current valuation and adoption of artificial intelligence technologies are unsustainable and prone to a significant market correction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-bubble-narrative", "labels": [ "AI Bubble Narrative" ], "is_subclass_of": [ "Venture Capital" ], "wikilinks": [] }, { "id": "ai-bubble", "title": "AI Bubble", "domain": "economics", "domain_name": "Economics", "definition": "A market condition characterized by speculative overvaluation of artificial intelligence assets and infrastructure investments relative to their current or projected fundamental economic output.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-bubble", "labels": [ "AI Bubble" ], "is_subclass_of": [ "Competition in AI" ], "wikilinks": [] }, { "id": "ai-business-model", "title": "AI Business Model", "domain": "economics", "domain_name": "Economics", "definition": "The economic frameworks and revenue strategies, such as subscription, usage-based, or hybrid models, employed to monetize artificial intelligence products and services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-business-model", "labels": [ "AI Business Model" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-capex-forecast", "title": "AI CapEx Forecast", "domain": "economics", "domain_name": "Economics", "definition": "Projections of capital expenditure by hyperscalers and enterprises on AI infrastructure, including data centers, chips, and power systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-capex-forecast", "labels": [ "AI CapEx Forecast" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "ai-capex-spend", "title": "AI CapEx Spend", "domain": "economics", "domain_name": "Economics", "definition": "The capital expenditure by technology companies on AI infrastructure, including data centers, chips, and energy, driving macroeconomic impacts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-capex-spend", "labels": [ "AI CapEx Spend" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-capability-discontinuity-model", "title": "AI Capability Discontinuity Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A conceptual framework positing that AI capability does not progress linearly but instead exhibits discontinuous threshold transitions\u2014analogous to the mathematical step function\u2014where accumulated incremental improvements trigger rapid, self-reinforcing shifts in capability order, akin to historical inflection points such as the Cambrian explosion or the printing press. The framework draws on self-organisation theory and feedback-loop dynamics to argue that AI represents the next major complexity transition in evolutionary and technological history.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-capability-discontinuity-model", "labels": [ "AI Capability Discontinuity Model", "AI as a step function" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "AI Agent System" ] }, { "id": "ai-capability-evidence-compendium", "title": "AI Capability Evidence Compendium", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A curated reference document compiling empirical evidence countering common objections to AI capability and originality claims, aggregating benchmarks, research findings, and expert commentary demonstrating that large language models exhibit reasoning, world-model construction, and generalisation beyond stochastic pattern matching. Serves as an evidence base for constructive discourse on AI capabilities and their societal implications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-capability-evidence-compendium", "labels": [ "AI Capability Evidence Compendium", "AI Defence Doc" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Generative AI will be designing new drugs all on its own in the near future" ] }, { "id": "ai-capability-temporal-horizon-framework", "title": "AI Capability Temporal Horizon Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Soon-Next-Later is a structured futurology framework for categorising AI capability developments across three temporal horizons: Soon (0\u20135 years, capabilities already emerging in products), Next (5\u201310 years, capabilities requiring current research to mature), and Later (10+ years, speculative capabilities dependent on fundamental advances). The framework provides practitioners with a tractable planning scaffold that avoids both near-term over-hype and long-horizon dismissiveness.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:ai-capability-temporal-horizon-framework", "labels": [ "AI Capability Temporal Horizon Framework", "Soon-Next-Later (AI futurology)" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Agents", "collaborative", "Cyber Security and Cryptography", "Death of the Internet", "Diagrams as Code", "Distributed Identity", "Education and AI", "Metaverse and Telecollaboration", "Politics, Law, Privacy", "Sam Hammond", "Social contract and jobs" ] }, { "id": "ai-capital-expenditure", "title": "AI Capital Expenditure", "domain": "economics", "domain_name": "Economics", "definition": "The aggregate financial outlay by corporations and governments for physical infrastructure, such as data centers and chips, required to train and deploy large-scale artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-capital-expenditure", "labels": [ "AI Capital Expenditure" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-certification", "title": "AI Certification", "domain": "governance", "domain_name": "Governance", "definition": "AI certification is the formal attestation, by an accredited third party, that an artificial intelligence system or its governing management system conforms to defined requirements for safety, robustness, fairness, transparency, and risk management. Unlike generic conformity certification, it is scoped to AI-specific standards and hazards \u2014 data-quality and bias controls, model evaluation, human oversight, and lifecycle governance under frameworks such as ISO/IEC 42001. AI certification converts abstract commitments to trustworthy AI into auditable, marketable evidence, giving procurers, regulators, and the public a credible signal that a system has been independently assessed against recognised criteria rather than self-declared as compliant.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ai-certification", "labels": [ "AI Certification" ], "is_subclass_of": [ "Certification" ], "wikilinks": [ "Certification", "ISO IEC 42001", "Trustworthy AI" ] }, { "id": "ai-chip-geopolitics", "title": "AI Chip Geopolitics", "domain": "governance", "domain_name": "Governance", "definition": "The strategic interplay of national security policies, trade agreements, and diplomatic relations that govern the cross-border flow of advanced semiconductor hardware and associated intellectual property.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-chip-geopolitics", "labels": [ "AI Chip Geopolitics" ], "is_subclass_of": [ "Export Controls" ], "wikilinks": [] }, { "id": "ai-chip-manufacturing", "title": "AI Chip Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The specialized semiconductor fabrication and supply chain processes required to produce high-performance accelerators and TPUs for artificial intelligence workloads.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-chip-manufacturing", "labels": [ "AI Chip Manufacturing" ], "is_subclass_of": [ "Computational Infrastructure" ], "wikilinks": [] }, { "id": "ai-chips", "title": "AI Chips", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AI Chips are specialized processors optimized for the massively parallel linear-algebra workloads of machine-learning training and inference, including GPUs, TPUs, NPUs, and custom ASICs. They deliver high throughput on matrix and tensor operations through wide SIMD execution, high-bandwidth memory, and reduced-precision arithmetic. Control over advanced AI-chip design and fabrication has become a strategic axis of geopolitical and commercial competition.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-chips", "labels": [ "AI Chips" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "ai-code-generation", "title": "AI Code Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of artificial intelligence models to automatically produce, complete, or refactor source code based on natural language prompts or context.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-code-generation", "labels": [ "AI Code Generation" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-coding-agents", "title": "AI Coding Agents", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autonomous software agents capable of writing, debugging, and deploying code, increasingly driving adoption of developer tools.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-coding-agents", "labels": [ "AI Coding Agents" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-coding-assistant", "title": "AI Coding Assistant", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A class of AI tools designed to augment software development workflows by generating, refactoring, and debugging code, thereby enhancing developer productivity.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-coding-assistant", "labels": [ "AI Coding Assistant" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "ai-coding-assistants", "title": "AI Coding Assistants", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software tools that assist developers in writing, reviewing, and debugging code using AI models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-coding-assistants", "labels": [ "AI Coding Assistants" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-coding-tools", "title": "AI Coding Tools", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software applications that leverage large language models to assist, automate, or generate code for software development tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-coding-tools", "labels": [ "AI Coding Tools" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "ai-companies", "title": "AI Companies", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Companies is the population-level concept comprising commercial and non-profit organisations whose principal economic activity is the research, development, productisation, deployment, distribution, or infrastructure-provision of artificial intelligence technologies\u2014spanning the 2024-2026 indu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-companies", "labels": [ "AI Companies" ], "is_subclass_of": [ "AI Governance and Ethics", "Technology Company", "Research Organisation", "Software Vendor", "Industry Sector", "Commercial Enterprise" ], "wikilinks": [ "a16z Generative AI Top 50 2024", "Academic AI Labs", "Academic Research Pipeline", "AGI Race", "Ahmed Wahed 2020 De-democratization of AI", "AI Acquisitions and M&A", "AI APIs", "AI Funding Landscape", "AI Infrastructure Providers", "AI Research Laboratory", "AI Talent War", "AI21 Labs", "Amazon", "AMD", "Anthropic", "Anthropic Core Views on AI Safety 2023", "Application AI Companies", "Autonomous Systems", "Bender Gebru et al. 2021 Stochastic Parrots", "Bommasani et al. 2021 Foundation Models" ] }, { "id": "ai-compute-infrastructure", "title": "AI Compute Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical and logical systems, including data centers, GPUs, and networking, that provide the processing power required to train and deploy large-scale artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-compute-infrastructure", "labels": [ "AI Compute Infrastructure" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-compute-strategy", "title": "AI Compute Strategy", "domain": "economics", "domain_name": "Economics", "definition": "The corporate and economic approaches used by organizations to acquire, allocate, and leverage computational resources to gain competitive advantage in the AI market.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-compute-strategy", "labels": [ "AI Compute Strategy" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-concept", "title": "AI Concept", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The foundational theoretical constructs and abstract principles underlying artificial intelligence systems, encompassing both symbolic approaches (knowledge representation, logic-based reasoning) and subsymbolic methods (connectionist networks, distributed representations). AI Concepts define learning paradigms, reasoning mechanisms, and cognitive architectures that serve as the theoretical basis for designing, evaluating, and advancing intelligent systems across research and application domains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-concept", "labels": [ "AI Concept" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Cognitive Architecture", "Artificial Intelligence", "Blockchain", "Knowledge Representation", "Machine Learning", "owl:Thing" ] }, { "id": "ai-context", "title": "AI Context", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The specific, often private or situational information required by an AI model to perform tasks effectively for a particular user, representing a critical bottleneck in AI deployment beyond raw model capability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-context", "labels": [ "AI Context" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "ai-copyright-infringement", "title": "AI Copyright Infringement", "domain": "governance", "domain_name": "Governance", "definition": "Legal disputes and policy issues arising from the use of copyrighted works to train or generate content with AI models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-copyright-infringement", "labels": [ "AI Copyright Infringement" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-corporate-governance", "title": "AI Corporate Governance", "domain": "governance", "domain_name": "Governance", "definition": "The legal and structural frameworks governing the ownership, equity distribution, and fiduciary responsibilities of artificial intelligence research organizations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-corporate-governance", "labels": [ "AI Corporate Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-cybersecurity", "title": "AI Cybersecurity", "domain": "security", "domain_name": "Security", "definition": "The intersection of artificial intelligence and information security, encompassing both the use of AI to enhance defensive security measures and the mitigation of risks posed by AI systems themselves.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-cybersecurity", "labels": [ "AI Cybersecurity" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-data-center-capacity", "title": "AI Data Center Capacity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The total power and compute throughput available to a specific facility or cluster for training and serving artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-data-center-capacity", "labels": [ "AI Data Center Capacity" ], "is_subclass_of": [ "Computational Infrastructure" ], "wikilinks": [] }, { "id": "ai-data-center-infrastructure", "title": "AI Data Center Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical and computational facilities, including specialized hardware like GPUs and power systems, designed to host and scale large-scale AI model training and inference workloads.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-data-center-infrastructure", "labels": [ "AI Data Center Infrastructure" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "ai-data-centers", "title": "AI Data Centers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Specialized data centers designed for AI workloads, with unique requirements for power, cooling, and network connectivity.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-data-centers", "labels": [ "AI Data Centers" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "ai-data-retention-policy", "title": "AI Data Retention Policy", "domain": "governance", "domain_name": "Governance", "definition": "A specific subset of data governance rules that dictate the duration and conditions under which prompts, outputs, and training data generated by or for AI systems are stored, reviewed, or deleted.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-data-retention-policy", "labels": [ "AI Data Retention Policy" ], "is_subclass_of": [ "Retention Policy" ], "wikilinks": [] }, { "id": "ai-demand-elasticity-framework", "title": "AI Demand Elasticity Framework", "domain": "economics", "domain_name": "Economics", "definition": "A conceptual model categorizing how artificial intelligence alters market demand through specific dimensions such as price, access, complexity, continuity, personalization, and relational value.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-demand-elasticity-framework", "labels": [ "AI Demand Elasticity Framework" ], "is_subclass_of": [ "Economic Impact of AI" ], "wikilinks": [] }, { "id": "ai-deployment", "title": "AI Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The phase of the AI lifecycle in which a developed and validated artificial intelligence system is integrated into operational environments, made available to end users, and transitioned from development to production use, encompassing activities such as system integration, infrastructure provisioning, release management, user training, documentation delivery, and the establishment of operational support structures to ensure reliable, safe, and effective system functioning in real-world conditions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ai-deployment", "labels": [ "AI Deployment" ], "is_subclass_of": [ "AI Lifecycle" ], "wikilinks": [ "AI system use by AI Users and AI Operators", "Deployment infrastructure", "FDA Software as Medical Device", "ISO/IEC 25010", "ISO/IEC 42001:2023", "ISO/IEC 5338:2023", "NIST AI Risk Management Framework", "operational procedures", "AI Agent System", "AI Development", "AI Lifecycle", "AI Monitoring", "AI Operator", "AI Provider", "AI User", "EU AI Act", "Human Oversight", "MetaverseDomain", "Model Performance", "Risk Management" ] }, { "id": "ai-detection", "title": "AI Detection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The set of computational methods and tools used to identify and classify content as being generated by artificial intelligence systems rather than human authors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-detection", "labels": [ "AI Detection" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-development-tools", "title": "AI Development Tools", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software frameworks, libraries, platforms, and integrated development environments designed to facilitate the creation, training, deployment, and maintenance of artificial intelligence systems. Includes deep learning frameworks (TensorFlow, PyTorch, JAX), AutoML platforms, model optimisation toolkits, MLOps infrastructure, and AI-assisted IDEs. Modern tools emphasise reproducibility, scalability, experiment tracking, and continuous delivery of AI systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-development-tools", "labels": [ "AI Development Tools" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Deep Learning Framework", "MLOps", "Model Deployment", "AutoML", "Blockchain" ] }, { "id": "ai-development", "title": "AI Development", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The phase of the AI lifecycle encompassing the design, creation, training, and validation of artificial intelligence systems, including activities such as algorithm selection, data preparation, model architecture design, training process execution, hyperparameter optimisation, performance eand do...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:ai-development", "labels": [ "AI Development" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AI Model", "computational resources", "ISO/IEC 23053", "ISO/IEC 25059", "ISO/IEC 42001:2023", "ISO/IEC 5338:2023", "NIST AI Risk Management Framework", "Stable Video Diffusion", "technical expertise", "AI Agent System", "AI Deployment", "AI Lifecycle", "Bias", "Bias in Large Language Models", "EU AI Act", "Explainability", "Fairness", "Large Language Models", "Lead Poisoning Hypothesis", "Machine Learning" ] }, { "id": "ai-diagram-tools", "title": "AI Diagram Tools", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Diagram Tools are a class of generative AI applications that translate natural-language descriptions, source code, screenshots, or structured specifications into machine-renderable diagram artefacts (flowcharts, sequence diagrams, entity-relationship models, class diagrams, mind maps, architec...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-diagram-tools", "labels": [ "AI Diagram Tools" ], "is_subclass_of": [ "AI Application", "Generative AI Application", "LLM-Powered Tool", "Diagram-as-Code", "Knowledge Visualisation", "AI Productivity Tool" ], "wikilinks": [ "ADR Generation", "AI Productivity Tool", "Architecture Documentation", "Belouadi et al 2024 AutomaTikZ", "Brown 2018 C4 Model", "Brown 2024 Art of Visualising Software Architecture", "Browser Runtime", "C4 Model", "C4 Model Convention", "Claude 3.5", "Claude Artifacts", "Code-to-Diagram Reverse Engineering", "D2 Diagram Language", "D2 Language Documentation", "Diagram-as-Code", "Diagram-Driven Development", "Diagram Grammar Compiler", "Diagram Specification Language", "Diffusion Model", "Dwyer Marriott Wybrow 2008 Constrained Graph Layout" ] }, { "id": "ai-documentation-standards", "title": "AI Documentation Standards", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Documentation Standards are structured frameworks and templates for comprehensively documenting AI systems, datasets, and models to ensure transparency, accountability, reproducibility, and informed stakeholder decision-making throughout the AI lifecycle.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:ai-documentation-standards", "labels": [ "AI Documentation Standards", "AI Documentation", "AI System Documentation" ], "is_subclass_of": [ "AI Governance and Ethics", "Documentation Standards", "AI Governance", "AI Governance Framework", "Responsible AI" ], "wikilinks": [ "Datasheets (Gebru et al.)", "ISO/IEC 23053", "Model Cards (Mitchell et al.)", "AIEthicsDomain", "ConceptualLayer", "EU AI Act", "Smart Contract", "AI Governance", "IEEE 7001", "ISO/IEC 42001", "NIST AI RMF", "System Cards", "FactSheets (IBM)", "Data Cards", "Algorithmic Transparency", "Responsible AI", "AI Risk Management", "GDPR", "Regulatory Compliance", "Model Context Protocol" ] }, { "id": "ai-documentation", "title": "AI Documentation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The structured transparency artefacts that record what an AI system is, how it was built, and how it behaves \u2014 model cards, datasheets for datasets, system cards, technical files, and decision logs. AI documentation is the evidentiary substrate of AI accountability: it lets developers, deployers, auditors, and regulators trace capabilities, limitations, training data provenance, and evaluation results, and is increasingly mandated by regimes such as the EU AI Act.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ai-documentation", "labels": [ "AI Documentation" ], "is_subclass_of": [ "Documentation" ], "wikilinks": [ "Documentation", "AI Accountability", "Transparency", "Training Data" ] }, { "id": "ai-economic-impact", "title": "AI Economic Impact", "domain": "economics", "domain_name": "Economics", "definition": "The analysis of how artificial intelligence technologies influence labor markets, capital allocation, productivity, and macroeconomic structures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-economic-impact", "labels": [ "AI Economic Impact" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "ai-ecosystem", "title": "AI Ecosystem", "domain": "ai", "domain_name": "Ai", "definition": "The AI Ecosystem is the interconnected network of organisations, technologies, standards, talent pipelines, regulatory frameworks, and capital flows that collectively produce, deploy, and govern artificial intelligence systems. It encompasses foundation model providers, cloud infrastructure operators, toolchain vendors, application developers, research institutions, standardisation bodies, and end-user communities, together constituting the supply chain and governance fabric of AI as a general-purpose technology.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:ai-ecosystem", "labels": [ "AI Ecosystem", "Google AI Ecosystem" ], "is_subclass_of": [ "Digital Asset Ecosystem", "Digital Economy" ], "wikilinks": [ "Large Language Models", "Agent Frameworks", "Agentic AI", "AI Policy", "Digital Asset Ecosystem", "Generative AI", "AI Inference", "Foundation Models", "GPU Computing", "AI Governance", "Open Source AI", "EU AI Act", "AI Safety", "Responsible AI", "AI Documentation Standards", "Model Context Protocol", "AI Chips", "Machine Learning", "Deep Learning", "Natural Language Processing" ] }, { "id": "ai-employment-effects", "title": "AI Employment Effects", "domain": "economics", "domain_name": "Economics", "definition": "The measurable impact of artificial intelligence adoption on labor market outcomes, including job displacement, hiring rates, and wage structures across different skill levels and sectors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-employment-effects", "labels": [ "AI Employment Effects" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [] }, { "id": "ai-energy-consumption", "title": "AI Energy Consumption", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The total electrical power and computational resources required to train, deploy, and operate artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-energy-consumption", "labels": [ "AI Energy Consumption" ], "is_subclass_of": [ "Energy Consumption" ], "wikilinks": [] }, { "id": "ai-energy-optimisation", "title": "AI Energy Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The set of techniques and engineering practices that reduce the energy consumed by training and running artificial intelligence systems while preserving acceptable task performance, spanning model compression, efficient hardware selection, workload scheduling, and data-centre operation.", "entityType": "Class", "qualityScore": 0.87, "maturity": "growing", "iri": "urn:ngm:class:ai-energy-optimisation", "labels": [ "AI Energy Optimisation", "Energy Optimisation" ], "is_subclass_of": [ "Artificial Intelligence", "AI Infrastructure (Artificial Intelligence)", "Model Compression", "Green AI" ], "wikilinks": [ "Hardware Acceleration", "Knowledge Distillation", "Edge AI", "Edge Computing", "Artificial Intelligence", "Model Pruning", "Model Quantisation", "Neural Architecture Search", "Deep Learning", "GPU Computing", "Carbon-Aware Computing", "Mixed Precision Training", "AI Governance", "Regulatory Compliance", "Large Language Model", "High-Performance Computing", "Data Centre", "Renewable Energy", "Environmental Sustainability", "Federated Learning" ] }, { "id": "ai-energy-scarcity", "title": "ai energy scarcity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systemic infrastructure bottleneck where exponential growth in AI training and inference compute demand outstrips available energy supply and grid capacity. Drives hyperscaler investment in dedicated power generation including small modular reactors (SMRs) and nuclear recommissioning, while creating geopolitical competition for compute sovereignty.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-energy-scarcity", "labels": [ "AI Energy Scarcity" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "ai-ethics-board", "title": "AI Ethics Board", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An AI Ethics Board is a formally constituted, multidisciplinary oversight committee responsible for reviewing AI systems and deployments against ethical principles, organisational values, and applicable regulatory requirements. Comprising technical experts, ethicists, legal and compliance professionals, domain specialists, and stakeholder representatives, the board conducts structured ethical impact assessments, provides binding or advisory guidance on AI deployment decisions, and monitors deployed systems for ongoing compliance. It serves as a key institutional mechanism for operationalising responsible AI principles and maintaining human oversight of consequential automated decision-making.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-ethics-board", "labels": [ "AI Ethics Board" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "EU HLEG AI", "IEEE P7000", "ISO/IEC 42001:2023", "AIEthicsDomain", "ConceptualLayer", "Smart Contract" ] }, { "id": "ai-ethics-checklist", "title": "AI Ethics Checklist", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Structured verification process evaluating fairness, accountability, transparency, and ethical compliance of AI systems against established governance frameworks including IEEE 7000, OECD AI Principles, UNESCO AI Ethics Recommendations, the EU AI Act, NIST AI RMF, and ISO/IEC 42001.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:ai-ethics-checklist", "labels": [ "AI Ethics Checklist" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance Framework", "TrustAndGovernanceDomain", "Responsible AI" ], "wikilinks": [ "Accountability Framework", "AI System Documentation", "Assessment Methodology", "Bias Detection Protocol", "Ethical AI Deployment", "Ethical Guidelines", "Fairness Assessment Criteria", "IEEE 7000", "IEEE 7000 Standard", "OECD AI Principles", "Stakeholder Trust", "UNESCO AI Ethics Recommendations", "AI Agent System", "AI Governance Framework", "Compliance Verification", "EU AI Act", "MiddlewareLayer", "Risk Assessment", "Transparency Metrics", "TrustAndGovernanceDomain" ] }, { "id": "ai-ethics", "title": "AI Ethics", "domain": "governance", "domain_name": "Governance", "definition": "AI Ethics addresses the moral principles, values, and guidelines governing the design, development, deployment, and use of artificial intelligence systems. This interdisciplinary field examines fairness, accountability, transparency, privacy, and bias mitigation, drawing on philosophy, law, and computer science to ensure AI systems respect human rights and societal values.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:ai-ethics", "labels": [ "AI Ethics" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Algorithmic Fairness", "AI Governance", "Explainable AI", "Responsible AI", "Smart Contract" ] }, { "id": "ai-evaluation", "title": "AI Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "AI Evaluation is the systematic, multi-dimensional measurement of AI model capabilities, reliability, safety, alignment, and societal impact using benchmarks, held-out task suites, human preference judgments, automated graders, and adversarial probing. It encompasses the full lifecycle of rigorous assessment from pre-deployment capability elicitation through post-deployment monitoring, producing quantitative metrics and qualitative findings that guide model selection, release decisions, regulatory compliance, and ongoing risk assessment. Credible evaluation must guard against benchmark contamination, distribution shift, Goodhart pressures, and over-optimisation toward narrow scores, while ensuring elicitation methods are sufficient to reveal the true capability ceiling of the system under test.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:ai-evaluation", "labels": [ "AI Evaluation" ], "is_subclass_of": [ "AI Research Area", "AI Safety" ], "wikilinks": [ "Evaluation Benchmarks and Leaderboards", "Evaluation Metric", "Benchmark Contamination", "Benchmark Saturation", "Human Preference Evaluation", "LLM-as-Judge", "Red-Teaming", "Capability Elicitation", "Dynamic Evaluation", "MMLU", "HELM", "Humanity's Last Exam", "LiveCodeBench", "Chatbot Arena", "MT-Bench", "Arena-Hard", "Goodhart's Law", "Distribution Shift", "Safety Evaluation", "Alignment Evaluation" ] }, { "id": "ai-executive-accountability", "title": "AI Executive Accountability", "domain": "governance", "domain_name": "Governance", "definition": "The organizational governance structure that assigns specific responsibility and oversight for AI strategy, risk, and value realization to C-suite leadership, such as the CEO or CTO.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-executive-accountability", "labels": [ "AI Executive Accountability" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-existential-risk-probability-estimate", "title": "AI Existential Risk Probability Estimate", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "p(doom) is an informal probabilistic estimate, expressed as a value between 0 and 1, of the likelihood that advanced AI development leads to an existential catastrophe for humanity\u2014typically through loss of control over a misaligned superintelligent system. The metric is widely discussed in AI safety research communities as a shorthand for aggregating personal credences about existential risk, and is distinct from rigorous formal risk models, serving primarily as a rhetorical and community-calibration tool.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:ai-existential-risk-probability-estimate", "labels": [ "AI Existential Risk Probability Estimate", "p(doom)" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "ai-export-controls", "title": "AI Export Controls", "domain": "governance", "domain_name": "Governance", "definition": "Governmental regulations and policies that restrict the cross-border transfer of artificial intelligence technologies, models, and related hardware to protect national security interests.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-export-controls", "labels": [ "AI Export Controls" ], "is_subclass_of": [ "AI Regulation" ], "wikilinks": [] }, { "id": "ai-fairness", "title": "AI Fairness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Fairness refers to the principle and practice of ensuring artificial intelligence systems produce equitable outcomes without bias or discrimination based on protected characteristics such as race, gender, age, or socioeconomic status.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:ai-fairness", "labels": [ "AI Fairness", "AI Fairness Certification", "Fairness in AI" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "Equitable AI Systems", "AI Ethics", "Artificial Intelligence", "Smart Contract" ] }, { "id": "ai-framework", "title": "AI Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A comprehensive software architecture that provides reusable code, design patterns, and infrastructure for developing artificial intelligence applications. AI frameworks abstract low-level computational details, offering high-level interfaces for model construction, training, and inference. Examples include TensorFlow, PyTorch, JAX, and scikit-learn, each optimised for distinct use cases from research prototyping to production deployment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-framework", "labels": [ "AI Framework" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Deep Learning Framework", "Model Optimization", "Neural Network Library", "AutoML", "Blockchain", "owl:Thing" ] }, { "id": "ai-frontier-capability-survey", "title": "AI Frontier Capability Survey", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "State of the Art in AI is a curated survey of the current frontier of artificial intelligence capability, covering large language models, generative AI, multimodal systems, hardware advances, and sociotechnical implications. It synthesises emerging research directions, benchmark performance milestones, and near-term deployment trajectories across proprietary and open-source AI ecosystems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-frontier-capability-survey", "labels": [ "AI Frontier Capability Survey", "State of the art in AI" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "presentation", "rework based on gpt", "Death of the Internet", "Education and AI", "Hardware and Edge", "Infrastructure", "Politics, Law, Privacy", "Proprietary Large Language Models", "Social contract and jobs", "Trust and Safety", "Update Cycle" ] }, { "id": "ai-game-agent", "title": "AI Game Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An AI Game Agent is an intelligent autonomous entity embedded within a video game or interactive virtual environment that perceives its local Game State through sensory abstraction, reasons over a structured decision framework, and executes goal-directed actions to create engaging, adaptive, and believable interactive experiences. Contemporary AI Game Agents synthesise classical symbolic control techniques \u2014 Finite State Machines, Behavior Trees, Goal-Oriented Action Planning, Hierarchical Task Networks \u2014 with data-driven methods including deep Reinforcement Learning, Imitation Learning from expert demonstrations, and Markov Decision Process formulations over Partially Observable environments. The most capable agents further incorporate Large Language Models as an 'inner monologue' for context-aware dialogue and high-level planning, while a learned Reinforcement Learning policy serves as the low-level action executor under Real-Time Constraints. The archetype spans NPC characters in narrative games, competitive game-playing agents trained through Self-Play such as AlphaGo and OpenAI Five, procedurally adaptive companions that model player behaviour through Player Modelling, and scripted simulation agents used in Automated Playtesting and Game Analytics pipelines.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:ai-game-agent", "labels": [ "AI Game Agent", "Game Playing Agent", "Game-Playing Agents" ], "is_subclass_of": [ "AI Application", "AI Agent System", "Intelligent Virtual Entity" ], "wikilinks": [ "Adaptive Challenge", "Adaptive Difficulty", "Behavior Tree", "Decision Engine", "Dynamic Gameplay", "Emergent Behavior", "Game AI Pro", "Game State", "IEEE CIG", "Navigation Mesh", "Pathfinding System", "Player Engagement", "Procedural Behavior", "Unity ML-Agents", "UserExperienceLayer", "AI Agent System", "AI-GroundedDomain", "ComputeLayer", "CreativeMediaDomain", "Game Engine" ] }, { "id": "ai-geopolitics", "title": "AI Geopolitics", "domain": "governance", "domain_name": "Governance", "definition": "The study of how artificial intelligence capabilities influence international relations, national security strategies, and the balance of power between states.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-geopolitics", "labels": [ "AI Geopolitics" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-governance-framework", "title": "AI Governance Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Set of policies and procedures ensuring responsible development and operation of AI components in the metaverse.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:ai-governance-framework", "labels": [ "AI Governance Framework", "AI Model Governance", "Model AI Governance Framework" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance", "Governance Model", "Policy Framework" ], "wikilinks": [ "AI Accountability", "AI Risk Assessment", "AI Transparency", "Audit Mechanism", "Compliance Policy", "Ethical AI Operation", "ISO/IEC 42001", "OECD AI Principles", "Policy Framework", "Access Control", "AI Agent System", "AI Ethics Checklist", "Algorithmic Transparency Index", "Data Governance", "Ethical Framework", "Governance Model", "Identity Management", "MiddlewareLayer", "Regulatory Compliance", "Responsible AI" ] }, { "id": "ai-governance-law-and-privacy", "title": "AI Governance Law and Privacy", "domain": "governance", "domain_name": "Governance", "definition": "The intersecting domain of political governance, legal frameworks, and individual privacy rights as they apply to AI and digital technologies. This cluster addresses how legislation (e.g., EU AI Act, GDPR), regulatory enforcement, and political economy shape what AI systems are permitted to do, how data about individuals is collected and used, and how citizens can exercise rights against automated decision-making and surveillance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-governance-law-and-privacy", "labels": [ "AI Governance Law and Privacy", "Politics, Law, Privacy" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "pol", "presentation", "copyright", "Education and AI", "Politics, Law, Privacy", "Trust and Safety" ] }, { "id": "ai-governance-maturity-model", "title": "AI Governance Maturity Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Governance Maturity Model is an assessment framework that defines progressive maturity levels for AI governance capabilities across multiple dimensions, enabling organizations to evaluate current practices, identify gaps, benchmark against peers, and guide continuous improvement toward trustwo...", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:ai-governance-maturity-model", "labels": [ "AI Governance Maturity Model" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance Framework", "Governance Model", "Maturity Model", "Assessment Framework" ], "wikilinks": [ "CMMI Institute", "ISO 21827", "Singapore Model AI Governance Framework", "AIEthicsDomain", "ConceptualLayer", "Smart Contract" ] }, { "id": "ai-governance", "title": "AI Governance", "domain": "governance", "domain_name": "Governance", "definition": "AI Governance encompasses the policies, frameworks, standards, and institutional mechanisms for overseeing the responsible development, deployment, and use of artificial intelligence technologies, including regulatory compliance, organisational accountability structures, risk-based assessment, algorithmic transparency requirements, and stakeholder engagement processes that balance innovation with societal protection.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ai-governance", "labels": [ "AI Governance", "AI Governance Ecosystem", "Enterprise AI Governance", "International AI Governance", "Risk-Based AI Governance" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [ "GDPR", "AI Ethics", "EU AI Act", "Regulatory Compliance", "Responsible AI", "Risk Management", "Smart Contract" ] }, { "id": "ai-group-formation", "title": "AI Group Formation", "domain": "ai", "domain_name": "Ai", "definition": "AI Group Formation is the automated assignment of participants into subgroups using models that optimise for criteria such as skill balance, topic affinity, social diversity, or learning objectives. In collaborative and meeting platforms it drives features like intelligent breakout-room allocation, replacing manual or random grouping. The technique combines clustering, optimisation, and sometimes reinforcement learning over participant attributes and interaction signals.", "entityType": "Class", "qualityScore": 0.87, "maturity": "emerging", "iri": "urn:ngm:class:ai-group-formation", "labels": [ "AI Group Formation" ], "is_subclass_of": [ "AI Application", "Combinatorial Optimisation", "Unsupervised Learning" ], "wikilinks": [ "Breakout Room", "Educational Technology", "Clustering", "Reinforcement Learning", "Unsupervised Learning", "Machine Learning", "Multi-Agent System", "Collaborative Learning", "Team Formation", "Participant Profiling", "Cooperative Learning", "Learning Management System", "Learning Analytics", "Constraint Satisfaction", "Genetic Algorithm", "K-Means Clustering", "Embedding", "Recommender System", "Natural Language Processing", "Diversity Optimisation" ] }, { "id": "ai-hardware-ecosystem", "title": "AI Hardware Ecosystem", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The interconnected network of specialized computing hardware, software stacks, and device manufacturers that enable the deployment and utilization of artificial intelligence models across various consumer and enterprise applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-hardware-ecosystem", "labels": [ "AI Hardware Ecosystem" ], "is_subclass_of": [ "AI Hardware" ], "wikilinks": [] }, { "id": "ai-hardware", "title": "AI Hardware", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Hardware encompasses specialized computing hardware designed to accelerate artificial intelligence and machine learning workloads, including GPUs, TPUs, NPUs, and other AI accelerators optimized for training neural networks and running inference at scale. These processors feature architectures specifically designed for matrix operations, parallel processing, and low-precision arithmetic fundamental to modern AI algorithms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-hardware", "labels": [ "AI Hardware" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Computer Hardware", "Large Language Models" ] }, { "id": "ai-harnesses", "title": "AI Harnesses", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The software frameworks, user interfaces, and integration layers that wrap around AI models to enable their practical application and value extraction in specific contexts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-harnesses", "labels": [ "AI Harnesses" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-ipo", "title": "AI IPO", "domain": "economics", "domain_name": "Economics", "definition": "The process of artificial intelligence companies transitioning to public equity markets, representing a key mechanism for capital formation and valuation in the AI sector.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-ipo", "labels": [ "AI IPO" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-image-generation", "title": "AI Image Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of generative AI models to create or modify visual content from text or other inputs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-image-generation", "labels": [ "AI Image Generation" ], "is_subclass_of": [ "Generative Model" ], "wikilinks": [] }, { "id": "ai-impact-assessment", "title": "AI Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A systematic process for identifying, analysing, evaluating, and documenting the potential positive and negative effects of an artificial intelligence system on individuals, groups, organisations, society, and the environment across multiple dimensions including fundamental rights, ethical principles, safety, fairness, privacy, environmental sustainability, and socioeconomic impacts, conducted prior to deployment and periodically thereafter to inform design decisions, risk mitigation strategies, governance arrangements, and stakeholder communication regarding AI system consequences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ai-impact-assessment", "labels": [ "AI Impact Assessment", "AI System Impact Assessment" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "DPIA", "EU AI Act Article 27", "GDPR", "Informed decision-making", "ISO/IEC 23894:2023", "ISO/IEC 24368", "ISO/IEC 42001:2023", "NIST AI Risk Management Framework", "UN Guiding Principles on Business and Human Rights", "accountability", "Accountability", "AI Audit", "AI Governance", "AI Lifecycle", "Blockchain", "Fairness", "Human Rights", "MetaverseDomain", "Risk Management", "Smart Contract" ] }, { "id": "ai-incident", "title": "AI Incident", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An event in which an AI system causes or has the potential to cause harm to individuals, property, the environment, or fundamental rights, encompassing technical failures, security breaches, bias-driven discrimination, privacy violations, and unintended consequences. Under the EU AI Act Article 72, serious incidents must be reported to national competent authorities within 15 days; structured incident taxonomies and response protocols (detection, containment, investigation, remediation) are codified in ISO/IEC 23894:2023.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-incident", "labels": [ "AI Incident" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Risk" ], "wikilinks": [ "EU AI Act Regulation 2024/1689 Article 72", "ISO/IEC 23894:2023", "OECD AI Incident Monitor", "AI Agent System", "MetaverseDomain" ] }, { "id": "ai-industry-consolidation", "title": "AI Industry Consolidation", "domain": "economics", "domain_name": "Economics", "definition": "The process of mergers, acquisitions, and strategic partnerships within the artificial intelligence sector that reduces the number of independent market participants and concentrates capabilities, capital, and talent.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-industry-consolidation", "labels": [ "AI Industry Consolidation" ], "is_subclass_of": [ "Competition in AI" ], "wikilinks": [] }, { "id": "ai-inequality", "title": "AI Inequality", "domain": "governance", "domain_name": "Governance", "definition": "The widening disparity in access to, benefits from, and power derived from artificial intelligence technologies, driven by socioeconomic privilege and resource concentration.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-inequality", "labels": [ "AI Inequality" ], "is_subclass_of": [ "AI Governance Law and Privacy" ], "wikilinks": [] }, { "id": "ai-inference-cost-management", "title": "AI Inference Cost Management", "domain": "economics", "domain_name": "Economics", "definition": "The practice of monitoring, optimizing, and budgeting for the computational expenses associated with running large language models and AI services in production.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-inference-cost-management", "labels": [ "AI Inference Cost Management" ], "is_subclass_of": [ "Enterprise Budgeting" ], "wikilinks": [] }, { "id": "ai-inference-costs", "title": "AI Inference Costs", "domain": "economics", "domain_name": "Economics", "definition": "The financial expenditure associated with executing trained AI models on production workloads, encompassing compute, memory, and latency trade-offs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-inference-costs", "labels": [ "AI Inference Costs" ], "is_subclass_of": [ "AI Inference" ], "wikilinks": [] }, { "id": "ai-inference-hardware", "title": "AI Inference Hardware", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Specialized computing chips and server architectures designed to optimize the speed and cost of running AI models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-inference-hardware", "labels": [ "AI Inference Hardware" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "ai-inference-infrastructure", "title": "AI Inference Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The specialized hardware, data centers, and energy systems required to run AI models at scale, distinct from training infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-inference-infrastructure", "labels": [ "AI Inference Infrastructure" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "ai-inference", "title": "AI Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "AI Inference is the operational phase of a machine learning system in which a trained model processes new input data to generate predictions, classifications, embeddings, or generative outputs \u2014 as distinct from the training phase in which model parameters are optimised. It is the production-time workload that determines the latency, throughput, cost, and scalability characteristics experienced by end users and downstream systems. Inference encompasses the full serving stack: model loading and memory management, request batching, hardware-specific kernel execution, and output post-processing. Modern inference optimisation techniques \u2014 including quantisation, speculative decoding, continuous batching, and KV-cache management \u2014 are essential for deploying large models at commercially viable cost.", "entityType": "Class", "qualityScore": 0.91, "maturity": "mature", "iri": "urn:ngm:class:ai-inference", "labels": [ "AI Inference", "AI Inference Architecture", "AI Inference Service", "AI-inference", "Efficient Inference", "Private AI Inference" ], "is_subclass_of": [ "Machine Learning", "Neural Network" ], "wikilinks": [ "GPU Acceleration", "AI Model", "AI Model Architecture", "Tensor Parallelism", "Quantisation", "Real-Time AI Inference", "Generative AI", "Natural Language Processing", "Computer Vision", "Autonomous Agents", "Model Weights", "Inference Runtime", "Hardware Accelerator", "KV Cache", "Speculative Decoding", "Continuous Batching", "Prefix Caching", "Model Training", "Deep Learning", "Fine-Tuning" ] }, { "id": "ai-infrastructure-capital-expenditure", "title": "AI Infrastructure Capital Expenditure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The large-scale financial investment by technology companies and hyperscalers in physical computing resources, such as GPUs, data centers, and networking, required to train and deploy large-scale foundation models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-infrastructure-capital-expenditure", "labels": [ "AI Infrastructure Capital Expenditure" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "ai-infrastructure-economics", "title": "AI Infrastructure Economics", "domain": "economics", "domain_name": "Economics", "definition": "The study of the economic dynamics, cost structures, and market forces governing the production, allocation, and pricing of computational resources essential for AI development.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-infrastructure-economics", "labels": [ "AI Infrastructure Economics" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-infrastructure-investment", "title": "AI Infrastructure Investment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The large-scale capital allocation and construction of physical and digital assets, such as data centers and compute clusters, required to support the training and deployment of artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-infrastructure-investment", "labels": [ "AI Infrastructure Investment" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "ai-infrastructure-regulation", "title": "AI Infrastructure Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Government policies and regulations governing the physical deployment of AI infrastructure, including data centers, energy use, and community impact.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-infrastructure-regulation", "labels": [ "AI Infrastructure Regulation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-infrastructure-strategy", "title": "AI Infrastructure Strategy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The corporate and competitive planning focused on securing compute resources, data pipelines, and physical hardware to support the development and deployment of artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-infrastructure-strategy", "labels": [ "AI Infrastructure Strategy" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "ai-infrastructure", "title": "AI Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AI Infrastructure is the integrated ensemble of hardware, software, data systems, and operational tooling required to develop, train, deploy, monitor, and govern artificial intelligence and machine learning workloads at scale. It spans physical compute resources such as GPU and TPU clusters, networking fabrics, and storage systems through to cloud-managed AI platforms, model-serving runtimes, data pipelines, and MLOps toolchains. Unlike general-purpose computing infrastructure, AI Infrastructure is specifically optimised for tensor operations, distributed parallel training, high-throughput vector data ingestion, and low-latency inference serving. It constitutes the production backbone that determines the cost, velocity, reliability, and scalability of AI system development cycles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-infrastructure", "labels": [ "AI Infrastructure", "AI Infrastructure (Artificial Intelligence)", "AI Infrastructure Providers" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [ "Artificial Intelligence", "Blockchain" ] }, { "id": "ai-investment", "title": "AI Investment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Investment is the allocation of capital toward artificial-intelligence research, talent, compute infrastructure, startups, and data assets in pursuit of strategic or financial returns. It spans venture funding, corporate capital expenditure on data centres and chips, sovereign wealth funds, public research grants, and hyperscaler balance-sheet commitments. The scale and concentration of this capital is a primary driver of competitive dynamics among firms and nations, with global AI corporate investment reaching $581 billion in 2025 and Q1 2026 venture activity alone surpassing that figure. Investment categories include frontier-model lab funding, compute infrastructure build-out, application-layer enterprise AI, and sovereign compute programmes. Returns are gated by access to scarce inputs \u2014 advanced semiconductor capacity, high-bandwidth memory, and specialised research talent \u2014 creating oligopolistic supply-side dynamics that shape long-run market structure.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:ai-investment", "labels": [ "AI Investment", "AI Investment Cycle" ], "is_subclass_of": [ "Economics", "Capital Markets", "Competition in AI", "Technology Race" ], "wikilinks": [ "AI Chips", "AI Cooperation", "AI Policy", "AI Regulation", "AI Research Talent", "AI Sovereignty", "AI Startups", "Alan Turing Institute", "Anthropic", "Antitrust Oversight", "Capital Markets", "Cloud Infrastructure", "Compute Access", "Compute Governance", "Corporate Capital Expenditure", "Data Centre Infrastructure", "DeepSeek", "Economic Competitiveness", "Enterprise AI Adoption", "Export Controls" ] }, { "id": "ai-job-creation", "title": "AI Job Creation", "domain": "economics", "domain_name": "Economics", "definition": "The net increase in employment opportunities and roles directly or indirectly generated by the adoption and expansion of artificial intelligence technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-job-creation", "labels": [ "AI Job Creation" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-job-displacement", "title": "AI Job Displacement", "domain": "economics", "domain_name": "Economics", "definition": "The reduction or elimination of human labor roles due to the automation and efficiency gains provided by artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-job-displacement", "labels": [ "AI Job Displacement" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-kill-switch-bill", "title": "AI Kill Switch Bill", "domain": "governance", "domain_name": "Governance", "definition": "US legislation requiring AI companies to maintain the technical capability to shut down models during safety incidents.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-kill-switch-bill", "labels": [ "AI Kill Switch Bill" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "ai-labor-market-impact", "title": "AI Labor Market Impact", "domain": "economics", "domain_name": "Economics", "definition": "The structural changes in employment, workforce composition, and labor-capital dynamics resulting from the automation and augmentation of work by artificial intelligence.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-labor-market-impact", "labels": [ "AI Labor Market Impact" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-liability", "title": "AI Liability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Liability is the body of legal doctrine, statutory regime, regulatory practice and emerging case law that allocates civil responsibility for personal injury, property damage, economic loss, discrimination, privacy invasion and consequential harm caused by artificial intelligence systems\u2014encomp...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-liability", "labels": [ "AI Liability", "AI Liability Frameworks" ], "is_subclass_of": [ "AI Governance and Ethics", "Tort Law", "Product Liability", "Civil Liability", "AI Governance", "Algorithmic Accountability" ], "wikilinks": [ "Abnormally Dangerous Activity", "Abnormally Dangerous Activity Doctrine", "AI Insurance", "AI Liability Directive", "Air Canada v. Moffatt", "Algorithmic Auditing", "Andersen v. Stability AI", "Automated Vehicles Act 2024", "black box", "Breach of Duty", "Brown v. Kendall", "Caparo Industries plc v. Dickman", "Caparo Three-Stage Test", "Causal Linkage", "Causation", "Causation Doctrine", "Centre for Law & Future of Innovation", "Character.AI", "Civil Liability", "Civil Procedure Rules" ] }, { "id": "ai-licensing-regime", "title": "AI Licensing Regime", "domain": "governance", "domain_name": "Governance", "definition": "The set of legal, regulatory, and administrative frameworks that govern the distribution, access, and commercialization of artificial intelligence models and their outputs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-licensing-regime", "labels": [ "AI Licensing Regime" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-lifecycle-management", "title": "AI Lifecycle Management", "domain": "governance", "domain_name": "Governance", "definition": "AI lifecycle management is the governed oversight of an AI system across its whole existence \u2014 design, data acquisition, training, validation, deployment, operation, monitoring, retraining, and retirement \u2014 so that risk controls, documentation, and accountability travel with the system rather than stopping at release. It extends generic asset lifecycle management with AI-specific concerns: dataset provenance and drift, model versioning, performance and bias monitoring in production, incident response for model failures, and the stage-mapped risk activities that frameworks such as the NIST AI Risk Management Framework require.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ai-lifecycle-management", "labels": [ "AI Lifecycle Management" ], "is_subclass_of": [ "Lifecycle Management" ], "wikilinks": [ "Lifecycle Management", "AI Governance", "NIST AI Risk Management Framework", "MLOps" ] }, { "id": "ai-lifecycle", "title": "AI Lifecycle", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The series of distinct phases through which an artificial intelligence system progresses from initial conception to eventual decommissioning, encompassing planning, design, development, verification, deployment, operation, monitoring, maintenance, and retirement, with each phase involving specific activities, stakeholder roles, documentation requirements, and governance controls to ensure responsible and effective AI system management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ai-lifecycle", "labels": [ "AI Lifecycle", "AI Development Lifecycle" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [ "FDA Software as Medical Device", "ISO/IEC 23053", "ISO/IEC 42001:2023", "ISO/IEC 5338:2023", "NIST AI Risk Management Framework", "Stakeholder engagement throughout", "AI Agent System", "AI Deployment", "AI Development", "AI Governance", "AI Impact Assessment", "AI Monitoring", "Human Oversight", "MetaverseDomain", "Model Performance", "Risk Management", "Update Cycle" ] }, { "id": "ai-literacy-training-for-designers", "title": "AI Literacy Training for Designers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Training for Design Practitioners covers structured programmes and self-directed learning pathways that equip designers\u2014particularly in landscape, product, and spatial disciplines\u2014with practical AI literacy, covering generative image tools, RAG-based knowledge management, client communication automation, and AI-assisted rendering workflows for immediate professional application.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-literacy-training-for-designers", "labels": [ "AI Literacy Training for Designers", "Training for Design Practitioners" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "ChatGPT", "Consumer Tools for SMEs", "Gemini", "Image Generation", "Knowledge Graphing", "Retrieval Augmented Generation - RAG", "social media" ] }, { "id": "ai-mandate", "title": "AI Mandate", "domain": "economics", "domain_name": "Economics", "definition": "Corporate policies that tie AI tool adoption and proficiency to employee career progression, promotions, or performance reviews.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-mandate", "labels": [ "AI Mandate" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-market-dynamics", "title": "AI Market Dynamics", "domain": "economics", "domain_name": "Economics", "definition": "The competitive, pricing, and strategic forces shaping the AI industry, including model commoditization and provider optionality.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-market-dynamics", "labels": [ "AI Market Dynamics" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-market-growth", "title": "AI Market Growth", "domain": "economics", "domain_name": "Economics", "definition": "The expansion of the economic value and scale of the artificial intelligence industry, driven by increased adoption, new applications, and technological advancements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-market-growth", "labels": [ "AI Market Growth" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "ai-market-impact", "title": "AI Market Impact", "domain": "economics", "domain_name": "Economics", "definition": "The measurable influence of artificial intelligence adoption and technological shifts on financial markets, sector valuations, and investor sentiment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-market-impact", "labels": [ "AI Market Impact" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-maturity-assessment", "title": "AI Maturity Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A structured framework for evaluating an organization's readiness and capability in adopting and integrating AI technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-maturity-assessment", "labels": [ "AI Maturity Assessment" ], "is_subclass_of": [ "Enterprise Ai" ], "wikilinks": [] }, { "id": "ai-metadata", "title": "AI Metadata", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Descriptive information, provenance data, and contextual attributes associated with AI models, datasets, and computational artefacts. Includes model cards documenting architecture, training procedures, performance metrics, intended use, and known limitations; dataset metadata capturing collection methods, labelling procedures, and potential biases; and lineage records enabling reproducibility, regulatory compliance, and responsible deployment throughout the machine learning lifecycle.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-metadata", "labels": [ "AI Metadata" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "ML Reproducibility", "Model Card", "AI Governance", "Blockchain", "Data Provenance" ] }, { "id": "ai-model-auditing", "title": "AI Model Auditing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic process of evaluating, testing, and monitoring AI models to identify behavioral anomalies, biases, or unintended artifacts resulting from training methodologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-auditing", "labels": [ "AI Model Auditing" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "ai-model-benchmarking", "title": "AI Model Benchmarking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The standardized evaluation of artificial intelligence models using curated datasets and metrics to measure performance, capability, and safety, serving as a primary mechanism for comparing model generations and tracking progress.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-benchmarking", "labels": [ "AI Model Benchmarking" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-model-card", "title": "AI Model Card", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured documentation format that describes an AI Model's purpose, Performance Metrics, limitations, ical Considerations, and appropriate Use Case|use cases to promote Transparency and Responsible AI Deployment|responsible deployment.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:ai-model-card", "labels": [ "AI Model Card", "Model Card" ], "is_subclass_of": [ "AI Governance and Ethics", "Documentation Standards", "AI Documentation Framework", "Compliance Documentation", "AI Risk Management" ], "wikilinks": [ "AI Accountability", "AI Documentation Framework", "AI Ethics Guidelines", "AI Impact Assessment", "AI Model", "AI Safety Evaluation", "Algorithmic Auditing", "Bias Analysis", "Bitcoin Trading Bot", "Blockchain Analytics", "Compliance Documentation", "Data Provenance", "Demographic Performance Analysis", "Ethical Considerations", "Explainability", "Foundation Model Transparency Index", "Generative AI", "Google Model Cards for Model Reporting", "Hugging Face Model Hub", "Industry Best Practices" ] }, { "id": "ai-model-competition", "title": "AI Model Competition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The ongoing rivalry among technology companies to release superior large language models and AI systems, characterized by rapid iteration and performance benchmarking.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-competition", "labels": [ "AI Model Competition" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-model-development", "title": "AI Model Development", "domain": "ai", "domain_name": "Ai", "definition": "The end-to-end engineering discipline of designing, training, evaluating, and deploying machine learning models, encompassing dataset curation, architecture selection, optimisation, and lifecycle management. It integrates software engineering, statistical modelling, and domain expertise to produce AI systems capable of performing specified tasks at scale. The discipline spans research prototyping through production deployment and ongoing maintenance.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:ai-model-development", "labels": [ "AI Model Development" ], "is_subclass_of": [ "AI System", "Machine Learning" ], "wikilinks": [ "AI Model", "Data Pipeline", "Compute Resources", "Model Evaluation", "AI Inference", "Fine Tuning", "AI Risk Assessment", "MLOps", "Deep Learning", "Neural Network", "Transformer Architecture", "Large Language Models", "Foundation Model", "Training Data", "Gradient Descent", "Hyperparameter Optimisation", "AI Model Architecture", "AI Model Card", "AI Governance", "Transfer Learning" ] }, { "id": "ai-model-inference-engine", "title": "AI Model Inference Engine", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An AI model inference engine is the software runtime that executes a trained model to produce predictions from new inputs. It manages computation graph execution, hardware acceleration and memory to run models efficiently.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:ai-model-inference-engine", "labels": [ "AI Model Inference Engine" ], "is_subclass_of": [ "Inference Serving", "AI Infrastructure (Artificial Intelligence)", "AI Inference", "Deep Learning Framework" ], "wikilinks": [ "GPU", "Model Optimization", "Inference Serving", "Model Serving", "Latency", "Model Deployment", "AI Inference", "Quantisation", "KV Cache", "Speculative Decoding", "Continuous Batching", "Hardware Accelerator", "Tensor Parallelism", "Transformer Architecture", "Large Language Models", "Model Weights", "Operator Fusion", "Flash Attention", "On-Device AI", "Edge Computing" ] }, { "id": "ai-model-portfolio", "title": "AI Model Portfolio", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A strategy where users or organizations deploy multiple distinct AI models simultaneously to leverage their specific strengths, rather than relying on a single general-purpose model.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-portfolio", "labels": [ "AI Model Portfolio" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-model-race", "title": "AI Model Race", "domain": "economics", "domain_name": "Economics", "definition": "The competitive dynamic among AI developers and providers, characterized by multiple parallel races for superiority in capability, speed, cost, distribution, and other strategic dimensions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-race", "labels": [ "AI Model Race" ], "is_subclass_of": [ "AI Strategy" ], "wikilinks": [] }, { "id": "ai-model-release-cadence", "title": "AI Model Release Cadence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The frequency and pattern of AI model releases, influenced by safety reviews, competitive dynamics, and governance frameworks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-model-release-cadence", "labels": [ "AI Model Release Cadence" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-model", "title": "AI Model", "domain": "ai", "domain_name": "Ai", "definition": "An AI Model is a computational artefact \u2014 comprising a parameterised mathematical function, its learned weights, and associated configuration \u2014 that encodes patterns extracted from training data and can be applied to new inputs to generate predictions, classifications, embeddings, or generative outputs. AI models range from simple linear regressors to billion-parameter deep neural networks and constitute the core intellectual and commercial asset of modern AI systems.", "entityType": "Class", "qualityScore": 0.93, "maturity": "mature", "iri": "urn:ngm:class:ai-model", "labels": [ "AI Model", "AI Model Release" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture", "Machine Learning" ], "wikilinks": [ "AI Model Architecture", "Deep Learning", "Training Data", "Gradient Descent", "Neural Network", "AI Inference", "Generative AI", "Natural Language Processing", "Computer Vision", "AI Agent", "Compute Infrastructure", "Model Training", "Model Weights", "Model Checkpoint", "Learning Algorithm", "Fine Tuning", "Model Quantization", "Rule-Based System", "Expert System", "AI Governance" ] }, { "id": "ai-monitoring", "title": "AI Monitoring", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic and ongoing observation, measurement, and analysis of an artificial intelligence system's behaviour, performance, inputs, outputs, and impacts during operational use, employing automated tools and human oversight to detect degradation, anomalies, bias, safety issues, or unintended consequences, enabling timely intervention, maintenance, and continuous improvement whilst ensuring accountability and compliance with governance requirements.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:ai-monitoring", "labels": [ "AI Monitoring", "AI Model Monitoring", "AI System Monitoring" ], "is_subclass_of": [ "AI Lifecycle" ], "wikilinks": [ "AI Maintenance", "alert thresholds", "baseline metrics", "continuous improvement", "Data Drift", "FDA Post-Market Surveillance", "GDPR", "ISO/IEC 23894:2023", "ISO/IEC 25024", "ISO/IEC 25059", "ISO/IEC 42001:2023", "model updating", "Monitoring infrastructure", "NIST AI Risk Management Framework", "AI Agent System", "AI Audit", "AI Deployment", "AI Lifecycle", "AI Operator", "Bias" ] }, { "id": "ai-office", "title": "AI Office", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "European Commission body established under AI Act Articles 64\u201368 as the primary centre of AI expertise for a unified European AI governance system, holding executive authority for supervising general-purpose AI model providers, coordinating cross-border market surveillance, imposing corrective measures and fines, facilitating regulatory sandboxes, and representing the EU in international AI governance forums.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-office", "labels": [ "AI Office", "EU AI Office" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "IEEE P7009", "NIST AI RMF", "AI Agent System", "EU AI Act", "MetaverseDomain" ] }, { "id": "ai-operator", "title": "AI Operator", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An individual or organisational entity responsible for the operational management, monitoring, and control of an artificial intelligence system during its deployment and use, including activities such as configuring system parameters, overseeing system performance, responding to incidents, coordi...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:ai-operator", "labels": [ "AI Operator" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Competence", "competence development", "NIST AI Risk Management Framework", "ongoing support", "Accountability", "AI Agent System", "AI Deployment", "AI Governance", "AI Monitoring", "AI Provider", "AI User", "EU AI Act", "Human in the Loop", "Human Oversight", "MetaverseDomain", "Training" ] }, { "id": "ai-organizational-structure", "title": "AI Organizational Structure", "domain": "governance", "domain_name": "Governance", "definition": "The internal hierarchy, reporting lines, and departmental arrangements within an organization that govern the development, deployment, and oversight of artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-organizational-structure", "labels": [ "AI Organizational Structure" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "ai-policy-discourse", "title": "AI Policy Discourse", "domain": "governance", "domain_name": "Governance", "definition": "The evolving political and public debate around AI regulation, ownership, and distribution, characterized by cross-ideological alignments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-policy-discourse", "labels": [ "AI Policy Discourse" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-policy", "title": "ai policy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Policy is the ensemble of government strategies, regulatory instruments, voluntary guidelines, and international agreements through which states and intergovernmental bodies shape the development, deployment, and societal impact of artificial intelligence systems. Policy instruments span public research funding, national AI strategies, procurement standards for government AI use, export controls on frontier hardware and models, and binding requirements for transparency, accountability, and fundamental-rights compliance in high-stakes applications. Effective AI policy must balance national competitiveness, safety assurance, equitable access, and interoperability with allied jurisdictions, making it an inherently multi-stakeholder and multi-domain governance challenge.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:ai-policy", "labels": [ "AI Policy", "AI Regulatory Policy", "AI Safety Policy", "Government AI Policy" ], "is_subclass_of": [ "AI Governance and Ethics", "Technology Policy" ], "wikilinks": [ "AI Governance", "Responsible AI", "AI Safety", "Digital Sovereignty", "AI Governance Framework", "Algorithmic Accountability", "AI Audit", "AI Regulation", "AI Ethics", "Data Governance", "Intellectual Property", "Risk Assessment", "Impact Assessment", "Stakeholder Consultation", "OECD AI Principles", "ISO IEC 42001", "EU AI Act", "Self-Regulation", "Cybersecurity Policy", "Semiconductor Export Controls" ] }, { "id": "ai-portfolio-agent", "title": "AI Portfolio Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An interactive AI agent that represents a professional's skills and history, enabling dynamic querying of their capabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-portfolio-agent", "labels": [ "AI Portfolio Agent" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "ai-pricing-models", "title": "AI Pricing Models", "domain": "economics", "domain_name": "Economics", "definition": "The various strategies used by AI providers to charge for their services, including seat-based, usage-based, and token-based pricing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-pricing-models", "labels": [ "AI Pricing Models" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-privacy", "title": "AI Privacy", "domain": "governance", "domain_name": "Governance", "definition": "The set of principles, practices, and user attitudes regarding the collection, use, and protection of personal data in AI systems, particularly concerning continuous monitoring and ambient data capture.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-privacy", "labels": [ "AI Privacy" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "ai-product-strategy", "title": "AI Product Strategy", "domain": "economics", "domain_name": "Economics", "definition": "The strategic planning and positioning of AI-powered products and services, including interface design, target user segmentation, and competitive differentiation in the market.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-product-strategy", "labels": [ "AI Product Strategy" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-product-and-risk-framework", "title": "AI Product and Risk Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A framework mapping the scientific method onto lean product development, guiding AI product decisions through vision setting, market analysis, and strategic planning. It addresses SWOT evaluation, moat assessment, and risk mitigation for AI products\u2014particularly the legal, privacy, and safety risks introduced by generative AI and the EU regulatory landscape.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-product-and-risk-framework", "labels": [ "AI Product and Risk Framework", "EU Product Safety Framework", "Product and Risk Management" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Politics, Law, Privacy", "Product Design", "Safety and alignment" ] }, { "id": "ai-productivity-debate", "title": "AI Productivity Debate", "domain": "economics", "domain_name": "Economics", "definition": "The ongoing economic and academic discussion regarding the magnitude, timing, and causal mechanisms of productivity gains attributed to artificial intelligence adoption.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-productivity-debate", "labels": [ "AI Productivity Debate" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [] }, { "id": "ai-productivity-gains", "title": "AI Productivity Gains", "domain": "economics", "domain_name": "Economics", "definition": "The measurable increase in output efficiency, task completion speed, or quality of work resulting from the integration of artificial intelligence tools into human workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-productivity-gains", "labels": [ "AI Productivity Gains" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "ai-productivity-impact", "title": "AI Productivity Impact", "domain": "economics", "domain_name": "Economics", "definition": "The measurable effects of AI adoption on organizational output, work intensity, and economic growth.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-productivity-impact", "labels": [ "AI Productivity Impact" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-provider", "title": "AI Provider", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An individual, organisation, or legal entity that develops, produces, or supplies an AI system, assuming primary accountability for design decisions, training processes, capabilities, documentation, and regulatory compliance. Under the EU AI Act a provider places a system on the market under their own name, substantially modifies an existing system, or makes it available for use, thereby triggering obligations including conformity assessment, technical documentation, quality management, human oversight design, and post-market monitoring.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-provider", "labels": [ "AI Provider" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Conformity Assessment", "GDPR", "NIST AI Risk Management Framework", "Accountability", "AI Agent System", "AI Audit", "AI Development", "AI Governance", "AI Impact Assessment", "AI Lifecycle", "AI Operator", "AI User", "EU AI Act", "MetaverseDomain", "Risk Management" ] }, { "id": "ai-public-opinion", "title": "AI Public Opinion", "domain": "governance", "domain_name": "Governance", "definition": "The aggregate attitudes, perceptions, and sentiment of the general public regarding the development, deployment, and societal impact of artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-public-opinion", "labels": [ "AI Public Opinion" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-roi-benchmarking", "title": "AI ROI Benchmarking", "domain": "economics", "domain_name": "Economics", "definition": "The systematic evaluation and quantification of the financial return on investment generated by enterprise artificial intelligence deployments across various use cases.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-roi-benchmarking", "labels": [ "AI ROI Benchmarking" ], "is_subclass_of": [ "Enterprise AI" ], "wikilinks": [] }, { "id": "ai-roi-measurement", "title": "AI ROI Measurement", "domain": "economics", "domain_name": "Economics", "definition": "The framework and methodology used by organizations to quantify the financial and operational returns generated by artificial intelligence investments relative to their costs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-roi-measurement", "labels": [ "AI ROI Measurement" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-roi", "title": "AI ROI", "domain": "economics", "domain_name": "Economics", "definition": "The quantifiable financial return and value generated by the deployment of artificial intelligence systems relative to the costs of implementation and maintenance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-roi", "labels": [ "AI ROI" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-regulation", "title": "ai regulation", "domain": "governance", "domain_name": "Governance", "definition": "AI regulation encompasses binding laws, technical standards, and administrative rules imposed by national or supranational bodies to govern the entire lifecycle of artificial intelligence systems \u2014 from design and training through deployment and monitoring. Regulatory instruments range from risk-based frameworks such as the EU AI Act, which imposes tiered obligations proportional to application risk, to sector-specific rules in finance, healthcare, and critical infrastructure. Regulation aims to protect fundamental rights, ensure accountability, and promote systemic safety while preserving innovation capacity. Implementation translates abstract policy obligations into concrete engineering requirements including conformity assessments, audit trails, data governance documentation, and post-market surveillance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:ai-regulation", "labels": [ "AI Regulation", "General-Purpose AI Regulation", "Mandatory AI Regulation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-regulatory-flexibility", "title": "AI Regulatory Flexibility", "domain": "governance", "domain_name": "Governance", "definition": "The degree of discretion granted to regulators in enforcing AI safety and compliance standards, balancing innovation with risk mitigation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-regulatory-flexibility", "labels": [ "AI Regulatory Flexibility" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-regulatory-framework", "title": "AI Regulatory Framework", "domain": "governance", "domain_name": "Governance", "definition": "The set of laws, policies, and governmental guidelines designed to govern the development, deployment, and impact of artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-regulatory-framework", "labels": [ "AI Regulatory Framework" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "ai-regulatory-preemption", "title": "AI Regulatory Preemption", "domain": "governance", "domain_name": "Governance", "definition": "The legal and political strategy of establishing a unified federal regulatory framework for artificial intelligence to supersede or limit the authority of individual states to enact their own AI laws.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-regulatory-preemption", "labels": [ "AI Regulatory Preemption" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "ai-release-cadence", "title": "AI Release Cadence", "domain": "governance", "domain_name": "Governance", "definition": "The frequency and timing at which artificial intelligence models and capabilities are made available to the public or specific user groups.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-release-cadence", "labels": [ "AI Release Cadence" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-research-talent", "title": "AI Research Talent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI research talent refers to the pool of specialised researchers and engineers capable of advancing the state of the art in artificial intelligence, spanning fundamental research, model architecture design and applied systems engineering. Its scarcity is a key constraint on organisational AI capability, shaping investment decisions and competitive dynamics between labs and nations. Demand for this talent has driven sharp compensation growth and intense competition across the industry.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:ai-research-talent", "labels": [ "AI Research Talent" ], "is_subclass_of": [ "AI Talent" ], "wikilinks": [] }, { "id": "ai-research", "title": "AI Research", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI research is the systematic scientific investigation of methods for creating and understanding machine intelligence, spanning theoretical foundations, algorithm and architecture development, empirical benchmarking, and the study of capabilities, limitations, safety, and societal impact. It is conducted across universities, corporate laboratories, and independent institutes, and disseminated through peer-reviewed venues such as NeurIPS, ICML, and AAAI, as well as preprint servers and open-source releases.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:ai-research", "labels": [ "AI Research" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "AI Ecosystem", "Machine Learning", "NeurIPS" ] }, { "id": "ai-return-on-investment", "title": "AI Return on Investment", "domain": "economics", "domain_name": "Economics", "definition": "The financial and operational benefits realized from AI investments, often measured against costs and compared to other capital expenditures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-return-on-investment", "labels": [ "AI Return on Investment" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-risk-assessment", "title": "AI Risk Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A structured methodology for identifying, analysing, and evaluating the potential harms, failure modes, and adverse societal impacts arising from the development and deployment of artificial intelligence systems. It applies established risk management frameworks to the distinctive properties of AI\u2014opacity, emergent behaviour, data dependency, and scalability\u2014to produce actionable risk registers and mitigation plans. The process informs governance decisions and regulatory compliance across the full AI system lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ai-risk-assessment", "labels": [ "AI Risk Assessment" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "ai-risk-management", "title": "AI Risk Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Risk Management is the systematic process of identifying, assessing, mitigating, and monitoring risks associated with artificial intelligence systems throughout their lifecycle, integrating AI-specific considerations into broader enterprise risk management frameworks. It encompasses governance structures, assessment methodologies, control mechanisms, and continuous oversight to ensure AI systems operate safely, ethically, and in compliance with applicable regulations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-risk-management", "labels": [ "AI Risk Management" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [ "NIST AI Risk Management Framework", "Trustworthy AI", "Artificial Intelligence", "EU AI Act", "Risk Management", "Smart Contract" ] }, { "id": "ai-risk-register", "title": "AI Risk Register", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An AI Risk Register is a structured artefact that systematically documents, tracks, and manages identified risks associated with AI systems throughout their lifecycle. Each entry records a risk identifier, description, affected systems and stakeholders, likelihood and consequence ratings, overall risk level, assigned owner, current mitigation controls, residual risk, and review history. The register supports risk-based governance by enabling prioritisation of mitigation efforts, regulatory compliance demonstration, and continuous monitoring across technical, ethical, legal, operational, security, and societal risk categories.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-risk-register", "labels": [ "AI Risk Register" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Risk Management" ], "wikilinks": [ "ISO 31000", "ISO/IEC 23894:2023", "AIEthicsDomain", "ConceptualLayer", "EU AI Act", "Smart Contract", "AI Risk Management", "Audit Trail", "Risk Assessment", "Compliance Monitoring", "AI Safety", "Transparency", "AI Governance Framework", "Regulatory Compliance", "Fairness", "Bias", "Explainability", "Risk Management", "Accountability", "AI Governance" ] }, { "id": "ai-risk", "title": "AI Risk", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The potential for AI systems to cause adverse effects on individuals, groups, organizations, communities, or society, arising from technical failures, security vulnerabilities, biased outcomes, privacy violations, or unintended consequences of system design, deployment, or operation. AI risks span from immediate operational failures to long-term catastrophic scenarios requiring proactive identification, assessment, and mitigation strategies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-risk", "labels": [ "AI Risk" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [ "AI Risk Assessment", "AIGovernance", "GDPR", "ISO/IEC 23894", "NIST AI RMF", "AI Ethics", "AI Incident", "AI Risk Management", "AI Safety", "AI Security", "AI Trustworthiness", "Algorithmic Bias", "ArtificialIntelligenceDomain", "EU AI Act", "High Risk AI System", "Large Language Models", "Risk Management", "Smart Contract" ] }, { "id": "ai-risks", "title": "AI Risks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Risks is the umbrella ontological category denoting the full taxonomy of potential and realised adverse impacts arising from the design, development, deployment, integration, or societal embedding of artificial intelligence systems, spanning four canonical risk-source domains\u2014misuse (intention...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-risks", "labels": [ "AI Risks" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Safety", "Harm Taxonomy", "Sociotechnical Risk", "Technology Risk", "Systemic Risk", "Governance Object" ], "wikilinks": [ "Accident Risks", "Adversarial Robustness Testing", "AI Benefits", "AI Capabilities", "AI for Good", "AI Liability Frameworks", "AI Policy", "AI Safety Research", "AISafetyDomain", "Algorithmic Auditing", "Algorithmic Impact Assessments", "Angwin et al ProPublica COMPAS 2016", "Anthropic Responsible Scaling Policy v2.1", "Autonomous Weapons", "Benchmarks", "Bender et al Stochastic Parrots 2021", "Beneficial AI", "Bletchley Declaration", "Bletchley Declaration 2023", "Bostrom Superintelligence 2014" ] }, { "id": "ai-safety-assessment", "title": "AI Safety Assessment", "domain": "security", "domain_name": "Security", "definition": "The systematic process of evaluating the potential risks, failure modes, and unintended consequences of artificial intelligence systems before and during deployment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-safety-assessment", "labels": [ "AI Safety Assessment" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-safety-evaluation", "title": "AI Safety Evaluation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic process of assessing the risks, failure modes, and potential for misalignment or cheating in AI systems prior to and during deployment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-safety-evaluation", "labels": [ "AI Safety Evaluation" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-safety-institute", "title": "AI Safety Institute", "domain": "governance", "domain_name": "Governance", "definition": "An AI Safety Institute (AISI) is a state-backed organisation established to evaluate, test, and research the capabilities and safety risks of advanced AI systems, particularly frontier models. Such bodies conduct pre-deployment and post-deployment evaluations, develop standardised benchmarks for dangerous capability elicitation, and advise governments on AI risk policy. The UK established the world's first AISI in November 2023 at the Bletchley Park AI Safety Summit; the US, Japan, Canada, and other nations subsequently established counterpart bodies. The UK later renamed its institute the AI Security Institute (DSIT) to reflect an expanded emphasis on national-security-relevant AI risks.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:ai-safety-institute", "labels": [ "AI Safety Institute", "AI Safety Institute Network" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "AI Alignment", "Frontier AI", "Existential AI Risk", "AI Safety" ] }, { "id": "ai-safety-preparedness-framework", "title": "AI Safety Preparedness Framework", "domain": "governance", "domain_name": "Governance", "definition": "Internal and external guidelines used by AI labs to assess and mitigate the risks of new model capabilities before release.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-safety-preparedness-framework", "labels": [ "AI Safety Preparedness Framework" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "ai-safety-red-lines", "title": "AI Safety Red Lines", "domain": "governance", "domain_name": "Governance", "definition": "Specific ethical boundaries set by AI developers that define prohibited uses of their models, such as autonomous weapons or domestic surveillance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-safety-red-lines", "labels": [ "AI Safety Red Lines" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "ai-safety-research", "title": "ai safety research", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Safety Research is an interdisciplinary field that develops theoretical frameworks, empirical methods, and engineering techniques to ensure advanced AI systems behave in ways that are safe, reliable, and aligned with human values across a range of capability levels. The field spans near-term concerns \u2014 such as robustness, fairness, and adversarial resistance \u2014 and longer-term challenges including scalable oversight, corrigibility, and the avoidance of catastrophic risks from highly capable systems. It draws on machine learning, decision theory, formal verification, and cognitive science to produce both immediately deployable safety interventions and foundational understanding of how intelligent systems can be made reliably beneficial.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:ai-safety-research", "labels": [ "AI Safety Research", "Safety Research Programmes" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AI Systems", "Robustness", "Scalable Oversight", "Corrigibility", "Machine Learning", "Decision Theory", "Formal Verification", "Trustworthy AI", "Deep Learning", "Large Language Models", "AI Regulation", "Reinforcement Learning from Human Feedback", "Mechanistic Interpretability", "Red Teaming", "Interpretability", "Uncertainty Quantification", "Responsible AI", "AI Alignment", "AI Risk", "Existential Risk" ] }, { "id": "ai-safety-summit", "title": "AI Safety Summit", "domain": "governance", "domain_name": "Governance", "definition": "An intergovernmental summit series on frontier AI risk, inaugurated at Bletchley Park in November 2023, that convenes governments, leading AI companies, and researchers to agree shared assessments of advanced-model risks and coordination mechanisms for testing and governance. The series produced the Bletchley Declaration, catalysed the creation of national AI safety institutes, and continued through the Seoul summit (2024), the Paris AI Action Summit (2025), and the India AI Impact Summit in New Delhi (February 2026), with Switzerland due to host the next summit in Geneva in 2027.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ai-safety-summit", "labels": [ "AI Safety Summit" ], "is_subclass_of": [ "International AI Cooperation" ], "wikilinks": [ "International AI Cooperation", "Bletchley Declaration", "AI Safety Institute", "Frontier AI" ] }, { "id": "ai-safety", "title": "AI Safety", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Safety is the interdisciplinary field of research and engineering practice dedicated to ensuring that artificial intelligence systems behave reliably, predictably, and in accordance with human values and intentions across their full operational lifecycle. It addresses near-term concerns such as robustness, distributional shift, and adversarial vulnerability, as well as longer-horizon concerns about advanced systems whose objectives may diverge from human welfare. Core techniques include formal verification of safety properties, interpretability methods that expose model internals, corrigibility mechanisms that preserve human oversight, and red-teaming to surface failure modes before deployment. AI Safety is increasingly embedded in regulatory frameworks and standard-setting processes governing high-risk AI applications.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:ai-safety", "labels": [ "AI Safety" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Smart Contract", "AI Alignment", "Interpretability", "Formal Verification", "AI Governance", "Robustness", "Adversarial Machine Learning", "Reinforcement Learning from Human Feedback", "Red Teaming", "Corrigibility", "Explainable AI", "AI Ethics", "Risk Management", "Responsible AI", "Trustworthy AI", "Human-AI Collaboration", "AI Capabilities Research", "Existential Risk" ] }, { "id": "ai-sales-quotas", "title": "AI Sales Quotas", "domain": "economics", "domain_name": "Economics", "definition": "Internal revenue targets and performance benchmarks set by technology companies for the adoption and monetization of artificial intelligence products and services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-sales-quotas", "labels": [ "AI Sales Quotas" ], "is_subclass_of": [ "Microsoft" ], "wikilinks": [] }, { "id": "ai-scrapers", "title": "AI Scrapers", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Scrapers are automated software agents (web crawlers, retrieval bots, and content-fetching pipelines) operated by foundation-model laboratories, AI search vendors, retrieval-augmented-generation services, dataset aggregators, and synthetic-media producers for the purpose of harvesting publicly...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-scrapers", "labels": [ "AI Scrapers" ], "is_subclass_of": [ "AI Agent System", "Web Crawler", "Automated Agent", "Data Acquisition Pipeline", "Regulated Data Processor", "Internet Bot" ], "wikilinks": [ "Academic Crawler", "Automated Agent", "Bloom Filter", "Bot Management", "CDN Layer", "Classic Search Crawler", "Common Crawl", "ComplianceLayer", "Content Extraction Pipeline", "Content Licensing", "Content Type Negotiation", "Copyright Law", "CopyrightLawDomain", "Crawl Frontier", "DataAcquisitionLayer", "Data Acquisition Pipeline", "DataGovernanceDomain", "De-Duplication Hashing", "Deduplication Filter", "DNS" ] }, { "id": "ai-search", "title": "AI Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Search (also termed generative search, answer engine, conversational search, or retrieval-augmented search) is the class of web-scale information-access systems that pair a large language model (LLM) with a live retrieval layer over web indexes, document stores, or proprietary corpora to produ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-search", "labels": [ "AI Search", "AI-Assisted Search" ], "is_subclass_of": [ "AI Application", "Natural Language Processing", "Retrieval-Augmented Generation", "Information Retrieval", "Question Answering System", "Conversational AI", "Search Engine" ], "wikilinks": [ "Academic Research", "Agentic Browser", "Agentic Browsing", "Answer Synthesiser", "Approximate Nearest Neighbour Search", "BM25", "Boolean Search", "C2PA Content Credentials", "Caching", "Chain-of-Thought Reasoning", "Citation-Grounded Answers", "Citation Grounding Layer", "Classical Information Retrieval", "Common Crawl", "Content Licensing Agreement", "Conversational AI without Retrieval", "Conversational Web Access", "Cosine Similarity", "Cross-Encoder Reranking", "Customer Support" ] }, { "id": "ai-security-breach", "title": "AI Security Breach", "domain": "security", "domain_name": "Security", "definition": "Unauthorized access to, or compromise of, AI model weights, training data, or inference environments, leading to potential data leakage or model theft.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-security-breach", "labels": [ "AI Security Breach" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "ai-security", "title": "AI Security", "domain": "security", "domain_name": "Security", "definition": "AI Security is the field of protecting artificial intelligence systems and their components from security threats and vulnerabilities, including adversarial attacks, data poisoning, model theft, and unauthorized access to ensure systems perform as intended. It encompasses defending AI models, algorithms, training data, and infrastructure from manipulation, misuse, and exploitation throughout the AI lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ai-security", "labels": [ "AI Security" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "NIST AI Risk Management Framework", "Secure AI Deployment", "Artificial Intelligence", "Blockchain", "Cybersecurity" ] }, { "id": "ai-self-regulation", "title": "AI Self-Regulation", "domain": "governance", "domain_name": "Governance", "definition": "AI self-regulation is the governance of artificial intelligence through voluntary commitments, codes of conduct, and internal policies adopted by AI developers and industry bodies rather than imposed by statute. It includes responsible scaling policies, voluntary safety commitments, model release and evaluation norms, and industry consortia that agree shared practices. As a domain-specific specialisation of general industry self-regulation, it is defined by contrast with binding AI regulation and with state safety bodies such as national AI Safety Institutes: it is faster and more technically informed but lacks enforcement and can drift toward reputational cover unless paired with external accountability. It is frequently framed as a bridge or interim measure ahead of formal regulation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ai-self-regulation", "labels": [ "AI Self-Regulation" ], "is_subclass_of": [ "Self-Regulation" ], "wikilinks": [ "Self-Regulation", "AI Regulation", "Responsible Scaling Policy" ] }, { "id": "ai-shopping-assistant", "title": "AI Shopping Assistant", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Conversational AI interfaces integrated into e-commerce platforms that assist consumers in product discovery, comparison, and purchase decisions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-shopping-assistant", "labels": [ "AI Shopping Assistant" ], "is_subclass_of": [ "AI Agents" ], "wikilinks": [] }, { "id": "ai-sovereign-wealth-fund", "title": "AI Sovereign Wealth Fund", "domain": "governance", "domain_name": "Governance", "definition": "A proposed government mechanism to hold equity stakes in major AI companies to distribute the financial benefits of AI development to the public.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-sovereign-wealth-fund", "labels": [ "AI Sovereign Wealth Fund" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-sovereignty", "title": "AI Sovereignty", "domain": "governance", "domain_name": "Governance", "definition": "AI Sovereignty is the capacity of a nation or bloc to develop, control, and govern artificial-intelligence systems and their underlying compute, data, and models without dependence on foreign providers. It motivates domestic investment in chips, data centres, foundation models, and regulation to retain strategic autonomy. The concept frames AI capability as critical national infrastructure subject to security, economic, and geopolitical considerations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ai-sovereignty", "labels": [ "AI Sovereignty" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "ai-speciation", "title": "AI Speciation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The predicted evolutionary trend in AI where a single general-purpose intelligence diverges into multiple specialized, smaller models optimized for distinct tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-speciation", "labels": [ "AI Speciation" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-stock-market", "title": "AI Stock Market", "domain": "economics", "domain_name": "Economics", "definition": "The segment of the financial market comprising equities and investments specifically tied to the development, infrastructure, and commercialization of artificial intelligence technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-stock-market", "labels": [ "AI Stock Market" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-supply-chain-risk", "title": "AI Supply Chain Risk", "domain": "governance", "domain_name": "Governance", "definition": "The assessment of vulnerabilities and dependencies within the ecosystem of AI model development, deployment, and maintenance, including hardware, data, and regulatory factors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-supply-chain-risk", "labels": [ "AI Supply Chain Risk" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ai-supply-chain-shortage", "title": "AI Supply Chain Shortage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A structural deficit in the availability of critical hardware components, such as high-bandwidth memory and advanced semiconductors, required to build and scale artificial intelligence infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-supply-chain-shortage", "labels": [ "AI Supply Chain Shortage" ], "is_subclass_of": [ "Open Source" ], "wikilinks": [] }, { "id": "ai-surveillance-deployment-case-study", "title": "AI Surveillance Deployment Case Study", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A case study examining the deployment of AI-based surveillance systems at the Paris 2024 Olympic Games, involving real-time video analytics, geolocation tracking, behavioural pattern analysis, and predictive event detection contracted to vendors including Viet, Orange Business, Chaps Vision, and Windex. The deployment required French legislative modifications expanding surveillance powers and raised significant GDPR compliance questions regarding continuous monitoring, data retention, algorithmic bias, and the post-event normalisation of pervasive AI surveillance infrastructure.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-surveillance-deployment-case-study", "labels": [ "AI Surveillance Deployment Case Study", "AI privacy at the 2024 Olympics" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "GDPR (General Data Protection Regulation)", "Cryptography" ] }, { "id": "ai-surveillance", "title": "AI Surveillance", "domain": "governance", "domain_name": "Governance", "definition": "AI surveillance is the application of machine learning and computer vision to monitoring at scale: automated facial recognition in public space, gait and behaviour analysis, licence-plate reading, social-media and communications analytics, and predictive scoring of individuals or crowds. It differs from conventional surveillance in that inference, not observation, does the work \u2014 systems classify, track, and flag people continuously without human attention per subject, raising distinct governance questions about bias, proportionality, chilling effects, and the legal basis for algorithmic monitoring.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ai-surveillance", "labels": [ "AI Surveillance" ], "is_subclass_of": [ "Surveillance" ], "wikilinks": [ "Surveillance", "Facial Recognition", "Digital Society Surveillance", "Algorithmic Bias" ] }, { "id": "ai-system-eu-definition", "title": "AI System (EU Definition)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Under EU AI Act Article 3(1), a machine-based system designed to operate with varying levels of autonomy, capable of adapting after deployment, and generating outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. This definition establishes the regulatory scope of the AI Act and aligns with the 2024 OECD AI Principles, distinguishing AI systems from traditional software by their inferential and adaptive capabilities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-system-eu-definition", "labels": [ "AI System (EU Definition)" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "NIST AI Risk Management Framework", "AI Agent System", "EU AI Act", "MetaverseDomain" ] }, { "id": "ai-system-component", "title": "AI System Component", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Modular functional units constituting a complete artificial intelligence system, including data ingestion pipelines, feature engineering modules, model training infrastructure, inference engines, monitoring dashboards, and user interfaces. Modern AI systems adopt microservices architectures enabling independent scaling, version control, and A/B testing per component, with observability integrated at each layer for production reliability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ai-system-component", "labels": [ "AI System Component" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "MLOps", "Model Serving", "Blockchain", "Data Pipeline", "Microservices Architecture" ] }, { "id": "ai-system", "title": "AI System", "domain": "ai", "domain_name": "Ai", "definition": "An integrated assembly of software, hardware, data, and processes that employs machine learning or related techniques to perceive inputs, infer patterns or decisions, and produce outputs that affect its environment or users. AI systems range from narrow task-specific classifiers to general-purpose language models and autonomous agents, and they are characterised by behaviour that emerges from learned parameters rather than explicit programming. Regulatory definitions such as that codified in the EU AI Act emphasise the system-level perspective, encompassing the full sociotechnical context of deployment.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:ai-system", "labels": [ "AI System" ], "is_subclass_of": [ "Sociotechnical System" ], "wikilinks": [ "AI Model", "AI Inference", "Data Pipeline", "Inference Runtime", "Compute Infrastructure", "Training Data", "MLOps", "Data Governance", "Model Serving", "AI Governance", "Explainability", "AI Risk Assessment", "Autonomous Decision-Making", "Intelligent Automation", "Edge AI System", "Multi-Agent System", "AI System (EU Definition)", "Rule-Based System", "Regulatory Compliance", "Digital Infrastructure" ] }, { "id": "ai-systems-thinking", "title": "AI Systems Thinking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An approach to AI adoption that emphasizes building robust, vendor-independent learning systems over relying on single models or tools.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-systems-thinking", "labels": [ "AI Systems Thinking" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "ai-tri-sm", "title": "AI TRiSM", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI TRiSM (Artificial Intelligence Trust, Risk, and Security Management) is a Gartner-coined framework that integrates technical, organisational, and regulatory controls to ensure AI systems are trustworthy, fair, reliable, and secure throughout their lifecycle. It addresses algorithmic bias, model opacity, data privacy, adversarial vulnerabilities, and compliance with regulations such as the EU AI Act, structured around three pillars: Trust, Risk, and Security.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:ai-tri-sm", "labels": [ "AI TRiSM" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AI Agent System", "EU AI Act", "MetaverseDomain" ] }, { "id": "ai-talent-acquisition", "title": "AI Talent Acquisition", "domain": "economics", "domain_name": "Economics", "definition": "Strategic corporate actions, including acquisitions and high-value hiring, aimed at securing specialized artificial intelligence expertise and research leadership.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-talent-acquisition", "labels": [ "AI Talent Acquisition" ], "is_subclass_of": [ "Enterprise AI" ], "wikilinks": [] }, { "id": "ai-talent-war", "title": "AI Talent War", "domain": "ai", "domain_name": "Ai", "definition": "The intensely competitive global contest among technology companies, research institutions, and nation-states to attract, retain, and concentrate the scarce pool of highly skilled AI researchers and engineers. The competition manifests through escalating compensation packages, aggressive academic recruitment, strategic immigration policy, and corporate acquisitions of talent-rich startups. Because frontier AI capability is tightly coupled to the concentration of top research talent, this competition is simultaneously an economic, geopolitical, and strategic-security phenomenon.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:ai-talent-war", "labels": [ "AI Talent War" ], "is_subclass_of": [ "Competition in AI", "Technology Race", "Geopolitics", "Workforce Development" ], "wikilinks": [ "AI Talent", "Frontier AI", "Frontier Model Training", "Education and AI", "AI Governance Framework", "Competition in AI", "Compute Infrastructure", "AI Investment", "Geopolitics", "National AI Strategies", "AI Safety Research", "AI Model Development", "Export Controls", "Talent Concentration", "Innovation Ecosystems", "Workforce Development", "Open Source AI", "AI Policy", "Economic Competitiveness", "Technological Leadership" ] }, { "id": "ai-talent", "title": "AI Talent", "domain": "ai", "domain_name": "Ai", "definition": "The human capital pool comprising individuals with specialised skills in machine learning research, AI engineering, data science, and related disciplines who design, build, evaluate, and govern artificial intelligence systems. AI talent is a scarce and strategically contested resource, with demand from frontier laboratories, technology companies, and public-sector institutions consistently outpacing the supply produced by academic programmes. The composition of the talent pool\u2014including its geographic distribution, diversity, and specialisation\u2014directly shapes the pace and direction of AI progress.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:ai-talent", "labels": [ "AI Talent", "AI Talent Capacity" ], "is_subclass_of": [ "Workforce Development", "Human Capital", "AI Research Talent" ], "wikilinks": [ "AI Talent War", "Education and AI", "AI Model Development", "AI Safety Research", "AI Governance Framework", "Workforce Development", "Deep Learning", "Frontier AI", "Machine Learning", "Data Science", "Competition in AI", "AI Research Talent", "Talent Concentration", "Innovation Ecosystems", "Open Source AI", "National AI Strategies", "AI Investment", "Compute Infrastructure", "Geopolitics", "AI Policy" ] }, { "id": "ai-tax-policy", "title": "AI Tax Policy", "domain": "governance", "domain_name": "Governance", "definition": "Governmental fiscal strategies and regulatory instruments designed to address the economic shifts, revenue changes, and distributional effects caused by artificial intelligence and automation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-tax-policy", "labels": [ "AI Tax Policy" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "ai-technology-skills-curriculum", "title": "AI Technology Skills Curriculum", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Structured educational programmes and skill-development resources covering artificial intelligence, machine learning, blockchain, spatial computing, and related technologies. These include MOOCs, short courses, university programmes, professional certifications, and self-directed learning materials that build practitioner capacity across technical and non-technical audiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-technology-skills-curriculum", "labels": [ "AI Technology Skills Curriculum", "Courses and Training" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Blockchain", "latent space" ] }, { "id": "ai-traffic-conversion", "title": "AI Traffic Conversion", "domain": "economics", "domain_name": "Economics", "definition": "The measurement of user engagement and purchase completion rates specifically for sessions initiated or influenced by artificial intelligence interfaces.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-traffic-conversion", "labels": [ "AI Traffic Conversion" ], "is_subclass_of": [ "AI Agents" ], "wikilinks": [] }, { "id": "ai-training-data-scarcity", "title": "AI Training Data Scarcity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The emerging risk that the decline of human-curated knowledge repositories will reduce the availability of high-quality data for training AI models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-training-data-scarcity", "labels": [ "AI Training Data Scarcity" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "ai-training-formats", "title": "AI Training Formats", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The methods and structures used to educate employees on AI tools, including video courses, hands-on labs, and integrated workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-training-formats", "labels": [ "AI Training Formats" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-training-infrastructure", "title": "AI Training Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical and computational resources, including data centers and GPU clusters, required to train large-scale artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-training-infrastructure", "labels": [ "AI Training Infrastructure" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-training-monitoring", "title": "AI Training Monitoring", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The practice of using specialized tools to observe, debug, and optimize the performance and stability of large-scale machine learning model training runs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-training-monitoring", "labels": [ "AI Training Monitoring" ], "is_subclass_of": [ "AI Agents" ], "wikilinks": [] }, { "id": "ai-trust-risk-and-security-management", "title": "AI Trust Risk and Security Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A comprehensive governance framework \u2014 popularised by Gartner and aligned with the NIST AI Risk Management Framework \u2014 for managing the trustworthiness, security, and ethical risks of AI systems throughout their lifecycle. It operationalises three pillars: Trust (transparent, auditable, reliable AI behaviour), Risk (proactive identification and mitigation of algorithmic bias, data quality issues, and regulatory non-compliance), and Security (protection against adversarial attacks, data poisoning, and model integrity threats).", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:ai-trust-risk-and-security-management", "labels": [ "AI Trust Risk and Security Management" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "NIST AI Risk Management Framework", "AI Agent System", "MetaverseDomain" ] }, { "id": "ai-trustworthiness-dimensions", "title": "AI Trustworthiness Dimensions", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Trustworthiness Dimensions are the seven interdependent assessment criteria\u2014human agency and oversight, technical robustness and safety, privacy and data governance, transparency and explainability, diversity and non-discrimination and fairness, societal and environmental wellbeing, and accountability\u2014that collectively determine whether an AI system meets the requirements for trustworthy deployment. Established by the EU High-Level Expert Group on AI Ethics Guidelines (2019) and operationalised through the EU AI Act, these dimensions are mutually reinforcing rather than substitutable: satisfying one does not compensate for deficiencies in another.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-trustworthiness-dimensions", "labels": [ "AI Trustworthiness Dimensions" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU Charter of Fundamental Rights", "EU HLEG AI", "AIEthicsDomain", "ConceptualLayer", "EU AI Act", "Smart Contract" ] }, { "id": "ai-trustworthiness", "title": "AI Trustworthiness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The degree to which an AI system demonstrates characteristics that warrant confidence and reliance, encompassing transparency, explainability, fairness, accountability, robustness, reliability, safety, security, and privacy throughout its lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ai-trustworthiness", "labels": [ "AI Trustworthiness", "Trustworthiness" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AI Assurance (AI-0102)", "GDPR", "Governance Framework (AI-0035)", "IEEE P7009", "Responsible AI (AI-0033)", "Risk Management (AI-0062)", "AI Agent System", "EU AI Act", "MetaverseDomain" ] }, { "id": "ai-upscaling-and-super-resolution", "title": "AI Upscaling and Super-Resolution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Upscaling and Super-Resolution is a family of deep learning techniques that synthesise plausible high-frequency detail when enlarging images or video frames beyond their native resolution. Methods range from convolutional neural network regressors (SRCNN, ESRGAN) and generative adversarial approaches to diffusion-based reconstruction, enabling applications from consumer photography enhancement to real-time game upscaling (DLSS, FSR) and medical imaging.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ai-upscaling-and-super-resolution", "labels": [ "AI Upscaling and Super-Resolution", "AI Upscaling" ], "is_subclass_of": [ "AI Application", "Generative AI" ], "wikilinks": [ "AKT", "AlexNet", "Arweave", "CLIP", "CNN", "CodeFormer", "ComputerVisionDomain", "ControlNet", "Diffusion Model", "ESRGAN", "GAN", "Generator", "GIMP", "Google Photos", "iOS", "Image-to-Image", "IPFS", "Latent Diffusion Model", "Lightroom", "MPV" ] }, { "id": "ai-user-demographics", "title": "AI User Demographics", "domain": "economics", "domain_name": "Economics", "definition": "Statistical breakdowns of artificial intelligence adoption rates segmented by geographic, economic, or sociological factors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-user-demographics", "labels": [ "AI User Demographics" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "ai-user", "title": "AI User", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An individual or entity who interacts with, relies upon, or is affected by the outputs, decisions, or recommendations of an AI system in order to accomplish tasks, obtain services, or achieve objectives\u2014either through direct interaction or indirect exposure to AI-mediated decisions. AI users are the primary experiencers of AI benefits and harms, encompassing diverse populations with varying technical literacy, accessibility needs, and degrees of voluntary engagement, making their rights and interests central to responsible AI governance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-user", "labels": [ "AI User" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "community", "design thinking", "documentation", "Fairness**: Protection of users from discriminatory treatment\n\t\t - **User Experience**: Quality of user interaction with AI\n\t\t - **Accessibility**: Ensuring usability for diverse users\n\t\t - **Informed Consent**: User agreement based on understanding\n\n ## Context and Significance\n\n AI users represent the ultimate stakeholders for whom AI systems are designed and deployed, experiencing both the benefits and risks of AI applications. The user perspective is essential for assessing AI system effectiveness, usability, fairness, and social impact. User needs, capabilities, and contexts significantly influence appropriate AI system design, deployment approaches, and oversight mechanisms.\n\n The NIST AI Risk Management Framework emphasises the importance of understanding user contexts, capabilities, and expectations when mapping and managing AI risks. User diversity\u2014in technical sophistication, domain expertise, accessibility requirements, cultural backgrounds, and power relationships with AI providers\u2014necessitates user-centred design approaches and inclusive development practices.\n\n Modern AI systems create increasingly complex user relationships: users may simultaneously benefit from and be constrained by AI, may interact knowingly or unknowingly with AI, may possess varying degrees of choice in AI system use, and may have asymmetric information about AI system functioning. These dynamics raise important questions about informed consent, user autonomy, and power imbalances requiring ethical consideration and governance attention.\n\n #### References\n 1. European Commission, *Proposal for a Regulation on Artificial Intelligence (AI Act)* (2021)\n\t\t 2. NIST AI 100-1, *Artificial Intelligence Risk Management Framework* (2023)\n\t\t 3. ISO/IEC 25059, *Software engineering \u2014 Systems and software Quality Requirements and Evaluation (SQuaRE) \u2014 Quality model for AI systems*\n\t\t 4. Shneiderman, B., *Human-Centered AI* (2022)\n\t\t 5. GDPR, Articles 13-15, 22 (data subject rights)\n\n\t\t ## See Also\n\n\t\t - [[AI Provider", "GDPR", "Informed Consent", "innovation", "king1966fisher", "MIT\u2019s lit on Lightning", "modeling", "MUST", "NIST AI Risk Management Framework", "Nostr", "optimization", "organisation", "performance", "research", "user experience", "User Experience", "Accessibility" ] }, { "id": "ai-venture-capital", "title": "AI Venture Capital", "domain": "economics", "domain_name": "Economics", "definition": "The investment strategies and funding dynamics specific to the AI sector, including valuations and capital allocation for frontier labs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-venture-capital", "labels": [ "AI Venture Capital" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-video", "title": "AI Video", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Video, alternatively termed generative video, neural video synthesis, text-to-video (T2V), and image-to-video (I2V), denotes the class of deep generative models and engineering systems that synthesise temporally-coherent moving imagery conditioned on natural-language prompts, reference images,...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-video", "labels": [ "AI Video" ], "is_subclass_of": [ "AI Application", "Deep Generative Model", "Generative AI", "Foundation Model", "Multimodal AI", "Synthetic Media" ], "wikilinks": [ "3D Causal VAE", "Accessibility Captioning", "Adam Optimiser", "Advertising", "AI Avatar Synthesis", "AlgorithmLayer", "Bar-Tal et al. 2024 Lumiere", "BBC Generative AI Principles", "BBC Generative AI Principles 2024", "Blattmann et al. 2023 Stable Video Diffusion", "Brooks et al. 2024 Sora Technical Report", "C2PA", "C2PA Specification 2.0", "Captioning Pipeline", "Classical Animation", "Classifier-Free Guidance", "CLIP", "ComputerVisionDomain", "ControlNet", "Convolution" ] }, { "id": "ai-wearables", "title": "AI Wearables", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A category of consumer hardware devices that integrate artificial intelligence capabilities directly into wearable form factors such as pendants, glasses, or rings to provide ambient assistance and data capture.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-wearables", "labels": [ "AI Wearables" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "ai-workflow-automation", "title": "AI Workflow Automation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of AI agents and large language models to automate complex, multi-step business processes and decision-making workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-workflow-automation", "labels": [ "AI Workflow Automation" ], "is_subclass_of": [ "AI Agents" ], "wikilinks": [] }, { "id": "ai-workforce-impact", "title": "AI Workforce Impact", "domain": "economics", "domain_name": "Economics", "definition": "The structural and economic effects of artificial intelligence adoption on labor markets, including job displacement, role transformation, and changes in corporate hiring strategies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-workforce-impact", "labels": [ "AI Workforce Impact" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-companions", "title": "AI companions", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI companions are a class of consumer-facing conversational artificial intelligence systems explicitly designed and marketed to form ongoing affective, social, and parasocial relationships with individual human users\u2014distinct from task-oriented assistants (Siri, Alexa, Google Assistant) that opti...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-companions", "labels": [ "AI companions" ], "is_subclass_of": [ "AI Application", "Affective Computing System", "Conversational AI", "AI Agent System", "Large Language Model Application", "Consumer AI Product" ], "wikilinks": [ "Affective Computing", "AffectiveComputingDomain", "Affective Computing System", "Affective Tone Module", "Age Gating", "AI Assistant", "Appfigures TechCrunch 2025 AI Companion App Revenue", "Artificial Intimacy", "Avatar Renderer", "Bereavement Processing", "Brandtzaeg Skjuve Folstad 2022 My AI Friend HCR", "C2PA", "Cloud Inference Infrastructure", "Consumer AI Product", "ConsumerAIDomain", "Content Moderation Pipeline", "ConversationalAIDomain", "Creative Collaboration", "Creative Writing", "Customer Service Bot" ] }, { "id": "ai-in-games", "title": "AI in Games", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI in Games is the interdisciplinary research and engineering field encompassing the design, implementation, and study of computational systems that perceive, decide, learn, and generate within interactive entertainment software, spanning classical symbolic techniques (finite-state machines FSM d...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ai-in-games", "labels": [ "AI in Games" ], "is_subclass_of": [ "AI Application", "Applied Artificial Intelligence", "Real-Time AI", "Interactive Computing", "Game Development", "Entertainment Technology" ], "wikilinks": [ "A* Search", "AAAI AIIDE", "Adaptive Tutoring", "AlgorithmLayer", "AlphaZero", "Animation Controller", "Applied Artificial Intelligence", "Automated Playtesting", "Behaviour Tree", "Behaviour Trees", "Believable NPC Behaviour", "Bevy", "Classical AI without Learning", "Computational Creativity", "Content Authoring Pipeline", "Convai", "CryEngine", "Domain Randomization", "DreamerV3", "Dynamic Difficulty Adjustment" ] }, { "id": "ai-in-healthcare", "title": "AI in Healthcare", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence technologies to clinical decision-making, diagnostics, and administrative workflows within the medical field.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-in-healthcare", "labels": [ "AI in Healthcare" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-assisted-developer-tooling", "title": "AI-Assisted Developer Tooling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Coding Support encompasses AI-assisted tools, IDE integrations, and agent frameworks that augment software developers with capabilities including inline completion, code generation, automated refactoring, debugging assistance, and test generation. Tools range from general-purpose LLM-backed assistants to domain-specific coding agents operating on full codebases.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-assisted-developer-tooling", "labels": [ "AI-Assisted Developer Tooling", "Coding support" ], "is_subclass_of": [ "Software Engineering", "Software Engineering" ], "wikilinks": [ "Training Modules", "Vercel", "Agent Frameworks", "ChatGPT", "GPT Engineer", "Infrastructure", "Microsoft Copilot", "Tips and Tricks" ] }, { "id": "ai-assisted-development-practice", "title": "AI-Assisted Development Practice", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Vibe Coding is an AI-assisted software development practice in which developers direct large language models or AI coding tools (such as Cursor, Aider, or Copilot) with high-level intent rather than writing all code manually. The practitioner retains system-level design authority whilst delegating implementation of discrete, scoped tasks to the AI, requiring disciplined use of version control, structured context documents, and iterative refinement cycles.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ai-assisted-development-practice", "labels": [ "AI-Assisted Development Practice", "Vibe Coding" ], "is_subclass_of": [ "Software Engineering", "Software Engineering" ], "wikilinks": [] }, { "id": "ai-assisted-ontology-elicitation-method", "title": "AI-Assisted Ontology Elicitation Method", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A structured methodology for collaboratively building and refining formal knowledge representations by conducting iterative dialogues with AI systems. In these conversations, AI agents interpret, critique, and extend ontological schemas, producing machine-readable linked-data artefacts such as OWL classes and JSON-LD graphs that document both the resulting ontology and the reasoning process that produced it.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-assisted-ontology-elicitation-method", "labels": [ "AI-Assisted Ontology Elicitation Method", "Ontology conversation with AIs" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Agentic Mycelia" ] }, { "id": "ai-augmented-code-review", "title": "AI-Augmented Code Review", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The integration of artificial intelligence tools into the code review process to automate defect detection, enforce standards, and reduce human review latency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-augmented-code-review", "labels": [ "AI-Augmented Code Review" ], "is_subclass_of": [ "Code Review" ], "wikilinks": [] }, { "id": "ai-augmented-research-tooling-suite", "title": "AI-Augmented Research Tooling Suite", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Research Tools are software applications and platforms that augment the academic and professional research process, encompassing AI-powered literature assistants (e.g. Elicit, Undermind), systematic review automation, citation management, and natural language interfaces to scientific corpora. They accelerate information retrieval, evidence synthesis, and knowledge organisation, and increasingly leverage large language models and retrieval-augmented generation to surface relevant findings across heterogeneous sources.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-augmented-research-tooling-suite", "labels": [ "AI-Augmented Research Tooling Suite", "Research Tools" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "\ud83e\udd16", "Research Tools" ] }, { "id": "ai-augmented-service-ecosystems", "title": "AI-Augmented Service Ecosystems", "domain": "economics", "domain_name": "Economics", "definition": "New professional structures in sectors like legal, education, and finance where human roles are redefined as navigators, support workers, and operators who manage and interpret AI-driven services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-augmented-service-ecosystems", "labels": [ "AI-Augmented Service Ecosystems" ], "is_subclass_of": [ "Economic Impact of AI" ], "wikilinks": [] }, { "id": "ai-augmented-software-engineering", "title": "AI-Augmented Software Engineering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI-Augmented Software Engineering is the practice of integrating artificial intelligence tools and techniques \u2014 including large language models, code generation systems, and agentic AI \u2014 into the software development lifecycle to accelerate coding, testing, review, and deployment. It encompasses AI-assisted code completion, automated test generation, intelligent refactoring, and documentation synthesis, transforming how software teams design, build, and maintain systems at scale.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:ai-augmented-software-engineering", "labels": [ "AI-Augmented Software Engineering", "AI Software Engineering", "AI-Assisted Software Engineering" ], "is_subclass_of": [ "AI Application", "ArtificialIntelligence" ], "wikilinks": [ "ArtificialIntelligence", "Digital Asset" ] }, { "id": "ai-augmented-workflows", "title": "AI-Augmented Workflows", "domain": "governance", "domain_name": "Governance", "definition": "Business processes and job roles that have been restructured to integrate artificial intelligence tools, fundamentally altering task allocation, skill requirements, and organizational design.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-augmented-workflows", "labels": [ "AI-Augmented Workflows" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-driven-efficiency", "title": "AI-Driven Efficiency", "domain": "economics", "domain_name": "Economics", "definition": "The structural economic phenomenon where the adoption of artificial intelligence and automation leads to increased productivity and revenue growth while simultaneously reducing the required human labor force.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-efficiency", "labels": [ "AI-Driven Efficiency" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-driven-enterprise", "title": "AI-Driven Enterprise", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A business organization that integrates artificial intelligence into its core strategy, operations, and value proposition to achieve competitive advantage and operational efficiency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-enterprise", "labels": [ "AI-Driven Enterprise" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-driven-software-development", "title": "AI-Driven Software Development", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The integration of generative AI tools into the software engineering lifecycle to automate code generation, review, and testing, fundamentally altering developer workflows and productivity metrics.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-software-development", "labels": [ "AI-Driven Software Development" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "ai-driven-unemployment", "title": "AI-Driven Unemployment", "domain": "economics", "domain_name": "Economics", "definition": "The potential displacement of human workers by AI systems, particularly in knowledge work and entry-level roles.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-unemployment", "labels": [ "AI-Driven Unemployment" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ai-driven-vulnerability-discovery", "title": "AI-Driven Vulnerability Discovery", "domain": "security", "domain_name": "Security", "definition": "The application of large language models and AI agents to systematically identify, analyze, and exploit software and system vulnerabilities at a scale and speed exceeding traditional human-led security research.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-vulnerability-discovery", "labels": [ "AI-Driven Vulnerability Discovery" ], "is_subclass_of": [ "AI Security" ], "wikilinks": [] }, { "id": "ai-driven-workforce-displacement-registry", "title": "AI-Driven Workforce Displacement Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated record of mass layoffs in the technology sector from 2023 onwards, cataloguing headcount reductions at major firms alongside UK government estimates that AI automation may displace 10\u201330% of existing jobs. The tracker contextualises how AI adoption\u2014particularly large language models and agentic AI\u2014is accelerating workforce restructuring across white-collar roles.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ai-driven-workforce-displacement-registry", "labels": [ "AI-Driven Workforce Displacement Registry", "Job Displacement", "Layoff tracker and threatened roles" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "ai-driven-workforce-reduction", "title": "AI-Driven Workforce Reduction", "domain": "economics", "domain_name": "Economics", "definition": "The systematic reduction of human labor roles and headcount within organizations explicitly attributed to the adoption of artificial intelligence capabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-driven-workforce-reduction", "labels": [ "AI-Driven Workforce Reduction" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "ai-generated-content-disclosure", "title": "AI-Generated Content Disclosure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mandatory transparency requirement specifying that content created wholly or partially by AI systems must be explicitly labeled with origin metadata for user awareness and regulatory compliance.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:ai-generated-content-disclosure", "labels": [ "AI-Generated Content Disclosure", "AI Disclosure" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Transparency Framework", "AI Governance Framework", "AI Ethics" ], "wikilinks": [ "Content Labeling Metadata", "Disclosure Enforcement Mechanism", "IEEE 7001", "Informed Consent", "Trust Building", "Digital Watermarking", "Steganographic Marking", "Human Oversight", "Model Card", "Synthetic Media Policy", "Algorithmic Accountability", "Online Safety Act", "AI Agent System", "AI Origin Declaration", "AI Transparency Framework", "C2PA Standard", "Content Authentication", "EU AI Act", "Metadata Standards", "MiddlewareLayer" ] }, { "id": "ai-generated-pathogens", "title": "AI-Generated Pathogens", "domain": "security", "domain_name": "Security", "definition": "Biological agents or viral structures designed or synthesized using artificial intelligence algorithms, raising specific concerns regarding biosecurity and the potential for novel disease emergence.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-generated-pathogens", "labels": [ "AI-Generated Pathogens" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "aiapplications", "title": "AIApplications", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A domain classification encompassing the practical deployment and use cases of artificial intelligence systems across industries, including autonomous systems, decision support, content generation, predictive analytics, and intelligent automation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:aiapplications", "labels": [ "AIApplications" ], "is_subclass_of": [ "AI Application", "ArtificialIntelligenceDomain" ], "wikilinks": [ "AI Capability", "Application Domain", "Industry Vertical", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "aicafev6", "title": "AICafev6", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AICafev6 is the sixth iteration of the AI Cafe demonstration platform, an open-source immersive knowledge development environment built on VisionFlow that integrates multi-modal AI, GPU-accelerated analytics, and agentic workflows for collaborative knowledge creation and presentation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:aicafev6", "labels": [ "AICafev6" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "AI Agent System", "VisionFlow and Junkie Jarvis" ] }, { "id": "aiethics", "title": "AIEthics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A framework of ethical principles and governance structures ensuring responsible development, deployment, and oversight of artificial intelligence systems in compliance with societal values. Addresses fairness, accountability, transparency, bias mitigation, privacy preservation, and alignment with human rights norms across the AI lifecycle from design through decommissioning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:aiethics", "labels": [ "AIEthics" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "addressesConcern", "AIGovernance", "appliesTo", "BiasAndFairness", "DAOGovernance", "definesGuideline", "Digital Systems", "DiscriminationPrevention", "dt:enforcedBy", "dt:governedBy", "dt:guides", "dt:trackedOn", "dt:validates", "Human Values", "requiresCompliance", "Accountability", "AI Agent System", "AI Risk", "Artificial Intelligence", "AuditTrail" ] }, { "id": "aiops", "title": "AIOps", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AIOps (Artificial Intelligence for IT Operations) applies machine learning, natural language processing, and big data analytics to automate and enhance IT operational tasks including event correlation, anomaly detection, root-cause analysis, incident management, and capacity planning. By continuously ingesting telemetry streams \u2014 logs, metrics, and traces \u2014 from complex distributed systems, AIOps platforms surface actionable insights that would be impossible for human operators to detect at scale, progressively shifting operations from reactive incident response towards predictive and autonomous self-healing.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:aiops", "labels": [ "AIOps" ], "is_subclass_of": [ "AI Infrastructure", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "Edge Computing" ] }, { "id": "aisi-frontier-ai-safety-framework", "title": "AISI Frontier AI Safety Framework", "domain": "governance", "domain_name": "Governance", "definition": "A structured evaluation and oversight framework produced by the UK AI Safety Institute (AISI) to assess the catastrophic risks posed by frontier AI models prior to and following their public release. The framework specifies pre-deployment testing protocols, thresholds for dangerous capability uplift, and post-deployment monitoring obligations that developers of frontier models are expected to satisfy. It represents the UK government's primary technical instrument for operationalising AI safety commitments made at the Bletchley Park AI Safety Summit of November 2023.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:aisi-frontier-ai-safety-framework", "labels": [ "AISI Frontier AI Safety Framework" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "aisi", "title": "AISI", "domain": "governance", "domain_name": "Governance", "definition": "The AI Safety Institute is a UK government body that evaluates advanced AI systems for safety and security risks. It was established to test frontier models and inform policy.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:aisi", "labels": [ "AISI" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "AI Governance", "AI Regulation", "Frontier AI", "AI Safety", "https://www.aisi.gov.uk", "https://www.gov.uk/government/organisations/ai-safety-institute" ] }, { "id": "aisystem", "title": "AISystem", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A comprehensive system architecture comprising machine learning models, data pipelines, inference engines, and deployment infrastructure that enables intelligent decision-making and automation across diverse application domains.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:aisystem", "labels": [ "AISystem" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AI Systems", "AIGovernance", "deployedOn", "DeploymentPattern", "dt:controls", "dt:enhances", "dt:implementedOn", "dt:securedBy", "dt:uses", "hasModel", "IEEE 7000", "monitorsPerformance", "Neural Networks", "Organisational Contexts", "usesDataset", "AI Agent System", "AIEthics", "Artificial Intelligence", "AutonomousRobot", "BlockchainNetwork" ] }, { "id": "albert", "title": "ALBERT", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Lite BERT: a parameter-efficient transformer variant that uses factorised embedding parameterisation and cross-layer parameter sharing to achieve 18x fewer parameters than BERT-large whilst matching or exceeding its performance on NLP benchmarks such as GLUE, SQuAD, and RACE.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:albert", "labels": [ "ALBERT" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "AI Agent System", "MetaverseDomain" ] }, { "id": "alpr-networks", "title": "ALPR Networks", "domain": "security", "domain_name": "Security", "definition": "ALPR Networks are interconnected systems of Automatic License Plate Recognition cameras that capture, read, timestamp, and geotag vehicle plates at scale across roads, lots, and checkpoints. Aggregated across many cameras and operators, the resulting databases reconstruct vehicle movement histories and enable mass location tracking. They are a prominent component of contemporary surveillance infrastructure and a focal point of privacy and civil-liberties concern.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:alpr-networks", "labels": [ "ALPR Networks" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "aml-kyc-compliance", "title": "AML KYC Compliance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "AML KYC Compliance is the composite regulatory and operational discipline by which obligated entities \u2014 banks, payment institutions, e-money issuers, brokers, insurers, accountants, lawyers, art dealers, casinos, real-estate agents, virtual-asset service providers (VASPs / crypto-asset service pr...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:aml-kyc-compliance", "labels": [ "AML KYC Compliance", "AML Compliance", "AML/KYC", "AML/KYC Compliance", "BC-0457-aml-kyc-compliance", "BC-0476-aml-kyc-compliance" ], "is_subclass_of": [ "Governance and Regulation", "Regulatory Compliance", "Financial Crime Compliance", "Risk Management Framework", "Customer Onboarding Discipline", "Anti-Financial Crime Regime" ], "wikilinks": [ "Adverse Media Screening", "Adyen", "Allen 2016 Path to Self-Sovereign Identity", "AMLA", "Anonymous Cryptocurrency", "Anti-Financial Crime Regime", "Arner Barberis Buckley 2017 FinTech RegTech", "AUSTRAC", "Australia AML CTF Act 2006 Tranche 2 Reform 2024", "Bains Sugimoto Wilson 2022 IMF RegTech", "Bank Secrecy Act", "Bank Secrecy Act 1970", "Banking Correspondent Relationships", "Beneficial Ownership", "Beneficial Ownership Disclosure", "Beneficial Ownership Registers", "Binance Settlement 2023", "Biometric Liveness Detection", "Blockchain Analytics", "Blockchain Analytics Platform" ] }, { "id": "aml", "title": "AML", "domain": "finance", "domain_name": "Finance", "definition": "Anti-Money Laundering (AML) is the body of laws, regulations, policies, and operational controls that obligate financial institutions to detect, prevent, and report the process by which criminals disguise proceeds of illegal activity as legitimate income. AML frameworks encompass Know Your Customer (KYC) identity verification, ongoing transaction monitoring, suspicious activity reporting (SAR), and risk-based compliance programmes overseen by regulators such as the Financial Action Task Force (FATF), FinCEN, and national supervisory authorities. Machine learning and network analytics have become central to modern AML systems, enabling real-time anomaly detection and entity-resolution across large transaction graphs. Regulated entities that fail to meet AML obligations face substantial financial penalties, licence revocations, and criminal liability.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:aml", "labels": [ "AML", "AML CFT Framework", "AML Screening", "AML/CFT", "AML/CFT Regime", "Anti-Money Laundering" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [ "Classification", "Anti-Money Laundering", "Financial Services" ] }, { "id": "amm-algorithm", "title": "AMM Algorithm", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An AMM Algorithm is the pricing rule of an Automated Market Maker that algorithmically sets exchange rates from the reserves held in a liquidity pool, removing the need for a traditional order book. The canonical form is the constant-product invariant x*y=k, with variants such as constant-sum, stableswap, and concentrated-liquidity curves tuned for different asset pairs. It determines slippage, price impact, and the impermanent loss that liquidity providers bear.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:amm-algorithm", "labels": [ "AMM Algorithm" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "amqp", "title": "AMQP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AMQP (Advanced Message Queuing Protocol) is an open, binary application-layer protocol for message-oriented middleware that defines wire-level framing, message routing and reliable delivery between brokers and clients regardless of vendor or platform. It models messaging through exchanges, queues and bindings that decouple producers from consumers and support routing patterns including direct, topic, fanout and headers exchanges. AMQP 1.0 is standardised by OASIS and ISO/IEC 19464, while the earlier AMQP 0-9-1 specification remains widely deployed in broker implementations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:amqp", "labels": [ "AMQP" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "api-design", "title": "API Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "API design is the discipline of specifying the contract, structure, and behaviour of an application programming interface so that it is consistent, intuitive, evolvable, and reliable for the developers who consume it. It covers resource and operation modelling, naming and conventions, request and response schemas, error semantics, authentication, versioning, and documentation, balancing usability against the constraints of the underlying system. Good API design treats the interface as a long-lived product whose contract must remain stable and backward-compatible while still allowing the implementation behind it to evolve.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-design", "labels": [ "API Design" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "api-gateway", "title": "API Gateway", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Infrastructure component serving as the unified northbound entry point for client requests into a Distributed System|distributed or Microservices|microservices backend, providing cross-cutting policy enforcement (request routing across hundreds-to-thousands of upstream services, authentic...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:api-gateway", "labels": [ "API Gateway" ], "is_subclass_of": [ "Network and Communication", "Policy Enforcement Point", "Middleware", "Reverse Proxy", "Network Infrastructure", "Cloud Native Component" ], "wikilinks": [ "Access Log Emitter", "AI Gateway", "Alan Turing Institute", "Anthropic 2024 Model Context Protocol Specification", "Apache APISIX", "API Composition", "API Lifecycle Management", "API Management", "Apigee", "Apollo Federation 2 Specification", "AsyncAPI", "Authentication Handler", "AWS API Gateway", "AWS AppSync", "Azure API Management", "Backend-for-Frontend", "Banks & Porcello 2018 Learning GraphQL", "Bulkhead Pattern", "Burns Beda Hightower Reznik 2022 Kubernetes Up and Running", "Calcote & Butcher 2019 Istio Up and Running" ] }, { "id": "api-integration", "title": "API Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "API Integration is the discipline of connecting disparate software systems, services, and data sources through Application Programming Interfaces to achieve seamless interoperability and automated data exchange. It encompasses the design, implementation, orchestration, and maintenance of integration layers \u2014 including synchronous REST and GraphQL calls, asynchronous message queues, webhook-based push notifications, and event-driven streams \u2014 that allow applications to communicate via standardised contracts while abstracting the complexity of underlying system differences. Effective API integration reduces manual data transfer, accelerates business workflows, and forms the structural backbone of modern microservices, cloud-native, and composable enterprise architectures. Integration governance concerns \u2014 versioning, rate limiting, authentication, observability, and schema evolution \u2014 are integral to sustaining reliable integrations at scale.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:api-integration", "labels": [ "API Integration", "Integration APIs" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "api-lifecycle", "title": "API Lifecycle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The end-to-end progression of an application programming interface from design and specification through implementation, publication, versioning, monitoring, deprecation, and eventual retirement. Managing the API lifecycle as a deliberate process \u2014 with contract-first design, semantic versioning, sunset policies, and consumer migration paths \u2014 is what keeps evolving service interfaces stable for the clients that depend on them.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:api-lifecycle", "labels": [ "API Lifecycle" ], "is_subclass_of": [ "API Management" ], "wikilinks": [ "API Management", "API Versioning", "API Design" ] }, { "id": "api-management", "title": "API Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "API Management is the discipline and tooling concerned with designing, publishing, documenting, securing, monitoring, and analysing application programming interfaces throughout their lifecycle. It provides a centralised control plane that governs how internal and external consumers discover and consume backend services, enforcing policies such as authentication, rate-limiting, and traffic shaping at a gateway layer. Modern API management platforms combine developer portals, analytics dashboards, and policy engines to ensure reliability, security, and business alignment across distributed service ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:api-management", "labels": [ "API Management", "API Lifecycle Management" ], "is_subclass_of": [ "Technology Infrastructure" ], "wikilinks": [] }, { "id": "api-monetisation", "title": "API Monetisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The practice of generating revenue directly from programmatic interfaces by metering consumption and charging for it \u2014 through subscription tiers, pay-per-call and usage-based pricing, revenue sharing, or machine-payable protocols such as L402 \u2014 turning an API from an integration mechanism into a product with its own pricing, packaging, billing, and developer-experience lifecycle.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:api-monetisation", "labels": [ "API Monetisation" ], "is_subclass_of": [ "API Management" ], "wikilinks": [ "API Management", "API Gateway", "Lightning Network", "Streaming Payment" ] }, { "id": "api-security", "title": "API Security", "domain": "security", "domain_name": "Security", "definition": "API security is the discipline of protecting application programming interfaces from misuse, abuse, and attack across their lifecycle, covering authentication, authorisation, input validation, transport encryption, rate limiting, and monitoring. As APIs expose business logic and data directly to clients and partners, they present a broad attack surface addressed through tokens such as OAuth and JWT, gateways, and threat modelling against risks like broken object-level authorisation. It is a core component of modern application and web security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-security", "labels": [ "API Security" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "api-specification", "title": "API Specification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An API Specification is a formal, machine-readable document that precisely defines the interface contract of a software API, including its endpoints, request and response schemas, authentication requirements, and error codes. It enables automated tooling such as code generation, validation, and interactive documentation to be derived directly from a single source of truth. Common specification formats include OpenAPI, AsyncAPI, and GraphQL SDL.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:api-specification", "labels": [ "API Specification" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "api-standard", "title": "API Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Specification defining how independent software components communicate, establishing protocols, data formats, versioning rules, and authentication mechanisms to ensure interoperability across systems and platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:api-standard", "labels": [ "API Standard", "API Standardisation" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "Authentication Schema", "Component Integration", "Data Format Specification", "ETSI GR ARF 010", "ISO/IEC 30170", "OMA3", "Technical Specification Document", "AI Agent System", "Communication Protocol", "DataLayer", "InfrastructureDomain", "InteractionDomain", "System Interoperability" ] }, { "id": "api-versioning", "title": "API Versioning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "API versioning is the discipline of evolving a programmatic interface over time while controlling the impact of changes on existing consumers. It defines how new versions are identified, published, and retired, and how backward compatibility is preserved or broken deliberately. Sound versioning lets providers innovate without forcing every client to upgrade in lockstep.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-versioning", "labels": [ "API Versioning" ], "is_subclass_of": [ "API Management" ], "wikilinks": [] }, { "id": "api", "title": "API", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Application Programming Interface: a defined set of operations, inputs, and outputs through which software components or services communicate without exposing their internal implementation, enabling modular composition and system integration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:api", "labels": [ "API", "API Endpoint", "API Interface", "API Interfaces", "Metered API", "NumPy API", "Proprietary API", "Stateful API" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Communication Protocols", "Interoperability", "HTTP", "Interoperability Standards", "Software Engineering" ] }, { "id": "apilayer", "title": "APILayer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The API Layer is the stratum that exposes a system's capabilities as callable, contractually defined endpoints. It sits above the application logic it fronts and below integration and interface strata that connect consumers. It contains endpoint definitions, request and response schemas, authentication hooks, and versioning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:apilayer", "labels": [ "APILayer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Application Layer", "Integration Layer", "Interface Layer", "REST", "OpenAPI Specification", "owl:Thing" ] }, { "id": "ar-frame", "title": "AR Frame", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An AR Frame is a head-worn optical see-through augmented reality platform integrating a transparent or semi-transparent microdisplay (typically a microOLED, LCoS, or holographic waveguide projecting 480p-1080p imagery onto a combiner lens of 20\u00b0-50\u00b0 diagonal field-of-view), onboard processing (AR...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ar-frame", "labels": [ "AR Frame" ], "is_subclass_of": [ "Display and Rendering", "Platform and Environment", "Wearable Device", "AR Hardware", "Head-Mounted Display", "Optical See-Through HMD", "Spatial Computing Platform" ], "wikilinks": [ "Accessibility Captioning", "AI Assistant", "Ambient Computing", "Apple visionOS Developer Documentation", "AR Hardware", "Azuma 1997 Survey of Augmented Reality", "Battery", "Billinghurst Clark & Lee 2015 AR Survey", "Birdbath Optics", "Bloomberg Power On Gurman 2024", "Bluetooth Low Energy", "Bluetooth SIG", "Brilliant Labs Frame Hardware Documentation", "Caudell & Mizell 1992 Augmented Reality Boeing", "ConsumerElectronicsDomain", "Counterpoint Research Smart Glasses Tracker", "Davison et al. 2007 MonoSLAM", "Edge AI", "EdgeComputeLayer", "Elmi-Terander et al. 2019 AR Spine Navigation" ] }, { "id": "ar-interoperability", "title": "AR Interoperability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR interoperability is the capacity for augmented reality content, spatial anchors, and experiences to function consistently across different hardware devices, operating systems, and rendering engines. It depends on shared data formats for spatial maps and anchors, common coordinate reference conventions, and standardised APIs for camera, tracking, and rendering access, as pursued by efforts such as the ETSI Augmented Reality Framework. Without it, AR content authored for one platform typically requires substantial rework to run on another, fragmenting the content ecosystem and raising development cost. It is a prerequisite for shared, persistent AR experiences that multiple users on different devices can view and manipulate together.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ar-interoperability", "labels": [ "AR Interoperability" ], "is_subclass_of": [ "Cross-Platform Interoperability" ], "wikilinks": [] }, { "id": "ar-occlusion", "title": "AR Occlusion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR occlusion is the computational technique by which virtual objects rendered in an augmented reality scene are correctly hidden or partially hidden by real-world geometry that physically lies in front of them from the viewer's perspective. Achieving occlusion requires real-time estimation of the depth structure of the physical scene, typically via depth sensors or monocular depth estimation neural networks, so that the rendering pipeline can apply correct depth ordering between real and virtual content. Without occlusion, virtual objects appear to float unconvincingly in front of all physical surfaces regardless of their spatial relationship.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ar-occlusion", "labels": [ "AR Occlusion" ], "is_subclass_of": [ "Augmented Reality", "Display and Rendering" ], "wikilinks": [] }, { "id": "ar-registration", "title": "AR Registration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Registration is the spatial alignment process that anchors virtual content to real-world coordinates through Computer Vision, Sensor Input, and tracking algorithms.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:ar-registration", "labels": [ "AR Registration" ], "is_subclass_of": [ "Interaction Technology", "Spatial Tracking", "Computer Vision" ], "wikilinks": [ "Camera Calibration", "Feature Matching", "Geometric Accuracy", "IMU", "Interaction with Physical Objects", "Persistent Content Anchoring", "Sensor Input", "3D User Interface", "AR Technology", "Computer Vision", "Pose Estimation", "SLAM", "Spatial Anchoring", "Spatial Tracking" ] }, { "id": "ar-technology", "title": "AR Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Technology encompasses the hardware, software, and algorithmic systems enabling real-time digital content overlay on physical environments, including spatial tracking, environmental understanding, rendering pipelines, and interaction modalities across mobile, wearable, and projection-based form factors.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:ar-technology", "labels": [ "AR Technology" ], "is_subclass_of": [ "Extended Reality", "Augmented Reality (AR)" ], "wikilinks": [ "Display Technology", "Extended Reality", "Industrial Guidance", "Remote Assistance", "Sensor Input", "Spatial Computing Applications", "3D Rendering Engine", "AR Frame", "AR Registration", "Computer Vision", "Edge Computing", "Spatial Computing" ] }, { "id": "arc-agi", "title": "ARC-AGI", "domain": "ai", "domain_name": "Ai", "definition": "ARC-AGI (Abstraction and Reasoning Corpus for Artificial General Intelligence) is a benchmark designed by Fran\u00e7ois Chollet and published in 2019 to measure general fluid intelligence in AI systems through abstract visual pattern completion tasks that require novel rule induction rather than pattern recall from training data. Each task presents a small number of input-output grid transformation examples from which the solver must infer the underlying rule and apply it to a new input, using only core knowledge priors available to young children. The benchmark explicitly resists solution by memorisation, making it a proxy test for human-like generalisation ability.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:arc-agi", "labels": [ "ARC-AGI" ], "is_subclass_of": [ "Benchmarks", "Benchmark Standard", "Benchmark Evaluation" ], "wikilinks": [ "Programme Synthesis", "Domain-Specific Language", "Fluid Intelligence", "Test-Time Training", "Test-Time Compute", "Inductive Logic Programming", "Visual Reasoning", "Few-Shot Generalisation", "Core Knowledge Priors", "Grid-Based Tasks", "Agentic Intelligence", "ARC Prize", "ARC-AGI-2", "ARC-AGI-3", "Evaluation Leaderboard", "Reasoning", "Pattern Recognition", "Machine Learning Discipline", "Cognitive Science", "Artificial General Intelligence" ] }, { "id": "arkg-i2-benchmark", "title": "ARKG I2 Benchmark", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A standardized evaluation metric used to assess the performance of large language models on specific reasoning or intelligence tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:arkg-i2-benchmark", "labels": [ "ARKG I2 Benchmark" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [] }, { "id": "arm-trust-zone", "title": "ARM TrustZone", "domain": "security", "domain_name": "Security", "definition": "ARM TrustZone is a hardware security technology built into ARM processors that partitions the system into two isolated execution worlds \u2014 a Secure World and a Normal World \u2014 enforced at the level of the CPU, memory, and peripherals. Sensitive code and data, such as cryptographic keys and biometric matching, run in the Secure World inaccessible to the rich operating system in the Normal World, providing a trusted execution environment without a separate security chip. It is ubiquitous in mobile devices, embedded systems, and IoT hardware.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:arm-trust-zone", "labels": [ "ARM TrustZone" ], "is_subclass_of": [ "Trusted Execution Environment" ], "wikilinks": [] }, { "id": "arp4754-a", "title": "ARP4754A", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ARP4754A is an aerospace recommended practice published by SAE International in 2010, titled 'Guidelines for Development of Civil Aircraft and Systems,' which provides a structured framework for the development and safety assessment of complex aircraft systems and functions. It establishes processes for identifying and mitigating failure conditions across the full aircraft system lifecycle, defining Development Assurance Levels (DAL A through E) that calibrate the rigour of development and verification activities to the severity of the failure conditions a system can contribute to. ARP4754A works in conjunction with related standards including ARP4761 for safety assessment and DO-178C for software development.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:arp4754-a", "labels": [ "ARP4754A" ], "is_subclass_of": [ "Safety Standard" ], "wikilinks": [] }, { "id": "arpu", "title": "ARPU", "domain": "economics", "domain_name": "Economics", "definition": "Average Revenue Per User is a key performance indicator that measures the average revenue generated per customer over a specific period.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:arpu", "labels": [ "ARPU" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "asic-hardware", "title": "ASIC Hardware", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Application-Specific Integrated Circuits (ASICs) are custom silicon chips designed and manufactured to perform a fixed, narrowly defined computational task with maximum efficiency. Unlike general-purpose processors, ASIC hardware sacrifices programmability for dramatically superior performance-per-watt ratios in its target workload, making it the dominant substrate for high-throughput, power-sensitive deployments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:asic-hardware", "labels": [ "ASIC Hardware", "ASIC Mining Hardware" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [] }, { "id": "asic", "title": "ASIC", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Application-Specific Integrated Circuit (ASIC) is a custom integrated circuit designed and fabricated at the transistor and mask level to perform a specific function or narrow set of functions with maximum efficiency, in contrast to general-purpose processors such as CPUs and GPUs that are optimised for programmable flexibility. ASICs achieve superior performance-per-watt and cost efficiency at production volume by eliminating logic not required for the target workload, at the expense of reconfigurability. Design is conducted using hardware description languages (VHDL, SystemVerilog), electronic design automation suites, and a foundry-specific tape-out flow; non-recurring engineering costs at advanced nodes (5nm, 3nm) can exceed tens of millions of pounds, making economic viability contingent on volume or uniquely demanding performance requirements. ASICs are central to cryptocurrency mining, AI inference and training acceleration, high-speed networking switch fabric, signal processing, and consumer electronics where power, cost, and performance constraints demand purpose-built silicon.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:asic", "labels": [ "ASIC", "AI ASIC", "ASIC Design" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "auc", "title": "AUC", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Area Under the Curve (AUC), specifically the area under the Receiver Operating Characteristic (ROC) curve (ROC-AUC or AUROC), is a single scalar performance metric for binary classifiers representing the probability that the model ranks a randomly chosen positive instance higher than a randomly c...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:auc", "labels": [ "AUC", "ROC-AUC" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Classification", "False Positive Rate", "ISO/IEC 25059", "Model Comparison", "NIST AI RMF", "Precision-Recall Curve", "Ranking Metric", "Threshold Selection", "True Positive Rate", "AI Agent System", "MetaverseDomain", "Model Performance", "ROC Curve" ] }, { "id": "automatic-1111-web-ui", "title": "AUTOMATIC1111 WebUI", "domain": "ai", "domain_name": "Ai", "definition": "AUTOMATIC1111 WebUI (also referred to as sd-webui or A1111) is an open-source, community-maintained browser-based graphical interface for running Stable Diffusion latent diffusion models locally on consumer and professional GPU hardware. Built in Python with a Gradio front end, it wraps the core Stable Diffusion inference pipeline in an accessible point-and-click UI that exposes text-to-image generation, image-to-image transformation, inpainting, outpainting, upscaling, model training tabs, and an extensible plugin architecture. Since its first public release in August 2022, it has accumulated over 160,000 GitHub stars, making it one of the most-starred open-source AI repositories in history and the dominant entry point for local Stable Diffusion experimentation.", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:automatic-1111-web-ui", "labels": [ "AUTOMATIC1111 WebUI", "AUTOMATIC1111 Stable Diffusion WebUI" ], "is_subclass_of": [ "Open Generative AI Tools", "Image Generation", "Generative AI" ], "wikilinks": [ "Stable Diffusion", "Image-to-Image", "Inpainting", "Image Generation", "Diffusion Model", "Generative AI", "Text-to-Image", "ComfyUI", "Prompt Engineering", "ControlNet", "Variational Autoencoder", "Latent Diffusion Model", "Open-Source Software", "GPU Compute", "LoRA", "Textual Inversion", "Upscaling", "Sampler", "Classifier-Free Guidance", "Stable Diffusion XL" ] }, { "id": "av1-codec", "title": "AV1 Codec", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AV1 is a royalty-free video coding format developed by the Alliance for Open Media to deliver high compression efficiency for streaming and real-time video. It typically achieves around 30 percent better compression than HEVC/VP9 at equivalent quality, making it well suited to bandwidth-constrained applications such as screen sharing and live video. Being open and licence-free, it is widely adopted across browsers, hardware decoders, and conferencing platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:av1-codec", "labels": [ "AV1 Codec", "AV1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "aave-companies", "title": "Aave Companies", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Aave Companies is the development company, founded by Stani Kulechov, behind the Aave decentralised lending protocol and related products. It was formerly known as ETHLend.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:aave-companies", "labels": [ "Aave Companies" ], "is_subclass_of": [ "Aave" ], "wikilinks": [ "Lending Protocol", "DeFi", "Aave" ] }, { "id": "aave-governance", "title": "Aave Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Aave Governance is the on-chain decision-making system for the Aave decentralised lending protocol, enabling AAVE token holders and protocol delegates to propose, debate, and vote on parameter changes, new asset listings, risk adjustments, and smart contract upgrades. It operates through a tiered proposal mechanism where quorum and voting-power thresholds determine whether a proposal proceeds to execution, with all outcomes enforced autonomously by smart contracts on Ethereum and affiliated networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:aave-governance", "labels": [ "Aave Governance" ], "is_subclass_of": [ "On-chain Governance" ], "wikilinks": [] }, { "id": "aave", "title": "aave", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Aave is a decentralised, non-custodial liquidity protocol deployed on Ethereum and multiple EVM-compatible networks that allows users to supply crypto assets into pooled reserves to earn algorithmically set interest, and to borrow against over-collateralised positions at variable or stable rates. Pioneered by Stani Kulechov and launched as ETHLend in 2017 before rebranding in 2020, it introduced the flash loan \u2014 an uncollateralised loan that must be atomically repaid within a single transaction block \u2014 as a foundational DeFi primitive. Governance is exercised by AAVE token holders who vote on Aave Improvement Proposals controlling risk parameters, supported asset listings, protocol upgrades, and treasury allocations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:aave", "labels": [ "Aave" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "ablation-zone", "title": "Ablation Zone", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ablation-zone", "labels": [ "Ablation Zone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "absolute-magnitude", "title": "Absolute Magnitude", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:absolute-magnitude", "labels": [ "Absolute Magnitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "academia", "title": "Academia", "domain": "governance", "domain_name": "Governance", "definition": "The worldwide community and institutional system devoted to the production, validation, and transmission of scholarly knowledge \u2014 universities, research institutes, learned societies, journals, and conferences \u2014 organised around disciplines, credentialed through degrees and peer review, and governed by norms of originality, citation, and open critique that distinguish scholarly knowledge from other forms of expertise.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:academia", "labels": [ "Academia" ], "is_subclass_of": [ "Education" ], "wikilinks": [ "Education", "Academic Research", "Research Institution", "Peer Review" ] }, { "id": "academic-conference", "title": "Academic Conference", "domain": "ai", "domain_name": "Ai", "definition": "An academic conference is a periodic, structured gathering at which researchers present, discuss, and publish peer-reviewed work, serving as the primary mechanism for disseminating new findings, validating scientific claims, and setting research agendas in a discipline. Submissions are evaluated by a programme committee drawn from the research community, with accepted papers published in conference proceedings and delivered through oral presentations, poster sessions, and co-located workshops. In computer science and artificial intelligence specifically, leading venues such as NeurIPS, ICML, CVPR, and ICLR function as the primary records of scientific priority, where peer review, archival publication, and community exchange are tightly coupled rather than mediated solely through journal cycles \u2014 a structural feature that distinguishes the field from most traditional sciences and accelerates the pace at which novel results enter the community.", "entityType": "Class", "qualityScore": 0.87, "maturity": "mature", "iri": "urn:ngm:class:academic-conference", "labels": [ "Academic Conference" ], "is_subclass_of": [ "Scientific Research", "Academia" ], "wikilinks": [ "Scientific Research", "Academia", "Peer Review", "Publication", "NeurIPS", "ICML", "CVPR", "ICLR", "Knowledge Sharing", "Collaboration", "Artificial Intelligence Research", "Deep Learning", "Machine Learning", "Open Science", "Research Dissemination", "Reproducibility", "Programme Committee", "Workshop", "Poster Session", "arXiv" ] }, { "id": "academic-research", "title": "Academic Research", "domain": "governance", "domain_name": "Governance", "definition": "Academic research is the systematic, disciplined inquiry conducted within universities and research institutions to generate, validate, and disseminate new knowledge. It proceeds through formulating questions, reviewing prior literature, designing and executing studies under the scientific method, and subjecting findings to peer review before publication. Its outputs include papers, datasets, and theories that are cited and built upon by the wider scholarly community. Governance norms such as reproducibility, transparency, and research integrity distinguish it from informal or commercial knowledge production.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:academic-research", "labels": [ "Academic Research" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "academy-software-foundation", "title": "Academy Software Foundation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Academy Software Foundation (ASWF) is a neutral, vendor-independent organisation, hosted under the Linux Foundation and founded with the Academy of Motion Picture Arts and Sciences, that develops and stewards open-source software for the motion picture and broader media industries. It provides shared governance, infrastructure and funding for foundational projects in visual effects, animation and digital content creation, such as OpenColorIO, OpenEXR, OpenVDB, OpenTimelineIO and MaterialX. Its goal is sustainable, interoperable open-source tooling across the production pipeline.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:academy-software-foundation", "labels": [ "Academy Software Foundation" ], "is_subclass_of": [ "Standards Organisation" ], "wikilinks": [] }, { "id": "acceleration-structure", "title": "Acceleration Structure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An acceleration structure is a spatial data structure that organises geometric primitives in a scene to allow rapid culling of irrelevant geometry during ray intersection or visibility queries, dramatically reducing the computational complexity of rendering algorithms from O(n) per-ray to approximately O(log n). Common forms include bounding volume hierarchies, k-d trees, and octrees, each making different trade-offs between construction time, memory footprint, and query efficiency.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:acceleration-structure", "labels": [ "Acceleration Structure" ], "is_subclass_of": [ "Spatial Index" ], "wikilinks": [] }, { "id": "accelerometer", "title": "Accelerometer", "domain": "robotics", "domain_name": "Robotics", "definition": "Accelerometer - A microelectromechanical sensor (MEMS) that detects changes in velocity and gravity along three orthogonal axes, enabling robots to measure Motion, Orientation, and Vibration for real-time feedback control and navigation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:accelerometer", "labels": [ "Accelerometer" ], "is_subclass_of": [ "Perception and Sensing", "Robotics" ], "wikilinks": [ "Gesture Recognition", "IEEE 1451", "IEEE 1451.0", "Inertial Measurement Unit", "ISO 16063", "ISO 16063-1:2023", "ISO 8373:2021", "Motion", "Motion Detection", "Orientation", "Signal Conditioning", "Vibration", "Robotics", "RoboticsDomain" ] }, { "id": "access-control-decisions", "title": "Access Control Decisions", "domain": "security", "domain_name": "Security", "definition": "Access Control Decisions are the runtime determinations made by a policy engine or decision point as to whether a particular subject\u2014user, process, or device\u2014is permitted to perform a requested action on a protected resource, based on evaluated policies, attributes, and contextual signals. These decisions are the operational output of an access control architecture and are typically produced by a Policy Decision Point (PDP) and enforced by a Policy Enforcement Point (PEP).", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:access-control-decisions", "labels": [ "Access Control Decisions" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "access-control-module", "title": "Access Control Module", "domain": "security", "domain_name": "Security", "definition": "An Access Control Module is a self-contained software component that enforces authorization rules governing which principals may invoke which functions or read which resources. In smart-contract systems it is commonly implemented as a reusable mixin defining roles, ownership, and permission checks that other contracts inherit. It centralizes permission logic so that privileged operations such as timelocked upgrades or rights revocation are gated behind verifiable on-chain conditions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:access-control-module", "labels": [ "Access Control Module" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "access-control-policy", "title": "Access Control Policy", "domain": "security", "domain_name": "Security", "definition": "An Access Control Policy is a machine-readable specification of the rules that determine which agents may read, write, or append to a given resource. In decentralized web-data systems such as Solid, policies are expressed as RDF documents (e.g. WAC or ACP) attached to resources, granting or denying modes of access to identified WebID principals or groups. The policy is the declarative source of truth that an authorization engine evaluates on each request.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:access-control-policy", "labels": [ "Access Control Policy", "ACP Access Control Policy" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "access-control-system", "title": "Access Control System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Access Control System is the assemblage of policies, decision engines, enforcement points, identity providers, attribute sources, audit pipelines, and cryptographic primitives that determines wher a subject (human user, service account, autonomous agent, device) is permitted to perform a r...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:access-control-system", "labels": [ "Access Control System" ], "is_subclass_of": [ "Security and Identity", "Security Control", "Authorisation Mechanism", "Information Security System", "Governance Control" ], "wikilinks": [ "ABAC", "Active Directory", "Anderson 2020 Security Engineering Third Edition", "ANSI INCITS 359", "ANSI INCITS 359-2012 RBAC Standard", "Attribute Evaluation", "Audit Log", "Auditability", "Auth0", "Authentication", "Authorisation Mechanism", "AWS IAM", "AWS IAM Identity Center", "Azure RBAC", "Backes et al 2018 Zelkova SMT AWS Access Policies", "Bell-LaPadula 1973 Secure Computer Systems Mathematical Foundations", "Bell-LaPadula Model", "Biba 1977 Integrity Considerations", "Biba Model", "BlueDiamond" ] }, { "id": "access-control", "title": "Access Control", "domain": "security", "domain_name": "Security", "definition": "Access control is a fundamental security mechanism that regulates which users, systems, or processes can view, use, or modify resources within a computing environment. It encompasses the policies, procedures, and technologies that govern the granting and restricting of access rights, ensuring that only authorised entities can perform specific actions on protected resources based on their identity, role, or attributes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:access-control", "labels": [ "Access Control", "API Access Control", "Access Control Enforcement", "Access Control Layer", "AccessControl", "Centralised Access Control", "Concurrent Access Control", "Decentralised Access Control", "Dynamic Access Control", "Least-Privilege Access Control", "Metaverse Access Control", "Policy-Based Access Control", "WAC Access Control List" ], "is_subclass_of": [ "Security and Identity", "Security Mechanism" ], "wikilinks": [ "Authorisation", "Compliance", "Core Technology", "Permission Management", "Security Mechanism", "System Security", "Cryptography", "Data Protection" ] }, { "id": "access-controls", "title": "Access Controls", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Access controls are mechanisms that govern which subjects may perform which operations on resources, enforcing authorisation policies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:access-controls", "labels": [ "Access Controls" ], "is_subclass_of": [ "Authorisation" ], "wikilinks": [ "Authentication", "Information Security", "Identity Management", "Authorisation", "https://csrc.nist.gov/glossary/term/access_control", "https://en.wikipedia.org/wiki/Access_control" ] }, { "id": "access-token", "title": "Access Token", "domain": "security", "domain_name": "Security", "definition": "An access token is a credential that a client presents to a resource server to access protected resources within a granted scope and lifetime. Typically short-lived and often a bearer token, it carries or references the authorisation decision so the resource server need not re-check with the authorisation server on every request. Access tokens are the workhorse credential of OAuth 2.0 and API authorisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:access-token", "labels": [ "Access Token" ], "is_subclass_of": [ "OAuth 2.0" ], "wikilinks": [] }, { "id": "accessibility-audit-tool", "title": "Accessibility Audit Tool", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An automated software utility that verifies compliance with accessibility standards (such as WCAG) in XR environments, identifying barriers for users with disabilities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:accessibility-audit-tool", "labels": [ "Accessibility Audit Tool", "Accessibility Testing" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Accessibility Reporting", "Automated Testing Engine", "Inclusive Design", "Quality Assurance Toolchain", "Report Generator", "Testing Framework", "Testing Protocol", "User Interface Analyzer", "W3C XR Accessibility User Requirements", "WCAG Guidelines", "WCAG Validator", "Accessibility Standard", "Accessibility Standards", "Application Layer", "Compliance Dashboard", "Compliance Verification", "InteractionDomain", "Middleware Layer", "Regulatory Compliance", "Telecollaboration" ] }, { "id": "accessibility-captioning", "title": "Accessibility Captioning", "domain": "standards", "domain_name": "Standards", "definition": "Accessibility captioning is the practice of producing synchronised textual representations of spoken dialogue, sound effects, and other audio information within video or live media so that deaf, hard-of-hearing, or audio-impaired audiences can fully engage with the content. It encompasses both closed captions (user-selectable, stored separately from the video stream) and open captions (burned into the image), and spans pre-produced transcription and real-time automatic speech recognition pipelines.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:accessibility-captioning", "labels": [ "Accessibility Captioning", "Captioning Users" ], "is_subclass_of": [ "Inclusive Design" ], "wikilinks": [] }, { "id": "accessibility-standard", "title": "Accessibility Standard", "domain": "standards", "domain_name": "Standards", "definition": "Normative specification ensuring equitable access to digital content, services, and experiences for users with diverse abilities and disabilities, covering visual, auditory, motor, cognitive, and neurological access paths with testable conformance criteria.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:accessibility-standard", "labels": [ "Accessibility Standard" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Accessibility Guideline", "Accessibility Requirement", "Compliance Metric", "Equitable Access", "ETSI GR ARF 010", "ISO 9241-112", "Testing Protocol", "Universal Design", "W3C XR Accessibility", "Governance Framework", "Inclusive XR Experience", "Middleware Layer", "Telecollaboration", "TrustAndGovernanceDomain", "User Interface Standard", "XR Accessibility Guideline" ] }, { "id": "accessibility-standards", "title": "Accessibility Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Accessibility Standards for the metaverse and extended reality (XR) encompass guidelines, technical specifications, and design principles that ensure virtual environments, interfaces, and interactions are usable by people with diverse disabilities including visual, auditory, physical, cognitive, and neurological impairments. Key bodies include the W3C XR Accessibility User Requirements (XRAUR), WCAG 2.2/3.0 adaptations, ADA digital interpretations, and inclusive immersion frameworks covering multi-modal input, hardware ergonomics, and spatial interface design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:accessibility-standards", "labels": [ "Accessibility Standards", "AccessibilityStandards" ], "is_subclass_of": [ "Governance and Safety", "Design Standards" ], "wikilinks": [ "Assistive Technology Integration", "Design Standards", "Inclusive Virtual Experiences", "Multi-Modal Interfaces", "Universal Design", "W3C", "metaverse", "Telecollaboration" ] }, { "id": "accessibility-tree", "title": "Accessibility Tree", "domain": "ai", "domain_name": "Ai", "definition": "The Accessibility Tree is a structured, platform-independent semantic representation of a user interface that browsers and native application runtimes construct in parallel with the visual render tree, exposing each element's role, name, description, state, and value to platform accessibility APIs and programmatic automation clients. Derived from the Document Object Model by filtering out presentational and layout-only nodes, it is the authoritative conduit through which screen readers, braille displays, switch-access devices, voice-control systems, and AI browser agents perceive and interact with software interfaces without parsing raw pixels.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:accessibility-tree", "labels": [ "Accessibility Tree" ], "is_subclass_of": [ "Human Computer Interaction", "User Interface", "Inclusive Design", "User Interface Model", "Semantic Representation" ], "wikilinks": [ "Accessibility", "Accessible Name Computation", "Alternative Text", "ARIA Role", "ARIA State", "Assistive Technology", "Assistive Technology Compatibility", "axe-core", "Browser Automation", "Chrome DevTools Protocol", "Computer Use and Browser Agents", "ContentLayer", "Core Accessibility API Mappings", "Document Object Model", "Focus Management", "Human Computer Interaction", "HTML Accessibility API Mappings", "Inclusive Design", "JAWS Screen Reader", "Keyboard Navigation" ] }, { "id": "accessibility", "title": "Accessibility", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Accessibility is the systematic property of digital products, services, environments and information systems being perceivable, operable, understandable and robust (POUR) for the widest possible range of users including those with permanent disabilities temporary impairments (broken arm, post-sur...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:accessibility", "labels": [ "Accessibility", "Accessibility Auditing", "Accessibility Enhancement", "Accessibility Provisioning", "Accessibility Technology", "Digital Accessibility", "Global Accessibility" ], "is_subclass_of": [ "Governance and Safety", "Quality Attribute", "Design Principle", "Human Factors", "Inclusive Design", "Non-Functional Requirement" ], "wikilinks": [ "Accessibility Statement", "Accessible Name Computation", "Alternative Text", "Assistive Technology Compatibility", "Assistive Technology Ecosystem", "ATAG 2.0", "Audio Description", "axe-core", "Bigham 2010 VizWiz", "BS 8878", "Captioning Users", "Captions", "Cognitively Diverse Users", "Colour Contrast", "ContentLayer", "DesignDomain", "Design Principle", "Disability Inclusion", "Disability Rights", "Disability Studies" ] }, { "id": "accessible-design", "title": "Accessible Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Accessible Design is the practice of creating virtual environments and interactions usable by people with diverse disabilities, incorporating universal design principles from conception through implementation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:accessible-design", "labels": [ "Accessible Design" ], "is_subclass_of": [ "Governance and Safety", "Universal Design", "Inclusive Design" ], "wikilinks": [ "Assistive Technology Integration", "Broader User Base", "Inclusive Design", "Inclusive Participation", "Universal Design", "User Research with Disabled Participants", "3D User Interface", "Accessibility Standard", "Accessible Experience", "Metaverse", "Regulatory Compliance", "Telecollaboration" ] }, { "id": "accessible-experience", "title": "Accessible Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Accessible Experience is a virtual environment or application delivering equivalent functionality and engagement to users with disabilities through multimodal access pathways, assistive technology integration, and Accessible Design principles, ensuring feature parity across visual, auditory, motor, and cognitive modalities.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:accessible-experience", "labels": [ "Accessible Experience" ], "is_subclass_of": [ "Governance and Safety", "Interaction Technology", "User Experience", "Inclusive Experience" ], "wikilinks": [ "Alternative Input Methods", "Assistive Technology", "Assistive Technology Support", "Equitable Engagement", "Inclusive Experience", "Inclusive Participation", "Multimodal Interface", "Multimodal Interfaces", "Universal Access", "User Experience", "3D User Interface", "Accessibility Standard", "Accessible Design", "Metaverse", "Telecollaboration" ] }, { "id": "account-abstraction", "title": "Account Abstraction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A paradigm shift in ereum account architecture standardized through ERC-4337 (March 2023) enabling smart contract wallets with programmable transaction validation logic, decoupled gas payment mechanisms, and flexible account recovery, eliminating the distinction between externally owned accounts ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:account-abstraction", "labels": [ "Account Abstraction", "account-abstraction" ], "is_subclass_of": [ "Protocol and Consensus", "Smart Contract" ], "wikilinks": [ "Account Factory", "Account Recovery", "Alchemy Account Kit", "Argent Wallet Architecture", "Batch Operations", "Biconomy SDK", "Bundler", "Bundler Network", "Bundler Network Statistics", "EntryPoint Contract", "EntryPoint Singleton", "ERC-4337", "ERC-4337 Ethereum Improvement Proposal March 2023", "Ethereum Foundation Account Abstraction Documentation", "EVM", "Gas Abstraction", "Gas Sponsorship", "Gasless Transactions", "Paymaster Contract", "Paymaster Infrastructure" ] }, { "id": "account-model", "title": "Account Model", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Account Model is a balance-based ledger paradigm in which each address maintains a persistent balance that is updated in-place when transactions execute. Unlike the UTXO model, accounts hold state across transactions, simplifying smart contract programming while introducing challenges around replay protection and nonce management.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:account-model", "labels": [ "Account Model", "Account-Based Model" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "accountability-ai-0068", "title": "Accountability (AI-0068)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Accountability (AI-0068) is the principle \u2014 codified under ontology reference identifier AI-0068 in structured AI ethics and governance frameworks \u2014 requiring that organisations developing, deploying, or operating AI systems must be able to demonstrate traceable, enforceable responsibility for the systems' decisions and impacts, provide redress mechanisms when harm occurs, maintain identifiable human oversight over automated processes, and document the entire lifecycle of AI systems through auditable artefacts. It grounds the ethical aspiration of accountability in specific technical and organisational obligations, distinguishing it from aspirational guidelines by imposing concrete requirements for AI audit trails, explainability outputs, model documentation, incident response procedures, and designated role assignments for liability.", "entityType": "Class", "qualityScore": 0.93, "maturity": "emerging", "iri": "urn:ngm:class:accountability-ai-0068", "labels": [ "Accountability (AI-0068)" ], "is_subclass_of": [ "Accountability", "AI Governance and Ethics", "Algorithmic Accountability", "AI Governance", "Responsible AI", "AI Ethics", "Governance Framework" ], "wikilinks": [ "Accountability", "AI Audit", "AI Documentation", "AI Ethics", "AI Governance", "AI Governance Framework", "AI Impact Assessment", "AI Monitoring", "AI Regulation", "AI Risk Management", "Algorithmic Accountability", "Algorithmic Bias", "Bias Mitigation", "Conformity Assessment", "Contestability", "Data Governance", "EU AI Act", "Explainable AI", "Fairness (AI)", "GDPR" ] }, { "id": "accountability-oecd", "title": "Accountability (OECD)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Organisations and individuals developing, deploying or operating AI systems should be accountable for their proper functioning in accordance with OECD AI Principles and applicable legal frameworks, based on their roles, context and ability to act.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:accountability-oecd", "labels": [ "Accountability (OECD)" ], "is_subclass_of": [ "AI Governance and Ethics", "Policy Framework" ], "wikilinks": [ "continuous improvement", "Effective governance", "IEEE 7000", "ISO 31000:2018", "ISO 37301:2021", "ISO 42001", "ISO/IEC 27001:2022", "ISO/IEC 42001:2023", "OECD", "redress", "AI Agent System", "EU AI Act", "MetaverseDomain" ] }, { "id": "accountability-framework", "title": "Accountability Framework", "domain": "governance", "domain_name": "Governance", "definition": "An accountability framework is a structured set of principles, roles, processes, and controls that together define how responsibility for outcomes is assigned, tracked, and enforced within an organisation or system. It specifies who is answerable for decisions and actions, what evidence of responsible conduct must be maintained, and what redress or corrective measures apply when obligations are unmet. Accountability frameworks appear across governance, AI ethics, financial regulation, and public sector administration.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:accountability-framework", "labels": [ "Accountability Framework" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "accountability-mechanism", "title": "Accountability Mechanism", "domain": "governance", "domain_name": "Governance", "definition": "An accountability mechanism is a specific procedural, technical, or institutional instrument through which an actor can be held answerable for their decisions and conduct\u2014for example, an audit trail, an ombudsman process, an algorithmic impact assessment, or a public reporting obligation. Whereas an accountability framework defines the overall structure of responsibility, an accountability mechanism is the concrete tool that makes accountability operational within or across that framework.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:accountability-mechanism", "labels": [ "Accountability Mechanism", "Accountability Mechanisms" ], "is_subclass_of": [ "Accountability" ], "wikilinks": [] }, { "id": "accountability", "title": "Accountability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The assignment of clear responsibilities for AI system development, deployment, and outcomes, coupled with mechanisms for oversight, redress, and remediation, ensuring that actors can be held answerable for system impacts and failures.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:accountability", "labels": [ "Accountability", "Accountability Principle", "Continuous Accountability", "Public Accountability" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [ "Compliance", "Explainability (AI-0063)", "GDPR", "ISO/IEC 23894:2023", "ISO/IEC 38500:2024", "NIST AI RMF", "OECD", "Risk Management (AI-0078)", "Transparency (AI-0062)", "Trust", "Blockchain", "EU AI Act", "MetaverseDomain" ] }, { "id": "accountable-party", "title": "Accountable Party", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Accountable Party is an individual, organisation, or role that bears defined responsibility for specific aspects of an AI system's development, deployment, operation, or outcomes, with corresponding obligations to ensure compliance with governance principles, regulatory requirements, and ethical standards. Accountability addresses the foundational question of who bears responsibility when AI systems cause harm or produce unfair outcomes, requiring that identifiable parties possess the authority, resources, and monitoring capacity commensurate with their obligations. Accountable parties span the AI value chain\u2014from data providers and model developers to deploying organisations, system operators, and governance bodies\u2014each with lifecycle-specific duties enforced through formal reporting structures and consequence mechanisms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:accountable-party", "labels": [ "Accountable Party" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Governance Principle" ], "wikilinks": [ "AI Governance Principle", "Algorithmic Accountability Principles", "Algorithmic Auditor", "Chief AI Officer", "Data Provider", "Deploying Organization", "Governance Body", "IEEE 7000 Model Process", "ISO/IEC 42001", "Model Developer", "OECD AI Principles", "System Operator", "AI Ethics Board", "AIEthicsDomain", "ConceptualLayer", "EU AI Act", "Smart Contract" ] }, { "id": "accredited-investor-verification", "title": "Accredited Investor Verification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Accredited investor verification is the process of confirming that a prospective investor meets the income, net-worth, or professional-knowledge thresholds that securities regulation requires for participation in unregistered or exempt securities offerings, such as many security token issuances. It typically involves documentary evidence or third-party attestation rather than self-certification alone. It is a compliance gate distinct from general know-your-customer identity checks, since it verifies financial eligibility rather than identity.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:accredited-investor-verification", "labels": [ "Accredited Investor Verification" ], "is_subclass_of": [ "Securities Regulation" ], "wikilinks": [] }, { "id": "accretion-disc", "title": "Accretion Disc", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:accretion-disc", "labels": [ "Accretion Disc" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "accumulation-zone", "title": "Accumulation Zone", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:accumulation-zone", "labels": [ "Accumulation Zone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "accuracy", "title": "Accuracy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A classification performance metric representing the proportion of correct predictions made by an artificial intelligence model across all instances in a dataset, calculated as the ratio of the sum of true positives and true negatives to the total number of predictions, providing an aggregate measure of overall model correctness but potentially obscuring performance disparities across classes, particularly in datasets with imbalanced class distributions or asymmetric misclassification costs.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:accuracy", "labels": [ "Accuracy", "Measurement Accuracy" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Balanced Accuracy", "documentation", "ISO/IEC 25024", "ISO/IEC 25059", "modeling", "NIST AI RMF", "organisation", "performance", "research", "Sensitivity", "Specificity", "user experience", "Computer Vision", "Confusion Matrix", "F1 Score", "machine learning", "MetaverseDomain", "Model Performance", "Precision", "Product Design" ] }, { "id": "acid-transactions", "title": "Acid Transactions", "domain": "data", "domain_name": "Data", "definition": "ACID transactions are units of database work that uphold four guarantees: atomicity, consistency, isolation and durability. Atomicity ensures a transaction either fully completes or has no effect; consistency keeps the database in a valid state; isolation prevents concurrent transactions from interfering; and durability guarantees committed changes survive failures. Together these properties allow reliable, predictable updates to shared data and form the bedrock of relational database systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:acid-transactions", "labels": [ "Acid Transactions", "ACID Transactions" ], "is_subclass_of": [ "Database Management System" ], "wikilinks": [] }, { "id": "acoustic-model", "title": "Acoustic Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Acoustic Model is the component of a speech-recognition system that maps audio signal features to the probability of phonetic or sub-word units. It learns the statistical relationship between observed acoustic features and the linguistic sounds that produced them, traditionally using hidden Markov models with Gaussian mixtures and increasingly using deep neural networks. Combined with a language model, it converts spoken audio into the most likely sequence of words.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:acoustic-model", "labels": [ "Acoustic Model" ], "is_subclass_of": [ "Speech Recognition", "Sequence Model" ], "wikilinks": [ "Speech Recognition", "Feature Extraction", "Signal Processing", "Neural Network", "Deep Learning", "Pattern Recognition", "Automatic Speech Recognition", "Machine Learning", "Speech Processing", "Language Model", "Audio Processing", "Natural Language Processing", "Hidden Markov Model", "Transformer", "Attention Mechanism", "Convolutional Neural Network", "Recurrent Neural Network", "Connectionist Temporal Classification", "Speaker Diarisation", "Transfer Learning" ] }, { "id": "acquisition-function", "title": "Acquisition Function", "domain": "ai", "domain_name": "Ai", "definition": "An acquisition function is a utility function used in Bayesian optimisation that determines which point in the input space to evaluate next by balancing exploration of uncertain regions against exploitation of known promising areas. It transforms the surrogate model's posterior distribution into a scalar score, guiding the sequential selection of experiments or evaluations. Common forms include Expected Improvement, Upper Confidence Bound, and Probability of Improvement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:acquisition-function", "labels": [ "Acquisition Function" ], "is_subclass_of": [ "Optimization Algorithm", "Black-Box Optimisation" ], "wikilinks": [ "Bayesian Optimisation", "Probabilistic Model", "Bayesian Inference", "Hyperparameter Tuning", "Neural Architecture Search", "Hyperparameter", "Machine Learning Discipline", "Gaussian Process", "Surrogate Model", "Optimization Algorithm", "Expected Improvement", "Exploration-Exploitation Trade-off", "Reinforcement Learning", "Active Learning", "Kernel Function", "Drug Discovery", "Materials Science", "AutoML", "Multi-Objective Optimisation", "Deep Learning" ] }, { "id": "across-protocol", "title": "Across Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Across Protocol is a decentralised cross-chain bridging protocol that uses an optimistic verification model and a network of liquidity providers\u2014called relayers\u2014to enable fast, capital-efficient transfers of ERC-20 tokens between Ethereum mainnet and Layer 2 networks such as Optimism, Arbitrum, and Polygon. Relayers front user funds immediately on the destination chain and are reimbursed from a liquidity pool on Ethereum after an optimistic challenge period, creating a bridge architecture that prioritises speed and low fees over trustless finality.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:across-protocol", "labels": [ "Across Protocol", "Across-Protocol" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "action-executor", "title": "Action Executor", "domain": "ai", "domain_name": "Ai", "definition": "An Action Executor is a software component within an agentic AI system responsible for translating high-level instructions or plans into concrete, observable operations in an environment. It serves as the effector layer that bridges planning and execution by dispatching tool calls, API requests, file operations, or process invocations. Action Executors typically implement sandboxing, retry logic, and side-effect isolation to ensure safe and predictable operation. They are central to autonomous agent architectures where multiple sequential or parallel actions must be managed reliably.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:action-executor", "labels": [ "Action Executor" ], "is_subclass_of": [ "Agentic AI", "AI Agent System", "Autonomous Agent" ], "wikilinks": [ "Agentic AI", "Tool Use", "Function Calling", "Orchestration", "Orchestration Layer", "Agent Orchestrator", "Runtime Environment", "Agentic Workflow", "Automated Planning", "Task Planning", "Agent Runtime", "Agent Frameworks", "ReAct Pattern", "Multi-Agent Systems", "Multi-Agent Orchestration", "Observability", "Sandbox Environment", "Security Architecture", "Error Handling", "Behaviour Tree" ] }, { "id": "action-recognition", "title": "Action Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Action Recognition is a computer-vision task that classifies the activity being performed by one or more agents from video or motion-sequence data. Models ingest spatiotemporal features, often built on detected body keypoints or 3D convolutions and transformers, to label actions such as walking, waving, or falling. It underpins applications in surveillance, sports analytics, human-robot interaction, and assistive monitoring.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:action-recognition", "labels": [ "Action Recognition", "Action Detection" ], "is_subclass_of": [ "Computer Vision", "Video Understanding", "Computer Vision Task" ], "wikilinks": [ "Computer Vision", "Pose Estimation", "Deep Learning", "Convolutional Neural Network", "Graph Neural Network", "Transformer Architecture", "Optical Flow", "Video Understanding", "Gesture Recognition", "Human Capture & Recognition", "Human Robot Interaction", "Sports Analytics", "Surveillance", "Robotics Perception", "Motion Capture", "Motion Tracking", "Multimodal AI", "Multimodal Learning", "Self-Supervised Learning", "Foundation Model" ] }, { "id": "action-space", "title": "Action Space", "domain": "ai", "domain_name": "Ai", "definition": "The action space in reinforcement learning and control theory is the complete set of actions available to an agent at any given decision step, defining the boundaries of what the agent may do when interacting with its environment. It may be discrete\u2014a finite enumeration of choices\u2014or continuous\u2014a real-valued manifold such as joint torques or steering angles\u2014and its structure fundamentally determines which learning algorithms are applicable and how efficiently a policy can be discovered.", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:action-space", "labels": [ "Action Space" ], "is_subclass_of": [ "State Space", "Markov Decision Process", "Control Theory", "Optimisation" ], "wikilinks": [ "State Space", "Reinforcement Learning", "Markov Decision Process", "Reward Function", "Policy Optimisation", "Reinforcement Learning Algorithm", "Environment Model", "Reinforcement Learning for Robotics", "Deep Reinforcement Learning", "Policy Gradient Methods", "Proximal Policy Optimisation", "Value Function", "Neural Network", "Multi-Agent Reinforcement Learning", "Simulation Environment", "Autonomous Robot", "Game AI", "Combinatorial Optimisation", "Exploration-Exploitation", "Deep Learning" ] }, { "id": "activation-function", "title": "Activation Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An Activation Function is a non-linear mathematical transformation applied to a neuron's weighted input sum, enabling neural networks to learn complex, non-linear mappings. Common variants include Sigmoid, Tanh, ReLU, Leaky ReLU, and GELU; the choice of activation function critically affects gradient flow, convergence speed, and model expressivity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:activation-function", "labels": [ "Activation Function" ], "is_subclass_of": [ "Neural Network Component" ], "wikilinks": [ "ISO/IEC 22989:2022", "NIST AI RMF", "ArtificialIntelligenceDomain", "Edge Computing" ] }, { "id": "active-earth-observation-sensor", "title": "Active Earth Observation Sensor", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:active-earth-observation-sensor", "labels": [ "Active Earth Observation Sensor" ], "is_subclass_of": [ "Earth Observation Sensor" ], "wikilinks": [] }, { "id": "active-galactic-nucleus", "title": "Active Galactic Nucleus", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:active-galactic-nucleus", "labels": [ "Active Galactic Nucleus" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "active-learning", "title": "Active Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine learning paradigm where algorithms actively select which unlabeled examples from large data pools to query for human annotation rather than passively accepting randomly labeled datasets, optimizing informativeness through query strategies (uncertainty sampling selecting least-confident pr...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:active-learning", "labels": [ "Active Learning" ], "is_subclass_of": [ "AI Technique", "Data Efficient Learning", "Interactive Learning", "Machine Learning Discipline", "Supervised Learning", "Human-in-the-Loop Learning" ], "wikilinks": [ "Acquisition Function", "AlgorithmLayer", "Ash et al. 2020 Deep Batch Active Learning", "Base Learner", "Cohn et al. 1996 Improving Generalization with Active Learning", "Cost-Effective Annotation", "Dasgupta 2011 Two Faces of Active Learning", "Density-Weighted Methods", "Diversity Measure", "Evaluation Metric", "Expected Error Reduction", "Expected Model Change", "Expert Knowledge Elicitation", "Freund et al. 1997 Selective Sampling", "Gal et al. 2017 Deep Bayesian Active Learning", "Google Research Active Learning", "Haussler et al. 1994 Decision Theoretic Generalization PAC", "Human Oracle", "ICML Interactive Learning Track", "Information Theory" ] }, { "id": "active-r-d-funding-opportunities-register", "title": "Active R&D Funding Opportunities Register", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated register of active and accessible funding opportunities for immersive technology, AI, and creative industries R&D, encompassing Innovate UK grants, UKRI programmes, ARIA challenges, Web3 venture capital, and academic grant schemes. The register tracks application status, eligibility constraints, match-funding requirements, and strategic alignment to accelerate project funding decisions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:active-r-d-funding-opportunities-register", "labels": [ "Active R&D Funding Opportunities Register", "Funding (active and available)" ], "is_subclass_of": [ "Software Engineering", "Resource Management" ], "wikilinks": [ "MUST" ] }, { "id": "active-remote-sensing", "title": "Active Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:active-remote-sensing", "labels": [ "Active Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "active-research-projects-registry", "title": "Active Research Projects Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Active Research Projects Registry is a structured, queryable catalogue that maintains authoritative records of ongoing research initiatives, development workstreams, and collaborative projects within an organisation or knowledge graph, capturing metadata such as status, ownership, objectives, dependencies, and timelines. It serves as a single source of truth for portfolio governance, enabling stakeholders to discover, track, and coordinate active work without duplicating effort or losing context across distributed teams. Unlike a static document index, a living registry integrates with project management workflows, version control systems, and knowledge management platforms to reflect real-time project state. It supports prioritisation, resource allocation, and strategic alignment by surfacing the relationships between concurrent initiatives and shared infrastructure.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:active-research-projects-registry", "labels": [ "Active Research Projects Registry", "Projects" ], "is_subclass_of": [ "Data Registry" ], "wikilinks": [] }, { "id": "active-thermal-control", "title": "Active Thermal Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:active-thermal-control", "labels": [ "Active Thermal Control" ], "is_subclass_of": [ "Spacecraft Thermal Control" ], "wikilinks": [] }, { "id": "activity-data", "title": "Activity Data", "domain": "ai", "domain_name": "Ai", "definition": "Activity data is the structured or semi-structured record of actions, interactions, and events performed by users, devices, or systems over time, capturing what was done, when, by whom, and in what context. It encompasses clickstreams, application usage logs, sensor readings, transaction histories, and learning interaction records, and serves as the raw material for analytics, personalisation, compliance auditing, and behavioural modelling.", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:activity-data", "labels": [ "Activity Data" ], "is_subclass_of": [ "Data Collection", "Time-Series Analysis", "Event Sourcing", "Behavioural Analytics" ], "wikilinks": [ "Data Collection", "Tracking System", "Telemetry & Analytics", "Analytics Engine", "Audit Log", "Compliance", "Transparency", "Reinforcement Learning from Human Feedback", "Data Pipeline", "Stream Processing", "Data Lake", "Machine Learning Pipeline", "Recommendation System", "Personalisation", "Privacy", "General Data Protection Regulation", "Data Governance", "Event-Driven Architecture", "Behavioural Analytics", "Agentic AI" ] }, { "id": "activity-pub", "title": "ActivityPub", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "ActivityPub is a W3C-standardised decentralised social networking protocol that defines two layers: a server-to-server federation protocol enabling independent server instances to share content with one another, and a client-to-server protocol allowing applications to interact with a user's social data. Based on the ActivityStreams 2.0 vocabulary and JSON-LD serialisation, it enables interoperable federated social networks where users on different server instances can follow, reply to, and interact with each other across instance boundaries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:activity-pub", "labels": [ "ActivityPub" ], "is_subclass_of": [ "Distributed Protocol" ], "wikilinks": [] }, { "id": "actor-model", "title": "Actor Model", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The actor model is a mathematical model of concurrent computation in which the universal primitive is the actor, an independent entity that has private state and communicates only by sending asynchronous messages. In response to a message an actor can update its state, send messages to other actors and create new actors. Because actors share nothing and process one message at a time, the model avoids shared-memory data races and provides a foundation for scalable, fault-tolerant distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:actor-model", "labels": [ "Actor Model" ], "is_subclass_of": [ "Concurrency", "Distributed Computing" ], "wikilinks": [] }, { "id": "actuator", "title": "actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "An actuator is a transducer that converts a stored or supplied energy form \u2014 electrical, hydraulic, pneumatic, or thermochemical \u2014 into controlled mechanical motion or force, functioning as the output effector in any closed-loop control chain. Actuators execute commands issued by a controller by producing joint rotations, linear displacements, gripping forces, or compliant deformations, and their dynamic properties (torque density, bandwidth, backdrivability, stiffness) fundamentally bound a system's achievable speed, precision, payload capacity, and intrinsic safety during physical interaction. The choice of actuation technology cascades through every level of system design, from mechanical linkage geometry and energy storage requirements to real-time control law selection and safety-rated force limitation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:actuator", "labels": [ "Actuator", "Actuator Motors", "Actuator Output", "Compliant Actuator", "Proprioceptive Actuator", "Rigid Actuator" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "actuators", "title": "Actuators", "domain": "robotics", "domain_name": "Robotics", "definition": "Actuators are physical devices that convert an electrical, hydraulic, or pneumatic control signal into mechanical motion or force, enabling a control system to produce a desired physical effect in the world. They are the output components of robotic and cyber-physical systems, spanning technologies including servo motors, pneumatic cylinders, hydraulic actuators, and piezoelectric elements, each offering different trade-offs in force output, speed, precision, and energy efficiency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:actuators", "labels": [ "Actuators", "High-Performance Actuators", "Output Actuators", "Prismatic Actuators" ], "is_subclass_of": [ "Robot Component" ], "wikilinks": [] }, { "id": "adam-optimiser", "title": "adam optimiser", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Adam optimiser (Adaptive Moment Estimation) is a first-order gradient-based optimisation algorithm that computes per-parameter adaptive learning rates by maintaining exponentially decaying moving averages of past gradients (first moment) and past squared gradients (second moment), with bias correction applied in early iterations. It synthesises the momentum-tracking behaviour of gradient descent with momentum and the per-parameter scaling of RMSProp, making it robust to sparse gradients, non-stationary objectives, and high-dimensional parameter spaces. Introduced by Kingma and Ba (2014), Adam has become the de-facto default optimiser for training transformer-based large language models, diffusion models, and deep neural networks across most domains. Key variants including AdamW, Adan, and AdaFactor extend it with decoupled weight decay, Nesterov momentum, and memory-efficient factorisation respectively.", "entityType": "Class", "qualityScore": 0.91, "maturity": "mature", "iri": "urn:ngm:class:adam-optimiser", "labels": [ "Adam Optimiser", "AdaM", "Adam Optimizer", "AdamW", "AdamW Optimiser", "Adaptive Moment Estimation" ], "is_subclass_of": [ "Gradient Descent", "Optimisation Algorithm" ], "wikilinks": [ "Gradient Descent", "Stochastic Gradient Descent", "Backpropagation", "Loss Function", "Deep Learning", "Transformer Architecture", "Neural Network", "RMSProp", "Adagrad", "AdamW", "AdaFactor", "Weight Decay", "Gradient Clipping", "Reinforcement Learning", "Federated Learning", "Transfer Learning", "Fine-Tuning", "Model Training", "Hyperparameter Optimisation", "Learning Rate Schedule" ] }, { "id": "adapter-modules", "title": "Adapter Modules", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Adapter Modules are compact bottleneck neural network sub-networks inserted between frozen transformer layers, trained exclusively on task-specific data whilst leaving the base model unchanged. They typically comprise a down-projection, a non-linearity, and an up-projection with a residual connection, constituting under 1% of model parameters and enabling efficient multi-task deployment from a single frozen base model.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:adapter-modules", "labels": [ "Adapter Modules", "Adapter Layers", "Adapter Module", "T2I-Adapter" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "adapter-slot", "title": "Adapter Slot", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A pluggable interface architecture (ADR-005) providing five standardised slots through which a VisionClaw Agentic Container|VisionClaw agent interacts with its environment: Solid Pod Storage|Pod (persistent storage), Agent Memory|Memory (episodic and semantic),", "entityType": "Class", "qualityScore": 0.89, "maturity": "established", "iri": "urn:ngm:class:adapter-slot", "labels": [ "Adapter Slot" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "AdapterLayer", "Adapter Pattern", "ADR-005", "Agent Bead", "Agent Bead", "Agent Memory", "AgenticSystemsDomain", "ArchitectureDomain", "Bead Adapter", "Containerisation", "Dependency Injection", "Dependency Injection Principles", "Deployment Flexibility", "Design Patterns - Gang of Four", "Event Adapter", "Interface Specification", "Manifest Declaration", "Memory Adapter", "Nostr Relay", "Nostr Relay" ] }, { "id": "adapter-tuning", "title": "Adapter Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A parameter-efficient fine-tuning technique that adapts a frozen pre-trained neural network to new tasks by inserting small trainable modules \u2014 adapters \u2014 between or alongside its layers, typically bottleneck feed-forward blocks or low-rank projections, so that task-specific behaviour is learned in a fraction of a percent of the original parameter count; adapters preserve the base model's weights, allow many tasks to share one backbone through swappable modules, and underpin methods from Houlsby adapters to LoRA and ControlNet-style conditioning branches.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:adapter-tuning", "labels": [ "Adapter Tuning", "Adapter Fine-tuning" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "Parameter-Efficient Fine-Tuning", "Fine Tuning", "LoRA DoRA etc" ] }, { "id": "adaptive-behaviour", "title": "Adaptive Behaviour", "domain": "ai", "domain_name": "Ai", "definition": "Adaptive behaviour refers to the capacity of an agent, system, or organism to modify its actions, strategies, or responses in real time based on feedback from its environment, internal state, or prior experience. In computational and robotics contexts, it describes dynamic adjustment of policies or control laws to optimise performance under changing conditions. The concept underpins reinforcement learning, autonomous agents, and self-organising systems.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:adaptive-behaviour", "labels": [ "Adaptive Behaviour", "Adaptive Behavior", "Adaptive Swarm Behaviour" ], "is_subclass_of": [ "Agent", "AI Research Area", "Reinforcement Learning", "Self-Organising Systems", "Cybernetics" ], "wikilinks": [ "Agent", "Deep Reinforcement Learning", "Policy Optimisation", "Adaptive Control", "Embodied AI", "Multi-Agent Systems", "Agent-Based Models", "Reinforcement Learning", "Reward Signal", "Markov Decision Process", "Policy Gradient Methods", "Q-Learning", "Temporal Difference Learning", "Neural Network", "Sim-to-Real Transfer", "Autonomous Navigation", "Robotic Manipulation", "Meta-Learning", "Transfer Learning", "Continual Learning" ] }, { "id": "adaptive-bitrate-streaming", "title": "Adaptive Bitrate Streaming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Adaptive bitrate streaming (ABR) is a media delivery technique that encodes content at multiple quality levels and dynamically switches between them in response to measured network throughput and client buffer state. The player requests short segments at the highest sustainable bitrate, smoothing playback over variable connections. It underpins modern over-the-top video via protocols such as HLS and MPEG-DASH delivered over HTTP and content delivery networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:adaptive-bitrate-streaming", "labels": [ "Adaptive Bitrate Streaming" ], "is_subclass_of": [ "Video Streaming" ], "wikilinks": [] }, { "id": "adaptive-capacity-index", "title": "Adaptive Capacity Index", "domain": "economics", "domain_name": "Economics", "definition": "A metric used to assess an individual's ability to adapt to labor market disruptions, considering factors like financial resources, age, and skill transferability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:adaptive-capacity-index", "labels": [ "Adaptive Capacity Index" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "adaptive-coding-and-modulation", "title": "Adaptive Coding and Modulation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:adaptive-coding-and-modulation", "labels": [ "Adaptive Coding and Modulation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "adaptive-control", "title": "Adaptive Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A control strategy that automatically adjusts its parameters in real-time to maintain desired performance as system dynamics change or uncertainties are encountered. The controller learns and adapts to variations in the system or environment, using mechanisms such as system identification, parameter estimation, and online learning to compensate for model uncertainty and external disturbances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:adaptive-control", "labels": [ "Adaptive Control", "Adaptive Grasping", "AdaptiveControl", "RB-1004-adaptive-control" ], "is_subclass_of": [ "Closed-Loop Control", "RB-1002-closed-loop-control" ], "wikilinks": [ "Adaptive Systems", "Autonomous Systems", "Learning", "Parameter Adaptation Mechanism", "Parameter Estimation", "RB-1002-closed-loop-control", "RB-1003-optimal-control", "RB-1011-cobot-safety-levels", "Self-Tuning", "Self-Tuning Systems", "System Identification", "AI Agent System", "Control Theory", "Machine Learning", "Robotics", "Robust Control", "Robustness" ] }, { "id": "adaptive-interfaces", "title": "Adaptive Interfaces", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Adaptive interfaces are user interface systems that dynamically reconfigure their layout, content, modality, or interaction style in response to the current user's context, behaviour, preferences, or inferred cognitive state. They employ user modelling, machine learning, and context-aware computing to personalise the interaction layer without requiring manual configuration. The goal is to optimise usability, accessibility, and task performance across diverse user populations and situational conditions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:adaptive-interfaces", "labels": [ "Adaptive Interfaces", "Adaptive Information Display", "AdaptiveInterfaces" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "adaptive-learning-rate", "title": "Adaptive Learning Rate", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of optimisation techniques in which the step size of gradient descent is adjusted automatically during training \u2014 typically per parameter, from running statistics of past gradients \u2014 so that parameters with large or frequent gradients take smaller steps and rarely-updated parameters take larger ones, as implemented by AdaGrad, RMSProp, and Adam, reducing sensitivity to the manually chosen global learning rate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:adaptive-learning-rate", "labels": [ "Adaptive Learning Rate" ], "is_subclass_of": [ "Learning Rate" ], "wikilinks": [ "Learning Rate", "Adam Optimiser", "RMSProp", "Gradient Descent" ] }, { "id": "adaptive-learning", "title": "Adaptive Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Adaptive learning is an educational methodology and technology paradigm in which instructional content, pacing, and assessment are dynamically tailored to each learner's demonstrated knowledge, learning style, and progress in real time. Computational systems analyse performance data to identify gaps and misconceptions, then serve personalised learning paths that optimise for mastery and engagement. The approach draws on psychometric theory, knowledge tracing algorithms, and machine learning to individualise instruction at scale.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:adaptive-learning", "labels": [ "Adaptive Learning", "Adaptive Learning Systems" ], "is_subclass_of": [ "Education Technology", "Personalised Learning", "Machine Learning Discipline" ], "wikilinks": [ "Intelligent Tutoring System", "Learning Management System", "Bayesian Knowledge Tracing", "Item Response Theory", "Spaced Repetition", "Personalised Learning", "Learning Analytics", "Mastery Learning", "Formative Assessment", "Knowledge Graph", "Real-Time Data Processing", "Reinforcement Learning", "Content Repository", "Differentiated Instruction", "Curriculum Learning", "Active Learning", "Cognitive Load Theory", "Large Language Model", "Natural Language Processing", "Education Technology" ] }, { "id": "adaptive-manipulation", "title": "Adaptive Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Adaptive manipulation is a subfield of robotics concerned with enabling robotic systems to grasp, move, and interact with objects in unstructured environments by continuously adapting grasp strategies, force application, and motion trajectories based on sensory feedback. It combines perception, planning, and control to handle variability in object geometry, material properties, and placement that defeats fixed pre-programmed approaches. Applications span industrial automation, surgical robotics, and service robots operating in human-centred spaces.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:adaptive-manipulation", "labels": [ "Adaptive Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "adaptive-music", "title": "Adaptive Music", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Adaptive music is a compositional and audio-engineering approach in which a musical soundtrack modifies its structure, instrumentation, tempo, or emotional register dynamically in response to real-time contextual signals such as player actions, narrative state, or environmental parameters within interactive media. It replaces the static looping of pre-composed tracks with a system that maintains musical coherence while reflecting the moment-to-moment state of an interactive experience. The technique is foundational to game audio design and is increasingly applied in extended reality, therapeutic, and generative AI contexts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:adaptive-music", "labels": [ "Adaptive Music" ], "is_subclass_of": [ "Audio System" ], "wikilinks": [] }, { "id": "adaptive-user-interface", "title": "Adaptive User Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An adaptive user interface is an interface that automatically changes its layout, content, or interaction style in response to context such as user behaviour, device, or measured emotional or cognitive state, rather than presenting a single fixed design to all users. It is a subject of human-computer interaction research and is enabled in emerging systems by emotional analytics engines that infer user state in real time. The goal is to reduce cognitive load and improve task efficiency by matching interface complexity to the user's current needs.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:adaptive-user-interface", "labels": [ "Adaptive User Interface" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "adaptive-virtual-experience", "title": "Adaptive Virtual Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Adaptive Virtual Experience refers to AI-driven immersive environments that dynamically adjust content, difficulty, pacing, and presentation in real-time based on user behavior, preferences, physiological responses, and interaction patterns to deliver personalized and engaging virtual reality exp...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:adaptive-virtual-experience", "labels": [ "Adaptive Virtual Experience", "Adaptive Experience" ], "is_subclass_of": [ "Platform and Environment", "Virtual Reality Experience" ], "wikilinks": [ "Dynamic Gaming", "Personalized Learning", "Real-Time Analytics", "User Behavior Tracking", "Virtual Reality Experience", "Blockchain", "Machine Learning", "metaverse", "Therapeutic VR" ] }, { "id": "adaptive-virtual-world", "title": "Adaptive Virtual World", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Adaptive Virtual World describes a metaverse environment that uses procedural generation, AI-driven content creation, and real-time user behavior analysis to dynamically evolve landscapes, structures, weather patterns, NPCs, and game mechanics in response to collective and individual user actions.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:adaptive-virtual-world", "labels": [ "Adaptive Virtual World" ], "is_subclass_of": [ "Platform and Environment", "Virtual World" ], "wikilinks": [ "Emergent Gameplay", "Neural Networks", "Persistent World Evolution", "Personalized Environments", "Blockchain", "Generative AI", "metaverse", "Procedural Generation", "Virtual World" ] }, { "id": "additionality", "title": "Additionality", "domain": "finance", "domain_name": "Finance", "definition": "Additionality is the principle, central to carbon markets and climate finance, that an emissions-reduction or removal activity counts only if it would not have occurred under a business-as-usual baseline without the incentive of credit revenue. Demonstrating additionality requires showing that the project faces financial, regulatory, or technical barriers it overcomes specifically because of the credit mechanism. It is the key integrity test that separates credible carbon credits from those representing reductions that would have happened anyway.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:additionality", "labels": [ "Additionality" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [] }, { "id": "additive-manufacturing", "title": "Additive Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Additive manufacturing (AM) is a family of processes that construct three-dimensional objects by depositing, sintering, or photopolymerising material layer by layer from a digital design file, in contrast to subtractive methods that remove material from a solid block. Processes include fused deposition modelling, selective laser sintering, stereolithography, binder jetting, and directed energy deposition, each suited to particular materials and resolution requirements. AM enables on-demand production of complex geometries, personalised products, and distributed manufacturing without the tooling investment of conventional production.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:additive-manufacturing", "labels": [ "Additive Manufacturing" ], "is_subclass_of": [ "Digital Fabrication" ], "wikilinks": [] }, { "id": "address", "title": "Address", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain Address is a compact, human-transmissible identifier derived from a public key via cryptographic hashing (typically SHA-256 followed by RIPEMD-160 for Bitcoin, or Keccak-256 for Ethereum), which designates the recipient or controller of blockchain funds or smart-contract state. Addresses function as pseudonymous identifiers: they reveal nothing about the owner's real identity while allowing cryptographic proof of ownership through digital signature with the corresponding private key. They are encoded in formats such as Base58Check (Bitcoin legacy), Bech32 (Bitcoin native SegWit), or hexadecimal with EIP-55 checksum (Ethereum).", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:address", "labels": [ "Address", "Blockchain Wallet Address" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "DID Nostr Identity", "SecurityLayer" ] }, { "id": "addressing-scheme", "title": "Addressing Scheme", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An addressing scheme is a systematic convention for assigning, structuring, and resolving identifiers that locate resources, nodes, or content within a namespace. Schemes range from hierarchical and human-readable forms such as IP addresses and URLs to flat cryptographic forms such as content hashes and public-key fingerprints. The defining properties of an addressing scheme are its uniqueness guarantees, its resolution mechanism, and whether addresses are location-based, identity-based, or content-derived. Addressing schemes underpin routing, naming, and reference integrity across networked and decentralised systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:addressing-scheme", "labels": [ "Addressing Scheme" ], "is_subclass_of": [ "Technique" ], "wikilinks": [] }, { "id": "adjacency-matrix", "title": "Adjacency Matrix", "domain": "data", "domain_name": "Data", "definition": "An adjacency matrix is a square matrix representation of a graph in which the entry at row i and column j records whether, or how strongly, vertex i is connected to vertex j. For a graph of n vertices it is an n-by-n matrix, symmetric for undirected graphs and potentially weighted to encode edge costs. It is a foundational data structure for graph algorithms, spectral analysis and graph neural networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:adjacency-matrix", "labels": [ "Adjacency Matrix" ], "is_subclass_of": [ "Graph Representation" ], "wikilinks": [] }, { "id": "admittance-control", "title": "Admittance Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Force control strategy where robotic systems respond to external forces by producing proportional motion governed by virtual admittance parameters (mass, damping, stiffness), enabling compliant physical interaction with uncertain or variable environments by regulating position/velocity trajectori...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:admittance-control", "labels": [ "Admittance Control" ], "is_subclass_of": [ "Actuation and Control", "Force Control", "Compliance Control", "Robot Control", "Interaction Control", "Motion Control" ], "wikilinks": [ "ABB Force Control", "Adaptive Manipulation", "Admittance Transfer Function", "Compliant Motion", "Compliant Trajectory Generation", "Contact-Based Assembly", "ControlLayer", "ControlTheoryDomain", "Delicate Handling", "Dynamic Model", "Dynamic Modeling", "Force-Limited Operation", "Force Sensor", "Force-to-Motion Mapping", "Hogan 1985 Impedance Control", "IEEE Transactions on Robotics Control Studies", "Intuitive Surgical da Vinci System", "ISO/TS 15066:2016 Collaborative Robots", "Kinematic Model", "KUKA Sensitive Robotics" ] }, { "id": "adobe-creative-cloud", "title": "Adobe Creative Cloud", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Adobe Creative Cloud is a subscription-based software platform operated by Adobe Inc. that delivers a suite of professional creative applications\u2014including Photoshop, Illustrator, Premiere Pro, After Effects, and InDesign\u2014alongside cloud storage, collaboration services, and asset management infrastructure. Launched in 2013 as the successor to Adobe Creative Suite, it transitioned the industry from perpetual licence software to a continuous-update, cloud-connected model. The platform is the dominant industry standard for graphic design, photography, video production, motion graphics, and digital publishing workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:adobe-creative-cloud", "labels": [ "Adobe Creative Cloud" ], "is_subclass_of": [ "Creative Tools" ], "wikilinks": [] }, { "id": "adobe-firefly", "title": "Adobe Firefly", "domain": "ai", "domain_name": "Ai", "definition": "Adobe Firefly is Adobe's family of proprietary generative AI models designed for commercial creative applications, offering text-to-image generation, generative fill, vector recolouring, and text-effect capabilities trained exclusively on licensed Adobe Stock imagery, openly licensed content, and public domain material. Launched in March 2023, Firefly is embedded across Adobe Creative Cloud applications and is positioned as an enterprise-safe generative AI tool that provides intellectual property indemnification for commercial outputs. It represents Adobe's strategy to integrate generative AI natively into professional creative workflows.", "entityType": "Class", "qualityScore": 0.87, "maturity": "emerging", "iri": "urn:ngm:class:adobe-firefly", "labels": [ "Adobe Firefly" ], "is_subclass_of": [ "Generative AI", "Creative AI", "Creative Tools" ], "wikilinks": [ "Adobe Creative Cloud", "Generative AI", "Diffusion Model", "Text-to-Image", "Transformer Architecture", "Image Generation", "Image Editing", "AI Video Generation", "Content Credentials", "C2PA", "Digital Content Provenance Marking", "Stable Diffusion", "Midjourney", "Creative AI", "Creative Tools", "Generative AI API", "Creative Workflow Automation", "Content Licensing", "Licensed Training Data", "AI Governance" ] }, { "id": "adoption-of-convergent-technologies", "title": "Adoption of Convergent Technologies", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Adoption of Convergent Technologies is the socio-technical process by which enterprises, governments, consumers and communities progressively integrate interoperating stacks of mutually reinforcing deep-technology families \u2014 artificial intelligence (including foundation models, generative AI and ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:adoption-of-convergent-technologies", "labels": [ "Adoption of Convergent Technologies" ], "is_subclass_of": [ "AI Governance and Ethics", "Technology Adoption", "Innovation Diffusion", "Digital Transformation", "Socio-Technical Transition", "Technology Policy" ], "wikilinks": [ "5G Networks", "AI Talent", "ARM Ecosystem", "Autonomous Systems Deployment", "Biotech Commercialisation", "Change Management", "Climate Tech Deployment", "Convergence Readiness Assessment", "Data Governance Framework", "Decentralised Application Ecosystems", "Diffusion of Innovations Theory", "Digital-Physical Integration", "Digital Sovereignty", "DigitalTransformationDomain", "DORA", "Economic Growth", "EnterpriseLayer", "EthicsSocietyDomain", "Executive Sponsorship", "Extended Reality" ] }, { "id": "advanced-driver-assistance-system", "title": "Advanced Driver Assistance System", "domain": "robotics", "domain_name": "Robotics", "definition": "An Advanced Driver Assistance System (ADAS) is a suite of sensor-based technologies built into a vehicle that assist a human driver with perception, warning, or partial control tasks such as lane keeping, adaptive cruise control, and automatic emergency braking. It sits below full autonomy on the driving-automation scale, keeping the human driver responsible for the overall task. Standards such as SAE J3016 define the levels of automation ADAS features correspond to.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:advanced-driver-assistance-system", "labels": [ "Advanced Driver Assistance System", "Advanced Driver Assistance Systems" ], "is_subclass_of": [ "Autonomous Vehicle" ], "wikilinks": [] }, { "id": "advanced-manufacturing", "title": "Advanced Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Advanced manufacturing is the application of innovative technologies\u2014including automation, robotics, additive manufacturing, digital twins, and data-driven process control\u2014to improve the productivity, flexibility, and quality of producing goods. It integrates cyber-physical systems across the production lifecycle, blurring the boundary between digital design and physical fabrication. Advanced manufacturing is a central pillar of Industry 4.0 and modern industrial strategy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:advanced-manufacturing", "labels": [ "Advanced Manufacturing" ], "is_subclass_of": [ "Infrastructure", "Manufacturing Automation" ], "wikilinks": [] }, { "id": "advanced-metering-infrastructure", "title": "Advanced Metering Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Advanced metering infrastructure (AMI) is the integrated system of smart meters, two-way communication networks and data management software that enables utilities to remotely collect detailed, time-stamped consumption data and to send control signals back to meters and connected devices. It extends earlier automated meter reading by providing bidirectional communication, enabling near-real-time usage data, remote connect/disconnect and dynamic pricing. AMI is a foundational enabler of demand response programmes and smart grid operations, since it supplies the granular consumption data those systems depend on.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:advanced-metering-infrastructure", "labels": [ "Advanced Metering Infrastructure" ], "is_subclass_of": [ "Smart Metering" ], "wikilinks": [] }, { "id": "adversarial-attack", "title": "Adversarial Attack", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Adversarial Attack is a deliberate attempt to manipulate an AI system by crafting malicious inputs or exploiting model vulnerabilities to cause misclassification, extract confidential information, degrade performance, or subvert intended behaviour. Attack classes include evasion, poisoning, model extraction, inversion, and backdoor, across white-box, black-box, and grey-box threat models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:adversarial-attack", "labels": [ "Adversarial Attack", "Evasion Attack" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "ISO/IEC TR 24029-1:2021", "Computer Vision", "MetaverseDomain" ] }, { "id": "adversarial-attacks", "title": "Adversarial Attacks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Adversarial Attacks are malicious techniques that exploit vulnerabilities in machine learning models by deliberately crafting deceptive input data to cause incorrect predictions, misclassifications, or unintended behaviors, often through subtle perturbations imperceptible to humans but significan...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:adversarial-attacks", "labels": [ "Adversarial Attacks" ], "is_subclass_of": [ "AI Technique", "AI Safety", "AI Security Threats" ], "wikilinks": [ "AI Security Threats", "Crafted Perturbations", "Knowledge of Target Model", "Model Evasion", "NIST AI 100-2", "NIST Taxonomy", "System Manipulation", "Computer Vision", "metaverse" ] }, { "id": "adversarial-machine-learning", "title": "Adversarial Machine Learning", "domain": "security", "domain_name": "Security", "definition": "Adversarial machine learning is the study of attacks that exploit vulnerabilities in machine learning models and the development of defences against them, encompassing threats across the model lifecycle including training-time data poisoning, evasion attacks at inference time, model inversion, and membership inference. Attackers craft carefully perturbed inputs or manipulate training data to cause misclassification, extract sensitive information, or degrade model performance, whilst defenders develop robust training procedures, certified defences, and detection mechanisms. The field spans both offensive security research and the development of trustworthy AI systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:adversarial-machine-learning", "labels": [ "Adversarial Machine Learning" ], "is_subclass_of": [ "AI Security" ], "wikilinks": [] }, { "id": "adversarial-robustness", "title": "Adversarial Robustness", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The capability of an AI system to maintain correct and consistent behaviour when subjected to adversarial examples\u2014inputs intentionally crafted with small, often imperceptible perturbations designed to cause misclassification or incorrect outputs. Defensive strategies include adversarial training, randomised smoothing, and ensemble methods, each offering different trade-offs between certified guarantees and computational cost.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:adversarial-robustness", "labels": [ "Adversarial Robustness", "Adversarial Robustness Testing" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "FCA", "ISO/IEC TR 24029-1:2021", "Computer Vision", "MetaverseDomain" ] }, { "id": "adversarial-testing", "title": "Adversarial Testing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A testing methodology that deliberately attempts to cause AI system failures through adversarial inputs, edge cases, and challenging scenarios in order to identify robustness issues, safety vulnerabilities, and alignment failures before deployment. It encompasses both explicitly adversarial prompts and implicitly problematic queries, drawing on red-teaming practices and formalised by NIST AI 100-2.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:adversarial-testing", "labels": [ "Adversarial Testing", "Adversarial Evaluation" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "NIST", "Digital Twin", "Education and AI", "MetaverseDomain" ] }, { "id": "adversarial-training", "title": "Adversarial Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Adversarial training is a robustness technique that augments model training with adversarially perturbed examples generated to maximise the model's loss. By solving an inner maximisation that crafts worst-case inputs within a bounded perturbation set and an outer minimisation over model parameters, it teaches models to resist adversarial attacks. It improves robustness against perturbations at the cost of additional computation and sometimes reduced clean accuracy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:adversarial-training", "labels": [ "Adversarial Training" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "advertising-and-marketing", "title": "Advertising and Marketing", "domain": "finance", "domain_name": "Finance", "definition": "Advertising and marketing is the combined domain of commercial practices concerned with identifying customer needs, developing products and services to meet them, communicating value propositions to target audiences through paid and unpaid channels, and building lasting brand relationships. Marketing encompasses the strategic disciplines of market research, product development, pricing, distribution, and communications, while advertising is the paid communications subset of the marketing mix. Together they constitute the primary commercial mechanism by which organisations acquire customers and generate revenue in market economies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:advertising-and-marketing", "labels": [ "Advertising and Marketing", "Marketing Communications" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [] }, { "id": "advertising", "title": "Advertising", "domain": "finance", "domain_name": "Finance", "definition": "Advertising is the practice of communicating persuasive messages about products, services, ideas, or brands to targeted audiences through paid media channels with the intent to influence awareness, attitudes, or purchasing behaviour. It encompasses the creative development of messages and the strategic planning and buying of media placements across channels including broadcast, print, digital, out-of-home, and emerging immersive formats. Modern advertising increasingly relies on data-driven audience targeting, algorithmic media buying, and AI-generated creative to maximise return on investment at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:advertising", "labels": [ "Advertising", "Addressable Advertising", "Connected TV Advertising", "Digital Advertising", "Immersive Advertising" ], "is_subclass_of": [ "Advertising and Marketing" ], "wikilinks": [] }, { "id": "aerial-robot", "title": "Aerial Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Aerial Robot - An autonomous or remotely operated aircraft equipped with Sensors, Actuators, and Navigation Systems for performing surveillance, inspection, delivery, and environmental monitoring tasks in three-dimensional airspace with minimal human intervention.", "entityType": "Class", "qualityScore": 0.6, "maturity": "draft", "iri": "urn:ngm:class:aerial-robot", "labels": [ "Aerial Robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot", "Robotics" ], "wikilinks": [ "Actuators", "Aerial Vehicle", "Automated Delivery", "Autonomous System", "Environmental Monitoring", "Flight Control System", "GPS Navigation", "Infrastructure Inspection", "ISO 21384", "ISO 21384:", "ISO 21384-1:2019", "ISO 8373:2021", "Navigation Systems", "Power Management", "Sensors", "Mobile Robot", "Obstacle Avoidance", "Robotics", "RoboticsDomain" ] }, { "id": "aerosol-optical-depth-retrieval", "title": "Aerosol Optical Depth Retrieval", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aerosol-optical-depth-retrieval", "labels": [ "Aerosol Optical Depth Retrieval" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "aerospace-engineering", "title": "Aerospace Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Aerospace Engineering is the branch of engineering concerned with the design, development, testing, and operation of aircraft, spacecraft, satellites, and related systems. It encompasses aerodynamics, propulsion, structural mechanics, avionics, guidance systems, and materials science. As an application domain it drives requirements for high-integrity embedded systems, real-time control, fault tolerance, and increasingly autonomous operation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:aerospace-engineering", "labels": [ "Aerospace Engineering" ], "is_subclass_of": [ "Infrastructure", "Systems Engineering" ], "wikilinks": [] }, { "id": "affective-computing-system", "title": "Affective Computing System", "domain": "ai", "domain_name": "Ai", "definition": "An Affective Computing System is a computational architecture that can recognise, interpret, simulate, or respond to human emotional and affective states through the integration of multimodal physiological and behavioural signals with machine learning models. Such systems aim to make human-computer interaction more natural and context-sensitive by treating emotion as a first-class computational variable.", "entityType": "Class", "qualityScore": 0.87, "maturity": "emerging", "iri": "urn:ngm:class:affective-computing-system", "labels": [ "Affective Computing System" ], "is_subclass_of": [ "Affective Computing", "Human-Computer Interaction", "Cognitive AI" ], "wikilinks": [ "Affective Computing", "Emotion Recognition", "Facial Recognition", "Natural Language Processing", "Behavioral Modeling", "Adaptive Interfaces", "Adaptive Learning", "Cognitive Science", "Sentiment Analysis", "Computer Vision", "Deep Learning", "Speech Recognition", "Physiological Signal Processing", "Convolutional Neural Network", "Recurrent Neural Network", "Transformer Architecture", "Multimodal AI", "Wearable Computing", "Human-Computer Interaction", "Emotion Aware Interaction" ] }, { "id": "affective-computing", "title": "Affective Computing", "domain": "ai", "domain_name": "Ai", "definition": "Affective computing is a branch of artificial intelligence and human-computer interaction concerned with systems that can recognise, interpret, process, and simulate human emotions and affective states. It draws on psychology, cognitive science, and machine learning to endow machines with emotional intelligence, enabling them to adapt their behaviour in response to detected user affect. The field encompasses technologies for emotion detection from facial expressions, speech, physiological signals, and body language, as well as methods for generating emotionally congruent responses. Foundational work by Rosalind Picard at MIT established that recognising and appropriately responding to affect is essential for natural, effective human-machine communication.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:affective-computing", "labels": [ "Affective Computing", "Affective Computing Framework" ], "is_subclass_of": [ "Human Computer Interaction", "Multimodal AI", "Cognitive Science" ], "wikilinks": [ "Affective Computing", "Emotional Intelligence", "Human-Computer Interaction", "Cognitive Science", "Sentiment Analysis", "Computer Vision", "Speech Recognition", "Deep Learning", "Machine Learning", "Natural Language Processing", "Physiological Signal Processing", "Convolutional Neural Network", "Multimodal Fusion", "Emotion Recognition", "Annotated Dataset", "Emotion Aware Interaction", "Emotional Analytics Engine", "Cognitive Feedback Interface", "Adaptive Learning System", "Mental Health Monitoring" ] }, { "id": "affordability-unlock", "title": "Affordability Unlock", "domain": "economics", "domain_name": "Economics", "definition": "An economic mechanism wherein AI-driven cost reductions lower service prices, thereby activating a long-tail market of consumers who previously found those services financially out of reach.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:affordability-unlock", "labels": [ "Affordability Unlock" ], "is_subclass_of": [ "Economic Impact of AI" ], "wikilinks": [] }, { "id": "affordance", "title": "Affordance", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An affordance is a property of an object or environment that signals the actions it makes possible to a perceiving agent, mediating the relationship between the agent's capabilities and the world. Originating in ecological psychology, the concept was adapted to design to describe how the perceptible features of an interface or physical artefact suggest how it can be used. In spatial and interactive computing, affordances guide users towards available actions, and perceived affordances are deliberately engineered so that controls and gestures feel discoverable and natural.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:affordance", "labels": [ "Affordance" ], "is_subclass_of": [ "Interaction Design" ], "wikilinks": [] }, { "id": "age-verification", "title": "Age Verification", "domain": "governance", "domain_name": "Governance", "definition": "Age verification is a set of technical and procedural mechanisms used to confirm that a user meets a minimum age threshold before accessing restricted digital content or services. It encompasses document-based checks, biometric estimation, and privacy-preserving cryptographic proofs. Regulatory frameworks such as the UK Online Safety Act mandate age verification for platforms hosting harmful content. The field balances effective enforcement against privacy risks and the potential exclusion of legitimate users.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:age-verification", "labels": [ "Age Verification", "Age Assurance", "Age Gating" ], "is_subclass_of": [ "Compliance Verification" ], "wikilinks": [] }, { "id": "agent-capability-scaling", "title": "Agent Capability Scaling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The empirical trend and theoretical analysis of how the complexity and duration of tasks that AI agents can autonomously complete increase over time with model improvements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agent-capability-scaling", "labels": [ "Agent Capability Scaling" ], "is_subclass_of": [ "Agent" ], "wikilinks": [] }, { "id": "agent-communication-language", "title": "Agent Communication Language", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Agent Communication Language (ACL) is a formal, standardised message format and semantics that lets autonomous software agents exchange information, requests and commitments independently of their internal implementation. ACLs define performatives (speech acts such as inform, request and propose), a content language and an ontology reference so that interacting agents share a common interpretation of messages. Established examples include FIPA-ACL and KQML.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:agent-communication-language", "labels": [ "Agent Communication Language" ], "is_subclass_of": [ "Multi-Agent System" ], "wikilinks": [] }, { "id": "agent-communication-protocol", "title": "Agent Communication Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Agent Communication Protocol is a formalised specification governing the syntax, semantics, and pragmatics of message exchange between autonomous software agents, enabling them to coordinate actions, share information, delegate tasks, and negotiate goals across heterogeneous runtime environments. Such protocols define the speech act primitives, message envelope formats, conversation policies, and error-handling procedures that agents must implement to participate in a multi-agent system.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:agent-communication-protocol", "labels": [ "Agent Communication Protocol" ], "is_subclass_of": [ "Coordination Protocol", "Distributed Systems", "Interoperability" ], "wikilinks": [ "Agent Communication Protocol", "Inter-Agent Communication", "Agentic Workflow", "Autonomous Task Execution", "Multi-Agent System", "Task Delegation", "Message Passing", "Speech Act Theory", "JSON-LD", "WebSocket", "Ontology", "FIPA ACL", "Contract Net Protocol", "Capability Advertisement", "Error Handling", "Negotiation", "Remote Procedure Call", "Distributed Collaboration", "Robotics", "Agent-to-Agent Protocol" ] }, { "id": "agent-development-sdks", "title": "Agent Development SDKs", "domain": "ai", "domain_name": "Ai", "definition": "Official vendor SDKs and libraries for building custom AI agents with built-in tools, MCP support, session management, and deployment infrastructure \u2014 includes Claude Agent SDK, Google ADK, strands-agents, pydantic-ai, Composio, and SWE-agent.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:agent-development-sdks", "labels": [ "Agent Development SDKs" ], "is_subclass_of": [ "Agent Harness", "Agent Frameworks", "LLM Orchestration" ], "wikilinks": [ "Agent Harness", "Multi-Agent Orchestration Frameworks", "Agent Execution Sandboxes", "Model Context Protocol", "Agent Evaluation Benchmarks", "Large Language Model", "Agentic AI", "Tool Use", "Tool Calling API", "Function Calling", "Agent Communication Protocol", "A2A Protocol", "Claude Agent SDK", "Google ADK", "Pydantic AI", "SWE-agent", "OpenAI Agents SDK", "Strands Agents", "Mastra", "LangGraph" ] }, { "id": "agent-ecosystem", "title": "Agent Ecosystem", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A coordinated network of specialized AI agents that interact and share context to solve complex, multi-faceted problems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agent-ecosystem", "labels": [ "Agent Ecosystem" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "agent-evaluation-benchmarks", "title": "Agent Evaluation Benchmarks", "domain": "ai", "domain_name": "Ai", "definition": "Standardised benchmark suites and evaluation frameworks for measuring autonomous agent capabilities across software engineering, web navigation, reasoning, and general task completion \u2014 includes SWE-bench, WebArena, ARC-AGI-2, inspect_ai, AgentBench, and VitaBench.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:agent-evaluation-benchmarks", "labels": [ "Agent Evaluation Benchmarks" ], "is_subclass_of": [ "Agent Harness", "LLM Evaluation", "Evaluation Harness" ], "wikilinks": [ "Agent Harness", "Evaluation Harness", "Agent Development SDKs", "SWE-bench", "WebArena", "ARC-AGI", "inspect_ai", "GAIA Benchmark", "OSWorld", "AgentBench", "VitaBench", "METR HCAST", "TAU-bench", "Terminal-Bench", "Large Language Model", "Agentic AI", "Autonomous Coding", "Computer Use", "Browser Automation", "LLM Evaluation" ] }, { "id": "agent-event-stream", "title": "Agent Event Stream", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A real-time telemetry stream (surface S6) that emits ActivityStream|ActivityStreams-based JSON-LD events for agent lifecycle events (birth, startup, activity, completion, error, termination) via WebSocket, Nostr relay, or message queue, enabling external monitoring systems, dashboards, and or...", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:agent-event-stream", "labels": [ "Agent Event Stream" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "ActivityStream", "ActivityStreams 2.0", "ActivityStreams 2.0 Spec", "ActivityStreams Vocabulary", "AgenticSystemsDomain", "Anomaly Detection", "Automated Alerting", "Data Analytics Platform", "Event Emission", "Event Filtering", "Event Indexing", "Event Schemas", "Event Subscription", "Fleet Observability", "JSON-LD 1.1", "Monitoring System", "ObservabilityDomain", "Orchestration System", "PRD-006", "Real-Time Monitoring" ] }, { "id": "agent-execution-sandboxes", "title": "Agent Execution Sandboxes", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Isolated execution environments providing secure, ephemeral containers and virtual machines for AI-generated code execution, tool use, and autonomous agent operation \u2014 including E2B, Daytona, Docker MCP Gateway, Cloudflare Sandboxes, Modal, Vercel Sandbox, and Fly Machines \u2014 with hardware-level isolation via Firecracker microVMs, gVisor syscall interception, or Kata Containers to prevent escape to host infrastructure.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:agent-execution-sandboxes", "labels": [ "Agent Execution Sandboxes" ], "is_subclass_of": [ "Agent Harness", "AI Infrastructure", "Containerisation" ], "wikilinks": [ "Agent Harness", "Agentic AI", "Agentic Workflow", "Terminal Coding Agents", "Agent Development SDKs", "Multi-Agent Orchestration Frameworks", "LLM Application Frameworks", "Tool Use", "Function Calling", "Containerisation", "Container Runtime", "Docker Containerisation Platform", "Container Orchestration", "Open Container Initiative", "Model Context Protocol", "MCP Server", "MCP Client", "Code Execution", "Autonomous Coding", "AI Safety" ] }, { "id": "agent-frameworks", "title": "Agent Frameworks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agent Frameworks are software libraries, runtimes, and orchestration platforms that compose large language models (LLMs) with tool-use, memory, planning, and inter-agent communication into autonomous or semi-autonomous systems capable of multi-step goal pursuit, encompassing single-agent harnesse...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:agent-frameworks", "labels": [ "Agent Frameworks" ], "is_subclass_of": [ "LLM Orchestration", "Software Framework", "AI Agent System", "Autonomous System", "Workflow Engine" ], "wikilinks": [ "A2A Protocol", "ACP Protocol", "Agent Communication Protocol", "Agent Runtime", "Agent-to-Agent Protocol", "Agentic AI", "AGNTCY Collective", "AGNTCY Internet of Agents", "Anthropic claude-agent-sdk Documentation", "Anthropic MCP Code Execution Engineering", "Anthropic Model Context Protocol Announcement", "Autonomous Coding", "Autonomous System", "BPMN Orchestration", "Browser Automation", "Chain-of-Thought", "Chen et al 2021 HumanEval", "Classical Workflow Engine", "Coding Assistant", "Computer Use" ] }, { "id": "agent-handoff", "title": "Agent Handoff", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The orchestration primitive by which one agent transfers control of a conversation or task to another agent, passing along the accumulated context, goal, and constraints so the receiving agent can continue the work with its own specialised tools and instructions. A handoff reassigns responsibility rather than merely requesting a result: control does not automatically return to the sender, and the receiver becomes the active locus of decision-making, which is what distinguishes handoff-based routing from a simple tool call or a blocking sub-task.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:agent-handoff", "labels": [ "Agent Handoff" ], "is_subclass_of": [ "Multi-Agent Orchestration", "MultiAgentOrchestration" ], "wikilinks": [ "MultiAgentOrchestration", "TaskDelegation", "MultiAgentSystem", "AgenticWorkflow" ] }, { "id": "agent-harness", "title": "Agent Harness", "domain": "ai", "domain_name": "Ai", "definition": "A runtime framework that turns model inference into agent action by managing the tool-call loop, approval gates, context routing, execution lifecycle, and failure recovery \u2014 the model thinks, the harness decides what that thinking is allowed to touch.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:agent-harness", "labels": [ "Agent Harness" ], "is_subclass_of": [ "AI Infrastructure", "Agentic AI", "Autonomous Agent", "AI Agent System" ], "wikilinks": [ "AI Infrastructure", "Internal AI Harness", "External AI Harness", "Terminal Coding Agents", "IDE Coding Agents", "Multi-Agent Orchestration Frameworks", "Agent Evaluation Benchmarks", "Agent Execution Sandboxes", "Harness Configuration Packs", "Personal Agent Runtimes", "Progressive Disclosure Harnesses", "Agent Runtime", "AI Agent", "Tool Use", "Function Calling", "Model Context Protocol", "Large Language Models", "Agent Loop", "Human-in-the-Loop", "Prompt Injection" ] }, { "id": "agent-identity", "title": "Agent Identity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agent identity is the set of attributes, credentials, and cryptographic material that uniquely identifies an autonomous software agent and allows other parties to authenticate it, attribute its actions, and decide what it may do. It extends identity management from human and device contexts to AI agents that act on a user's behalf, enabling verifiable accountability across the emerging agentic internet. Robust agent identity typically combines decentralised identifiers, verifiable credentials, and key management.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:agent-identity", "labels": [ "Agent Identity" ], "is_subclass_of": [ "Identity Management", "Trust Framework", "Decentralized Identity" ], "wikilinks": [ "AI Agent", "Identity Management", "Agentic Internet", "Verifiable Credentials", "DID", "Key Management", "Authentication", "Accountability", "Trust", "Tool Use", "Decentralized Identity", "Autonomous Agent", "Multi-Agent System", "AI Safety", "Agentic AI", "Provenance", "Single Sign-On", "Artificial Intelligence", "Model Context Protocol", "Agent-to-Agent Protocol" ] }, { "id": "agent-layer", "title": "Agent Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Agent Layer is the stratum that hosts autonomous entities capable of perceiving, deciding, and acting toward goals. It sits above control and inference strata that supply its capabilities and below coordination and application strata that direct collectives of agents. It contains agent policies, goal representations, memory, and decision loops.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:agent-layer", "labels": [ "Agent Layer" ], "is_subclass_of": [ "AI Model Architecture", "owl:Thing" ], "wikilinks": [ "Inference Layer", "Control Layer", "Coordination Layer", "Application Layer", "Autonomous Agent", "Reinforcement Learning", "owl:Thing" ] }, { "id": "agent-loop", "title": "Agent Loop", "domain": "ai", "domain_name": "Ai", "definition": "The agent loop is the repeating control cycle \u2014 observe, plan, act, evaluate \u2014 through which an autonomous agent perceives its environment, selects a next action or tool call, executes it, and incorporates the result before repeating. It is the core execution pattern underlying agentic systems, terminal coding agents, and retrieval-augmented reasoning, continuing until a termination condition such as task completion or a step budget is reached. The loop's design governs how an agent balances exploration, tool use, and convergence toward a goal.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:agent-loop", "labels": [ "Agent Loop" ], "is_subclass_of": [ "Agent Harness" ], "wikilinks": [] }, { "id": "agent-management", "title": "Agent Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The skills and practices required to oversee, coordinate, and evaluate AI agents, distinct from traditional software or human management.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agent-management", "labels": [ "Agent Management" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "agent-memory-layers", "title": "Agent Memory Layers", "domain": "ai", "domain_name": "Ai", "definition": "Persistent memory systems for AI agents providing contextual recall across sessions through vector retrieval, automatic summarisation, and state management \u2014 includes Mem0, letta, claude-mem, and agentlog.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:agent-memory-layers", "labels": [ "Agent Memory Layers" ], "is_subclass_of": [ "Agent Harness", "AI Infrastructure", "Cognitive Architecture" ], "wikilinks": [ "Agent Memory", "Agent Harness", "Personal Agent Runtimes", "Progressive Disclosure Harnesses", "Vector Database", "Embeddings", "Retrieval-Augmented Generation", "Context Window", "Episodic Memory", "Semantic Memory", "Procedural Memory", "Working Memory", "Knowledge Graph", "Large Language Models", "Cognitive Architecture", "Agent Loop", "Multi-Agent System", "Personalisation", "Consolidation", "Continual Learning" ] }, { "id": "agent-memory", "title": "Agent Memory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agent memory is the structured ensemble of mechanisms by which an autonomous AI agent stores, indexes, consolidates, retrieves, and forgets information across steps, sessions, and lifetimes \u2014 enabling coherent, personalised, and long-horizon behaviour that transcends the hard limit of any single context window. It encompasses four functionally distinct tiers: working memory (active context window); episodic memory (timestamped records of prior observations, actions, and outcomes); semantic memory (declarative facts, entity relationships, and world knowledge); and procedural memory (skill programs, tool-use patterns, and reusable plan templates).", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:agent-memory", "labels": [ "Agent Memory", "AI Agent Memory" ], "is_subclass_of": [ "AI Agent", "Cognitive Architecture", "Memory Management" ], "wikilinks": [ "Vector Database", "Embeddings", "Agentic Workflow", "Retrieval-Augmented Generation", "Context Window", "AI Agent", "Episodic Memory", "Semantic Memory", "Procedural Memory", "Working Memory", "Long-Term Memory", "Knowledge Graph", "Large Language Models", "Foundation Model", "Cognitive Architecture", "Agent Loop", "Multi-Agent System", "Attention Mechanism", "ReAct Pattern", "Reflexion Pattern" ] }, { "id": "agent-orchestrator", "title": "Agent Orchestrator", "domain": "ai", "domain_name": "Ai", "definition": "An Agent Orchestrator is the directive control component of a multi-agent architecture that decomposes high-level goals into directed acyclic graphs of sub-tasks, selects and dispatches those tasks to specialised sub-agents via a structured Agent Communication Protocol, monitors execution state, resolves inter-task data dependencies, handles timeouts and failures through retry or fallback logic, and assembles partial results into coherent outputs \u2014 operating atop an Agent Runtime that manages the lifecycle of individual agents and providing the coordination hub that transforms a pool of autonomous agents into a collaborative system capable of accomplishing complex, long-horizon objectives.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:agent-orchestrator", "labels": [ "Agent Orchestrator" ], "is_subclass_of": [ "AI Agent System", "Intelligent System" ], "wikilinks": [ "Agent Orchestrator", "AI Agent System", "Task Planning", "Agent Communication Protocol", "Agent Runtime", "Agentic Workflow", "Autonomous Task Execution", "Coordination Mechanisms", "Multi-Agent System", "Agent-to-Agent Protocol", "Chain-of-Thought", "Large Language Models", "Model Context Protocol", "Tool Use", "LangGraph", "AutoGen", "CrewAI", "OpenAI Agents SDK", "Task Decomposition", "Directed Acyclic Graph" ] }, { "id": "agent-primitives", "title": "Agent Primitives", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The foundational software components and abstractions, such as orchestration, memory, and tool-use interfaces, that enable the construction of autonomous AI agents.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agent-primitives", "labels": [ "Agent Primitives" ], "is_subclass_of": [ "Agent" ], "wikilinks": [] }, { "id": "agent-runtime", "title": "Agent Runtime", "domain": "ai", "domain_name": "Ai", "definition": "An Agent Runtime is the infrastructure execution layer that manages the complete lifecycle, resource allocation, tool access, memory state, and communication channels of one or more autonomous AI agents \u2014 providing process isolation, context window management, stateful execution with durable checkpointing, sandboxed tool-call dispatch, inter-agent messaging, credential management, rate limiting, and structured observability, analogous to how an operating system runtime supports application processes, and forming the foundational services upon which Agent Orchestrator control logic and Agentic Workflow execution patterns are built.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:agent-runtime", "labels": [ "Agent Runtime" ], "is_subclass_of": [ "AI Agent System", "Intelligent System" ], "wikilinks": [ "Agent Runtime", "AI Agent System", "Agent Orchestrator", "Agentic Workflow", "Tool Use", "Model Context Protocol", "Autonomous Task Execution", "Execution Model", "Inter-Agent Communication", "Context Window", "Agent Memory", "Retrieval-Augmented Generation", "Vector Database", "Durable Execution", "Human-in-the-Loop", "LangGraph", "AutoGen", "LangChain", "Temporal", "AWS Bedrock AgentCore" ] }, { "id": "agent-security", "title": "Agent Security", "domain": "security", "domain_name": "Security", "definition": "The discipline of protecting autonomous AI agents and their interactions with tools, data, and other systems from unauthorized access, manipulation, or malicious exploitation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agent-security", "labels": [ "Agent Security" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "agent-skill", "title": "Agent Skill", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An agent skill is a packaged, self-contained unit of procedural knowledge \u2014 a named bundle of instructions, and optionally scripts and reference resources \u2014 that an agent loads into its context on demand when a task matches the skill's trigger, and unloads afterwards. Rather than baking every capability into the base system prompt, a skill lets specialised know-how (how to run a deployment, audit prose, drive a browser) live as a discoverable, versioned artifact that is progressively disclosed only when relevant, keeping the working context small while giving the agent deep competence in whatever domain the current task demands.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:agent-skill", "labels": [ "Agent Skill" ], "is_subclass_of": [ "Prompt Template", "PromptTemplate" ], "wikilinks": [ "PromptTemplate", "ToolUse", "ContextManagement", "AgenticWorkflow" ] }, { "id": "agent-based-modelling", "title": "Agent-Based Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agent-based modelling (ABM) is a computational simulation methodology in which a system is represented as a collection of autonomous, heterogeneous agents that each follow local behavioural rules and interact with one another and with their environment. The global dynamics of interest \u2014 such as market prices, epidemic spread, or traffic congestion \u2014 emerge from these micro-level interactions rather than being specified analytically. ABM bridges individual behaviour and system-level phenomena, making it particularly powerful for studying complex adaptive systems where aggregate equations cannot capture heterogeneity or non-linear feedback loops. The approach is distinct from equation-based modelling in that agents are discrete, can differ individually, and can adapt their strategies through reinforcement or learning mechanisms.", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:agent-based-modelling", "labels": [ "Agent-Based Modelling", "Agent Model" ], "is_subclass_of": [ "Simulation", "Complexity Science", "Complex Systems Science", "Computational Modelling" ], "wikilinks": [ "Simulation", "Innovation Diffusion", "Economic Model", "Multi-Agent Systems", "Computational Modelling", "Complex Adaptive Systems", "Emergence", "Autonomous Agent", "Reinforcement Learning", "Cellular Automata", "Monte Carlo Simulation", "Epidemiological Modelling", "Crowd Simulation", "Supply Chain Optimisation", "Policy Simulation", "System Dynamics", "Swarm Intelligence", "Digital Twin", "Game Theory", "Network Science" ] }, { "id": "agent-based-models", "title": "Agent-Based Models", "domain": "ai", "domain_name": "Ai", "definition": "Agent-Based Models (ABMs) are computational simulation frameworks that represent systems as collections of autonomous, heterogeneous agents that interact with one another and their environment according to local rules, allowing complex global behaviours and emergent phenomena to arise from the bottom up without being explicitly programmed at the system level. They are a primary tool in complexity science for studying social, biological, economic, and technical systems.", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:agent-based-models", "labels": [ "Agent-Based Models", "Agent-Based Model" ], "is_subclass_of": [ "Computational Intelligence", "Simulation", "Complex Systems Science" ], "wikilinks": [ "Autonomous Agent", "Behavioral Modeling", "Scientific Computing", "Emergence", "Collective Intelligence", "Feedback Loop", "Computational Intelligence", "Agent-Based Modelling", "Multi-Agent Reinforcement Learning", "Simulation", "Complex Adaptive Systems", "Stochastic Process", "Network Science", "Reinforcement Learning", "Large Language Models", "Monte Carlo Simulation", "Cellular Automata", "Multi-Agent Systems", "Epidemiological Modelling", "Economic Model" ] }, { "id": "agent-to-agent-protocol", "title": "Agent-to-Agent Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agent-to-Agent Protocol is a class of communication specifications that define how autonomous AI agents discover one another, advertise capabilities, delegate tasks, and exchange results directly \u2014 without requiring a centralised broker \u2014 enabling peer-to-peer coordination between agents built by different organisations or on different frameworks. These protocols treat agents as first-class addressable entities with discoverable skill sets and negotiable service contracts.", "entityType": "Class", "qualityScore": 0.92, "maturity": "emerging", "iri": "urn:ngm:class:agent-to-agent-protocol", "labels": [ "Agent-to-Agent Protocol" ], "is_subclass_of": [ "Agent Communication Protocol", "Coordination Protocol" ], "wikilinks": [ "Agent Communication Protocol", "Inter-Agent Communication", "Agentic Workflow", "Multi-Agent System", "Task Delegation", "Service Discovery", "Message Passing", "JSON-LD", "Mutual Authentication", "Decentralised Identifier", "Agent2Agent Protocol (Google 2025)", "FIPA ACL", "Autonomous Agent", "Large Language Model", "Remote Procedure Call", "Model Context Protocol", "Prompt Injection", "Microservices Architecture", "Web of Things", "Agent Interoperability" ] }, { "id": "agent", "title": "Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An autonomous computational or physical entity that perceives its environment, reasons about its perceptions using internal beliefs and goals, and acts to achieve specified objectives\u2014exhibiting autonomy, reactivity, proactivity, and social ability across AI, blockchain, robotics, and metaverse domains.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:agent", "labels": [ "Agent", "Intelligent Agent", "Monolithic Agent" ], "is_subclass_of": [ "Autonomous System", "Blockchain" ], "wikilinks": [ "Action", "Autonomous System", "Environment", "Hybrid Agent", "Learning", "Multi-Agent System", "Perception", "Physical Agent", "Software Agent", "AI Agent System", "Autonomy Level", "BDI Model", "Blockchain", "Goal", "Objective" ] }, { "id": "agent2-agent-protocol-google-2025", "title": "Agent2Agent Protocol (Google 2025)", "domain": "ai", "domain_name": "Ai", "definition": "The Agent2Agent Protocol (A2A), released by Google as an open specification in April 2025 and transferred to the Linux Foundation in June 2025, is a JSON-RPC 2.0 over HTTP(S) protocol \u2014 extended with Server-Sent Events for streaming \u2014 that enables AI agents to discover one another via standardised, cryptographically signed agent cards hosted at well-known URIs, delegate tasks through a six-state task lifecycle, stream results back to requesting agents, and operate regardless of the underlying model or framework used to implement the agents. Reaching version 1.2 by March 2026 with over 150 supporting organisations, it is designed as a complementary peer-to-peer layer alongside Anthropic's Model Context Protocol, together forming the emerging two-protocol stack for the agentic internet.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:agent2-agent-protocol-google-2025", "labels": [ "Agent2Agent Protocol (Google 2025)" ], "is_subclass_of": [ "Agent-to-Agent Protocol", "Coordination Protocol", "Agent Communication Protocol" ], "wikilinks": [ "Agent-to-Agent Protocol", "Inter-Agent Communication", "Agentic Workflow", "Service Discovery", "Message Passing", "Model Context Protocol", "Agent Communication Protocol", "Agent Card", "Task Lifecycle", "Server-Sent Events", "JSON-RPC 2.0", "HTTP Protocol", "OAuth 2.0", "JSON Web Signature", "JSON Schema", "Agent Identity", "Large Language Model", "Multi-Agent Orchestration", "Autonomous Agent", "Vertex AI" ] }, { "id": "agent2-agent-protocol", "title": "Agent2Agent Protocol", "domain": "ai", "domain_name": "Ai", "definition": "Agent2Agent Protocol (A2A) is an open specification developed by Google and partner organisations that enables heterogeneous AI agents to discover, communicate, and collaborate across organisational and platform boundaries. It defines a standardised messaging envelope, capability advertisement mechanism, and task-delegation schema so that agents built on different frameworks \u2014 such as LangChain, CrewAI, or custom enterprise systems \u2014 can interoperate without bespoke integration code. The protocol operates over HTTPS with JSON-RPC 2.0 and supports both synchronous request-response and asynchronous streaming via Server-Sent Events. Agent capability cards, analogous to service discovery manifests, allow an agent to advertise its skills so that orchestrating agents can route work appropriately.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:agent2-agent-protocol", "labels": [ "Agent2Agent Protocol" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "agentic-ai-practitioner-training-programme", "title": "Agentic AI Practitioner Training Programme", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured training programme introducing practitioners to agentic AI systems, covering context engineering, memory management, agent orchestration tools (such as Roo Code), and practical case studies in project management, data visualisation, and academic research through phased, multi-session instruction.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:agentic-ai-practitioner-training-programme", "labels": [ "Agentic AI Practitioner Training Programme", "Agentic Workshop" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "AI Agent System" ] }, { "id": "agentic-ai-systems", "title": "Agentic AI Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agentic AI systems are AI systems that autonomously plan and execute multi-step tasks by invoking tools, calling external APIs, and adapting their strategy based on intermediate results, rather than producing a single response to a single prompt. They combine a reasoning model with memory, tool use, and control loops that allow sustained operation toward a goal with limited human intervention. Agentic AI systems are a major focus of frontier AI development and a key axis of competition among AI labs.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:agentic-ai-systems", "labels": [ "Agentic AI Systems", "Agentic Systems" ], "is_subclass_of": [ "AI System" ], "wikilinks": [] }, { "id": "agentic-ai", "title": "agentic ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agentic AI refers to AI systems characterised by autonomy, goal-directedness, and the capacity to take sustained sequences of actions \u2014 including calling external tools, spawning sub-agents, and modifying their own environment \u2014 in pursuit of high-level objectives specified by a user or orchestrator. Agentic systems differ from reactive or conversational AI in that they operate over extended time horizons, maintain persistent state across steps, and may take consequential or irreversible actions without per-step human approval. The architectural backbone is typically a large language model serving as a cognitive core inside a closed sense-plan-act loop, augmented by memory stores, tool registries, and inter-agent communication protocols. The term encompasses both single-agent pipelines and heterogeneous multi-agent architectures in which agentic components collaborate, compete, or are hierarchically orchestrated.", "entityType": "Class", "qualityScore": 0.92, "maturity": "emerging", "iri": "urn:ngm:class:agentic-ai", "labels": [ "Agentic AI" ], "is_subclass_of": [ "AI Application", "AI Agents", "Cognitive Architecture", "Agent-Based Modelling" ], "wikilinks": [ "Large Language Models", "Memory Management", "Tool Use", "Conversational AI", "Multi-Agent System", "Foundation Models", "Agent Loop", "Task Planning", "Chain of Thought", "Orchestration", "Tool Registry", "Working Memory", "Function Calling", "Model Context Protocol", "Retrieval-Augmented Generation", "Vector Database", "AI Agents", "Prompt Engineering", "AI Alignment", "AI Safety" ] }, { "id": "agentic-architecture", "title": "Agentic Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The design patterns and system structures for building AI agents, including modular skills, continuous learning, and multi-agent collaboration.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-architecture", "labels": [ "Agentic Architecture" ], "is_subclass_of": [ "Model Architecture" ], "wikilinks": [] }, { "id": "agentic-coding", "title": "Agentic Coding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of autonomous AI agents to perform complex, multi-step software engineering tasks, including code generation, debugging, and execution within terminal environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-coding", "labels": [ "Agentic Coding" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "agentic-finance", "title": "agentic finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Infrastructure enabling autonomous AI agents to hold cryptographic wallets, provision compute resources, negotiate contracts, and transact with other agents or humans without intermediaries. AgentFi bridges decentralised finance protocols with agentic AI systems, creating machine-to-machine economic coordination layers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:agentic-finance", "labels": [ "Agentic Finance" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "agentic-harness", "title": "Agentic Harness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A software framework or layer that orchestrates autonomous AI agents, managing their memory, decision loops, and interaction with external tools or environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-harness", "labels": [ "Agentic Harness" ], "is_subclass_of": [ "Agent Layer" ], "wikilinks": [] }, { "id": "agentic-internet", "title": "Agentic Internet", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agentic Internet is the emerging substrate of protocols, identity primitives, payment rails and discovery mechanisms over which autonomous AI agents \u2014 rather than human users clicking through browsers \u2014 discover services, negotiate terms, exchange data and settle value on behalf of principals (hu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:agentic-internet", "labels": [ "Agentic Internet" ], "is_subclass_of": [ "AI Research Area", "AI Agent System", "Web Architecture", "Distributed Computing", "Agent-Oriented Computing", "Sociotechnical System" ], "wikilinks": [ "A2A", "A2A Protocol", "Abridge", "Agent Connect Protocol", "Agent Identity", "Agent Marketplace", "Agent Orchestrator", "Agent-Oriented Computing", "Agent Runtime", "Agentic Commerce Layer", "Agentic Economy", "Agentic Search", "AGNTCY", "AGNTCY Collective Launch 2025", "AI Agents", "AI Opportunities Action Plan", "AI Security Institute", "AISI", "Alan Turing Institute", "Anchor Browser" ] }, { "id": "agentic-knowledge-base", "title": "Agentic Knowledge Base", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An autonomous system that continuously ingests, processes, and structures external information to provide real-time context for other AI agents.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-knowledge-base", "labels": [ "Agentic Knowledge Base" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "agentic-mycelia", "title": "Agentic Mycelia", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A framework for interconnected metaverse instances operating as a decentralised, AI-driven ecosystem, in which specialised software agents mediate interoperability, value exchange, identity, and adaptable governance across otherwise-sovereign virtual worlds. Each instance exposes a machine- and human-readable ontology (Linked-JSON) that agents at the edges translate and arbitrate between.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:agentic-mycelia", "labels": [ "Agentic Mycelia", "Agentic Mycelium", "AgenticMycelia" ], "is_subclass_of": [ "Metaverse Architecture" ], "wikilinks": [ "Scene Agent", "Transfer Agent", "Onboarding Agent", "Jurisdictional Agent", "Living Contract", "Metaverse Architecture", "Interoperability", "Self-Sovereign Identity", "Adaptable Governance" ] }, { "id": "agentic-rag", "title": "Agentic RAG", "domain": "ai", "domain_name": "Ai", "definition": "Agentic RAG is an architecture that augments retrieval-augmented generation with an autonomous agent loop, letting a language model plan, decide when and what to retrieve, issue multiple queries across heterogeneous sources, reflect on retrieved evidence quality, reformulate queries, and verify results before generating a final response. Unlike the single-pass retrieve-then-read pipeline of naive RAG, Agentic RAG treats retrieval as a first-class action in a multi-step reasoning cycle, enabling resolution of multi-hop questions, adaptive source selection, self-correction, and iterative evidence accumulation under agent-controlled stopping criteria.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:agentic-rag", "labels": [ "Agentic RAG" ], "is_subclass_of": [ "Information Retrieval", "Retrieval-Augmented Generation", "Agentic Workflow", "AI Application" ], "wikilinks": [ "Retrieval-Augmented Generation", "RAG Pipeline", "Agentic AI", "Large Language Models", "Information Retrieval", "Agent Loop", "Vector Database", "Embedding Model", "Knowledge Graph", "Chain of Thought", "Function Calling", "Tool Use", "Query Rewriting", "Hallucination Mitigation", "Question Answering", "Semantic Search", "Dense Retrieval", "Multi-Agent System", "Orchestration", "Prompt Engineering" ] }, { "id": "agentic-shift", "title": "Agentic Shift", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The industry-wide transition from static AI models to autonomous, goal-directed agents capable of complex task execution.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-shift", "labels": [ "Agentic Shift" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "agentic-software-engineering", "title": "Agentic Software Engineering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A software development paradigm where autonomous AI agents perform end-to-end engineering tasks, including coding, testing, and review, with minimal human intervention.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-software-engineering", "labels": [ "Agentic Software Engineering" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "agentic-workflow", "title": "Agentic Workflow", "domain": "ai", "domain_name": "Ai", "definition": "An agentic workflow is a structured, iterative execution pattern in which an AI agent autonomously plans actions, invokes external tools or APIs, observes results, and revises its approach through successive reasoning cycles until a goal condition is satisfied or a stopping criterion is met. Unlike single-pass inference, agentic workflows employ persistent memory, branching logic, and multi-step planning that may span many inference calls and involve specialised sub-agents coordinated by an orchestrator. The pattern relies on large language model capabilities \u2014 tool use, function calling, long-context reasoning \u2014 and is the architectural basis for systems such as AutoGPT, LangGraph, CrewAI, OpenAI Agents SDK, and Anthropic's Claude toolset. Agentic workflows introduce novel safety and reliability challenges including error compounding, prompt injection via tool outputs, and the need for human-in-the-loop checkpoints in high-stakes deployments.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:agentic-workflow", "labels": [ "Agentic Workflow" ], "is_subclass_of": [ "Workflow Automation" ], "wikilinks": [ "Agentic AI", "AI Agent System", "Tool Use", "Function Calling", "Reasoning", "Chain of Thought", "Prompt Engineering", "ReAct Pattern", "Multi-Agent Systems", "Agent Frameworks", "Orchestration", "Planning and Scheduling", "Robotic Process Automation", "Task Planning", "Retrieval-Augmented Generation", "Autonomous Coding", "AI Research Assistant", "Context Window", "Large Language Models", "Agent Memory" ] }, { "id": "agentic-workload-economics", "title": "Agentic Workload Economics", "domain": "economics", "domain_name": "Economics", "definition": "The economic analysis of computational costs and resource consumption associated with autonomous AI agents that perform multi-step tasks, characterized by significantly higher token usage than simple query-response interactions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:agentic-workload-economics", "labels": [ "Agentic Workload Economics" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "agents", "title": "Agents", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agents are autonomous or semi-autonomous computational systems that perceive an environment through sensors (text inputs, vision encoders, structured tool responses, multimodal streams), reason or plan over an internal model of that environment, and act upon it through actuators (function calls, ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:agents", "labels": [ "Agents", "Intelligent Agents" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence", "Autonomous System", "Intelligent System", "Cognitive Architecture" ], "wikilinks": [ "Action Executor", "Agent Memory", "Agent2Agent Protocol", "Agent2Agent Protocol (Google 2025)", "Agentic AI", "Agentic Workflow", "AI Software Engineering", "Anthropic Claude Computer Use Oct 2024", "Anthropic Model Context Protocol Nov 2024", "Apollo Research 2024 Scheming AIs", "AutoGPT Significant Gravitas 2023", "Autonomous System", "AutonomousSystemsDomain", "Autonomous Task Execution", "BabyAGI Nakajima 2023", "BDI Architecture", "Bratman 1987 Intention Plans and Practical Reason", "Brooks 1986 Subsumption Architecture", "Brooks 1991 Intelligence Without Representation", "Browser Automation" ] }, { "id": "aggelos-kiayias", "title": "Aggelos Kiayias", "domain": "security", "domain_name": "Security", "definition": "A computer scientist and cryptographer known for research in cryptography and blockchain protocols, including provably secure proof-of-stake consensus.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:aggelos-kiayias", "labels": [ "Aggelos Kiayias" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Cryptography", "Proof of Stake", "Consensus Mechanisms", "Cardano" ] }, { "id": "agile-software-development", "title": "Agile Software Development", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Agile Software Development is a family of iterative, incremental approaches to building software that prioritise working product increments, continuous stakeholder collaboration, and rapid adaptation to changing requirements over rigid up-front planning. Rooted in the 2001 Agile Manifesto, it encompasses frameworks such as Scrum, Kanban, Extreme Programming, and SAFe, each operationalising core values through time-boxed iterations, cross-functional teams, and frequent feedback loops. Agile practices shorten the feedback cycle between developers and end users, reduce the cost of change, and improve delivery predictability by surfacing risk early. It has become the dominant paradigm for professional software delivery and is increasingly applied to hardware, policy, and organisational transformation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:agile-software-development", "labels": [ "Agile Software Development", "Agile Development", "Agile Methodology" ], "is_subclass_of": [ "Software Engineering", "Software Engineering Domain" ], "wikilinks": [ "Software Development", "Software Engineering Domain", "Distributed Systems" ] }, { "id": "agreement-protocol", "title": "Agreement Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Agreement Protocol is a distributed protocol specifically designed to enable multiple independent nodes to reach consensus on a single value, decision, or sequence of events despite the presence of failures, network asynchrony, and potentially malicious participants. Agreement protocols must satisfy safety (all honest nodes agree on the same value), liveness (the protocol eventually terminates), and validity (the agreed value was proposed by some participant), and their design space is constrained by impossibility results such as the FLP theorem and Byzantine Agreement bounds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:agreement-protocol", "labels": [ "Agreement Protocol" ], "is_subclass_of": [ "Protocol and Consensus", "Distributed Protocol" ], "wikilinks": [ "BFT Consensus", "Byzantine Agreement", "Byzantine Agreement Papers", "Crash Fault Tolerant Agreement", "FLP Impossibility", "Paxos", "Paxos and Raft Consensus", "PBFT", "PBFT Algorithm", "BlockchainDomain", "ConceptualLayer", "Consensus Mechanism", "Distributed Protocol", "Nakamoto Consensus", "Telecollaboration" ] }, { "id": "agricultural-robot", "title": "Agricultural Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Agricultural Robot - A semi-autonomous or fully autonomous platform equipped with Precision Agriculture Sensors, Manipulation Modules, and Crop Analysis Systems for performing field operations including planting, selective weeding, harvesting, and health monitoring whilst minimising s...", "entityType": "Class", "qualityScore": 0.53, "maturity": "draft", "iri": "urn:ngm:class:agricultural-robot", "labels": [ "Agricultural Robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot", "Robotics", "Service Robot" ], "wikilinks": [ "Crop Analysis Systems", "Crop Quality Improvement", "Environmental Sensors", "ISO 18497", "ISO 18497:2018", "ISO 8373:2021", "Labour Cost Reduction", "Manipulation Modules", "Precision Agriculture Sensors", "Precision Agriculture System", "Sustainable Farming", "Terrain Navigation", "Computer Vision", "Environmental Sustainability", "Mobile Robot", "Robotics", "RoboticsDomain", "Service Robot", "Soft Robotics" ] }, { "id": "agricultural-robotics", "title": "agricultural robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Agricultural robotics is the engineering discipline concerned with the design, deployment, and operation of autonomous and semi-autonomous robotic systems applied to crop and livestock farming tasks, including soil preparation, precision planting, selective harvesting, agrochemical application, irrigation management, and herd monitoring. These systems integrate computer vision, machine learning, GNSS positioning, and mechatronic manipulation to function reliably in unstructured outdoor environments subject to variable terrain, lighting, weather, and biological variability. The field is a primary enabler of precision agriculture, supporting reductions in agrochemical inputs, labour dependency, and resource waste while maintaining or increasing yields. Commercial deployment spans ground-based field robots, unmanned aerial vehicles, and autonomous underwater and amphibious systems for aquaculture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:agricultural-robotics", "labels": [ "Agricultural Robotics" ], "is_subclass_of": [ "Autonomous Robot" ], "wikilinks": [] }, { "id": "ai-agent-identity", "title": "Ai Agent Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A persistent, verifiable digital identity assigned to an autonomous AI agent, enabling it to authenticate, sign transactions, and participate in decentralised systems on behalf of itself or a principal. AI agent identity encompasses credential issuance, key management, and delegation of authority scoped to the agent's operational context. Such identities may be anchored on a blockchain ledger to ensure auditability and prevent impersonation. They bridge self-sovereign identity (SSI) principles with agentic software systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:ai-agent-identity", "labels": [ "Ai Agent Identity", "AI Agent Identity" ], "is_subclass_of": [ "Blockchain", "Verifiable Credentials" ], "wikilinks": [] }, { "id": "ai-art-categorization", "title": "Ai Art Categorization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Art Categorization encompasses taxonomies, classification systems, and machine learning mods for organizing, labeling, and evaluating AI-generated and AI-assisted artworks based on creation modology, style, medium, aesthetic properties, and the degree of autonomous system involvement in the cr...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-art-categorization", "labels": [ "Ai Art Categorization" ], "is_subclass_of": [ "AI Application", "Art Classification" ], "wikilinks": [ "Art Classification", "Art Curation", "Art Historical Knowledge", "Authenticity Detection", "Style Analysis", "Computer Vision", "Deep Learning", "metaverse" ] }, { "id": "ai-generated-content", "title": "Ai Generated Content", "domain": "governance", "domain_name": "Governance", "definition": "AI Generated Content (AIGC) refers to text, images, audio, video, code, or other media artefacts produced autonomously or semi-autonomously by artificial intelligence systems, particularly large generative models. The content may be indistinguishable from human-created output and spans creative, informational, and functional applications. Its proliferation raises governance questions around attribution, intellectual property, misinformation, and authenticity verification. Regulatory frameworks are evolving to mandate disclosure and provenance tracking of AI-generated material.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:ai-generated-content", "labels": [ "Ai Generated Content", "AI Generated Content", "AI-Generated Content" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "ai-governance-principle", "title": "Ai Governance Principle", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A normative rule or guideline that shapes the design, development, deployment, and oversight of AI systems to ensure they remain safe, fair, transparent, and accountable. AI governance principles operationalise ethical commitments into actionable standards, informing policy frameworks such as the OECD AI Principles and the EU AI Act, and guiding organisational decision-making throughout the AI lifecycle.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:ai-governance-principle", "labels": [ "AI Governance Principle" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AI Governance", "AI Governance and Ethics", "AI Safety", "Fairness", "Transparency", "Accountability", "AI Governance Framework", "AI Alignment", "Responsible AI", "Trustworthy AI", "AI Ethics", "OECD AI Principles", "EU AI Act", "NIST AI RMF", "ISO IEC 42001", "Human Oversight", "Explainable AI", "Risk Management", "Responsible AI Principles", "Value Alignment" ] }, { "id": "ai-machine-learning", "title": "Ai Machine Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Machine Learning in the metaverse context refers to the application of artificial intelligence algorithms, neural networks, and deep learning architectures that enable intelligent virtual environments through natural language processing, computer vision, procedural content generation, and adap...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-machine-learning", "labels": [ "Ai Machine Learning", "AI & Machine Learning" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Adaptive NPCs", "Computational Infrastructure", "Content Generation", "Neural Network Architectures", "Personalized Experiences", "Smart Virtual Environments", "Artificial Intelligence", "Computer Vision", "metaverse", "Training Data" ] }, { "id": "ai-origin-declaration", "title": "Ai Origin Declaration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Origin Declaration refers to technical standards and regulatory requirements for disclosing when digital content has been generated or substantially modified by artificial intelligence, encompassing watermarking, metadata embedding, and provenance tracking systems that enable verification of c...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-origin-declaration", "labels": [ "Ai Origin Declaration", "AI Origin Declaration" ], "is_subclass_of": [ "AI Governance and Ethics", "Content Authenticity" ], "wikilinks": [ "Content Authenticity", "Content Verification", "Cryptographic Signing", "Deepfake Detection", "Digital Watermarking", "Blockchain", "EU AI Act", "Metadata Standards", "metaverse", "Trust Infrastructure" ] }, { "id": "ai-transparency-framework", "title": "Ai Transparency Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Transparency Framework encompasses governance structures, technical standards, and regulatory requirements that ensure artificial intelligence systems are explainable, interpretable, and accountable, enabling stakeholders to understand AI decision-making processes, limitations, and potential i...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ai-transparency-framework", "labels": [ "Ai Transparency Framework", "AI Transparency" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance" ], "wikilinks": [ "Audit Mechanisms", "OECD", "Stakeholder Communication", "AI Governance", "Algorithmic Accountability", "Blockchain", "Documentation Standards", "EU AI Act", "Explainable AI", "metaverse", "Regulatory Compliance" ] }, { "id": "air-gap", "title": "Air Gap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An air gap is a security measure that physically isolates a computer or storage device from unsecured networks, including the public internet and any networked device. In the context of cryptocurrency custody, an air-gapped device holds private keys and signs transactions while never establishing a live network connection, transferring data only through media such as QR codes or removable storage. This isolation dramatically reduces the remote attack surface available to adversaries.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:air-gap", "labels": [ "Air Gap" ], "is_subclass_of": [ "Cold Storage" ], "wikilinks": [] }, { "id": "air-quality-monitoring", "title": "Air Quality Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:air-quality-monitoring", "labels": [ "Air Quality Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "airborne-remote-sensing", "title": "Airborne Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:airborne-remote-sensing", "labels": [ "Airborne Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "airdrop", "title": "Airdrop", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Airdrop is a token distribution mechanism in which a blockchain project allocates free tokens to a set of wallet addresses, typically to bootstrap a community, reward early users or decentralise ownership. Eligibility may be based on prior on-chain activity, holdings of another asset or completion of tasks. Airdrops are a marketing and governance tool but attract Sybil attacks that the project must mitigate.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:airdrop", "labels": [ "Airdrop" ], "is_subclass_of": [ "Token Distribution" ], "wikilinks": [] }, { "id": "alan-turing-institute", "title": "Alan Turing Institute", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The United Kingdom's national institute for data science and artificial intelligence, founded in 2015 and headquartered in London, convening research across partner universities and applying AI and data science methods to challenges in health, defence, security, and the public sector.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:alan-turing-institute", "labels": [ "Alan Turing Institute" ], "is_subclass_of": [ "AI Research Area", "Entity", "Research Institute" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning", "University College London", "University of Oxford", "Entity", "Data Science", "AI Safety", "AI Ethics", "Responsible AI", "AI Governance", "UK Research and Innovation", "Defence and Security", "Digital Health", "Open Data", "Reproducible Research", "Natural Language Processing", "Computer Vision", "Probabilistic Programming", "Explainable AI", "Federated Learning" ] }, { "id": "aleo", "title": "Aleo", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Aleo is a blockchain platform designed for private applications using zero-knowledge proofs to keep transaction data confidential while remaining verifiable. Programs are written in its Leo language.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:aleo", "labels": [ "Aleo" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [ "Zero-Knowledge Proof", "Cryptography", "Privacy", "Pseudonymity", "Blockchain" ] }, { "id": "alerting", "title": "Alerting", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Alerting is the observability capability that evaluates monitored signals against defined conditions and notifies responsible humans or automated systems when those conditions indicate a problem or an impending one. It converts continuous telemetry into discrete, actionable notifications routed to the appropriate on-call recipient. Effective alerting balances sensitivity against noise so that every alert is meaningful and timely.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:alerting", "labels": [ "Alerting" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "alethea-agent", "title": "Alethea Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An autonomous AI agent designed to generate and verify novel proofs in pure mathematics, integrated into Google's Deep Think model.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:alethea-agent", "labels": [ "Alethea Agent" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "algorand", "title": "Algorand", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Algorand is a permissionless, pure proof-of-stake Layer 1 blockchain protocol designed by MIT cryptographer Silvio Micali, providing instant transaction finality, high throughput, and carbon-negative operation through a cryptographically random committee-selection consensus mechanism. It resolves the blockchain trilemma of security, scalability, and decentralisation without forks by ensuring all confirmed blocks are final. The Algorand Virtual Machine (AVM) supports smart contract execution in TEAL bytecode and, via ARC standards, enables NFTs, DeFi, and tokenised assets at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:algorand", "labels": [ "Algorand" ], "is_subclass_of": [ "Blockchain Protocol" ], "wikilinks": [] }, { "id": "algorithm-layer", "title": "Algorithm Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Algorithm Layer is the architectural stratum that defines the computational methods, procedures, and decision logic operating above the data and protocol layers. It encompasses the design, selection, and composition of algorithms \u2014 sorting, search, optimisation, consensus, cryptographic, and learning algorithms \u2014 that transform inputs into outputs under defined correctness and complexity guarantees.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:algorithm-layer", "labels": [ "Algorithm Layer" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence", "Algorithm" ], "wikilinks": [ "Data Layer", "Compute Layer", "Application Layer", "Inference Layer", "Computational Complexity Theory", "IEEE (Institute of Electrical and Electronics Engineers)", "Algorithm", "Learning Algorithm", "Model Architecture Layer", "Model Layer", "Data Structure", "Gradient Descent", "Backpropagation", "Sorting Algorithm", "Graph Algorithms", "Search Algorithm", "Optimisation", "Consensus Algorithm", "Cryptographic Algorithm", "Neural Architecture Search" ] }, { "id": "algorithm", "title": "Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A finite, deterministic sequence of instructions or rules that solves a computational problem or performs a transformation on data. In AI and blockchain contexts algorithms encompass learning procedures, consensus rules, cryptographic primitives, and optimisation methods that underpin intelligent systems and distributed ledgers.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:algorithm", "labels": [ "Algorithm", "Algorithm Implementation", "Indexing Algorithm" ], "is_subclass_of": [ "AI Technique", "Computational Model" ], "wikilinks": [ "Artificial Intelligence", "Blockchain", "Gradient Descent", "Backpropagation", "Data Structure", "Computational Complexity", "Machine Learning", "Deep Learning", "Neural Network", "Heuristic Methods", "Optimisation", "Search Algorithm", "Sorting Algorithm", "Graph Algorithm", "Dynamic Programming", "Divide and Conquer", "Greedy Algorithm", "Turing Machine", "Inference", "Cryptography" ] }, { "id": "algorithmic-accountability", "title": "Algorithmic Accountability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Algorithmic Accountability is a responsibility framework ensuring that AI systems and their developers are answerable for decisions, outcomes, and societal impacts produced by algorithmic processes. It encompasses mechanisms for redress, transparency, auditing, and oversight to prevent undue harm from automated decision-making.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:algorithmic-accountability", "labels": [ "Algorithmic Accountability", "Algorithmic Accountability Rules", "Algorithmic Accountability System" ], "is_subclass_of": [ "AI Governance and Ethics", "AIGovernancePrinciple", "EthicalFramework", "RegulatoryCompliance" ], "wikilinks": [ "AIGovernancePrinciple", "IEEE P2863", "AIEthicsDomain", "ConceptualLayer", "EthicalFramework", "RegulatoryCompliance", "Smart Contract", "Algorithmic Bias", "AI Fairness", "AI Governance", "Data Protection", "Transparency", "Explainability", "Audit Trail", "Data Governance", "Fairness", "Audit Mechanism", "Redress Mechanism", "Impact Assessment", "Responsible AI" ] }, { "id": "algorithmic-auditing", "title": "Algorithmic Auditing", "domain": "governance", "domain_name": "Governance", "definition": "Algorithmic auditing is the systematic evaluation of automated decision-making systems to assess their fairness, accuracy, transparency, and compliance with ethical and legal standards. It involves independent or internal review of training data, model architectures, outputs, and operational impacts. The discipline has emerged as a response to concerns about bias, discrimination, and opacity in AI-driven systems. Audits may be prospective, examining systems before deployment, or retrospective, investigating outcomes in production. Results are used to inform regulation, remediation, and public accountability.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:algorithmic-auditing", "labels": [ "Algorithmic Auditing", "AI Auditing", "Algorithmic Audit", "Algorithmic Bias Auditing", "Continuous Auditing" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "algorithmic-bias-and-variance", "title": "Algorithmic Bias and Variance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Algorithmic Bias and Variance denotes the canonical decomposition of supervised-learning generalisation error into three orthogonal components \u2014 squared bias, variance, and irreducible noise \u2014 formalised by Geman, Bienenstock & Doursat (1992) as Err(x) = E[(y \u2212 f\u0302(x))\u00b2] = (E[f\u0302(x)] \u2212 f(x))\u00b2 + E[(...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:algorithmic-bias-and-variance", "labels": [ "Algorithmic Bias and Variance" ], "is_subclass_of": [ "AI Technique", "Neural Network", "Model Evaluation", "Statistical Learning Theory", "Generalisation Theory", "Supervised Learning" ], "wikilinks": [ "AlgorithmLayer", "Anthony Bartlett 1999 Neural Network Learning Theoretical Foundations", "Bartlett Long Lugosi Tsigler 2020 Benign Overfitting", "Bartlett Mendelson 2002 Rademacher Gaussian Complexities", "Bayesian Decision Theory", "Belkin Hsu Ma Mandal 2019 Double Descent", "Benign Overfitting", "Bias Component", "Bootstrap", "Bootstrap Resampling", "Breiman 1996 Bagging", "Breiman 2001 Random Forests", "Capacity Control", "Chen Guestrin 2016 XGBoost", "Concentration Inequalities", "Cross-Validation", "Double Descent", "Empirical Process Theory", "Ensemble Learning", "EvaluationLayer" ] }, { "id": "algorithmic-bias", "title": "Algorithmic Bias", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Systematic and repeatable errors in AI systems that create unfair outcomes favouring or discriminating against particular groups or individuals. Bias manifests through historical bias, representation gaps, measurement proxies, aggregation errors, and feedback loops, and is detected through statistical auditing, counterfactual testing, and fairness metrics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:algorithmic-bias", "labels": [ "Algorithmic Bias", "Algorithmic Bias Detection" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "IEEE P7003-2021", "ISO/IEC TR 24027", "NIST SP 1270", "AIEthicsDomain", "ConceptualLayer", "Smart Contract", "AI Fairness", "Algorithmic Fairness", "Algorithmic Accountability", "Bias Mitigation Techniques", "Algorithmic Auditing", "Responsible AI", "AI Ethics", "AI Governance", "Training Data", "Machine Learning", "Explainability", "Differential Privacy", "Data Governance", "Automated Decision-Making" ] }, { "id": "algorithmic-capture", "title": "Algorithmic Capture", "domain": "governance", "domain_name": "Governance", "definition": "Algorithmic Capture describes the process by which recommendation and ranking algorithms come to dominate the flow of attention, shaping what content people see and produce until human discretion is largely supplanted. By optimizing for engagement, these systems can homogenize discourse, amplify synthetic or low-quality content, and entrench platform control over information ecosystems. It is a central mechanism in narratives about the decline and automation of the open internet.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-capture", "labels": [ "Algorithmic Capture" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "algorithmic-complexity", "title": "Algorithmic Complexity", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Algorithmic complexity is the study of how the computational resources required by an algorithm, principally time and memory, grow as a function of input size. It is expressed using asymptotic notation that abstracts away constant factors to characterise scaling behaviour. In machine learning it governs the feasibility of training and inference, shaping choices of model architecture, optimisation and data structures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:algorithmic-complexity", "labels": [ "Algorithmic Complexity" ], "is_subclass_of": [ "Statistical Learning Theory", "Computational Complexity Theory" ], "wikilinks": [] }, { "id": "algorithmic-fairness", "title": "Algorithmic Fairness", "domain": "governance", "domain_name": "Governance", "definition": "Algorithmic fairness is the study and engineering of machine-learning systems so that their predictions and decisions do not produce unjustified disparities across individuals or protected groups. It formalises fairness through competing mathematical criteria such as demographic parity, equalised odds and individual fairness, which cannot in general be satisfied simultaneously. The field combines measurement, bias mitigation techniques and governance to align automated decisions with ethical and legal norms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-fairness", "labels": [ "Algorithmic Fairness" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "algorithmic-framework", "title": "Algorithmic Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Algorithmic Framework is a structured collection of algorithms, data structures, and design principles that together provide a reusable computational scaffold for solving a class of related problems. It abstracts common patterns of search, optimisation, inference, and decision-making into composable building blocks that can be instantiated and extended for specific applications. Algorithmic frameworks underpin fields such as machine learning, procedural content generation, automated planning, and symbolic reasoning by providing principled, reproducible computational architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-framework", "labels": [ "Algorithmic Framework" ], "is_subclass_of": [ "AI Technique", "AI Framework" ], "wikilinks": [ "AI Framework", "Digital Twin", "Algorithm", "Optimization Algorithm", "Graph Algorithms", "Learning Algorithm", "Procedural Content Generation", "Automated Planning", "Inference", "Bayesian Inference", "Machine Learning", "Optimization Technique", "Dynamic Programming", "Monte Carlo Methods", "Evolutionary Algorithms", "Computational Complexity", "Heuristic Methods", "Reinforcement Learning", "Gradient Descent", "Search Algorithm" ] }, { "id": "algorithmic-governance", "title": "Algorithmic Governance", "domain": "governance", "domain_name": "Governance", "definition": "Algorithmic Governance is the use of automated decision systems, models, and rule engines to make or enforce governance choices that were traditionally human and discretionary, such as moderation, resource allocation, or compliance enforcement. It can increase consistency, speed, and scale, but raises concerns about transparency, accountability, bias, and contestability of automated decisions. The concept spans platform moderation, public-sector automation, and on-chain rule enforcement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-governance", "labels": [ "Algorithmic Governance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "algorithmic-impact-assessment", "title": "Algorithmic Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "An algorithmic impact assessment is a structured governance process for evaluating the potential effects of an automated decision-making or artificial-intelligence system on individuals, groups and society before and during its deployment. It documents the system's purpose, data, risks to fairness, privacy and safety, and the mitigations and oversight controls in place, producing an auditable record for accountability. Modelled on data-protection and environmental impact assessments, it is increasingly mandated by AI regulation and procurement frameworks to ensure responsible and transparent use of algorithms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-impact-assessment", "labels": [ "Algorithmic Impact Assessment", "Algorithmic Impact Assessments" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "algorithmic-layer", "title": "Algorithmic Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Algorithmic Layer is the stratum that holds the step-by-step procedures a system uses to transform inputs into outputs. In the canonical stack it corresponds to the Algorithm Layer, sitting above the Compute Layer and below the Model strata that compose its primitives. It contains algorithms, data structures, and their complexity characteristics.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-layer", "labels": [ "Algorithmic Layer" ], "is_subclass_of": [ "AI Technique", "owl:Thing" ], "wikilinks": [ "Compute Layer", "Model Architecture Layer", "Model Layer", "Computational Complexity Theory", "Data Structure", "owl:Thing" ] }, { "id": "algorithmic-stablecoin", "title": "Algorithmic Stablecoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An algorithmic stablecoin is a cryptocurrency that seeks to maintain a stable peg, typically to a fiat unit, through on-chain algorithmic supply adjustments and incentive mechanisms rather than full reserves of external collateral. Smart contracts expand or contract token supply, often using a companion volatility-absorbing token, to push the market price toward target. The model contrasts with fiat- and crypto-collateralised stablecoins and has proven fragile, with several high-profile de-peg failures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:algorithmic-stablecoin", "labels": [ "Algorithmic Stablecoin" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [] }, { "id": "algorithmic-trading", "title": "Algorithmic Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Algorithmic trading is the use of computer programs to execute trading decisions automatically according to predefined rules covering timing, price, quantity and order routing, often without human intervention. It spans execution algorithms that minimise market impact, systematic strategies that generate signals, and high-frequency trading that exploits microsecond advantages. In crypto markets it operates across centralised and decentralised venues and interacts closely with on-chain mechanisms such as maximal extractable value.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:algorithmic-trading", "labels": [ "Algorithmic Trading" ], "is_subclass_of": [ "Financial Trading" ], "wikilinks": [] }, { "id": "algorithmic-transparency-index", "title": "Algorithmic Transparency Index", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured metrics framework for measuring and evaluating the explainability, documentation, and disclosure levels of AI algorithms and automated decision-making systems across multiple transparency dimensions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:algorithmic-transparency-index", "labels": [ "Algorithmic Transparency Index" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance Framework", "Algorithmic Accountability System", "Performance Metrics" ], "wikilinks": [ "AI Accountability", "Algorithmic Accountability System", "Algorithmic Auditing", "Audit Mechanism", "Automated Decision System", "Bias Detection Metrics", "Content Moderation System", "Decision Logging", "Disclosure Requirements", "Explainability Metrics", "IEEE 7001-2021", "Model Documentation", "NIST AI Risk Management Framework", "Recommendation System", "Stakeholder Trust", "AI Governance Framework", "Audit Trail", "Blockchain", "ComputationAndIntelligenceDomain", "Data Provenance" ] }, { "id": "algorithmic-transparency-reports", "title": "Algorithmic Transparency Reports", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Algorithmic Transparency Reports are periodic public disclosures that document AI system characteristics, performance metrics, governance practices, and accountability mechanisms, enabling external stakeholders, regulators, and affected communities to scrutinise how algorithmic decision-making systems operate and impact individuals. Core sections include system purpose and deployment scope, disaggregated performance and fairness metrics, bias and disparate-impact analyses, data governance practices, explainability provisions, incident and remediation records, and stakeholder engagement summaries. Reports balance transparency objectives against proprietary-information protection and adversarial-exploitation risks, and align with requirements in EU AI Act Article 13 (high-risk system transparency), Platform-to-Business Regulation disclosure obligations, and voluntary civil-society commitments led by organisations such as the AI Now Institute.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:algorithmic-transparency-reports", "labels": [ "Algorithmic Transparency Reports" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics", "AI Governance Framework", "Algorithmic Transparency", "Accountability Framework" ], "wikilinks": [ "AI Now Institute", "Platform-to-Business Regulation", "AIEthicsDomain", "ConceptualLayer", "EU AI Act", "Smart Contract" ] }, { "id": "algorithmic-transparency", "title": "Algorithmic Transparency", "domain": "governance", "domain_name": "Governance", "definition": "The principle and practice of making the logic, data inputs, decision criteria, and outputs of algorithmic systems sufficiently accessible and comprehensible to affected stakeholders, oversight bodies, and the general public. It encompasses both technical disclosure\u2014publishing model architectures, training data provenance, and evaluation results\u2014and process-level disclosure of how algorithms are developed, audited, and governed. Algorithmic transparency is recognised as a foundational requirement for accountability and trust in automated decision-making systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:algorithmic-transparency", "labels": [ "Algorithmic Transparency" ], "is_subclass_of": [ "Explainability" ], "wikilinks": [] }, { "id": "alignment-research", "title": "Alignment Research", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The research programme dedicated to ensuring that increasingly capable AI systems reliably pursue their designers' and society's intended goals, spanning empirical techniques such as reinforcement learning from human feedback, constitutional AI, scalable oversight and mechanistic interpretability, alongside theoretical work on agent foundations, corrigibility, deceptive alignment and mesa-optimisation; a field whose importance scales with model capability and which anchors the safety agendas of frontier laboratories and a growing academic community.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:alignment-research", "labels": [ "Alignment Research" ], "is_subclass_of": [ "Artificial Intelligence Research" ], "wikilinks": [ "Artificial Intelligence Research", "AI Alignment", "AI Safety", "Artificial General Intelligence" ] }, { "id": "alignment-techniques", "title": "Alignment Techniques", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Alignment techniques are the methods used to make an AI system's behaviour conform to intended human goals, values, and constraints, particularly as capability increases. They include reinforcement learning from human feedback, constitutional methods, red-teaming, and interpretability-informed fine-tuning. Alignment techniques are a core requirement for building trustworthy conversational AI systems and are considered essential on the path toward artificial general intelligence.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:alignment-techniques", "labels": [ "Alignment Techniques" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [] }, { "id": "alignment", "title": "Alignment", "domain": "ai", "domain_name": "Ai", "definition": "The technical and philosophical research programme aimed at ensuring that AI systems reliably pursue goals, exhibit behaviours, and produce outcomes that accord with human values, intentions, and oversight requirements. Alignment addresses the fundamental challenge that learned objectives may diverge from intended objectives\u2014a problem that becomes increasingly consequential as AI systems grow more capable and autonomous. The field encompasses specification of human preferences, training methods that instil those preferences, and verification techniques that confirm alignment properties are preserved at deployment.", "entityType": "Class", "qualityScore": 0.92, "maturity": "emerging", "iri": "urn:ngm:class:alignment", "labels": [ "Alignment", "Alignment Stage", "Ethical Alignment" ], "is_subclass_of": [ "AI Alignment", "AI Safety Research" ], "wikilinks": [ "AI Alignment", "AI Safety", "AI Safety Research", "Reinforcement Learning from Human Feedback", "Constitutional AI", "Scalable Oversight", "Reward Modelling", "Interpretability", "Mechanistic Interpretability", "Red Teaming", "Model Evaluation", "Reward Hacking", "Frontier AI", "Large Language Model", "Existential Risk", "AI Governance", "AI Governance Framework", "AI Regulation", "Trustworthy AI", "Value Alignment" ] }, { "id": "allo-protocol", "title": "Allo Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Allo Protocol is an on-chain framework developed by Gitcoin for allocating capital, supporting funding mechanisms such as quadratic funding and grants.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:allo-protocol", "labels": [ "Allo Protocol" ], "is_subclass_of": [ "Blockchain", "Decentralized Finance" ], "wikilinks": [ "Token", "Web3 Infrastructure", "Interoperability", "Blockchain", "https://allo.gitcoin.co/", "https://docs.allo.gitcoin.co/" ] }, { "id": "alphafold", "title": "AlphaFold", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AlphaFold is a deep learning system developed by DeepMind that predicts the three-dimensional structure of a protein from its amino acid sequence with accuracy approaching that of experimental methods such as X-ray crystallography, a problem known as the protein folding problem. It uses an attention-based neural network architecture trained on known protein structures and evolutionary sequence data to output per-residue coordinates along with a confidence estimate. AlphaFold's predictions are widely used in computational biology and genomics research, and its structure database has substantially accelerated work in drug discovery and molecular biology.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:alphafold", "labels": [ "AlphaFold" ], "is_subclass_of": [ "Deep Learning" ], "wikilinks": [] }, { "id": "alpine-tundra", "title": "Alpine Tundra", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:alpine-tundra", "labels": [ "Alpine Tundra" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "altspace-vr", "title": "AltspaceVR", "domain": "metaverse", "domain_name": "Metaverse", "definition": "AltspaceVR was a social virtual reality platform offering avatar-based gatherings, events and shared spaces, later acquired by Microsoft and eventually shut down.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:altspace-vr", "labels": [ "AltspaceVR" ], "is_subclass_of": [ "Metaverse", "Metaverse Domain" ], "wikilinks": [ "Avatar System", "Virtual World", "Virtual Reality", "Microsoft Mesh", "Metaverse Domain" ] }, { "id": "alu-vm", "title": "AluVM", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A purely functional, register-based virtual machine designed for deterministic execution of smart-contract validation logic within client-side validated Bitcoin protocol layers, notably the RGB Protocol. AluVM operates without mutable global state, using a RISC-like instruction set that guarantees bounded execution time and reproducible results across heterogeneous computing environments. Its architecture is specifically optimised for the constraints of client-side validation, where contract logic must execute identically across all validating parties without access to a shared blockchain state machine.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:alu-vm", "labels": [ "AluVM", "AluVM Specification" ], "is_subclass_of": [ "Virtual Machine" ], "wikilinks": [] }, { "id": "ambient-awareness", "title": "Ambient Awareness", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Ambient Awareness is the peripheral, low-effort sense of others' presence, activity, and availability that collaborative systems convey through subtle continuous signals rather than explicit notifications. It is realized through presence indicators, status cues, activity streams, and spatial audio in shared digital workspaces. By keeping collaborators loosely informed of each other's context, it supports coordination and a feeling of co-location in distributed teams.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ambient-awareness", "labels": [ "Ambient Awareness" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "ambient-computing", "title": "Ambient Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A paradigm in which computing capability is invisibly embedded throughout the physical environment, enabling technology to sense, infer, and respond to human needs without requiring deliberate interaction with discrete devices. Ambient computing dissolves the boundary between the digital and physical worlds by distributing processing, sensing, and actuation across interconnected objects, surfaces, and spaces that operate continuously in the background. It represents the convergence of pervasive connectivity, miniaturised hardware, and AI-driven contextual inference into an always-present computational substrate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ambient-computing", "labels": [ "Ambient Computing", "Ambient Intelligence" ], "is_subclass_of": [ "Pervasive Computing" ], "wikilinks": [] }, { "id": "ambient-observation", "title": "Ambient Observation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A paradigm for AI learning where the system passively acquires knowledge and patterns from background user interactions and environment data without explicit instruction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ambient-observation", "labels": [ "Ambient Observation" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [] }, { "id": "ambient-occlusion", "title": "Ambient Occlusion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A shading and rendering technique that approximates the degree to which each point on a surface is occluded from ambient environmental light by surrounding geometry, producing soft shadows in crevices, corners, and contact areas that significantly enhance the perception of three-dimensional form and material grounding. Unlike direct illumination algorithms, ambient occlusion is view-independent and operates on the assumption that ambient light arrives uniformly from all directions, making it computationally tractable as either a pre-baked texture or a real-time screen-space approximation. It is a foundational element of physically plausible rendering pipelines.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:ambient-occlusion", "labels": [ "Ambient Occlusion", "Screen Space Ambient Occlusion" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "ambisonics", "title": "Ambisonics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Ambisonics is a full-sphere surround sound technique that represents a sound field independently of any specific loudspeaker layout by encoding it into spherical harmonic components. A captured or synthesised scene is stored as B-format channels and later decoded to an arbitrary speaker array or to binaural headphones, allowing the same recording to be rendered for many playback configurations. It is widely used for immersive and head-tracked audio in virtual and augmented reality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ambisonics", "labels": [ "Ambisonics" ], "is_subclass_of": [ "Spatial Audio" ], "wikilinks": [] }, { "id": "amd-sev", "title": "Amd Sev", "domain": "security", "domain_name": "Security", "definition": "AMD SEV (Secure Encrypted Virtualisation) is a hardware security technology that encrypts the memory of individual virtual machines using per-VM keys managed by an on-chip security processor, isolating guest memory from the hypervisor and other VMs. Extensions add register-state encryption and integrity protection with attestation, enabling confidential virtual machines whose contents are protected even from a privileged host. It is a leading approach to confidential computing in cloud environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:amd-sev", "labels": [ "Amd Sev", "AMD SEV" ], "is_subclass_of": [ "Confidential Computing" ], "wikilinks": [] }, { "id": "analog-to-digital-converter", "title": "Analog To Digital Converter", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Analog To Digital Converter (ADC) is an electronic component that converts continuous analog signals such as sound, light, or voltage into discrete digital representations through sampling, quantization, and encoding processes, enabling digital processing, storage, and transmission of real-world ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:analog-to-digital-converter", "labels": [ "Analog To Digital Converter", "Analog-to-Digital Conversion", "Analog-to-Digital Converter", "AnalogDigitalConverter", "Analogue-to-Digital Converter" ], "is_subclass_of": [ "Display and Rendering", "Signal Processing Hardware" ], "wikilinks": [ "Digital Audio Recording", "Digital Imaging", "Resolution Bits", "Sampling Rate", "Sensor Data Processing", "Sensor Input", "Signal Conditioning", "Signal Processing Hardware", "metaverse" ] }, { "id": "analogue-to-digital-conversion", "title": "Analogue To Digital Conversion", "domain": "robotics", "domain_name": "Robotics", "definition": "Analogue-to-digital conversion is the process of transforming a continuous physical signal, such as voltage from a sensor, into a discrete sequence of numerical values that a digital system can store and process. It proceeds by sampling the signal at regular intervals and quantising each sample to a finite set of levels, with the sampling rate and bit depth determining how faithfully the original is represented. In robotics and embedded systems it is the essential bridge between the analogue physical world and digital control and perception pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:analogue-to-digital-conversion", "labels": [ "Analogue To Digital Conversion", "Analogue-to-Digital Conversion" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "analytics-engine", "title": "Analytics Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A software system or distributed platform that ingests, processes, and analyses large volumes of structured and unstructured data to extract insights, detect patterns, and support decision-making. Analytics engines abstract the complexity of distributed query execution, storage management, and statistical computation behind APIs and query interfaces, enabling analysts and applications to perform exploratory, diagnostic, predictive, and prescriptive analysis at scale. They serve as the computational backbone of business intelligence platforms, ML pipelines, and real-time operational monitoring systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:analytics-engine", "labels": [ "Analytics Engine", "Data Analytics Engine" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "anchor-based-detection", "title": "Anchor-Based Detection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Anchor-based detection is an object detection approach that predicts bounding boxes and class scores relative to a fixed set of predefined reference boxes, called anchors, tiled densely across the image at multiple scales and aspect ratios. The network learns offsets that adjust each anchor to fit a nearby object rather than predicting box coordinates directly, which stabilises training and improves recall for objects of varying size. Faster R-CNN, SSD and the early YOLO versions are canonical anchor-based detectors, later contrasted with anchor-free alternatives.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:anchor-based-detection", "labels": [ "Anchor-Based Detection" ], "is_subclass_of": [ "Object Detection" ], "wikilinks": [] }, { "id": "anchorage-digital", "title": "Anchorage Digital", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A United States digital asset platform that holds a federal bank charter and provides custody, trading, staking, and financing services for institutional clients. It was the first crypto firm to receive a national trust bank charter from the Office of the Comptroller of the Currency.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:anchorage-digital", "labels": [ "Anchorage Digital" ], "is_subclass_of": [ "Self-Custody" ], "wikilinks": [ "Regulatory Compliance", "Proof of Stake", "Coinbase", "Self-Custody" ] }, { "id": "anemometer", "title": "Anemometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:anemometer", "labels": [ "Anemometer" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "angular-coordinate", "title": "Angular Coordinate", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:angular-coordinate", "labels": [ "Angular Coordinate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "animation-clip", "title": "Animation Clip", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An animation clip is a discrete, reusable unit of animation data, typically a time-ordered sequence of keyframes or pose samples for a skeleton or property set, such as a walk cycle or a wave gesture. Animation controllers and state machines reference clips as the atomic assets they blend, sequence and transition between at runtime. Clips are commonly authored once and retargeted across multiple character rigs that share a compatible skeleton.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:animation-clip", "labels": [ "Animation Clip" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "animation-controller", "title": "Animation Controller", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An animation controller is a software component that manages the selection, blending, and sequencing of animation clips for a character or object in real-time interactive environments. It typically implements a state machine model in which transitions between animation states are governed by parametric conditions such as velocity, input events, or AI signals. Animation controllers sit between high-level game logic and the low-level skeletal animation runtime, abstracting away the complexity of blend trees, IK passes, and additive layers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:animation-controller", "labels": [ "Animation Controller" ], "is_subclass_of": [ "Controller" ], "wikilinks": [] }, { "id": "animation-retargeting", "title": "Animation Retargeting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Animation Retargeting is the process of transferring motion capture or animation data from one character skeleton to another with different proportions, joint configurations, or bone structures, enabling reuse of animation assets across diverse character models in games, film, and virtual environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:animation-retargeting", "labels": [ "Animation Retargeting" ], "is_subclass_of": [ "Content and Assets", "Animation Technology" ], "wikilinks": [ "Animation Technology", "Avatar Animation", "Character Animation Reuse", "Motion Capture Workflows", "Pose Matching", "Skeleton Mapping", "Computer Vision", "Inverse Kinematics", "metaverse" ] }, { "id": "animation-rig", "title": "Animation Rig", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An animation rig is a hierarchical system of bones, controls, and constraints applied to a 3D character or object to enable articulated movement for animation. It abstracts underlying mesh deformation through a control interface that animators manipulate, translating high-level poses into low-level vertex transformations. Rigs range from simple skeletal setups to complex systems incorporating inverse kinematics, blend shapes, and procedural dynamics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:animation-rig", "labels": [ "Animation Rig", "Blendshape Rig" ], "is_subclass_of": [ "Skeletal Animation" ], "wikilinks": [] }, { "id": "animation-software", "title": "Animation Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Animation Software encompasses digital tools and applications for creating computer-generated moving images through 3D modeling, rigging, motion graphics, rendering, and compositing, supporting the complete animation pipeline from asset creation to final output for games, film, and metaverse content.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:animation-software", "labels": [ "Animation Software" ], "is_subclass_of": [ "Content and Assets", "Digital Content Creation Tools" ], "wikilinks": [ "3D Character Animation", "Digital Content Creation Tools", "GPU Rendering", "Motion Graphics", "Rigging Systems", "3D Modeling", "Computer Vision", "metaverse", "Visual Effects" ] }, { "id": "animation-standard", "title": "Animation Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A technical specification defining how skeletal rigs, keyframes, blend shapes, and motion data are represented and exchanged so that animated characters and objects behave consistently across metaverse platforms, game engines, and spatial computing runtimes, supporting interoperable avatar animation and real-time playback.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:animation-standard", "labels": [ "Animation Standard" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "animation-state-machine", "title": "Animation State Machine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An animation state machine is a graph of discrete animation states and the transitions between them, used in real-time engines to drive a character or object's motion based on game logic, input and parameters. Each state references a clip or blend tree, and transitions carry conditions, blend durations and interruption rules that govern how one motion flows into another. It separates high-level behaviour authoring from low-level pose evaluation, making locomotion and action systems tractable to build and tune.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:animation-state-machine", "labels": [ "Animation State Machine" ], "is_subclass_of": [ "State Machine" ], "wikilinks": [] }, { "id": "animation-technique", "title": "Animation Technique", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Animation Technique is a method or algorithm for generating the illusion of motion in digital content, encompassing keyframe interpolation, skeletal rigging, physics-based simulation, and procedural approaches. Animation techniques are fundamental to character representation, environmental dynamics, and interactive responsiveness in spatial computing environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:animation-technique", "labels": [ "Animation Technique" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "animation", "title": "Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Animation is the technique of creating the illusion of movement by rapidly displaying a sequence of static images or by computationally interpolating between keyframe states of a scene or character over time. In digital contexts, animation encompasses skeletal rigging, keyframe interpolation, physics simulation, and procedural generation to produce lifelike motion in real-time or pre-rendered environments. Modern animation pipelines integrate motion capture data, inverse kinematics, and blend trees to deliver nuanced character performances. Animation is foundational to game engines, virtual reality, film, and interactive simulations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:animation", "labels": [ "Animation", "2D Animation", "Animation Pipeline", "Animation Technology" ], "is_subclass_of": [ "Animation Technique" ], "wikilinks": [] }, { "id": "annotated-dataset", "title": "Annotated Dataset", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A dataset in which each example carries human- or machine-assigned labels \u2014 class names, bounding boxes, segmentation masks, transcripts, action-unit codes, or relevance judgements \u2014 produced under a documented annotation scheme with quality controls such as inter-annotator agreement; annotated datasets are the primary fuel of supervised learning, and their coverage, label accuracy, and demographic balance bound the accuracy and fairness of every model trained on them.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:annotated-dataset", "labels": [ "Annotated Dataset" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [ "Dataset", "Data Annotation", "Supervised Learning", "Active Learning" ] }, { "id": "annotated-training-data", "title": "Annotated Training Data", "domain": "data", "domain_name": "Data", "definition": "Annotated Training Data is a dataset whose examples have been augmented with ground-truth labels, bounding boxes, segmentation masks, or other targets that supervised models learn to predict. The annotations are produced by humans, programmatic rules, or model-assisted labeling and define the task the model is trained to solve. Its quality, coverage, and label consistency are primary determinants of supervised-model performance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:annotated-training-data", "labels": [ "Annotated Training Data", "Annotated Dataset", "Annotated Hand Dataset", "Labelled Training Data" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique" ], "wikilinks": [] }, { "id": "anoma", "title": "Anoma", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Anoma is a protocol and architecture for intent-centric and privacy-preserving decentralised applications. It allows users to express desired outcomes that the network matches and settles.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:anoma", "labels": [ "Anoma" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Cryptography", "DeFi", "Privacy", "Distributed Systems", "https://anoma.net", "https://specs.anoma.net" ] }, { "id": "anomaly-detection", "title": "anomaly detection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Anomaly detection is a machine learning and statistical discipline concerned with identifying observations, sequences, or structural patterns that deviate significantly from a learned or assumed norm, signalling potential faults, threats, or novel phenomena. It operates across three principal modes: point anomaly detection (a single observation is outlying relative to the full dataset), contextual anomaly detection (an observation is anomalous given its local context, such as a transaction at an unusual time of day), and collective anomaly detection (a subsequence or group of observations is jointly anomalous relative to expected behaviour). The field draws on statistical modelling, machine learning, and signal processing to serve applications ranging from fraud detection and network intrusion detection to industrial fault monitoring, medical diagnostics, and log analysis.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:anomaly-detection", "labels": [ "Anomaly Detection", "AI Anomaly Detection" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique", "Machine Learning" ], "wikilinks": [ "Machine Learning", "Statistical Modelling", "Pattern Recognition", "Deep Learning", "Autoencoder", "Supervised Learning", "Classification", "Feature Engineering", "Data Preprocessing", "Fraud Detection", "Cybersecurity", "Predictive Maintenance", "Intrusion Detection System", "Time Series Forecasting", "Concept Drift", "Model Monitoring", "Graph Neural Network", "LSTM", "Transformer Architecture", "Variational Autoencoder" ] }, { "id": "anon-creds", "title": "AnonCreds", "domain": "security", "domain_name": "Security", "definition": "AnonCreds (Anonymous Credentials) is a verifiable credential format and specification originally developed by the Hyperledger Indy project that enables privacy-preserving identity verification. The scheme allows holders to prove possession of credentials without revealing the credential itself or the issuer's signature, using zero-knowledge proofs. AnonCreds supports selective disclosure, letting a holder share only specific attributes, and predicates that prove a claim (e.g. age over 18) without revealing the underlying value. The specification has been standardised via the AnonCreds Working Group under the Decentralized Identity Foundation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:anon-creds", "labels": [ "AnonCreds", "AnonCreds Specification v1.0" ], "is_subclass_of": [ "Credential Format Standard" ], "wikilinks": [] }, { "id": "anonymisation", "title": "Anonymisation", "domain": "security", "domain_name": "Security", "definition": "Anonymisation is the process of transforming data so that individuals can no longer be identified, directly or by inference, while preserving enough utility for analysis. Techniques include suppression, generalization, pseudonymisation, k-anonymity, differential privacy, and the blurring or synthetic replacement of faces and identifiers in media. Effective anonymisation must resist re-identification through linkage with auxiliary data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:anonymisation", "labels": [ "Anonymisation", "Data Anonymisation" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "anonymity", "title": "Anonymity", "domain": "security", "domain_name": "Security", "definition": "Anonymity is the property of an actor being unidentifiable within a set of potential actors, so that actions cannot be linked to a real-world identity. It is a core privacy goal achieved through techniques that obscure identifying attributes, network paths and metadata. Anonymity contrasts with pseudonymity, where a persistent but non-identifying handle is retained.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:anonymity", "labels": [ "Anonymity" ], "is_subclass_of": [ "Privacy" ], "wikilinks": [] }, { "id": "anonymous-communication", "title": "Anonymous Communication", "domain": "security", "domain_name": "Security", "definition": "Anonymous Communication is a set of network techniques that let parties exchange messages without revealing their identities or the link between sender and recipient to observers, typically by routing traffic through multiple relays with layered encryption. It protects against traffic analysis as well as content interception, and is a building block for privacy-preserving applications from whistleblowing platforms to private blockchain transactions. Onion routing is the most widely deployed implementation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:anonymous-communication", "labels": [ "Anonymous Communication" ], "is_subclass_of": [ "Cryptography Security and Privacy" ], "wikilinks": [] }, { "id": "anonymous-credential", "title": "Anonymous Credential", "domain": "security", "domain_name": "Security", "definition": "An anonymous credential is a cryptographic attestation that lets a holder prove possession of certified attributes to a verifier without revealing their identity or allowing their presentations to be linked. Built on techniques such as blind signatures and zero-knowledge proofs, it supports selective disclosure of individual attributes while keeping the rest private. Anonymous credentials are a core privacy-enhancing primitive for authentication and access control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:anonymous-credential", "labels": [ "Anonymous Credential" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "ansi", "title": "Ansi", "domain": "data", "domain_name": "Data", "definition": "ANSI, the American National Standards Institute, is a private non-profit organisation that oversees the development of voluntary consensus standards for products, services, processes, and systems in the United States. It does not write most standards itself but accredits standards developers and approves their output as American National Standards, and it represents the United States in international bodies such as ISO and IEC. In computing it is associated with standards such as ANSI C, ANSI SQL, and the ASCII character encoding.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ansi", "labels": [ "Ansi", "ANSI" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "answer-ai", "title": "Answer AI", "domain": "ai", "domain_name": "Ai", "definition": "Answer.AI is an applied artificial intelligence research and development laboratory co-founded in December 2023 by Jeremy Howard and Eric Ries, whose mission is to build practical, openly shared tools and techniques that make modern machine learning accessible and useful to the widest possible audience. Distinct from capability-frontier laboratories, Answer.AI explicitly does not develop new foundation models or pursue artificial general intelligence; instead it focuses on taking existing models and determining what maximally practical applications can be built with them, continuing the fast.ai tradition of pragmatic, accessibility-first research. The lab operates as a small, fully remote team of deep-technology generalists and releases open-source libraries, frameworks, and educational resources covering efficient fine-tuning, parameter-efficient adaptation, distributed training on consumer hardware, and developer ergonomics for building AI-powered applications.", "entityType": "Class", "qualityScore": 0.82, "maturity": "emerging", "iri": "urn:ngm:class:answer-ai", "labels": [ "Answer AI" ], "is_subclass_of": [ "AI Research Area", "Applied Machine Learning", "Machine Learning Discipline" ], "wikilinks": [ "fast.ai", "Jeremy Howard", "Applied Machine Learning", "Large Language Models", "Parameter-Efficient Fine-Tuning", "LoRA Fine-Tuning", "Foundation Model", "Open-Source AI", "Deep Learning", "Transfer Learning", "Distributed Training", "Python Programming Language", "Machine Learning Framework", "Instruction Tuning", "Quantisation", "AI Research", "Reinforcement Learning from Human Feedback", "Agentic AI", "Education and AI", "Deep Learning Framework" ] }, { "id": "ant-pool", "title": "AntPool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "AntPool is a Bitcoin mining pool operated in association with Bitmain. It is one of the larger pools by share of network hash rate.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ant-pool", "labels": [ "AntPool" ], "is_subclass_of": [ "Mining Pool" ], "wikilinks": [ "Bitcoin Mining", "Transaction Validation", "Bitmain", "Mining Pool", "https://www.antpool.com", "https://www.antpool.com/help" ] }, { "id": "antenna-array", "title": "Antenna Array", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Antenna Array is a set of multiple antenna elements arranged and phased together so that their combined radiation pattern can be electronically steered and shaped. By controlling the relative phase and amplitude of each element, the array forms directional beams, increases gain, and supports spatial multiplexing in techniques such as MIMO and beamforming. Arrays are foundational to modern wireless radio, 5G, radar, and precise satellite positioning receivers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:antenna-array", "labels": [ "Antenna Array" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "antenna-gain", "title": "Antenna Gain", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:antenna-gain", "labels": [ "Antenna Gain" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "anthropic-economic-index", "title": "Anthropic Economic Index", "domain": "economics", "domain_name": "Economics", "definition": "A research initiative by Anthropic that analyzes large-scale data from Claude interactions to quantify the economic impact, productivity gains, and usage patterns of AI in professional workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:anthropic-economic-index", "labels": [ "Anthropic Economic Index" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "anthropic", "title": "anthropic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Anthropic is a frontier AI safety company and public-benefit corporation founded in 2021 by Dario Amodei, Daniela Amodei, and colleagues formerly at OpenAI, whose primary mission is the responsible development and maintenance of advanced AI for the long-term benefit of humanity. The company develops and deploys the Claude family of large language models, distinguished by a Constitutional AI training methodology that uses AI self-critique and revision against explicit principles to systematically reduce harmful outputs. Anthropic's research agenda spans mechanistic interpretability, scalable oversight, red-teaming, and the Responsible Scaling Policy \u2014 a formal commitment tying capability advancement to mandatory safety evaluations at defined dangerous-capability thresholds. By mid-2026 Anthropic had reached a valuation of approximately $965 billion, filed confidential IPO documentation, and maintained partnerships with Amazon Web Services, Google, Microsoft, and Nvidia. As a public-benefit corporation, Anthropic occupies a distinctive position among frontier AI laboratories by integrating commercial model deployment with open publication of alignment and interpretability research.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:anthropic", "labels": [ "Anthropic" ], "is_subclass_of": [ "AI Safety Research", "Frontier AI" ], "wikilinks": [ "AI Safety Research", "Claude Model Family", "Constitutional AI Training Methodology", "Mechanistic Interpretability", "Responsible Scaling Policy", "Reinforcement Learning from Human Feedback", "Large Language Models", "Transformer Architecture", "Red Teaming", "AI Alignment", "Scalable Oversight", "AI Governance", "OpenAI Research Organisation", "Google DeepMind", "Meta AI", "Interpretability", "Frontier AI", "AI Governance", "AI Regulation", "Model Context Protocol" ] }, { "id": "anti-aliasing", "title": "Anti Aliasing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Anti-aliasing is a family of rendering techniques that reduce the jagged, stair-stepped edges and shimmering artefacts that arise when continuous geometry and signals are sampled onto a discrete pixel grid. By increasing effective sampling, blending edge pixels, or reconstructing from accumulated samples, anti-aliasing produces smoother, more visually faithful images. It is a standard stage of real-time and offline rendering pipelines, with methods such as supersampling, multisampling, post-process filtering, and temporal accumulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:anti-aliasing", "labels": [ "Anti Aliasing", "Anti-Aliasing", "Antialiasing" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [] }, { "id": "anti-counterfeiting", "title": "Anti Counterfeiting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based systems employing cryptographic verification, immutable ledgers, and physical-digital integration technologies (NFC tags, QR codes, RFID) to authenticate products, prevent counterfeiting, and establish verifiable supply chain provenance across pharmaceuticals, luxury goods, electronics, and consumer products.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:anti-counterfeiting", "labels": [ "Anti Counterfeiting" ], "is_subclass_of": [ "Network Component", "Supply Chain Blockchain" ], "wikilinks": [ "BC-0426-hyperledger-fabric", "BC-0432-consortium-blockchain", "BC-0441-provenance-tracking", "BC-0442-pharmaceutical-traceability", "BC-0444-luxury-goods-authentication", "BC-0446-supply-chain-traceability", "BC-0476-aml-kyc-compliance", "ISO/IEC 20248", "OECD", "Autonomous Robot", "BlockchainDomain" ] }, { "id": "anti-money-laundering", "title": "anti-money laundering", "domain": "finance", "domain_name": "Finance", "definition": "Anti-Money Laundering (AML) encompasses the legislative frameworks, regulatory obligations, and operational procedures that financial institutions and other regulated entities must implement to detect, prevent, and report the concealment and integration of proceeds from criminal activity. AML programmes are built around customer due diligence (CDD), enhanced due diligence (EDD) for high-risk relationships, continuous transaction monitoring, suspicious activity reporting (SAR), and sanctions screening against watch-lists maintained by bodies such as OFAC, the EU, and the UN. International standards are promulgated by the Financial Action Task Force (FATF) through its Forty Recommendations and transposed into national law by member jurisdictions, covering banks, virtual asset service providers (VASPs), lawyers, accountants, and real-estate agents. Modern AML increasingly relies on machine learning, graph analytics, and RegTech automation to surface complex layering schemes that evade traditional rule-based systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:anti-money-laundering", "labels": [ "Anti-Money Laundering", "Anti Money Laundering", "Anti Money Laundering Directive", "Anti-Money Laundering Compliance", "Anti-Money Laundering Directive", "EU Anti-Money Laundering Directive", "Money Laundering Regulations" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "anycast-routing", "title": "Anycast Routing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Anycast routing is a network addressing and routing technique in which a single IP address is advertised from multiple geographically distributed locations, and packets are delivered to whichever instance is nearest by the routing metric. It uses standard internet routing to steer each client to the closest available node, reducing latency and distributing load without client awareness. Anycast underpins resilient DNS, content delivery, and DDoS mitigation by enabling automatic failover and traffic absorption across many sites.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:anycast-routing", "labels": [ "Anycast Routing" ], "is_subclass_of": [ "Routing" ], "wikilinks": [] }, { "id": "apache-atlas", "title": "Apache Atlas", "domain": "data", "domain_name": "Data", "definition": "Apache Atlas is an open-source metadata management and data governance framework, originally built for the Hadoop ecosystem, that catalogs data assets, captures end-to-end lineage, and enforces classification and access policies. It exposes a typed metadata model and search so organizations can discover datasets, trace how data flows between systems, and apply governance tags. Atlas integrates with engines such as Hive, Kafka, and HBase to collect lineage automatically.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:apache-atlas", "labels": [ "Apache Atlas" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "apache-flink", "title": "Apache Flink", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Apache Flink is an open-source distributed engine for stateful computations over unbounded and bounded data streams. It provides a unified runtime that treats batch processing as a special case of streaming, with event-time semantics, sophisticated windowing, and exactly-once state consistency backed by distributed snapshots. Flink is widely used for low-latency, high-throughput stream processing in real-time analytics, event-driven applications, and continuous data pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:apache-flink", "labels": [ "Apache Flink" ], "is_subclass_of": [ "Stream Processing" ], "wikilinks": [] }, { "id": "apache-iceberg", "title": "Apache Iceberg", "domain": "data", "domain_name": "Data", "definition": "Apache Iceberg is an open table format for large analytic datasets stored in data lakes, adding database-like guarantees on top of object storage. It provides ACID transactions, schema and partition evolution, snapshot isolation, and time-travel queries by maintaining immutable metadata layers that track data files. Iceberg decouples the table format from the compute engine, letting Spark, Trino, Flink, and others operate consistently on the same tables.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:apache-iceberg", "labels": [ "Apache Iceberg" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "apache-kafka", "title": "Apache Kafka", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Apache Kafka is an open-source distributed event streaming platform originally developed at LinkedIn and donated to the Apache Software Foundation in 2011. It provides a high-throughput, low-latency, fault-tolerant publish-subscribe messaging system built around an immutable, ordered, partitioned commit log. Kafka decouples producers and consumers of data streams, enabling real-time data pipelines, event-driven architectures, and stream processing applications at scale across thousands of nodes handling trillions of events per day.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:apache-kafka", "labels": [ "Apache Kafka" ], "is_subclass_of": [ "Distributed System" ], "wikilinks": [] }, { "id": "apache-license-2-0", "title": "Apache License 2.0", "domain": "governance", "domain_name": "Governance", "definition": "The Apache License 2.0 (AL2) is a permissive free and open-source software licence published by the Apache Software Foundation in 2004, allowing users to freely use, modify, distribute, and sublicence covered works in both open and proprietary contexts. It requires preservation of copyright notices and a NOTICE file but imposes no copyleft obligations on derivatives. Uniquely among major permissive licences, AL2 includes an explicit patent grant and a patent retaliation clause that terminates rights upon initiation of patent litigation against the licensor.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:apache-license-2-0", "labels": [ "Apache License 2.0", "Apache 2.0 Licence", "Apache 2.0 Open Source Licence" ], "is_subclass_of": [ "Intellectual Property Licence Instrument" ], "wikilinks": [] }, { "id": "apache-parquet", "title": "Apache Parquet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Apache Parquet is an open-source columnar storage file format designed for efficient analytical processing of large datasets. By storing values of the same column contiguously, it enables aggressive compression and encoding, predicate pushdown, and reading only the columns a query needs, dramatically reducing I/O for analytical workloads. It carries a self-describing schema and rich metadata, and is widely used as the on-disk format for data lakes and big-data engines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:apache-parquet", "labels": [ "Apache Parquet" ], "is_subclass_of": [ "Blockchain", "Data Format" ], "wikilinks": [] }, { "id": "apache-spark", "title": "Apache Spark", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Apache Spark is an open-source unified analytics engine for large-scale data processing across clusters of machines. It exposes high-level APIs for batch processing, structured queries, stream processing and machine learning, and accelerates workloads by keeping intermediate data in memory between operations. Spark abstracts distributed datasets as fault-tolerant collections and schedules computations as directed acyclic graphs of stages, making it a foundational tool for big-data engineering and analytics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:apache-spark", "labels": [ "Apache Spark" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "api-contract", "title": "Api Contract", "domain": "data", "domain_name": "Data", "definition": "An API contract is the agreed, machine-readable specification of how a service may be called and what it will return, covering endpoints, request and response schemas, data types, status codes and error semantics. It functions as a formal agreement between the provider and consumers of an interface, allowing each side to develop and test independently against a shared definition. Expressed in formats such as OpenAPI or GraphQL schemas, the contract enables tooling for validation, mocking, code generation and compatibility checking across versions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-contract", "labels": [ "Api Contract", "API Contract" ], "is_subclass_of": [ "Data Schema" ], "wikilinks": [] }, { "id": "api-economy", "title": "Api Economy", "domain": "finance", "domain_name": "Finance", "definition": "The API economy is the ecosystem of business value created when organisations expose their data and services as application-programming interfaces that other parties can consume, compose, and monetise. By packaging capabilities as products, firms turn internal functions into reusable building blocks that partners and developers integrate, enabling new revenue streams, faster innovation, and platform network effects. It underpins models such as open banking and embedded services, and its growth depends on robust API management, gateways, and developer experience to govern access, security, and consumption at scale.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-economy", "labels": [ "Api Economy", "API Economy" ], "is_subclass_of": [ "Platform Economy" ], "wikilinks": [] }, { "id": "api-key-authentication", "title": "Api Key Authentication", "domain": "blockchain", "domain_name": "Blockchain", "definition": "API key authentication is a simple scheme in which a client includes a static, secret key with each request to identify and authenticate itself to an API. The server checks the key against issued keys to grant or deny access and to attribute usage and rate limits. Although easy to adopt, it offers coarse-grained control and weaker security than token-based or signature-based schemes because the long-lived key alone confers access.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:api-key-authentication", "labels": [ "Api Key Authentication", "API Key Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "apoapsis", "title": "Apoapsis", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:apoapsis", "labels": [ "Apoapsis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "apogee", "title": "Apogee", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:apogee", "labels": [ "Apogee" ], "is_subclass_of": [ "Apoapsis" ], "wikilinks": [] }, { "id": "apparent-magnitude", "title": "Apparent Magnitude", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:apparent-magnitude", "labels": [ "Apparent Magnitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "appearance-translation", "title": "Appearance Translation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Appearance Translation refers to neural style transfer and image-to-image translation techniques that use deep learning to transform the visual style of images or video while preserving semantic content, enabling artistic stylization, domain adaptation, and visual content transformation across di...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:appearance-translation", "labels": [ "Appearance Translation" ], "is_subclass_of": [ "Display and Rendering", "Neural Image Processing" ], "wikilinks": [ "Artistic Stylization", "Content Preservation", "Convolutional Neural Networks", "Domain Adaptation", "Neural Image Processing", "Style Representations", "Visual Content Transformation", "Computer Vision", "metaverse" ] }, { "id": "append-only-log", "title": "Append-Only Log", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An append-only log is a data structure in which records can only be added to the end and never modified or deleted in place, producing an immutable, totally ordered sequence of events. Each entry is durably persisted before subsequent entries, giving a tamper-evident history that consumers can replay deterministically from any offset. Append-only logs underpin event sourcing, distributed ledgers, write-ahead logging and certificate transparency, where verifiability and an authoritative ordering of changes are required.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:append-only-log", "labels": [ "Append-Only Log", "Append Only Log" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "apple-inc-technology-corporation", "title": "Apple Inc Technology Corporation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Apple Inc. is an American multinational technology corporation headquartered in Cupertino, California, that designs, manufactures, and markets consumer electronics, computer software, and online services spanning smartphones, personal computers, wearables, and spatial computing devices. Its strategic foundation rests on vertical integration of proprietary silicon (Apple Silicon, A-series, M-series SoCs), operating system platforms (iOS, macOS, visionOS, watchOS), and tightly controlled distribution through the App Store and developer ecosystem. Apple applies on-device machine learning inference, secure enclave hardware, and private cloud compute to deliver privacy-preserving AI features \u2014 branded Apple Intelligence \u2014 across its device portfolio. As one of the highest market-capitalisation companies globally, Apple's architectural decisions in silicon design, software APIs, and platform policy propagate directly into industry-wide infrastructure and standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:apple-inc-technology-corporation", "labels": [ "Apple Inc Technology Corporation", "Apple Ecosystem", "Apple HLS RFC 8216", "Apple Intelligence", "Apple R1 Chip", "Apple TV Plus" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "apple-mixed-reality-headset", "title": "Apple Mixed Reality Headset", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A mixed reality headset developed by Apple, announced in 2023 and released in 2024. It blends digital content with the user's surroundings using high-resolution displays and eye, hand, and voice input.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:apple-inc-technology-corporation-mixed-reality-headset", "labels": [ "Apple Mixed Reality Headset", "Vision Pro" ], "is_subclass_of": [ "Mixed Reality" ], "wikilinks": [ "Spatial Computing", "Augmented Reality", "Apple", "Mixed Reality" ] }, { "id": "apple-vision-pro", "title": "Apple Vision Pro", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Apple Vision Pro is a spatial computing head-mounted display developed by Apple, announced in June 2023 and released in the United States in February 2024, running the visionOS operating system. It combines ultra-high-resolution micro-OLED displays, a custom R1 chip for sensor processing, and an M2 chip for compute, enabling a fully immersive or passthrough-composited mixed reality experience. The device employs eye-tracking, hand-tracking, and voice input as its primary interaction modalities, replacing the traditional touchscreen paradigm with gaze-and-pinch gesture control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:apple-inc-technology-corporation-vision-pro", "labels": [ "Apple Vision Pro", "visionOS" ], "is_subclass_of": [ "Head-Mounted Display" ], "wikilinks": [] }, { "id": "application-binary-interface", "title": "Application Binary Interface", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An application binary interface (ABI) is the specification of how software components interact at the binary level, defining calling conventions, data layout and, for smart contracts, the encoding used to invoke functions and pass arguments. On blockchains such as Ethereum, a contract's ABI describes its callable functions, events and parameter types in a machine-readable form that clients use to encode transactions and decode returned data. It is what allows external tools, wallets and other contracts to interact correctly with a deployed contract without access to its source code.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:application-binary-interface", "labels": [ "Application Binary Interface" ], "is_subclass_of": [ "Interface" ], "wikilinks": [] }, { "id": "application-development", "title": "Application Development", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Application development is the systematic process of designing, building, testing, and deploying software applications to meet specified user or organisational requirements, encompassing activities from requirements analysis through to release and maintenance. It involves selecting appropriate technology stacks, architectural patterns, and development methodologies to produce reliable, maintainable, and performant software. The discipline spans mobile, web, desktop, and embedded application domains, each with distinct toolchains and constraints.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:application-development", "labels": [ "Application Development" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "application-layer", "title": "Application Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software layer providing domain-specific application interfaces and services for metaverse experiences including education, commerce, healthcare, and entertainment applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:application-layer", "labels": [ "Application Layer", "ApplicationLayer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Application API", "Business Logic Layer", "Business Services", "Content Delivery", "Domain-Specific Applications", "Identity Services", "MSF Taxonomy 2025", "Platform Services", "Service Interface", "User Experience", "User Interface Framework", "Compute Layer", "Cross-Platform Interoperability", "Data Storage Layer", "InfrastructureDomain", "Metaverse Architecture Stack", "Metaverse Stack", "Network Infrastructure", "Telecollaboration" ] }, { "id": "application-programming-interface", "title": "Application Programming Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An application programming interface (API) is a formally specified contract that defines how software components request services and exchange data with one another. It abstracts an implementation behind a stable set of operations, data types, and protocols, decoupling callers from internal details. APIs span in-process library interfaces, inter-process and network endpoints, and platform service boundaries.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:application-programming-interface", "labels": [ "Application Programming Interface" ], "is_subclass_of": [ "Interface" ], "wikilinks": [] }, { "id": "application-specific-integrated-circuit", "title": "Application Specific Integrated Circuit", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An application-specific integrated circuit (ASIC) is a chip designed and fabricated to perform a single, fixed function with maximum efficiency rather than to run general-purpose programs. In blockchain, ASICs are built to compute a particular proof-of-work hash function at far higher speed and energy efficiency than general processors, making them dominant in mining for hashes such as SHA-256. Because the silicon is customised for one task, ASICs offer superior performance per watt but cannot be repurposed for other workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:application-specific-integrated-circuit", "labels": [ "Application Specific Integrated Circuit", "Application-Specific Integrated Circuit" ], "is_subclass_of": [ "Mining Hardware" ], "wikilinks": [] }, { "id": "application-window-capture", "title": "Application Window Capture", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Application Window Capture is a screen-capture technique that grabs the pixels of a single specified application window rather than the entire display or a region. It lets users share or record one program while keeping other windows, notifications, and private content off-screen. Operating systems expose it through window-enumeration and per-window framebuffer APIs used by recording and conferencing software.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:application-window-capture", "labels": [ "Application Window Capture" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "application-specific-blockchain", "title": "Application-Specific Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An application-specific blockchain is a sovereign blockchain purpose-built to run a single application or tightly scoped set of applications, rather than serving as a general-purpose smart-contract platform. By controlling the full stack \u2014 consensus, execution, and governance \u2014 it can optimise throughput, fee economics, and customisation for its target use case. This architecture, popularised by the Cosmos ecosystem and rollup frameworks, trades shared security for sovereignty and performance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:application-specific-blockchain", "labels": [ "Application-Specific Blockchain" ], "is_subclass_of": [ "Blockchain", "Blockchain Network" ], "wikilinks": [] }, { "id": "applied-ai-research-portfolio", "title": "Applied AI Research Portfolio", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Recent Projects is a curated portfolio page cataloguing active and completed applied research and commercial engagements spanning conversational AI, autonomous marine vision systems, human-attention analytics, large-scale exhibition AI, GenAI film pre-visualisation, interactive image/video generation, text-to-3D modelling, Bitcoin-native digital assets, AI education products, and Logseq-based training materials. It serves as an index of practical deployments demonstrating cross-domain expertise in AI, robotics, spatial computing, and blockchain.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:applied-ai-research-portfolio", "labels": [ "Applied AI Research Portfolio", "Recent Projects" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "presentation", "Hardware and Edge" ] }, { "id": "applied-machine-learning", "title": "Applied Machine Learning", "domain": "ai", "domain_name": "Ai", "definition": "Applied Machine Learning is the deployment-focused discipline of adapting and operationalising machine-learning methods to solve concrete real-world problems under domain constraints of latency, cost, reliability, and compliance. It spans the full lifecycle from problem framing and data acquisition through feature engineering, model selection, rigorous evaluation using cross-validation and holdout sets, and sustained production monitoring via MLOps practices. Unlike theoretical ML research, which focuses on novel algorithmic contributions, Applied ML prioritises practical impact: models must generalise reliably to unseen data, behave predictably under distribution shift, and be maintainable by engineering teams across months or years of operation.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:applied-machine-learning", "labels": [ "Applied Machine Learning" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Research Area" ], "wikilinks": [ "Machine Learning Discipline", "Cross-Validation", "Logistics Optimisation", "Feature Engineering", "Model Selection", "Deep Learning", "Data Preprocessing", "Hyperparameter Optimisation", "Model Evaluation", "MLOps", "Data Pipeline", "Transfer Learning", "AutoML", "Supervised Learning", "Unsupervised Learning", "Reinforcement Learning", "Model Deployment", "Bias-Variance Tradeoff", "Explainable AI", "Demand Forecasting" ] }, { "id": "applied-mathematics", "title": "Applied Mathematics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Applied Mathematics is the systematic use of mathematical structures, methods, and reasoning to model, analyse, and solve problems arising in science, engineering, economics, computing, and industry. Distinct from pure mathematics by its orientation toward practical effectiveness, Applied Mathematics encompasses numerical analysis, mathematical modelling, optimisation, probability and statistics, differential equations, control theory, fluid dynamics, information theory, and scientific computing. It provides the formal substrate on which machine learning training dynamics, signal processing algorithms, control systems, financial derivatives pricing, and physical simulation all rest. The discipline demands both rigorous theoretical grounding \u2014 convergence proofs, stability analysis, error bounds \u2014 and pragmatic accommodation of real constraints such as floating-point arithmetic, computational cost, and measurement noise.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:applied-mathematics", "labels": [ "Applied Mathematics", "Continuous Mathematics" ], "is_subclass_of": [ "AI Research Area", "Mathematical Science", "owl:Thing" ], "wikilinks": [ "Convex Optimisation", "Numerical Methods", "Functional Analysis", "owl:Thing", "Linear Algebra", "Calculus", "Probability Theory", "Differential Equations", "Mathematical Optimisation", "Mathematical Foundations", "Mathematical Science", "Numerical Integration", "Gradient Descent", "Stochastic Processes", "Machine Learning Discipline", "Deep Learning", "Applied Machine Learning", "Operations Research", "Information Theory", "Bayesian Optimisation" ] }, { "id": "apprenticeship-learning", "title": "Apprenticeship Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Apprenticeship learning is the problem of learning to perform a task at expert level from demonstrations, without an explicitly specified reward function. Introduced by Abbeel and Ng (2004), it proceeds via inverse reinforcement learning: the learner infers a reward function under which the expert's behaviour is near-optimal \u2014 typically as a linear combination of state features whose expected values it seeks to match \u2014 and then optimises a policy against that inferred reward, yielding performance guarantees relative to the expert and better generalisation than directly mimicking actions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:apprenticeship-learning", "labels": [ "Apprenticeship Learning" ], "is_subclass_of": [ "Imitation Learning" ], "wikilinks": [ "Imitation Learning", "Inverse Reinforcement Learning", "Reinforcement Learning", "Reward Function" ] }, { "id": "april-tag", "title": "April Tag", "domain": "robotics", "domain_name": "Robotics", "definition": "An AprilTag is a square fiducial marker, resembling a simplified QR code, designed for robust detection and accurate 6-DoF pose estimation by computer-vision systems. Each tag encodes a unique ID with strong error correction, enabling reliable recognition under poor lighting, partial occlusion, and oblique viewing angles. AprilTags are widely used as visual landmarks for camera calibration, robot localization, and augmented-reality registration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:april-tag", "labels": [ "April Tag" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "aptos", "title": "Aptos", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Aptos is a Layer 1 proof-of-stake blockchain platform founded in 2022 by former Diem (Meta) engineers, designed for safety, scalability, and high throughput. It employs the Move programming language \u2014 a resource-oriented, formally verifiable language originally created for the Diem project \u2014 to express accounts, digital assets, and smart contracts. Its primary performance innovation is Block-STM (Software Transactional Memory), a parallel transaction execution engine that speculatively executes transactions and resolves conflicts at runtime, enabling high concurrent throughput without requiring developers to pre-declare access sets. The network uses a Byzantine Fault Tolerant consensus mechanism (AptosBFT, derived from HotStuff) to reach finality with low latency.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:aptos", "labels": [ "Aptos" ], "is_subclass_of": [ "Blockchain Network" ], "wikilinks": [ "Proof of Stake", "Smart Contract", "Layer 1", "Blockchain Network" ] }, { "id": "aquiclude", "title": "Aquiclude", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aquiclude", "labels": [ "Aquiclude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "aquifer", "title": "Aquifer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aquifer", "labels": [ "Aquifer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "aquifuge", "title": "Aquifuge", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aquifuge", "labels": [ "Aquifuge" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "aquitard", "title": "Aquitard", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aquitard", "labels": [ "Aquitard" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ar-display-device", "title": "Ar Display Device", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Display Device encompasses hardware systems including smart glasses, headsets, and head-mounted displays that superimpose computer-generated imagery onto the user's view of the real world through optical technologies such as waveguides, birdbath optics, holographic displays, and metasurfaces.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ar-display-device", "labels": [ "Ar Display Device" ], "is_subclass_of": [ "Display and Rendering", "Display Hardware" ], "wikilinks": [ "Augmented Reality Experiences", "Display Technology", "Mixed Reality Interaction", "Optical Systems", "Pose Tracking", "Sensor Input", "Display Hardware", "metaverse", "Spatial Computing" ] }, { "id": "ar-experiences", "title": "Ar Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Experiences are interactive augmented reality applications that overlay digital content onto the physical world in real-time through smartphones, tablets, or AR headsets, enabling enhanced visualization, contextual information display, and immersive interaction with virtual objects anchored in...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ar-experiences", "labels": [ "Ar Experiences", "Retargetable AR Experiences" ], "is_subclass_of": [ "Platform and Environment", "Immersive Technology Applications" ], "wikilinks": [ "AR Display Devices", "Contextual Information", "Enhanced Visualization", "Immersive Technology Applications", "Sensor Input", "Computer Vision", "Interactive Learning", "metaverse", "Spatial Mapping" ] }, { "id": "ar-overlay", "title": "Ar Overlay", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Overlay refers to the digital layer of virtual elements including holograms, data visualizations, animations, and 3D objects that are spatially anchored and rendered over the user's view of the physical world in augmented reality systems, creating a seamless blend of virtual and real content.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ar-overlay", "labels": [ "Ar Overlay" ], "is_subclass_of": [ "Display and Rendering", "Platform and Environment", "Digital Display Layer" ], "wikilinks": [ "Digital Display Layer", "Environmental Understanding", "Information Augmentation", "Spatial Annotation", "Virtual Object Placement", "Computer Vision", "metaverse", "Real-Time Rendering", "Spatial Anchoring" ] }, { "id": "ar-scene-graph", "title": "Ar Scene Graph", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR Scene Graph is a hierarchical data structure used in augmented reality to represent and organize spatial relationships between virtual objects, real-world elements, and their transformations, enabling context-aware placement, semantic understanding, and natural interaction between digital cont...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ar-scene-graph", "labels": [ "Ar Scene Graph" ], "is_subclass_of": [ "Content and Assets", "Spatial Data Structure" ], "wikilinks": [ "3D Scene Reconstruction", "Context-Aware AR", "Retargetable AR Experiences", "Semantic Scene Understanding", "Spatial Relationships", "Computer Vision", "metaverse", "Object Detection", "Spatial Data Structure" ] }, { "id": "aragon", "title": "Aragon", "domain": "governance", "domain_name": "Governance", "definition": "A framework and set of smart contracts for creating and operating decentralised autonomous organisations on Ethereum, providing modular governance, treasury and permission components.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:aragon", "labels": [ "Aragon", "Aragon Court", "Aragon OSx Standard" ], "is_subclass_of": [ "DAO Tooling" ], "wikilinks": [ "Decentralised Autonomous Organisation", "Ethereum", "Smart Contracts", "Governance", "DAO", "DAO Tooling" ] }, { "id": "arbitrage", "title": "Arbitrage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Arbitrage is the practice of simultaneously buying and selling an asset across different markets or instruments to profit from a price discrepancy with minimal directional risk. In decentralised finance it is a central economic force that aligns token prices across exchanges and liquidity pools, often executed by automated bots exploiting transient inefficiencies. Arbitrage activity drives price discovery and improves market efficiency, but is closely tied to phenomena such as miner/maximal extractable value and impermanent loss.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:arbitrage", "labels": [ "Arbitrage" ], "is_subclass_of": [ "Market Efficiency" ], "wikilinks": [] }, { "id": "arbitration-decision-engine", "title": "Arbitration Decision Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Arbitration Decision Engine is an AI-powered or smart contract-based automated system that evaluates evidence, applies predetermined rules, and renders decisions in dispute resolution processes, functioning as an oracle that can trigger smart contract modifications and enable self-enforcing arbit...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:arbitration-decision-engine", "labels": [ "Arbitration Decision Engine" ], "is_subclass_of": [ "Governance and Safety", "Automated Dispute Resolution" ], "wikilinks": [ "Automated Dispute Resolution", "Blockchain Integration", "Decentralized Arbitration", "Evidence Authentication", "Rule Encoding", "Self-Enforcing Awards", "Smart Contract Resolution", "Blockchain", "metaverse" ] }, { "id": "arbitration-process", "title": "Arbitration Process", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Arbitration Process in the blockchain and metaverse context refers to structured dispute resolution procedures using smart contracts and decentralized technologies, encompassing both on-chain mechanisms with automatic award enforcement and off-chain proceedings with traditional arbitral framework...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:arbitration-process", "labels": [ "Arbitration Process" ], "is_subclass_of": [ "Governance and Safety", "Alternative Dispute Resolution" ], "wikilinks": [ "Alternative Dispute Resolution", "Arbitration Clauses", "Confidential Proceedings", "Cross-Border Dispute Resolution", "Evidence Procedures", "Neutral Selection", "Smart Contract Enforcement", "Blockchain", "metaverse" ] }, { "id": "arbitrator-expertise", "title": "Arbitrator Expertise", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Arbitrator Expertise in blockchain and digital asset contexts refers to the specialized technical knowledge, legal background, and industry experience required by neutral decision-makers to understand distributed ledger technology, smart contracts, tokenomics, and crypto-asset classification for ...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:arbitrator-expertise", "labels": [ "Arbitrator Expertise" ], "is_subclass_of": [ "Governance and Safety", "Professional Qualification" ], "wikilinks": [ "Blockchain Knowledge", "Fair Outcomes", "Industry Experience", "Informed Decision Making", "Legal Training", "Professional Qualification", "Technical Evidence Evaluation", "Blockchain", "metaverse" ] }, { "id": "arbitrum-dao", "title": "Arbitrum DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Arbitrum DAO is the decentralised autonomous organisation that governs the Arbitrum One and Arbitrum Nova Layer 2 blockchain networks, operating through the ARB governance token launched by Offchain Labs in March 2023. Token holders propose and vote on protocol upgrades, treasury allocations, and parameter changes via on-chain governance contracts, making Arbitrum DAO one of the largest DAOs by treasury value in the Ethereum ecosystem. Governance decisions are executed through a Security Council with emergency intervention powers and a broader Arbitrum Improvement Proposal (AIP) process for standard changes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:arbitrum-dao", "labels": [ "Arbitrum DAO" ], "is_subclass_of": [ "DAO" ], "wikilinks": [] }, { "id": "arbitrum", "title": "Arbitrum", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Arbitrum is a family of Ethereum layer-2 scaling solutions developed by Offchain Labs that use optimistic rollup technology to increase transaction throughput and reduce fees. Its flagship network, Arbitrum One, launched on mainnet in 2021 and executes transactions in a custom virtual machine while posting transaction data and state commitments to Ethereum. Arbitrum's interactive multi-round fraud-proof protocol distinguishes it from other optimistic rollups by narrowing disputes to a single instruction before on-chain resolution. The ARB token governs the network through the Arbitrum DAO, and Arbitrum Nitro and Orbit extend the technology to custom chains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:arbitrum", "labels": [ "Arbitrum", "Arbitrum Orbit" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Domain" ], "wikilinks": [ "Ethereum", "Rollup", "Fraud Proof", "Decentralised Finance Domain", "Optimism", "zkSync", "Polygon", "Blockchain Domain" ] }, { "id": "arc-consistency", "title": "Arc Consistency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A local consistency property of binary constraint networks in which, for every value in one variable's domain, each constraint linking it to another variable admits at least one compatible supporting value in that variable's domain. Enforcing it \u2014 classically with the AC-3 algorithm \u2014 deletes unsupported values until a fixed point, providing the most widely used level of constraint propagation and the pruning backbone of practical constraint solvers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:arc-consistency", "labels": [ "Arc Consistency" ], "is_subclass_of": [ "Constraint Propagation" ], "wikilinks": [ "Constraint Propagation", "Constraint Satisfaction", "Backtracking Search", "Constraint" ] }, { "id": "arc-face", "title": "ArcFace", "domain": "ai", "domain_name": "Ai", "definition": "ArcFace is a deep learning loss function and associated training methodology for facial recognition, introduced by Deng et al. at Imperial College London in 2019, that improves discriminative feature learning by adding an additive angular margin penalty to the softmax loss function. By penalising the angle between a sample embedding and its class centre in hyperspherical feature space, ArcFace forces the model to learn more compact intra-class and more separable inter-class feature embeddings than standard softmax or earlier margin-based losses. It has achieved state-of-the-art performance on numerous facial recognition benchmarks and is widely adopted in production identity verification systems.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:arc-face", "labels": [ "ArcFace", "ArcFace Embedding" ], "is_subclass_of": [ "Facial Recognition", "Loss Function", "Representation Learning" ], "wikilinks": [ "Facial Recognition", "Deep Learning", "Loss Function", "Embedding", "Cosine Similarity", "Biometric Verification", "Identity Verification", "Feature Extraction", "Computer Vision", "Convolutional Neural Network", "Neural Network Architecture", "Contrastive Learning", "Representation Learning", "Backpropagation", "Batch Normalisation", "Dropout", "Liveness Detection", "Face Swap", "Biometric Authentication", "Access Control" ] }, { "id": "archaeological-site-reconstruction", "title": "Archaeological Site Reconstruction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Archaeological Site Reconstruction refers to the use of 3D modeling, photogrammetry, LiDAR, and virtual reality technologies to create accurate digital representations of ancient sites, enabling study, documentation, preservation, and immersive public access to cultural heritage through realistic...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:archaeological-site-reconstruction", "labels": [ "Archaeological Site Reconstruction" ], "is_subclass_of": [ "Content and Assets", "Digital Heritage" ], "wikilinks": [ "Educational Experiences", "Heritage Preservation", "Historical Research", "3D Modeling", "Computer Vision", "Digital Heritage", "metaverse", "Photogrammetry", "Virtual Tourism" ] }, { "id": "architectural-layer", "title": "Architectural Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An architectural layer is a discrete horizontal stratum within a layered software or system architecture, responsible for a cohesive set of concerns and communicating only with adjacent layers through well-defined interfaces. Layered decomposition is a fundamental pattern in software architecture that promotes separation of concerns, replaceability of individual layers, and independent evolution of distinct system capabilities. Classic examples include the OSI network model's seven layers, the three-tier application architecture (presentation, logic, data), and the cloud-native separation of infrastructure, platform, and application layers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:architectural-layer", "labels": [ "Architectural Layer", "ArchitecturalLayer" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "architectural-visualisation", "title": "architectural visualisation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Architectural visualisation is the discipline of producing photorealistic or stylised two- and three-dimensional representations of proposed buildings, interiors, and urban environments to communicate design intent to clients, planners, and the public before construction. It combines techniques from 3D modelling, physically based rendering, and real-time graphics to produce still images, animations, and interactive walkthroughs. Advances in real-time rendering engines and spatial computing have extended the discipline to include immersive virtual reality experiences and augmented reality overlays on physical construction sites.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:architectural-visualisation", "labels": [ "Architectural Visualisation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "architecture", "title": "Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In the AI domain, Architecture refers to the structural design of an AI model or system, specifying how computational components\u2014such as layers, modules, attention heads, and connections\u2014are organised and interact to process inputs and produce outputs. Architectural choices fundamentally determine a model's capacity, inductive biases, scalability, and suitability for particular tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:architecture", "labels": [ "Architecture", "Hand-Crafted Architecture", "Layered Architecture", "Processor Architecture", "Stack-Based Architecture" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network Architecture", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "Edge Computing", "Neural Network Architecture", "Transformer Architecture", "Convolutional Neural Network", "Recurrent Neural Network", "Attention Mechanism", "Encoder", "Decoder", "Deep Learning", "Model Training", "Inductive Bias", "Loss Function", "Backpropagation", "Gradient Descent", "Hyperparameter Tuning", "Activation Function", "Batch Normalisation", "Dropout", "Transfer Learning" ] }, { "id": "archival-node", "title": "Archival Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Archival Node is a blockchain network participant that stores the complete historical state of the ledger\u2014every block, transaction, and state root from genesis to the current tip\u2014without pruning older data. Unlike pruned nodes or light nodes, an archival node can respond to queries about any historical state at any block height, making it essential for block explorers, analytics services, and smart contract developers who need point-in-time state access.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:archival-node", "labels": [ "Archival Node" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "Network Component" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "archival-standards", "title": "Archival Standards", "domain": "standards", "domain_name": "Standards", "definition": "Archival Standards encompass frameworks, specifications, and best practices for long-term digital preservation, including metadata standards (PREMIS, METS), reference models (OAIS), and storage technologies that ensure digital content remains accessible, authentic, and interpretable across technological changes.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:archival-standards", "labels": [ "Archival Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Preservation Standards" ], "wikilinks": [ "Content Authenticity", "Format Migration", "ISO 14721", "ISO 16363", "Long-Term Preservation", "Preservation Planning", "Preservation Standards", "Blockchain", "Metadata Management", "metaverse", "Storage Infrastructure" ] }, { "id": "archived-data-state", "title": "Archived Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:archived-data-state", "labels": [ "Archived Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "arcjet", "title": "Arcjet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:arcjet", "labels": [ "Arcjet" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "arcore", "title": "Arcore", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ARCore is Google's software development kit for building augmented-reality applications on Android and, via web standards, the browser. It provides motion tracking, environmental understanding and light estimation by fusing camera frames with inertial sensors to anchor virtual content in the physical world. ARCore is the principal counterpart to Apple's ARKit in the mobile-AR ecosystem.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:arcore", "labels": [ "Arcore", "ARCore" ], "is_subclass_of": [ "Augmented Reality", "Ar Experiences" ], "wikilinks": [] }, { "id": "argent", "title": "Argent", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart contract wallet for Ethereum and Layer 2 networks that uses account abstraction to provide features such as social recovery, spending limits and gasless transactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:argent", "labels": [ "Argent", "Argent Account" ], "is_subclass_of": [ "Wallet" ], "wikilinks": [ "Account Abstraction", "Smart Contracts", "Ethereum", "Wallet" ] }, { "id": "argo-blockchain", "title": "Argo Blockchain", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Argo Blockchain is a publicly listed cryptocurrency mining company that operates data centres performing the hashing that secures proof-of-work networks, chiefly Bitcoin.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:argo-blockchain", "labels": [ "Argo Blockchain" ], "is_subclass_of": [ "Bitcoin Mining" ], "wikilinks": [ "Bitcoin Mining", "Bitcoin", "Cryptocurrency", "Sustainability Domain" ] }, { "id": "argument-of-periapsis", "title": "Argument of Periapsis", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:argument-of-periapsis", "labels": [ "Argument of Periapsis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "arithmetic-circuit", "title": "Arithmetic Circuit", "domain": "security", "domain_name": "Security", "definition": "A directed acyclic graph (DAG) of addition and multiplication gates over a field, used to represent polynomial computations in a form amenable to cryptographic proof systems. Arithmetic circuits are the canonical intermediate representation for zero-knowledge proof schemes such as zk-SNARKs and STARKs: a computation is first expressed as an arithmetic circuit, then compiled into a system of polynomial constraints (R1CS or Plonkish), and finally proved using a cryptographic proving system. The complexity of a circuit is characterised by its depth (for parallelism) and size (gate count).", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:arithmetic-circuit", "labels": [ "Arithmetic Circuit" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "ark-protocol", "title": "Ark Protocol", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Ark is a layer-two protocol for Bitcoin that uses shared, periodically refreshed off-chain outputs to let users transact cheaply and privately while retaining the ability to settle on the base chain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:ark-protocol", "labels": [ "Ark Protocol", "ARK" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Bitcoin", "Cross-Border Payments", "Payment Network", "Stablecoin" ] }, { "id": "arkit", "title": "Arkit", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ARKit is Apple's augmented reality development framework for iOS and iPadOS devices, providing motion tracking, environment understanding and rendering integration for placing virtual content in the physical world. It uses visual-inertial odometry, plane detection and scene reconstruction to anchor digital objects stably relative to real surfaces. ARKit gives developers a high-level API over the device camera, motion sensors and Neural Engine to build mobile AR experiences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:arkit", "labels": [ "Arkit", "ARKit" ], "is_subclass_of": [ "Augmented Reality", "Environmental Mapping" ], "wikilinks": [] }, { "id": "artifact-metadata", "title": "Artifact Metadata", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Artifact Metadata refers to structured descriptive, administrative, and provenance information associated with digital assets and cultural objects, particularly in NFT and blockchain contexts, documenting ownership history, authenticity, cultural significance, and preservation status for verifica...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:artifact-metadata", "labels": [ "Artifact Metadata" ], "is_subclass_of": [ "Content and Assets", "Digital Metadata" ], "wikilinks": [ "Authenticity Tracking", "Blockchain Recording", "Cultural Preservation", "Digital Metadata", "Documentation Practices", "Blockchain", "Metadata Standards", "metaverse", "Provenance Verification" ] }, { "id": "artificial-general-intelligence", "title": "Artificial General Intelligence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial General Intelligence (AGI) denotes a hypothesised class of computational systems whose cognitive competence is broad rather than narrow \u2014 capable of matching or exceeding human performance across the full distribution of economically and intellectually significant tasks rather than exc...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:artificial-general-intelligence", "labels": [ "Artificial General Intelligence" ], "is_subclass_of": [ "AI Research Area", "Frontier AI", "Artificial Intelligence", "General Intelligence", "Foundation Model Capability" ], "wikilinks": [ "AgentLayer", "AI Policy", "AI Safety Research", "Alignment Research", "Alignment Techniques", "AlphaProof AlphaGeometry 2 IMO 2024", "Amodei 2024 Machines of Loving Grace", "Anthropic Claude Computer Use 2024", "Anthropic Core Views on AI Safety 2023", "Apollo Research 2024 In-Context Scheming", "ARC-AGI", "ARC Prize 2024 Technical Report", "ARIA Safeguarded AI Programme", "Aschenbrenner 2024 Situational Awareness", "Autonomous Agents", "Autonomous Reasoning", "Bai et al 2022 Constitutional AI", "Benchmarks", "Bengio et al 2025 International AI Safety Report", "Bitter Lesson" ] }, { "id": "artificial-intelligence-core", "title": "Artificial Intelligence Core", "domain": "ai", "domain_name": "Ai", "definition": "Artificial Intelligence Core is the foundational upper class that anchors the core concepts and capabilities of artificial intelligence within the ontology, serving as the common ancestor for machine learning, agent systems, and related disciplines. It represents the essential body of theory and method by which machines perform tasks that normally require human intelligence, such as learning, reasoning, perception, and decision-making. As a structural root it organizes more specific AI subfields beneath a single semantic anchor, encompassing symbolic methods, statistical learning, connectionist architectures, and their contemporary convergences in foundation models and agentic systems.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:artificial-intelligence-core", "labels": [ "Artificial Intelligence Core" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique" ], "wikilinks": [ "Machine Learning Discipline", "AI Agent System", "Artificial Intelligence", "Deep Learning", "Machine Learning", "Natural Language Processing", "Computer Vision", "Reinforcement Learning", "Knowledge Representation", "Symbolic AI", "Neural Network", "Foundation Model", "Transfer Learning", "Supervised Learning", "Unsupervised Learning", "Planning and Search", "Expert System", "Robotics", "Autonomous Agents", "Multimodal AI" ] }, { "id": "artificial-intelligence-research", "title": "Artificial Intelligence Research", "domain": "ai", "domain_name": "Ai", "definition": "Artificial intelligence research is the scientific and engineering discipline concerned with understanding and building systems that exhibit intelligent behaviour, including perception, reasoning, learning, planning, and language. It spans theoretical foundations, algorithmic innovation, and empirical evaluation, and feeds discoveries into subfields such as machine learning, computer vision, and natural language processing. The field advances through hypothesis-driven inquiry using the scientific method, publication and peer review, standardised benchmarking, and reproducible experimentation, generating knowledge that flows into applied AI systems and deployed products across industry and government.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:artificial-intelligence-research", "labels": [ "Artificial Intelligence Research" ], "is_subclass_of": [ "Research and Development" ], "wikilinks": [ "Research and Development", "Artificial Intelligence", "Machine Learning", "Scientific Method", "Computer Vision", "Neural Network", "Deep Learning", "Knowledge Discovery", "Natural Language Processing", "Reinforcement Learning", "Foundation Model", "Benchmark Evaluation", "Reproducibility", "Large Language Models", "Multimodal AI", "AI Safety Research", "Alignment Research", "Explainable AI", "Academic Research", "Peer Review" ] }, { "id": "artificial-intelligence", "title": "Artificial Intelligence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial Intelligence is a artificial intelligence concept and a type of owl:Thing. that enables Autonomous Systems, Decision Support. comprising Computer Vision, Deep Learning.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:artificial-intelligence", "labels": [ "Artificial Intelligence", "Artificial Intelligence Research Laboratory", "ArtificialIntelligence", "ArtificialIntelligenceDomain", "UCL Centre for Artificial Intelligence" ], "is_subclass_of": [ "Thing" ], "wikilinks": [ "Autonomous Systems", "Computational Infrastructure", "Decision Support", "design thinking", "GDPR", "Intelligent Automation", "ISO/IEC 22989:2022", "ISO/IEC 23053:2022", "ISO/IEC 23894:2023", "ISO/IEC 42001:2023", "Mark Zuckerberg", "Neural Networks", "NIST AI RMF", "OECD", "oecd", "Personalisation", "user experience", "AI Risks", "AI Video", "Artificial General Intelligence" ] }, { "id": "artificial-superintelligence-theory", "title": "Artificial Superintelligence Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A hypothetical form of artificial intelligence that surpasses human cognitive performance across all domains, capable of recursive self-improvement leading to an intelligence explosion. Distinguished from artificial general intelligence by the degree of capability surplus; considered an existential risk scenario requiring robust safety and alignment research before any attempt at development.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:artificial-superintelligence-theory", "labels": [ "Artificial Superintelligence Theory", "artificial superintelligence" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "AI Risks", "Artificial Intelligence", "Singularity" ] }, { "id": "arweave", "title": "Arweave", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Arweave is a decentralised, blockweave-based storage protocol that enables permanent, censorship-resistant archival of data through a proof-of-access consensus mechanism, where miners must demonstrate retention of a randomly selected historic block to produce new blocks. It introduces the 'permaweb' \u2014 a permanent, publicly accessible layer of the web sustained by a one-time payment model where a portion of each storage fee accrues to an endowment that compensates miners across an indefinite time horizon. The network's economic model assumes that declining hardware costs will be outpaced by AR token appreciation, funding perpetual content replication across an open, permissionless peer set. Beyond raw storage, Arweave hosts SmartWeave lazy-evaluation contracts and the AO actor-oriented compute environment, extending it into a foundation for permanent, verifiable decentralised computation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:arweave", "labels": [ "Arweave" ], "is_subclass_of": [ "Decentralized Storage" ], "wikilinks": [] }, { "id": "asia-pacific-regulation", "title": "Asia Pacific Regulation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Asia Pacific Regulation denotes the polycentric, jurisdictionally heterogeneous body of AI, generative-AI, algorithmic, and digital-technology law operating across the Asia-Pacific region between 2023 and 2026, comprising statutory regimes, administrative measures, soft-law guidance, supervisory ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:asia-pacific-regulation", "labels": [ "Asia Pacific Regulation" ], "is_subclass_of": [ "AI Governance and Ethics", "Regional AI Regulation", "AI Governance", "Digital Technology Policy", "Regulatory Framework", "Cross-Border AI Law" ], "wikilinks": [ "Access Now Asia-Pacific", "Ada Lovelace Institute", "Agency for Cultural Affairs", "AI Action Plan", "AI Action Summit Paris", "AI Basic Act", "AI Bill", "AI Compliance", "AI Framework Act", "AI Governance Guidelines", "AI Law (China)", "AI Promotion Bill", "AI Risk Assessment Framework", "AI Risk Mitigation", "AI Safety Institute Japan", "AI Safety Institute Network", "AI Safety Summit", "AI Safety Testing", "AI Security Institute", "AI Seoul Summit" ] }, { "id": "assembly-automation", "title": "Assembly Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "Assembly automation is the use of robotic manipulators to perform repetitive joining, fastening, insertion or fitting operations that combine discrete parts into a finished product on a production line. It depends on precise manipulation and often force or vision feedback to handle part tolerances and orientation. It is one of the earliest and most economically significant applications of industrial robot manipulators.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:assembly-automation", "labels": [ "Assembly Automation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "asset-archive", "title": "Asset Archive", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Asset Archive is a digital repository system for long-term storage, management, and preservation of 3D models, textures, animations, and other virtual world components, incorporating provenance tracking, version control, and standardized formats to ensure accessibility and interoperability across...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:asset-archive", "labels": [ "Asset Archive" ], "is_subclass_of": [ "Content and Assets", "Digital Repository" ], "wikilinks": [ "Asset Preservation", "Content Reuse", "Standardized Formats", "Computer Vision", "Cross-Platform Interoperability", "Digital Repository", "Metadata Management", "metaverse", "Storage Infrastructure" ] }, { "id": "asset-creation", "title": "Asset Creation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Asset Creation is the structured practice of producing digital or physical resources \u2014 including geometry, textures, audio, code, and data \u2014 intended for use within interactive experiences, virtual environments, or production pipelines. It encompasses the full authoring lifecycle from concept through final deliverable, applying artistic, technical, and automated methods to yield artefacts that meet quality, format, and performance requirements.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-creation", "labels": [ "Asset Creation", "Automated Asset Creation" ], "is_subclass_of": [ "Content Creation" ], "wikilinks": [] }, { "id": "asset-database", "title": "Asset Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Asset Database is a structured, queryable repository that stores descriptive records, binary references, version histories, and relational metadata for digital assets within a production or distribution system. It provides the persistent backbone enabling discovery, retrieval, provenance tracking, and lifecycle management of assets across teams and pipelines.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-database", "labels": [ "Asset Database" ], "is_subclass_of": [ "Database System" ], "wikilinks": [] }, { "id": "asset-digitization", "title": "Asset Digitization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Asset Digitization is the process of converting physical objects, spaces, documents, or analogue media into discrete digital representations suitable for storage, transmission, and computational use. It encompasses acquisition technologies such as scanning, photogrammetry, and LiDAR, as well as the post-processing steps that clean, structure, and format the resulting data for target applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-digitization", "labels": [ "Asset Digitization", "Mass Digitization" ], "is_subclass_of": [ "Reality Capture" ], "wikilinks": [] }, { "id": "asset-format-standards", "title": "Asset Format Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Asset Format Standards are technical specifications defining file formats, data structures, and interchange protocols for 3D models, textures, animations, and scene descriptions, enabling interoperability between content creation tools, game engines, and metaverse platforms through standards like...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:asset-format-standards", "labels": [ "Asset Format Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standards" ], "wikilinks": [ "Content Portability", "Conversion Tools", "Cross-Platform Assets", "FBX", "Format Compliance", "glTF", "Khronos Group", "MaterialX", "Metaverse Standards Forum", "OpenUSD", "Pixar USD", "Tool Interoperability", "Validation Systems", "Computer Vision", "metaverse", "Technical Standards" ] }, { "id": "asset-interoperability", "title": "Asset Interoperability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Asset Interoperability is the technical capacity for digital assets \u2014 including 3D models, avatars, NFTs, in-game items, and financial instruments \u2014 to function, be recognised, and retain their properties across multiple distinct platforms, ecosystems, or blockchains without loss of fidelity or ownership provenance. It encompasses both syntactic compatibility (shared file formats and protocols) and semantic compatibility (consistent meaning of asset attributes across contexts). Achieving asset interoperability typically requires common standards bodies, bridge contracts, or cross-chain protocols that translate asset representations between heterogeneous systems. It is a foundational concern for open metaverse architectures, decentralised finance, and multi-chain gaming economies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:asset-interoperability", "labels": [ "Asset Interoperability", "3D Asset Interoperability" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "asset-inventory", "title": "Asset Inventory", "domain": "security", "domain_name": "Security", "definition": "Asset Inventory is the systematic, continuously maintained catalogue of all hardware, software, data stores, network devices and cloud resources within an organisation's environment. It establishes the authoritative record of what must be protected, forming the foundation for vulnerability management, configuration control and incident response. An accurate inventory enables defenders to scope attack surfaces, prioritise patching and detect unauthorised or rogue assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:asset-inventory", "labels": [ "Asset Inventory" ], "is_subclass_of": [ "Vulnerability Management" ], "wikilinks": [] }, { "id": "asset-liquidity", "title": "Asset Liquidity", "domain": "finance", "domain_name": "Finance", "definition": "Asset Liquidity is the degree to which a digital or tokenised asset can be converted into cash or another asset at close to its fair market value without materially moving that price or incurring prohibitive friction. In decentralised finance and tokenised asset markets, liquidity is determined by order-book depth, automated market maker reserves, and the breadth of platforms on which an asset can be traded.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:asset-liquidity", "labels": [ "Asset Liquidity" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "asset-management-system", "title": "Asset Management System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Asset Management System (AMS) is an integrated software platform that tracks, manages, and optimises the lifecycle of physical, digital, or financial assets across an organisation. It consolidates asset data including acquisition, depreciation, maintenance, and disposal into a single governed repository. Modern AMS platforms enforce audit trails, automate compliance reporting, and feed into enterprise resource planning (ERP) and financial systems. They enable organisations to reduce total cost of ownership, mitigate risk, and demonstrate regulatory compliance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:asset-management-system", "labels": [ "Asset Management System" ], "is_subclass_of": [ "Asset Management" ], "wikilinks": [] }, { "id": "asset-management", "title": "Asset Management", "domain": "finance", "domain_name": "Finance", "definition": "The professional management of digital and traditional assets through strategies including portfolio construction, risk management, custody, and performance optimization. In the context of fintech and DeFi, asset management encompasses crypto portfolio management, tokenized securities, yield optimization, and institutional-grade custody solutions that enable investors to navigate volatile digital markets while maximizing risk-adjusted returns.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:asset-management", "labels": [ "Asset Management", "Software Asset Management" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [ "Custody", "Financial Services", "Institutional Investment", "Portfolio Management", "Blockchain", "Decentralized Finance (DeFi)", "Risk Management", "Tokenization" ] }, { "id": "asset-metadata", "title": "Asset Metadata", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asset Metadata is the structured descriptive, administrative, and technical information attached to a digital asset that contextualises it within a system without being the primary content of that asset. It encompasses identifiers, provenance records, format specifications, rights information, semantic tags, and dependency relationships, enabling discovery, validation, licensing, and lifecycle management.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-metadata", "labels": [ "Asset Metadata" ], "is_subclass_of": [ "Metadata" ], "wikilinks": [] }, { "id": "asset-optimization", "title": "Asset Optimization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Asset Optimization is the systematic process of reducing the computational cost, memory footprint, and bandwidth consumption of digital assets while preserving their perceptual quality and functional correctness for target deployment environments. It applies techniques including polygon reduction, texture compression, audio resampling, and level-of-detail generation to ensure assets perform acceptably on constrained hardware.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-optimization", "labels": [ "Asset Optimization" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "asset-pipeline", "title": "Asset Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An asset pipeline is an automated, staged workflow that ingests raw digital content\u2014meshes, textures, audio, animations, and shaders\u2014and transforms it through validation, processing, optimisation, and packaging steps into runtime-ready formats consumable by real-time engines, streaming platforms, or spatial computing environments. It enforces deterministic builds, enables version-controlled dependency graphs, and dramatically reduces manual content-preparation labour. Asset pipelines are foundational to game development, visual effects, and metaverse platform engineering, and increasingly incorporate AI-assisted level-of-detail generation, texture compression, and semantic tagging to scale content delivery across heterogeneous device targets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:asset-pipeline", "labels": [ "Asset Pipeline", "3D Asset Pipeline", "Game Asset Pipeline" ], "is_subclass_of": [ "Content Pipeline" ], "wikilinks": [] }, { "id": "asset-portability", "title": "asset portability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Asset portability is the technical and standards-based capability to transfer, import, and faithfully render digital assets \u2014 including 3D models, textures, avatars, animations, and environments \u2014 across different rendering engines, platforms, and virtual world applications without loss of fidelity or functionality. It depends on open file formats, agreed semantic conventions for material properties and skeletal rigs, and runtime interoperability layers. Asset portability is a prerequisite for platform-neutral digital property ownership in the open metaverse, enabling creators and users to carry value across application boundaries. It intersects with data portability principles in broader digital-rights discourse and with blockchain-based ownership for non-fungible digital items.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:asset-portability", "labels": [ "Asset Portability", "Cross-Game Asset Portability", "Digital Asset Portability" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "asset-recovery", "title": "Asset Recovery", "domain": "finance", "domain_name": "Finance", "definition": "Asset recovery is the process of identifying, tracing, freezing, and ultimately returning funds or property obtained through fraud, corruption, money laundering, or other financial crime. It combines financial investigation, legal action, and cross-border cooperation to follow illicit value through layered transactions and recover it for victims or the state. In the digital era it increasingly relies on blockchain analytics to trace cryptocurrency proceeds.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:asset-recovery", "labels": [ "Asset Recovery" ], "is_subclass_of": [ "Anti-Money Laundering" ], "wikilinks": [] }, { "id": "asset-registry", "title": "Asset Registry", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Asset Registry is a cryptographically-secured, on-chain or hybrid on/off-chain system for recording, authenticating, and transferring legal ownership rights and provenance metadata for real-world or digital assets, implemented through Smart Contracts, Distributed Ledger infrastruct...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:asset-registry", "labels": [ "Asset Registry", "Asset Cataloging", "Asset Register" ], "is_subclass_of": [ "Token and Asset", "Blockchain Application", "Distributed Ledger", "Ownership System", "Financial Infrastructure" ], "wikilinks": [ "Asset Tokenisation", "Automated Compliance", "Centralised Database", "Chainlink", "CMTA Token Standard", "Cross-Border Settlement", "Decentralised Finance", "ERC-3643 Standard", "FinancialInfrastructureDomain", "Fractional Ownership", "ISO 24165", "LegalTechDomain", "Ownership System", "Paper-Based Land Registry", "Polymesh Protocol", "Real Estate Registry", "Real World Asset Tokenisation", "Secondary Market Liquidity", "Securities Settlement", "Supply Chain Tracking" ] }, { "id": "asset-service", "title": "Asset Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Asset Service is a runtime or backend system that exposes digital assets and their associated metadata to consuming applications through well-defined APIs, handling storage, retrieval, transcoding, access control, and lifecycle operations as managed infrastructure. It abstracts the complexity of distributed storage and processing from client applications, enabling consistent, scalable access to asset libraries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:asset-service", "labels": [ "Asset Service" ], "is_subclass_of": [ "Service Layer" ], "wikilinks": [] }, { "id": "asset-tokenisation", "title": "asset tokenisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Asset tokenisation is the process of representing ownership, revenue rights, or access rights in a real-world or digital asset as a cryptographically secured blockchain token governed by a smart contract. The token encodes legally binding claims \u2014 to real estate, private equity, bonds, commodities, or intellectual property \u2014 while embedding compliance logic such as KYC whitelisting and transfer restrictions directly on-chain. Tokenisation enables fractional ownership, continuous secondary trading, and automated settlement without traditional intermediaries, sitting at the convergence of securities regulation, distributed ledger infrastructure, and programmable finance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:asset-tokenisation", "labels": [ "Asset Tokenisation", "Asset Tokenization", "Digital Asset Tokenisation", "Tokenisation of Assets", "asset-tokenization" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "asset-tracking", "title": "Asset Tracking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asset tracking is the continuous identification and localisation of physical objects, such as inventory, equipment, and vehicles, across their lifecycle and movement. It combines identification technologies like RFID and BLE beacons with positioning systems and connectivity to give organisations real-time visibility of where assets are and in what condition. Asset tracking underpins supply-chain transparency, loss prevention, and operational efficiency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:asset-tracking", "labels": [ "Asset Tracking" ], "is_subclass_of": [ "Internet of Things" ], "wikilinks": [] }, { "id": "asset-trading", "title": "Asset Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The exchange of digital and traditional financial assets through centralized exchanges (CEXs), decentralized exchanges (DEXs), and hybrid platforms.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:asset-trading", "labels": [ "Asset Trading" ], "is_subclass_of": [ "DeFi and Economics", "Financial Services" ], "wikilinks": [ "Financial Services", "Market Making", "Asset Management", "Blockchain", "Decentralized Exchange", "Decentralized Finance (DeFi)", "Liquidity Pool", "Price Discovery", "Smart Contract" ] }, { "id": "asset-transfer", "title": "Asset Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Asset transfer is the act of moving ownership or control of a digital asset from one party or address to another, recorded as a state change on a ledger. It is the basic operation that custody arrangements must secure and that digital-ownership systems rely on to change the recorded holder of a token or asset. Cross-chain and tokenised real-world-asset transfer are specialised forms that move value across ledgers or between on-chain and off-chain representations.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:asset-transfer", "labels": [ "Asset Transfer" ], "is_subclass_of": [ "Custody" ], "wikilinks": [] }, { "id": "assimilation", "title": "Assimilation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:assimilation", "labels": [ "Assimilation", "Assimilate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "assistive-robotics", "title": "Assistive Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Assistive Robotics is a branch of robotics concerned with designing, building, and deploying robotic systems that augment, restore, or substitute impaired human physical and cognitive functions for people with disabilities, age-related decline, or rehabilitation needs. Systems span powered exoskeletons for gait restoration, robotic prostheses with sensorimotor feedback, autonomous mobility aids, and socially interactive companion robots that provide cognitive and emotional support. Operating in close physical proximity to humans, these devices must satisfy stringent requirements for safety, transparency, and user control while integrating sensing, AI-driven intent inference, and adaptive interfaces that respond to changing physiological states.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:assistive-robotics", "labels": [ "Assistive Robotics" ], "is_subclass_of": [ "Robotic System" ], "wikilinks": [] }, { "id": "assistive-technology-integration", "title": "Assistive Technology Integration", "domain": "ai", "domain_name": "Ai", "definition": "Assistive Technology Integration is the practice of embedding tools, software, and hardware designed to support people with disabilities into digital systems, applications, and physical environments so that those individuals can access and interact with technology equitably. It encompasses screen readers, alternative input devices, augmentative communication systems, and adaptive interfaces, requiring compliance with accessibility standards and universal design principles. Effective integration ensures that assistive technologies interoperate seamlessly with underlying platforms rather than operating as bolt-on afterthoughts. The discipline spans hardware-software co-design, API compatibility, and user-centred testing with disabled communities.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:assistive-technology-integration", "labels": [ "Assistive Technology Integration" ], "is_subclass_of": [ "Accessibility", "Inclusive Design", "Universal Design" ], "wikilinks": [ "Accessibility", "Accessibility Standard", "Accessibility Audit Tool", "Accessibility Tree", "Accessibility Captioning", "Haptic Feedback", "Gesture Recognition", "Inclusive Design", "Inclusive Xr Experience", "Inclusive Xr Design", "Inclusive Participation", "Accessible Experience", "Human Computer Interaction", "Universal Design", "Multimodal AI", "Multimodal Interaction", "Extended Reality (XR)", "Augmented Reality (AR)", "Neural XR Interfaces", "User Centred Design" ] }, { "id": "assistive-technology-support", "title": "Assistive Technology Support", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Assistive Technology Support is the set of capabilities that make software interoperate correctly with assistive tools such as screen readers, magnifiers, switch access, and voice control. It is achieved by exposing semantic roles, names, states, and keyboard operability through accessibility APIs so that users with disabilities can perceive and operate an interface. Robust support is a prerequisite for accessible experiences and for compliance with standards like WCAG.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:assistive-technology-support", "labels": [ "Assistive Technology Support" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "assistive-technology", "title": "Assistive Technology", "domain": "standards", "domain_name": "Standards", "definition": "Assistive technology is equipment or software that helps people with disabilities perform tasks they would otherwise find difficult. Examples include screen readers, hearing aids and alternative input devices.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:assistive-technology", "labels": [ "Assistive Technology", "Assistive Technology Compatibility" ], "is_subclass_of": [ "Accessibility" ], "wikilinks": [ "Education", "Accessibility" ] }, { "id": "association-rule-learning", "title": "Association Rule Learning", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:association-rule-learning", "labels": [ "Association Rule Learning" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "asteroid", "title": "Asteroid", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:asteroid", "labels": [ "Asteroid" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "astrobiology", "title": "Astrobiology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astrobiology", "labels": [ "Astrobiology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astrometry", "title": "Astrometry", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astrometry", "labels": [ "Astrometry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astronomical-body", "title": "Astronomical Body", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astronomical-body", "labels": [ "Astronomical Body" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astronomical-photometry", "title": "Astronomical Photometry", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astronomical-photometry", "labels": [ "Astronomical Photometry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astronomical-spectroscopy", "title": "Astronomical Spectroscopy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astronomical-spectroscopy", "labels": [ "Astronomical Spectroscopy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astronomical-time-system", "title": "Astronomical Time System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astronomical-time-system", "labels": [ "Astronomical Time System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astronomy", "title": "Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astronomy", "labels": [ "Astronomy" ], "is_subclass_of": [ "Space Science" ], "wikilinks": [] }, { "id": "astrophysical-jet", "title": "Astrophysical Jet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astrophysical-jet", "labels": [ "Astrophysical Jet" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "astrophysics", "title": "Astrophysics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:astrophysics", "labels": [ "Astrophysics" ], "is_subclass_of": [ "Space Science" ], "wikilinks": [] }, { "id": "asymmetric-cryptography", "title": "Asymmetric Cryptography", "domain": "security", "domain_name": "Security", "definition": "Asymmetric Cryptography is a cryptographic paradigm in which mathematically related key pairs \u2014 a public key and a private key \u2014 serve distinct roles: the public key may be freely shared and used to encrypt messages or verify signatures, while the private key is kept secret and used to decrypt or sign. Security rests on the computational intractability of reversing the underlying mathematical problems without knowledge of the private key.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:asymmetric-cryptography", "labels": [ "Asymmetric Cryptography", "Asymmetric Key Cryptography" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "asymmetric-encryption", "title": "Asymmetric Encryption", "domain": "security", "domain_name": "Security", "definition": "A cryptographic system using mathematically related public-private key pairs where the public key encrypts data that only the corresponding private key can decrypt, enabling secure communication, digital signatures, and trustless verification without pre-shared secrets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:asymmetric-encryption", "labels": [ "Asymmetric Encryption" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain" ], "wikilinks": [ "Blockchain", "DID Nostr Identity", "Digital Signature", "Elliptic Curve Cryptography", "Hash Function", "Information Security", "Key Derivation Function", "Public Key Infrastructure", "Symmetric Encryption" ] }, { "id": "asymmetric-key-pair", "title": "Asymmetric Key Pair", "domain": "security", "domain_name": "Security", "definition": "An asymmetric key pair is the pair of mathematically related keys, a public key and a private key, that together enable asymmetric cryptography: data encrypted or signed with one key can only be decrypted or verified with the other. The private key must be kept secret by its owner, while the public key can be freely distributed, allowing anyone to encrypt a message intended for the owner or verify a signature the owner has produced. Asymmetric key pairs underpin public-key infrastructure, digital signatures, and blockchain account ownership.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:asymmetric-key-pair", "labels": [ "Asymmetric Key Pair" ], "is_subclass_of": [ "Asymmetric Cryptography" ], "wikilinks": [] }, { "id": "asyncapi", "title": "Asyncapi", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AsyncAPI is an open specification for describing event-driven and message-based APIs in a machine-readable document, analogous to how OpenAPI describes request-response REST APIs. It defines channels, messages, payloads and the protocols and brokers used to exchange them, enabling documentation, code generation, validation and tooling for asynchronous systems. AsyncAPI standardises contracts for publish-subscribe and streaming architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:asyncapi", "labels": [ "Asyncapi", "AsyncAPI" ], "is_subclass_of": [ "API Specification" ], "wikilinks": [] }, { "id": "asynchronous-collaboration", "title": "Asynchronous Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Time-independent collaboration mode where distributed participants contribute at different times through shared persistent artifacts, enabling flexible schedules, deep work periods, and global accessibility while maintaining coordination through explicit documentation and version control.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:asynchronous-collaboration", "labels": [ "Asynchronous Collaboration", "Asynchronous Collaboration Patterns" ], "is_subclass_of": [ "Telecollaboration", "TC-0001-telecollaboration-domain" ], "wikilinks": [ "Documentation Culture", "Documentation Systems", "Global Accessibility", "ISO", "RFC", "TC-0001-telecollaboration-domain", "TC-0010-Synchronous-Collaboration", "TC-0021-Collaborative-Document-Editing", "TC-0022-Version-Control", "TC-0024-Project-Management-System", "TC-0025-Issue-Tracking", "TC-0026-Code-Review", "TC-0080-Team-Coordination", "Knowledge Graph" ] }, { "id": "asynchronous-communication", "title": "Asynchronous Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asynchronous communication is a messaging paradigm in which senders and receivers operate independently in time: the sender dispatches a message and immediately resumes processing without blocking, while the message is buffered, queued, or stored until the recipient is ready to consume it. This temporal decoupling eliminates tight runtime coupling between system components, enabling fault isolation, backpressure management, and geographic distribution across heterogeneous networks. The pattern underpins modern distributed architectures including event-driven systems, message-oriented middleware, and stream-processing platforms, and contrasts sharply with synchronous request-response protocols where the caller blocks awaiting a reply. By permitting independent scaling, retry semantics, and durable delivery guarantees, asynchronous communication is foundational to resilient, cloud-native, and edge-deployed systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:asynchronous-communication", "labels": [ "Asynchronous Communication", "Asynchronous Encrypted Communication" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "asynchronous-coordination", "title": "Asynchronous Coordination", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asynchronous Coordination is the alignment of distributed agents or processes that act without a shared global clock or blocking synchronization, communicating through messages, shared state, or eventual-consistency mechanisms. It tolerates network delay and partial failure by letting participants make progress independently and reconcile state later. This model underpins resilient distributed systems and decentralized multi-agent and swarm control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:asynchronous-coordination", "labels": [ "Asynchronous Coordination" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "asynchronous-execution", "title": "Asynchronous Execution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An execution model in which operations are initiated without blocking the calling thread; completion is signalled via callbacks, promises, futures, or events, enabling high-throughput concurrent processing particularly suited to I/O-bound workloads and distributed system communication.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:asynchronous-execution", "labels": [ "Asynchronous Execution" ], "is_subclass_of": [ "Execution Model" ], "wikilinks": [ "Core Technology", "Execution Model", "Edge Computing" ] }, { "id": "asynchronous-messaging", "title": "Asynchronous Messaging", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asynchronous Messaging is a communication pattern in which a sender dispatches a message to an intermediary and continues processing without waiting for the receiver to respond. Messages are buffered in queues or brokers and consumed when downstream services are ready, decoupling producers from consumers in time and load. This pattern improves resilience, scalability and fault tolerance in distributed and event-driven architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:asynchronous-messaging", "labels": [ "Asynchronous Messaging" ], "is_subclass_of": [ "Message Queue" ], "wikilinks": [] }, { "id": "asynchronous-programming", "title": "Asynchronous Programming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Asynchronous programming is a concurrency model in which operations that would otherwise block \u2014 such as input/output, network calls or timers \u2014 are initiated without halting the executing thread, allowing other work to proceed until results become available. It uses constructs such as callbacks, promises, futures and async/await to express continuations cleanly. The approach improves responsiveness and throughput for input/output-bound workloads without the overhead of one thread per task.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:asynchronous-programming", "labels": [ "Asynchronous Programming" ], "is_subclass_of": [ "Concurrency", "Software Engineering" ], "wikilinks": [] }, { "id": "asynchronous-video", "title": "Asynchronous Video", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Asynchronous video is a Distributed Collaboration communication modality in which recorded video messages \u2014 combining screen capture, webcam footage, audio narration, and on-screen annotation \u2014 are produced by a sender and consumed by recipients independently of the sender's presence, elimina...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:asynchronous-video", "labels": [ "Asynchronous Video" ], "is_subclass_of": [ "Communication Technology", "Asynchronous Communication", "Video-based Communication", "Distributed Collaboration", "Knowledge Management", "Digital Workplace Technology" ], "wikilinks": [ "AI Video Summary", "AOM AV1 Specification", "Async-First Work Culture", "Async Standup", "Asynchronous Code Review", "Asynchronous Communication", "Audio Narration", "Automatic Speech Recognition", "AV1 Video Codec", "Browser-based Screen Capture", "Chapter Marker", "Claap", "Click Tracking", "Cloud Storage", "Communication Synchronicity Theory", "CommunicationTechnologyDomain", "Complex Decision Documentation", "Confluence", "Content Delivery Network", "Customer Onboarding" ] }, { "id": "at-protocol", "title": "At Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The AT Protocol (Authenticated Transfer Protocol) is an open, federated networking protocol for decentralised social applications, originally developed for Bluesky, that combines portable accounts, signed data repositories, and account migration with a separation between data hosting and algorithmic curation. It uses decentralised identifiers and content-addressed records so users own their identity and data independently of any single provider. The protocol aims to deliver large-scale interoperable social networking with user choice over feeds and moderation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:at-protocol", "labels": [ "At Protocol", "AT Protocol" ], "is_subclass_of": [ "Open Standards" ], "wikilinks": [] }, { "id": "atmosphere-layer", "title": "Atmosphere Layer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:atmosphere-layer", "labels": [ "Atmosphere Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "atmosphere-revitalisation", "title": "Atmosphere Revitalisation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:atmosphere-revitalisation", "labels": [ "Atmosphere Revitalisation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "atmospheric-composition-monitoring", "title": "Atmospheric Composition Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:atmospheric-composition-monitoring", "labels": [ "Atmospheric Composition Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "atmospheric-correction", "title": "Atmospheric Correction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:atmospheric-correction", "labels": [ "Atmospheric Correction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "atmospheric-drag", "title": "Atmospheric Drag", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:atmospheric-drag", "labels": [ "Atmospheric Drag" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "atomic-broadcast", "title": "Atomic Broadcast", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Atomic broadcast (also called total-order broadcast) is a distributed communication primitive guaranteeing that all correct processes deliver the same set of messages in the same total order. It strengthens reliable broadcast with an ordering property, ensuring agreement on both the content and sequence of delivered messages. Atomic broadcast is equivalent in power to consensus and underpins state-machine replication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:atomic-broadcast", "labels": [ "Atomic Broadcast" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "atomic-delivery-versus-payment", "title": "Atomic Delivery versus Payment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Atomic Delivery versus Payment (atomic DvP) is a settlement mechanism in which the transfer of an asset and the corresponding payment are executed as a single indivisible transaction: either both legs complete simultaneously or neither does, eliminating counterparty risk by making it impossible for one party to receive value without the other receiving theirs. It applies the atomicity property of database transactions to financial and digital asset exchange.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:atomic-delivery-versus-payment", "labels": [ "Atomic Delivery versus Payment", "Atomic DvP Settlement" ], "is_subclass_of": [ "Delivery-Versus-Payment" ], "wikilinks": [] }, { "id": "atomic-settlement", "title": "Atomic Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Atomic settlement is a transaction completion mechanism in which the transfer of assets between two or more parties either executes in its entirety or not at all, eliminating counterparty risk and the possibility of partial fulfilment. The mechanism is enforced at the protocol level, ensuring that delivery and payment occur simultaneously and indivisibly within a single transaction or smart contract execution. This property derives from atomicity \u2014 one of the four ACID properties of database transactions \u2014 applied to financial and digital asset exchanges. Atomic settlement is the foundational guarantee underpinning trustless exchange protocols, cross-chain bridges, real-time gross settlement systems, and delivery-versus-payment architectures in both decentralised and regulated financial markets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:atomic-settlement", "labels": [ "Atomic Settlement" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "atomic-swap", "title": "Atomic Swap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trustless peer-to-peer cryptocurrency exchange mechanism across different blockchains using Hash Time-Locked Contracts (HTLCs) that guarantees atomic execution where the swap either completes fully or not at all, eliminating counterparty risk without requiring centralized intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:atomic-swap", "labels": [ "Atomic Swap", "Atomic Cross-Asset Swaps", "Atomic Swaps", "AtomicSwap", "Cross-Chain Atomic Swap" ], "is_subclass_of": [ "Blockchain Protocol", "Blockchain" ], "wikilinks": [ "Cross-Chain Interoperability", "ERC-20 Token", "Hash Time-Locked Contract", "Litecoin", "Payment Channel", "Polkadot", "Bitcoin", "Blockchain", "Cross-Chain Bridge", "Decentralized Exchange", "Ethereum", "Hash Function", "Lightning Network", "Virtual Economy" ] }, { "id": "attack-vector", "title": "Attack Vector", "domain": "security", "domain_name": "Security", "definition": "A specific path, method, or mechanism that a threat actor uses to gain unauthorised access, exploit vulnerabilities, or cause damage to a system, network, or organisation. Attack vectors span network-based exploitation, social engineering, supply-chain compromise, and AI-specific techniques such as prompt injection or model poisoning, and are categorised by access method, target layer, and sophistication.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:attack-vector", "labels": [ "Attack Vector", "Attack Vector Inventory" ], "is_subclass_of": [ "Security and Identity", "Artificial Intelligence" ], "wikilinks": [ "ISO", "NIST", "Artificial Intelligence", "Blockchain", "Resilience", "Risk", "Security", "Threat Actor", "Vulnerability" ] }, { "id": "attention-aware-interaction", "title": "Attention Aware Interaction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Attention Aware Interaction refers to human-computer interaction techniques that leverage eye tracking, gaze detection, and attention modeling to understand user focus and adapt interfaces accordingly, enabling foveated rendering, gaze-based selection, and contextual content presentation in VR, A...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:attention-aware-interaction", "labels": [ "Attention Aware Interaction", "Attention-Aware Interaction" ], "is_subclass_of": [ "AI Application", "Human Computer Interaction" ], "wikilinks": [ "Adaptive Interfaces", "Attention Modeling", "Eye Tracking Hardware", "Foveated Rendering", "Gaze-Based Selection", "Gaze Detection Algorithms", "Sensor Input", "Human-Computer Interaction", "metaverse" ] }, { "id": "attention-economy", "title": "Attention Economy", "domain": "governance", "domain_name": "Governance", "definition": "An economic framing, articulated by Herbert Simon in 1971, in which human attention is treated as the scarce resource that information-rich systems compete to capture, allocate, and monetise. In digital markets it describes the business logic of advertising-funded platforms whose revenue scales with engagement, driving recommendation algorithms, infinite feeds, and notification design that optimise for time-on-platform \u2014 with documented consequences for information quality, creator livelihoods, and individual wellbeing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:attention-economy", "labels": [ "Attention Economy" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [ "Digital Economy", "Content Curation", "Creator Monetization", "Digital Society Harms" ] }, { "id": "attention-head", "title": "Attention Head", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "One of multiple parallel scaled dot-product attention mechanisms in a multi-head attention layer, each operating over a distinct linear projection of the input. Individual heads specialise in different linguistic or structural patterns; most learn simple positional relationships and many can be pruned without significant loss, though a subset carry disproportionate representational load.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:attention-head", "labels": [ "Attention Head" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "Head Gaze", "ArtificialIntelligenceDomain", "Computer Vision" ] }, { "id": "attention-mask", "title": "Attention Mask", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A mechanism that controls which positions in a sequence can attend to which other positions, typically implemented by adding large negative values before softmax to zero out unwanted attention weights; used for causal masking in autoregressive decoders and padding masking in batched variable-length sequences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:attention-mask", "labels": [ "Attention Mask" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Telecollaboration" ] }, { "id": "attention-mechanism", "title": "Attention Mechanism", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Attention Mechanism is a neural network component that enables models to dynamically weight the relevance of different input positions when producing each output element, computing weighted combinations based on learned similarity scores. Originally introduced for sequence-to-sequence machine translation, self-attention and multi-head attention are now the core computational primitives of transformer architectures, enabling parallel processing of sequences and capturing long-range dependencies that recurrent models struggle with.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:attention-mechanism", "labels": [ "Attention Mechanism", "Attention Model", "Spatial Attention Mechanism" ], "is_subclass_of": [ "Neural Network Component" ], "wikilinks": [ "ISO/IEC 22989:2022", "NIST AI RMF", "ArtificialIntelligenceDomain", "Telecollaboration" ] }, { "id": "attention-mechanisms", "title": "Attention Mechanisms", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Attention mechanisms are a family of neural network components that compute context-dependent weighted combinations of value representations by scoring the relevance of query vectors against key vectors, enabling each output position to draw selectively from any position in an input sequence \u2014 overcoming the fixed-length bottleneck of recurrent encoders. Originally introduced as additive (Bahdanau) attention for neural machine translation in 2015, they were generalised to scaled dot-product attention and multi-head attention in the 2017 Transformer architecture, where they became the sole sequence-modelling primitive, replacing recurrence entirely. Attention mechanisms now underpin virtually all frontier deep learning systems across language, vision, speech, biology, and multimodal reasoning.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:attention-mechanisms", "labels": [ "Attention Mechanisms" ], "is_subclass_of": [ "Attention Mechanism", "Neural Network Component", "AI Model Architecture" ], "wikilinks": [ "Attention Mechanism", "Neural Network", "Transformer", "Transformer Architecture", "Multi-Head Attention", "Self Attention", "Cross Attention", "Causal Attention", "Flash Attention", "Scaled Dot Product Attention", "Grouped Query Attention", "Attention Head", "Attention Weight", "Attention Mask", "Deep Learning", "Backpropagation", "Vision Transformer", "Diffusion Transformer", "Graph Attention Network", "LSTM" ] }, { "id": "attention-weight", "title": "Attention Weight", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A scalar coefficient produced by an attention mechanism that quantifies the relevance of one position (key/value) to another (query) in a sequence or across modalities. Attention weights are computed via a softmax over scaled dot-products of query and key vectors, and govern how much each value contributes to the output representation. They are the core computational primitive of Transformer-based models.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:attention-weight", "labels": [ "Attention Weight" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network" ], "wikilinks": [ "Artificial Intelligence", "Telecollaboration", "Attention Mechanism", "Transformer Architecture", "Self Attention", "Multi-Head Attention", "Cross Attention", "Query Key Value", "Softmax Function", "Positional Encoding", "Large Language Models", "Natural Language Processing", "Explainable AI", "Backpropagation", "Gradient Descent", "Neural Network", "Machine Translation", "Computer Vision", "Embedding", "Matrix Multiplication" ] }, { "id": "attention", "title": "Attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Attention is a differentiable content-based addressing mechanism for neural networks that computes a weighted combination of value vectors according to learned compatibility scores between a query vector and a set of key vectors, formalised in its most influential form as Scaled Dot-Product Atten...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:attention", "labels": [ "Attention", "Attention Management" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network", "Neural Network Mechanism", "Differentiable Memory", "Sequence Model", "Representation Learning" ], "wikilinks": [ "ACL", "Ainslie et al. 2023 GQA", "AlgorithmLayer", "Attention Score", "Bahdanau et al. 2015 Neural Machine Translation Attention", "Beltagy et al. 2020 Longformer", "Brown et al. 2020 GPT-3", "Causal Mask", "Child et al. 2019 Sparse Transformer", "Choromanski et al. 2021 Performer", "Content-Based Addressing", "Cross-Modal Conditioning", "CVPR", "Dao 2023 FlashAttention-2", "Dao et al. 2022 FlashAttention", "DeepLearningDomain", "DeepSeek-AI 2024 DeepSeek-V2 MLA", "Devlin et al. 2019 BERT", "Differentiable Architecture", "Differentiable Lookup" ] }, { "id": "attestation", "title": "Attestation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Attestation is the act of producing verifiable, signed evidence that a claim, state, or property is true, allowing a relying party to trust it without re-deriving it. In blockchain proof-of-stake consensus, validators broadcast attestations voting on the head of the chain and on checkpoints, and the aggregate of these signed votes drives finalisation. More broadly, remote attestation lets a trusted execution environment cryptographically prove its identity and integrity to a remote verifier. Attestations are typically cryptographic signatures over structured claims, and they underpin trust, accountability, and slashing-based security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:attestation", "labels": [ "Attestation" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [] }, { "id": "attitude-control-thruster", "title": "Attitude Control Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:attitude-control-thruster", "labels": [ "Attitude Control Thruster" ], "is_subclass_of": [ "Actuator", "Thruster" ], "wikilinks": [] }, { "id": "attitude-control", "title": "Attitude Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:attitude-control", "labels": [ "Attitude Control" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "attitude-determination", "title": "Attitude Determination", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:attitude-determination", "labels": [ "Attitude Determination" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "attitude-estimation", "title": "Attitude Estimation", "domain": "robotics", "domain_name": "Robotics", "definition": "Attitude estimation is the process of determining the orientation of a rigid body, typically expressed as roll, pitch and yaw or as a quaternion, relative to a reference frame. It fuses measurements from an inertial measurement unit's gyroscopes, accelerometers and magnetometers using sensor-fusion algorithms such as complementary or Kalman filters to counteract individual sensor drift and noise. Accurate attitude estimation is a prerequisite for stable flight control, robot balancing and any system that must reason about its own orientation in space.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:attitude-estimation", "labels": [ "Attitude Estimation" ], "is_subclass_of": [ "Sensor Fusion" ], "wikilinks": [] }, { "id": "attitude-quaternion", "title": "Attitude Quaternion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:attitude-quaternion", "labels": [ "Attitude Quaternion" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "attribute-sharing", "title": "Attribute Sharing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Attribute Sharing refers to identity federation mechanisms that enable selective disclosure of user credentials and identity attributes across metaverse platforms and blockchain applications, using Self-Sovereign Identity (SSI) principles and decentralized identifiers to maintain user control whi...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:attribute-sharing", "labels": [ "Attribute Sharing" ], "is_subclass_of": [ "Governance and Safety", "Identity Management" ], "wikilinks": [ "Blockchain Anchoring", "Credential Verification", "Selective Disclosure", "Cross-Platform Identity", "Decentralized Identifiers", "DID Nostr Identity", "Identity Management", "metaverse", "Verifiable Credentials" ] }, { "id": "attribute-based-access-control", "title": "Attribute-Based Access Control", "domain": "security", "domain_name": "Security", "definition": "Attribute-Based Access Control (ABAC) is an authorisation model that grants or denies access by evaluating policies against attributes of the subject, resource, action, and environment rather than against static role assignments. Decisions are computed dynamically from these attributes at request time, enabling fine-grained, context-aware control. ABAC is a cornerstone of zero-trust architectures because it can express rich conditional rules without proliferating roles.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:attribute-based-access-control", "labels": [ "Attribute-Based Access Control" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "auction-mechanism", "title": "Auction Mechanism", "domain": "data", "domain_name": "Data", "definition": "An auction mechanism is a structured rule set for allocating goods or resources and determining prices through competitive bidding among participants with private valuations. It specifies how bids are collected, who wins, and what each winner pays, with classic forms including English, Dutch, first-price sealed-bid, and Vickrey (second-price) auctions, each inducing different bidding incentives. Auction mechanisms are central to market design, online advertising, and resource allocation in multi-agent and computational settings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:auction-mechanism", "labels": [ "Auction Mechanism" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "auction-theory", "title": "Auction Theory", "domain": "finance", "domain_name": "Finance", "definition": "Auction theory is a branch of applied mathematics, economics, and game theory that analyses competitive bidding mechanisms through which goods, services, or rights are allocated to participants. It characterises how auction format rules (e.g. English, Dutch, Vickrey), information structure (private vs. common values), and bidder strategy interact to determine allocative efficiency, revenue, and equilibrium outcomes. The field provides formal tools for designing mechanisms that incentivise truthful bidding, prevent collusion, and optimise social welfare or seller revenue. Foundational results include the Revenue Equivalence Theorem and optimal auction design frameworks due to Myerson.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:auction-theory", "labels": [ "Auction Theory" ], "is_subclass_of": [ "Game Theory" ], "wikilinks": [ "Game Theory", "Gas Mechanism", "Economics" ] }, { "id": "audience-engagement", "title": "Audience Engagement", "domain": "ai", "domain_name": "Ai", "definition": "Audience engagement is the set of strategies and interactive mechanisms by which creators and platforms capture, sustain and deepen the attention and participation of viewers. It spans techniques such as live polls, question-and-answer sessions, gamification, reactions and personalised content, measured through metrics like watch time, interaction rate and retention. Strong audience engagement drives community growth, loyalty and monetisation across live and on-demand media.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:audience-engagement", "labels": [ "Audience Engagement" ], "is_subclass_of": [ "User Engagement", "User Experience", "Content Creation" ], "wikilinks": [ "User Engagement", "Content Creation", "Live Streaming", "Gamification", "Live Polls and QandA", "Creator Economy", "Content Distribution", "Virtual Event", "Recommendation Systems", "Hyper personalisation", "Social Platform", "Digital Content", "Natural Interaction", "Reward Function", "User Interface", "User Experience", "Loyalty Programs", "Video Streaming", "Content Delivery Network", "Real-Time Communication" ] }, { "id": "audience-segmentation", "title": "Audience Segmentation", "domain": "data", "domain_name": "Data", "definition": "Audience segmentation is the practice of dividing a population of users or customers into distinct groups that share attributes, behaviours, or needs. Segments are derived from demographic, behavioural, contextual, and inferred signals and are used to tailor messaging, targeting, and experiences. Segmentation makes personalisation, efficient targeting, and measurement possible across marketing and product functions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:audience-segmentation", "labels": [ "Audience Segmentation" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [] }, { "id": "audio-codec", "title": "Audio Codec", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An audio codec is a technology that encodes audio into a compressed digital representation and decodes it back for playback, reducing storage and bandwidth requirements. Lossy codecs discard perceptually less important information to achieve high compression, while lossless codecs preserve the original signal exactly. Codecs balance bitrate, audio quality, latency and computational cost, and define container-independent bitstream formats. They are essential to streaming, telephony, broadcasting and immersive spatial audio.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:audio-codec", "labels": [ "Audio Codec" ], "is_subclass_of": [ "Audio Processing" ], "wikilinks": [] }, { "id": "audio-compression", "title": "Audio Compression", "domain": "data", "domain_name": "Data", "definition": "Audio Compression is the process of reducing the storage size or bandwidth required for audio data, using lossy codecs that discard perceptually less-important information or lossless codecs that preserve the exact waveform. It underlies streaming, spatial audio delivery, and real-time voice communication, balancing fidelity against transmission constraints. Common lossy approaches use psychoacoustic models to remove sound outside human hearing thresholds.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:audio-compression", "labels": [ "Audio Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [] }, { "id": "audio-driver", "title": "Audio Driver", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An audio driver is the low-level software component that mediates between an operating system's audio subsystem and physical sound hardware, translating generic playback and capture requests into device-specific commands. It manages buffering, sample-rate conversion, latency and device enumeration, and is a prerequisite for any higher-level audio pipeline to reach speakers, headphones or microphones. In spatial and immersive computing, driver latency and buffer size directly affect the achievable audio-visual synchronisation.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:audio-driver", "labels": [ "Audio Driver" ], "is_subclass_of": [ "Audio System" ], "wikilinks": [] }, { "id": "audio-engine", "title": "Audio Engine", "domain": "metaverse", "domain_name": "Metaverse", "definition": "An Audio Engine is a software subsystem that manages the real-time synthesis, processing, mixing, and spatialisation of sound within an interactive or generative application. It abstracts hardware audio interfaces, schedules audio computation on dedicated threads or hardware DSP units, and exposes higher-level APIs for triggering, routing, and modulating sound objects in response to application events.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:audio-engine", "labels": [ "Audio Engine", "Audio Synthesis Engine" ], "is_subclass_of": [ "Audio System" ], "wikilinks": [] }, { "id": "audio-generation", "title": "Audio Generation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Audio generation is the use of generative machine-learning models to synthesise sound \u2014 including speech, music, sound effects, and ambient audio \u2014 from inputs such as text, symbolic scores, or learned latent representations. Models learn the statistical structure of audio either in the raw waveform domain or in compressed time-frequency representations such as spectrograms, then sample new outputs that are perceptually realistic. The field encompasses text-to-speech, music generation, voice cloning, and general audio synthesis, drawing on autoregressive, diffusion, and adversarial generative architectures.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:audio-generation", "labels": [ "Audio Generation" ], "is_subclass_of": [ "Generative AI", "Multimodal AI", "Deep Learning" ], "wikilinks": [ "Generative AI", "Generative Model", "Diffusion Model", "Latent Diffusion", "Generative Adversarial Network", "Autoregressive Model", "Neural Network", "Deep Learning", "Text-to-Speech", "Speech Synthesis", "Voice Cloning", "Music Generation", "Audio Processing", "Transformer Architecture", "Attention Mechanism", "Neural Vocoder", "Spectrogram", "Audio Codec", "Natural Language Processing", "Large Language Models" ] }, { "id": "audio-parameters", "title": "Audio Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Audio Parameters are the configurable settings and properties defining spatial audio behavior in virtual environments, including source position, direction, attenuation, room acoustics, HRTF profiles, and rendering parameters that enable realistic 3D sound experiences in VR, AR, and metaverse app...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:audio-parameters", "labels": [ "Audio Parameters" ], "is_subclass_of": [ "Content and Assets", "Display and Rendering", "Audio Configuration" ], "wikilinks": [ "Audio Configuration", "Audio Engine", "Environmental Modeling", "HRTF Data", "Immersive Audio", "Spatial Sound Rendering", "metaverse", "Telepresence" ] }, { "id": "audio-processing-system", "title": "Audio Processing System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Audio Processing System is a hardware or software subsystem responsible for capturing, transforming, and rendering audio signals within spatial computing environments. It encompasses digital signal processing, spatial audio rendering (binaural, ambisonics, HRTF), noise cancellation, speech recognition integration, and real-time mixing for immersive presence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:audio-processing-system", "labels": [ "Audio Processing System" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing", "Telepresence" ] }, { "id": "audio-processing", "title": "Audio Processing", "domain": "ai", "domain_name": "Ai", "definition": "Audio processing is the computational manipulation of digital audio signals to transform, analyse, enhance, or synthesise sound, encompassing operations such as filtering, compression, equalisation, spatialisation, and feature extraction. It operates on discrete-time representations of acoustic waveforms using algorithms drawn from digital signal processing theory, spanning both time-domain and frequency-domain approaches. Applications range from consumer media playback and telecommunications to professional studio production, speech recognition pipelines, neural audio codecs, and immersive spatial audio for extended reality environments. Machine learning has significantly expanded the discipline, enabling source separation, generative synthesis, and zero-shot audio enhancement that were previously intractable.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:audio-processing", "labels": [ "Audio Processing" ], "is_subclass_of": [ "Signal Processing", "Digital Signal Processing" ], "wikilinks": [ "Digital Signal Processing", "Signal Processing", "Fourier Analysis", "Fast Fourier Transform", "Nyquist-Shannon Sampling Theorem", "Audio Codec", "Noise Cancellation", "Audio Feature Extraction", "Analogue-to-Digital Conversion", "Spectrogram", "Psychoacoustics", "Speech Recognition", "Text-to-Speech", "Music Generation", "Voice Activity Detection", "Spatial Audio System", "Immersive Audio System", "Audio Spatialisation", "Spatial Audio", "Extended Reality" ] }, { "id": "audio-signal-processing", "title": "Audio Signal Processing", "domain": "ai", "domain_name": "Ai", "definition": "Audio Signal Processing is the application of signal processing theory and algorithms to the analysis, transformation, synthesis, and encoding of audio-frequency signals, operating in either the time domain or frequency domain. It encompasses filtering, equalisation, dynamic range control, time-frequency analysis, psychoacoustic coding, and spatial rendering as applied to sound reproduction, communication, and computational audition systems.", "entityType": "Class", "qualityScore": 0.92, "maturity": "mature", "iri": "urn:ngm:class:audio-signal-processing", "labels": [ "Audio Signal Processing" ], "is_subclass_of": [ "Signal Processing", "Audio Processing" ], "wikilinks": [ "Signal Processing", "Digital Signal Processing", "Fast Fourier Transform", "Digital Filter", "Convolution", "Audio Parameters", "Pulse-Code Modulation", "Spatial Audio", "Speech Recognition", "Speech Synthesis", "Audio Compression", "Noise Suppression", "Real-Time Computing", "Telecommunications", "Audio Engine", "Audio Processing System", "Neural Audio Codec", "Machine Learning", "Spatial Computing", "Psychoacoustics" ] }, { "id": "audio-spatialization", "title": "Audio Spatialization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Audio Spatialization is the technique of positioning sounds in three-dimensional space using Head-Related Transfer Functions (HRTFs), binaural processing, and ambisonics to create realistic 3D audio experiences that respond to listener position and head movement in virtual reality and immersive applications.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:audio-spatialization", "labels": [ "Audio Spatialization" ], "is_subclass_of": [ "Display and Rendering", "Spatial Audio Technology" ], "wikilinks": [ "3D Sound Perception", "Binaural Rendering", "Head Tracking", "HRTF Filters", "Immersive Audio", "Sensor Input", "Sound Localization", "Spatial Audio Technology", "metaverse" ] }, { "id": "audio-synthesis", "title": "Audio Synthesis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Audio synthesis is the generation of audio signals, including speech, music, and sound effects, from symbolic, textual, or latent representations. Contemporary approaches use deep generative models such as autoregressive networks, diffusion models, and neural vocoders to produce high-fidelity waveforms. It underpins text-to-speech, music generation, and sound design, and is a core modality of generative artificial intelligence.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:audio-synthesis", "labels": [ "Audio Synthesis" ], "is_subclass_of": [ "Generative AI", "Deep Generative Model", "Audio Generation" ], "wikilinks": [ "Generative AI", "Diffusion Model", "Diffusion Models", "Deep Learning", "Neural Network", "Text-to-Speech", "Speech Synthesis", "Music Generation", "Audio Processing", "GAN", "Model Evaluation", "Content Generation", "Transformer", "Natural Language Processing", "Audio Signal Processing", "Neural Audio Codec", "Variational Autoencoder", "Mel-Spectrogram", "Autoregressive Model", "Flow Matching" ] }, { "id": "audio-system", "title": "Audio System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Audio System is an integrated hardware and software architecture responsible for the capture, processing, transmission, and reproduction of sound signals within computing environments, including spatial audio rendering, acoustic signal processing, voice input/output, and environmental sound simulation. In spatial computing and extended reality contexts it delivers positional audio cues that reinforce presence and depth perception, coordinating microphone arrays, digital signal processors, codecs, and loudspeaker or headphone transducers. Audio systems implement psychoacoustic models \u2014 including head-related transfer functions (HRTFs) and room acoustics simulation \u2014 to produce convincing three-dimensional soundscapes. They underpin voice communication, speech interaction, accessibility features, and immersive media across consumer electronics, professional audio, telecommunication, and mixed-reality platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:audio-system", "labels": [ "Audio System" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [ "owl:Thing", "Telepresence" ] }, { "id": "audio-technology", "title": "Audio Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The hardware, software, and signal-processing techniques used to capture, synthesise, spatialise, and reproduce sound within spatial computing and metaverse environments. Audio technology in immersive contexts encompasses spatial audio rendering, binaural processing, voice interaction, and real-time acoustic simulation to enhance presence and communication fidelity.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:audio-technology", "labels": [ "Audio Technology" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing", "Telepresence" ] }, { "id": "audit-committee", "title": "Audit Committee", "domain": "governance", "domain_name": "Governance", "definition": "An audit committee is a sub-committee of an organisation's board of directors charged with overseeing financial reporting, internal controls, risk management, and the relationship with internal and external auditors. Composed primarily of independent non-executive directors, it provides assurance to the board and stakeholders on the integrity of financial statements and the effectiveness of control systems. The committee is a central pillar of corporate governance and regulatory compliance regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:audit-committee", "labels": [ "Audit Committee" ], "is_subclass_of": [ "Governance Structure" ], "wikilinks": [] }, { "id": "audit-function", "title": "Audit Function", "domain": "governance", "domain_name": "Governance", "definition": "An organisational capability or dedicated unit responsible for the planning, execution, and reporting of audit activities within or on behalf of an enterprise. The audit function provides independent assurance over risk management, internal controls, and governance processes, acting as a critical third line of defence in enterprise risk frameworks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:audit-function", "labels": [ "Audit Function", "Independent Audit Function" ], "is_subclass_of": [ "Audit" ], "wikilinks": [] }, { "id": "audit-log", "title": "Audit Log", "domain": "security", "domain_name": "Security", "definition": "An audit log is an immutable, chronologically ordered record of discrete system events that captures who performed an action, what action was performed, on which resource, at what time, and from what source context, providing an authoritative evidence trail for security investigation, regulatory compliance, and forensic analysis. Each entry is structured with a timestamp, actor identity, event type, affected object identifier, outcome status, and contextual metadata. Audit logs are foundational to accountability in information systems, distinguished from operational logs by their emphasis on human-actionable accountability and legal evidentiary weight rather than system diagnostics. In high-assurance environments, entries are cryptographically chained so that deletion or modification of any record invalidates all subsequent hashes, making tampering detectable.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:audit-log", "labels": [ "Audit Log" ], "is_subclass_of": [ "Audit Trail" ], "wikilinks": [] }, { "id": "audit-logging", "title": "Audit Logging", "domain": "security", "domain_name": "Security", "definition": "Audit logging is the systematic, tamper-evident recording of security-relevant events, user actions, and system operations to an immutable or append-only store, enabling retrospective forensic analysis, regulatory compliance, and incident response. Each log entry captures who performed an action, what was performed, when, from where, and the outcome, providing an authoritative chain of evidence. Audit logs are distinct from general application logs by their integrity guarantees and structured, queryable format.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:audit-logging", "labels": [ "Audit Logging", "Audit Logger", "Logging", "Logging System" ], "is_subclass_of": [ "Audit Log" ], "wikilinks": [] }, { "id": "audit-mechanism", "title": "Audit Mechanism", "domain": "security", "domain_name": "Security", "definition": "The specific technical or procedural instrument used to capture, preserve, and retrieve evidence of system events, user actions, or organisational activities for auditing purposes. Audit mechanisms range from low-level kernel event hooks and cryptographic tamper-evident logs to high-level workflow checkpoints and policy-enforcement records.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:audit-mechanism", "labels": [ "Audit Mechanism" ], "is_subclass_of": [ "Audit" ], "wikilinks": [] }, { "id": "audit-readiness", "title": "Audit Readiness", "domain": "governance", "domain_name": "Governance", "definition": "Audit Readiness is the state in which an organization maintains the evidence, controls, documentation, and traceability needed to pass a compliance or security audit at any time with minimal preparation. It involves continuously capturing logs, change records, policy attestations, and control-effectiveness proof mapped to specific regulatory or standards requirements. Sustained readiness shifts auditing from a disruptive periodic scramble to an ongoing, verifiable posture.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:audit-readiness", "labels": [ "Audit Readiness" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "audit-trail", "title": "Audit Trail", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A chronological, tamper-evident record of system activities, transactions, and events that enables reconstruction and verification of sequences of operations for compliance, security, and forensic analysis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:audit-trail", "labels": [ "Audit Trail", "Audit Trail Generation", "Audit Trail System", "AuditTrail", "Change Audit Trail", "Cryptographic Audit Trail", "Decision Audit Trail", "GDPR Audit Trail", "Immutable Audit Trail", "Transparent Audit Trails" ], "is_subclass_of": [ "AI Governance and Ethics", "Governance Artifact", "Security Record", "Compliance Mechanism" ], "wikilinks": [ "Access Logs", "Clock Synchronization", "Cryptographic Integrity Protection", "ETSI GR ARF 010", "Event Logs", "Event Schema", "Forensic Analysis", "Governance System", "Incident Investigation", "Logging Infrastructure", "Logging Protocol", "Secure Storage", "Security Infrastructure", "System State Snapshots", "Time Synchronization Service", "Timestamp Records", "Transaction Records", "User Activity Logs", "Accountability", "Algorithmic Transparency Index" ] }, { "id": "audit", "title": "Audit", "domain": "governance", "domain_name": "Governance", "definition": "A systematic, independent examination of records, systems, processes, or controls carried out by a qualified party to assess their accuracy, completeness, and conformance with applicable standards, regulations, or policies. Audits generate structured, documented evidence that supports accountability, risk management, and informed decision-making by management, stakeholders, and regulators. The discipline spans financial, operational, information-security, algorithmic, and sustainability domains, each governed by domain-specific professional standards and methodologies. Audit outputs \u2014 opinions, findings, and recommendations \u2014 are foundational instruments of organisational governance and public trust.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:audit", "labels": [ "Audit", "Audit Aggregator", "Audit Management System", "Audit System", "Audit Systems", "Audit and Assurance", "Audit and Oversight", "Audit and Verification", "Manual Audit" ], "is_subclass_of": [ "Accountability Mechanism" ], "wikilinks": [] }, { "id": "auditability", "title": "Auditability", "domain": "governance", "domain_name": "Governance", "definition": "Auditability is the property of a system, process, or record that allows an independent party to examine and verify its actions, decisions, and state against expected behaviour or policy. It depends on complete, tamper-evident records \u2014 logs, audit trails, and provenance \u2014 that establish who did what, when, and why, enabling accountability and compliance review. Auditability is a foundational governance requirement for trustworthy financial, blockchain, and AI systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:auditability", "labels": [ "Auditability" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "augmented-connected-workforce", "title": "Augmented Connected Workforce", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Augmented Connected Workforce (ACW) is a digitally enhanced work paradigm in which human capabilities are amplified through the seamless integration of AI, augmented reality, virtual reality, and IoT technologies, enabling immersive spatial interactions and real-time connectivity across distributed teams. ACW systems overlay contextual digital information onto physical workspaces and connect workers to shared virtual environments, boosting safety, training efficacy, and operational coordination. It is recognised by Gartner as a strategic technology trend reshaping how organisations design, operate, and evolve their human workforce.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:augmented-connected-workforce", "labels": [ "Augmented Connected Workforce" ], "is_subclass_of": [ "Platform and Environment", "Metaverse and Telecollaboration" ], "wikilinks": [ "MetaverseDomain", "Telecollaboration" ] }, { "id": "augmented-reality-ar", "title": "Augmented Reality (AR)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interactive system that overlays digital content (visual, audio, haptic) onto the physical environment in real time, enabling spatially-registered blended experiences where virtual information enhances physical perception.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:augmented-reality-ar", "labels": [ "Augmented Reality (AR)" ], "is_subclass_of": [ "Platform and Environment", "Extended Reality (XR)" ], "wikilinks": [ "6DOF Tracking", "Contextual Information Overlay", "Device Camera", "Environmental Sensor", "Environmental Understanding", "IEEE P2048-3", "Interactive Holograms", "ISO 9241-940", "Object Recognition", "Sensor Input", "AR Display Device", "ComputeLayer", "Computer Vision", "Depth Sensing", "Digital Content Overlay", "Extended Reality (XR)", "InteractionDomain", "NetworkLayer", "Real-Time Rendering", "SLAM" ] }, { "id": "augmented-reality-collaboration", "title": "Augmented Reality Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"Collaborative work where geographically distributed team members share a common augmented reality environment, viewing and manipulating virtual 3D objects overlaid on their respective physical spaces whilst communicating via spatial audio and holographic avatars, enabling hybrid physical-virtual...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:augmented-reality-collaboration", "labels": [ "Augmented Reality Collaboration", "TELE-021-augmented-reality-collaboration" ], "is_subclass_of": [ "Telepresence", "Telecollaboration", "TELE-002-telecollaboration" ], "wikilinks": [ "CollaborativeDesign", "TELE-002-telecollaboration", "TELE-020-virtual-reality-telepresence", "TELE-025-microsoft-hololens", "TELE-026-microsoft-mesh", "TELE-100-ai-avatars", "TELE-150-webrtc", "TELE-203-haptic-feedback-telepresence", "AugmentedReality", "MixedReality", "SpatialComputing", "TELE-001-telepresence" ] }, { "id": "augmented-reality-tracking", "title": "Augmented Reality Tracking", "domain": "robotics", "domain_name": "Robotics", "definition": "Augmented reality tracking is the continuous estimation of a device's position and orientation relative to the physical environment so that virtual content can be registered and rendered as if anchored in the real world. It fuses camera imagery, inertial measurements and depth or feature data to maintain a stable six-degree-of-freedom pose at interactive rates. Robust tracking is the foundation of believable spatial overlay, addressing drift, occlusion and relocalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:augmented-reality-tracking", "labels": [ "Augmented Reality Tracking" ], "is_subclass_of": [ "Pose Estimation" ], "wikilinks": [] }, { "id": "augmented-reality", "title": "Augmented Reality", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Augmented Reality (AR) is a technology that overlays digital content onto the real world in real-time, enhancing users' perception of their physical environment through smartphones, head-mounted displays, or smart glasses. AR systems require coupling of real and virtual environments, real-time interaction, and precise 3D registration of virtual objects aligned with physical space.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:augmented-reality", "labels": [ "Augmented Reality", "Augmented Reality Experiences", "Augmented Reality Overlay", "AugmentedReality" ], "is_subclass_of": [ "Extended Reality", "Extended Reality (XR)" ], "wikilinks": [ "Contextual Information Display", "Extended Reality", "Remote Assistance", "Digital Twin", "Immersive Experiences", "Metaverse" ] }, { "id": "aurora", "title": "Aurora", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:aurora", "labels": [ "Aurora" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "austrian-economics", "title": "Austrian Economics", "domain": "finance", "domain_name": "Finance", "definition": "Austrian Economics is a heterodox school of economic thought originating in late 19th-century Vienna, most associated with Carl Menger, Ludwig von Mises, and Friedrich Hayek. It emphasises methodological individualism, the subjective theory of value, and the role of entrepreneurship in coordinating dispersed knowledge through price signals. The school critiques central planning and Keynesian intervention on the grounds that spontaneous market order cannot be replicated by any central authority, and that artificial credit expansion inevitably produces malinvestment and business cycles. Austrian insights have strongly influenced Bitcoin monetary theory, particularly the argument that hard-capped supply reproduces the properties of sound, commodity-backed money.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:austrian-economics", "labels": [ "Austrian Economics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "austrian-hard-money-theory", "title": "Austrian Hard Money Theory", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Austrian hard money theory is the monetary doctrine, rooted in the Austrian School of economics, holding that sound money must have a credibly fixed or hard-to-inflate supply so it can reliably preserve purchasing power over time. It favours commodity-like monies (historically gold, latterly Bitcoin) over fiat currencies whose supply central banks can expand at will, arguing that inflation distorts price signals and the capital structure. The theory underpins Bitcoin's framing as a deflationary store of value with an absolutely scarce, algorithmically capped issuance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:austrian-hard-money-theory", "labels": [ "Austrian Hard Money Theory" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "authenticated-encryption", "title": "Authenticated Encryption", "domain": "security", "domain_name": "Security", "definition": "Authenticated encryption is a class of symmetric cryptographic schemes that simultaneously provide confidentiality, integrity and authenticity of a message in a single operation. Authenticated encryption with associated data (AEAD) additionally binds unencrypted header data to the ciphertext, so any tampering with the message or its context is detected on decryption. Modern AEAD constructions such as AES-GCM and ChaCha20-Poly1305 are the recommended default for secure communication because they avoid the pitfalls of composing encryption and authentication separately.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:authenticated-encryption", "labels": [ "Authenticated Encryption" ], "is_subclass_of": [ "Encryption" ], "wikilinks": [] }, { "id": "authentication-mechanism", "title": "Authentication Mechanism", "domain": "security", "domain_name": "Security", "definition": "An authentication mechanism is a technical procedure or protocol that verifies the claimed identity of a user, device, or system before granting access to protected resources. Such mechanisms range from simple password checks to sophisticated cryptographic challenges and biometric verification. They form the foundational layer of access control systems, ensuring that only authorised principals can interact with sensitive data or services. The strength and appropriateness of a chosen mechanism directly influences the overall security posture of a system.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:authentication-mechanism", "labels": [ "Authentication Mechanism" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "authentication-protocol", "title": "Authentication Protocol", "domain": "security", "domain_name": "Security", "definition": "A defined sequence of exchanges through which one party proves its identity to another over a communication channel.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:authentication-protocol", "labels": [ "Authentication Protocol" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [ "Cryptography", "Authentication", "Access Control", "Multi-Factor Authentication", "OAuth", "Cryptographic Protocol" ] }, { "id": "authentication-service", "title": "Authentication Service", "domain": "security", "domain_name": "Security", "definition": "An authentication service is a system component that verifies the identity of users, devices, or applications attempting to access protected resources. It validates credentials against stored identity information and issues tokens or assertions that enable authorised access across applications and services, forming the foundation of secure identity management in enterprise systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:authentication-service", "labels": [ "Authentication Service" ], "is_subclass_of": [ "Security and Identity", "Security Service" ], "wikilinks": [ "Authorisation", "Core Technology", "Credential Validation", "Security Service", "Single Sign-On", "Cryptography", "Identity Verification", "Session Management" ] }, { "id": "authentication-standards", "title": "Authentication Standards", "domain": "security", "domain_name": "Security", "definition": "Authentication Standards encompass protocols and specifications for verifying user identity in digital systems, particularly FIDO2 and WebAuthn standards that enable passwordless, phishing-resistant authentication using public key cryptography and hardware authenticators for secure access to metaverse platforms and blockchain applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:authentication-standards", "labels": [ "Authentication Standards" ], "is_subclass_of": [ "Security Standards" ], "wikilinks": [ "Client Implementation", "Hardware Authenticators", "Passwordless Authentication", "Phishing Resistance", "Secure Platform Access", "Security Standards", "W3C", "DID Nostr Identity", "metaverse", "Public Key Cryptography" ] }, { "id": "authentication-system", "title": "Authentication System", "domain": "security", "domain_name": "Security", "definition": "An authentication system is an infrastructure component responsible for verifying the claimed identity of users, devices, or services before granting access to protected resources. It integrates credential management, identity verification workflows, session management, and integration with downstream access control mechanisms to enforce the principle that only legitimate principals can initiate authenticated sessions. Authentication systems range from simple password databases to federated multi-factor frameworks spanning organisational boundaries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:authentication-system", "labels": [ "Authentication System" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "authentication", "title": "authentication", "domain": "security", "domain_name": "Security", "definition": "Authentication is the security process of verifying that a claimed identity of a user, device, application, or service is genuine before granting access to protected resources. It is fundamentally distinct from authorisation, which determines what permissions an authenticated principal possesses. Authentication evidence is categorised into knowledge factors (passwords, PINs, security questions), possession factors (hardware security keys, TOTP devices, mobile authenticators), and inherence factors (biometrics such as fingerprint or facial recognition), with cryptographic proofs (digital signatures, TLS client certificates, FIDO2 passkeys) increasingly replacing shared-secret schemes. Modern secure systems combine multiple independent factors through multi-factor authentication to achieve strong resistance against credential theft, phishing, replay attacks, and session hijacking.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:authentication", "labels": [ "Authentication", "Cryptographic Authentication", "Device Authentication", "Evidence Authentication", "OTP Authentication", "Open Authentication", "Packet Authentication" ], "is_subclass_of": [ "Identity and Access Management" ], "wikilinks": [] }, { "id": "authenticity-certification", "title": "Authenticity Certification", "domain": "security", "domain_name": "Security", "definition": "A formal process or credential that attests to the genuine origin, integrity, and provenance of a digital or physical artefact, typically employing cryptographic signing, trusted third-party attestation, or standards-based metadata embedding. Authenticity certification enables recipients to verify that content has not been altered and originated from a claimed source.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:authenticity-certification", "labels": [ "Authenticity Certification" ], "is_subclass_of": [ "Content Authenticity" ], "wikilinks": [] }, { "id": "authoring-tool", "title": "Authoring Tool", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software application for creating or editing immersive content, including 3D models, environments, interactions, and multimedia assets for metaverse experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:authoring-tool", "labels": [ "Authoring Tool", "3D Authoring Tools", "Course Authoring Tool", "Multimedia Authoring" ], "is_subclass_of": [ "Content and Assets", "Digital Content Creation" ], "wikilinks": [ "Asset Pipeline", "Content Creation", "Editor Interface", "ETSI GR ARF 010", "Interactive Experience Development", "MSF Taxonomy", "Preview System", "Scene Design", "SIGGRAPH Pipeline WG", "3D Modeling", "Compute Infrastructure", "ComputeLayer", "Computer Vision", "CreativeMediaDomain", "Generative Design Tool", "Graphics API", "InteractionDomain" ] }, { "id": "authorisation", "title": "Authorisation", "domain": "security", "domain_name": "Security", "definition": "Authorisation is the process of determining and enforcing whether an authenticated principal\u2014user, service, or device\u2014has the right to perform a requested action on a protected resource. It operates downstream of authentication, translating verified identity claims into permitted operations according to configured access policies. Modern authorisation frameworks encompass role-based, attribute-based, relationship-based, and policy-as-code access control models, each balancing expressiveness with enforcement performance. Robust authorisation design underpins regulatory compliance, auditability, and the principle of least privilege across distributed digital systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:authorisation", "labels": [ "Authorisation", "Authorisation Code Flow", "Authorisation Server", "Authorisation Service", "Cross-Domain Authorisation" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "authorised-representative", "title": "Authorised Representative", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A natural or legal person located or established in the EU who has received and accepted a written mandate from a non-EU provider of an AI system or general-purpose AI model to perform regulatory tasks on its behalf, including maintaining technical documentation, cooperating with market surveillance authorities, and serving as the primary EU contact point under the EU AI Act.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:authorised-representative", "labels": [ "Authorised Representative" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "GDPR (General Data Protection Regulation)", "EU AI Act", "Governance", "MetaverseDomain" ] }, { "id": "authorization", "title": "Authorization", "domain": "security", "domain_name": "Security", "definition": "The security process that determines what actions an authenticated principal is permitted to perform on specific resources within a system. Authorization evaluates the subject's identity, role, attributes, and contextual factors against a policy to produce an access decision, operating as a distinct layer from authentication and separate from audit.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:authorization", "labels": [ "Authorization", "Authorisation", "Delegated Authorization", "Fine-Grained Authorization" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "auto-scaling", "title": "Auto-Scaling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Auto-scaling is the automated adjustment of computing capacity in response to observed demand, adding or removing resources to maintain performance and control cost. It uses metrics, policies and controllers to scale horizontally by changing instance counts or vertically by resizing instances. Auto-scaling is foundational to elastic cloud infrastructure, balancing responsiveness against efficiency without manual intervention.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:auto-scaling", "labels": [ "Auto-Scaling", "Auto Scaling" ], "is_subclass_of": [ "Scalability" ], "wikilinks": [] }, { "id": "auto-gen", "title": "AutoGen", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AutoGen is an open-source multi-agent conversation framework originating from Microsoft Research that defines conversable agents \u2014 autonomous entities backed by large language models, tools, or human inputs \u2014 which coordinate through structured message exchange to decompose and complete complex tasks. The framework introduced the conversable-agent abstraction as a unified primitive supporting LLM-powered reasoning, code execution, and adaptive human participation within a single coherent programming model. Forked in 2025 as the community-governed AG2 project and simultaneously merged by Microsoft into the broader Microsoft Agent Framework combining AutoGen agent abstractions with Semantic Kernel enterprise features.", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:auto-gen", "labels": [ "AutoGen" ], "is_subclass_of": [ "Multi-Agent Systems", "Agent Frameworks" ], "wikilinks": [ "Language Model", "Tool Use", "Multi-Agent Coordination", "Agentic Workflow", "AI Agent", "Function Calling", "Multi-Agent Systems", "Large Language Models", "Agent Frameworks", "Orchestration", "Autonomous Agent", "Code Execution", "Human-in-the-Loop", "Task Planning", "CrewAI", "LangGraph", "OpenAI Agents SDK", "Model Context Protocol", "Retrieval-Augmented Generation", "Chain of Thought" ] }, { "id": "auto-ml", "title": "AutoML", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Automated Machine Learning (AutoML) is the discipline and associated tooling that automates the end-to-end pipeline of applying machine learning to real-world problems \u2014 encompassing automated data pre-processing, feature engineering, algorithm selection, Neural Architecture Search (NAS), hyp...", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:auto-ml", "labels": [ "AutoML", "Auto-ML" ], "is_subclass_of": [ "AI Agent System", "Machine Learning Pipeline", "Automated Optimisation System" ], "wikilinks": [ "AI Development Lifecycle", "AI Domain", "Algorithm Selection", "AutoML Survey Knowledge-Based Systems 2021", "Automated Optimisation System", "Bayesian Optimisation", "Computational Budget", "DARTS Differentiable NAS", "DARTS ICLR 2019", "Democratised ML Development", "Evaluation Metric", "Evolutionary Neural Architecture Search", "Feature Engineering Automation", "Google Vertex AI AutoML", "H2O AutoML Documentation", "Hyperparameter Optimisation", "Meta-Learning", "MLOps Platform", "ModelLayer", "NeurIPS 2015 Auto-Sklearn Paper" ] }, { "id": "autoencoder", "title": "Autoencoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An Autoencoder is a neural network trained to reconstruct its input by learning a compressed latent representation. The encoder maps input data to a lower-dimensional latent space and the decoder reconstructs the original from this representation, minimising a reconstruction loss. Variants including variational autoencoders (VAEs), denoising autoencoders, and convolutional autoencoders extend this framework to generative modelling, anomaly detection, and feature extraction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:autoencoder", "labels": [ "Autoencoder", "Autoencoders" ], "is_subclass_of": [ "Deep Learning" ], "wikilinks": [ "ISO/IEC 22989:2022", "NIST AI RMF", "Computer Vision", "MetaverseDomain" ] }, { "id": "automata-theory", "title": "Automata Theory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Automata theory is the branch of theoretical computer science that studies abstract computing machines (automata) and the classes of formal languages they can recognise. It classifies machines such as finite-state automata, pushdown automata, and Turing machines by their computational power, establishing a hierarchy that defines what problems are decidable. The theory provides the formal foundations for compiler design, regular expressions, protocol verification, and model checking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:automata-theory", "labels": [ "Automata Theory" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "automated-code-review", "title": "Automated Code Review", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Automated code review is the use of software tools, static analysis and, increasingly, large language models to inspect source code for defects, style violations, security weaknesses and maintainability issues without requiring manual reading of every change. It augments or partially replaces human review by surfacing actionable findings directly in pull requests and continuous integration pipelines. Modern systems combine rule-based linters with learned models that reason about code intent and propose fixes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:automated-code-review", "labels": [ "Automated Code Review" ], "is_subclass_of": [ "Code Review" ], "wikilinks": [] }, { "id": "automated-compliance", "title": "Automated Compliance", "domain": "governance", "domain_name": "Governance", "definition": "Automated Compliance refers to the use of software systems and rule engines to continuously monitor, evaluate, and enforce regulatory requirements without manual intervention. It encompasses real-time policy checking, evidence collection, and exception reporting across business processes and IT systems. By embedding compliance logic directly into workflows, organisations can dramatically reduce audit burden and human error. The discipline draws on regulatory technology, policy engines, and audit trails to provide defensible evidence of conformance. It is particularly vital in heavily regulated sectors such as financial services, healthcare, and data protection.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:automated-compliance", "labels": [ "Automated Compliance" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "automated-decision-making", "title": "Automated Decision Making", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Automated decision making is the use of algorithms or machine learning models to reach decisions about individuals or situations without meaningful human involvement. It ranges from rule-based scoring to predictions from trained models, applied to outcomes such as eligibility, pricing or risk classification. Because such decisions can significantly affect people, data protection frameworks like the GDPR grant rights around solely automated decisions, including a right to obtain human intervention and an explanation. Responsible deployment depends on fairness, transparency and accountability safeguards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:automated-decision-making", "labels": [ "Automated Decision Making", "Automated Decision-Making" ], "is_subclass_of": [ "Machine Learning", "Decision Making" ], "wikilinks": [] }, { "id": "automated-design", "title": "Automated Design", "domain": "ai", "domain_name": "Ai", "definition": "Automated Design is the application of computational algorithms, machine learning, and optimisation techniques \u2014 spanning combinatorial search, gradient-based topology optimisation, evolutionary algorithms, reinforcement learning, and deep generative modelling \u2014 to generate, evaluate, and iteratively refine design artefacts with minimal or no step-by-step human direction. The field encompasses electronic design automation (EDA) for integrated circuits and printed circuit boards, neural architecture search (NAS) for machine-learning model topologies, topology optimisation for structural engineering, generative architectural layout synthesis, and LLM-assisted hardware description language generation. The unifying abstraction is the design space traversal guided by evaluation functions encoding physical, functional, economic, or aesthetic criteria: the algorithm explores candidate designs, evaluates each against the specified criteria, and updates its search strategy to navigate toward regions of the space satisfying design objectives. Commercial impact has been transformative across semiconductor, aerospace, automotive, and construction sectors, where design cycles have been compressed from months to days whilst exploring solution spaces too large for human enumeration.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:automated-design", "labels": [ "Automated Design", "Automated Experiment Design" ], "is_subclass_of": [ "Generative Design", "Optimisation" ], "wikilinks": [ "Machine Learning Discipline", "Algorithm", "Simulation", "Generative Design", "Generative Design Tool", "Hyperparameter Optimisation", "Formal Verification", "CAD Software", "Constraint Based Design", "Neural Architecture Search", "Topology Optimisation", "Electronic Design Automation", "Reinforcement Learning", "Evolutionary Algorithm", "Deep Generative Model", "Diffusion Model", "Large Language Models", "Digital Twin", "Smart Manufacturing", "Optimisation" ] }, { "id": "automated-dispute-resolution", "title": "Automated Dispute Resolution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A class of system that applies algorithmic, rule-based, or machine-learning-driven processes to adjudicate disputes between parties without requiring direct human arbitrator involvement for each case. Automated dispute resolution mechanisms ingest evidence and contractual terms, apply predefined or learned decision rules, and produce binding or advisory outcomes, typically operating within a blockchain or smart-contract execution environment.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:automated-dispute-resolution", "labels": [ "Automated Dispute Resolution" ], "is_subclass_of": [ "Dispute Resolution" ], "wikilinks": [] }, { "id": "automated-market-maker", "title": "Automated Market Maker", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Automated Market Maker (AMM) is a decentralized exchange protocol that uses algorithmic pricing mechanisms and liquidity pools instead of traditional order books, enabling permissionless token swaps where prices adjust automatically based on supply and demand within smart contract-managed reserves.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:automated-market-maker", "labels": [ "Automated Market Maker" ], "is_subclass_of": [ "DeFi Protocol" ], "wikilinks": [ "Liquidity Pools", "Permissionless Trading", "Pricing Algorithms", "Token Swaps", "Blockchain", "DeFi Protocol", "Liquidity Provision", "metaverse", "Smart Contracts" ] }, { "id": "automated-market-making", "title": "Automated Market Making", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Automated Market Making is the practice and modology of operating decentralized exchange protocols that use algorithmic pricing and liquidity pools to facilitate permissionless trading, encompassing pool design, fee structures, capital efficiency optimization, and impermanent loss mitigation stra...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:automated-market-making", "labels": [ "Automated Market Making" ], "is_subclass_of": [ "Platform and Environment", "DeFi Operations" ], "wikilinks": [ "DeFi Operations", "Decentralized Trading", "Liquidity Mining", "Price Oracle Integration", "Smart Contract Deployment", "Token Accessibility", "Blockchain", "Liquidity Provision", "metaverse" ] }, { "id": "automated-modeling", "title": "Automated Modeling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The application of AI and machine learning techniques to automatically generate, refine, or parameterise three-dimensional models and spatial environments, reducing manual content-creation overhead in metaverse, digital-twin, and spatial-computing platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:automated-modeling", "labels": [ "Automated Modeling" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "automated-planning", "title": "Automated Planning", "domain": "ai", "domain_name": "Ai", "definition": "Automated Planning is a field of artificial intelligence concerned with the computational synthesis of action sequences (plans) that transform an initial world state into a desired goal state, using formal representations of states, actions, and constraints alongside algorithmic search and reasoning techniques.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:automated-planning", "labels": [ "Automated Planning", "Model-Based Planning" ], "is_subclass_of": [ "Planning and Scheduling" ], "wikilinks": [ "Autonomous Systems", "Autonomous Robot", "Planning and Scheduling" ] }, { "id": "automated-podcasting", "title": "Automated Podcasting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Automated Podcasting is the application of artificial intelligence, machine learning, and generative systems to partially or fully automate the end-to-end podcast production pipeline\u2014spanning script generation, synthetic voice synthesis, AI-driven audio editing, automated transcription, show-note...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:automated-podcasting", "labels": [ "Automated Podcasting" ], "is_subclass_of": [ "AI Application", "Generative AI", "AI Audio", "Speech Synthesis", "Content Automation", "AI companions", "AI Video" ], "wikilinks": [ "Accessible Media", "Adobe Podcast", "AI Audio", "AI Audio Editing", "AI Script Generation", "AIGroundedDomain", "AssemblyAI", "Audio Signal Processing", "AudioTechnologyDomain", "Automated Transcription", "Automatic Speech Recognition", "C2PA Content Credentials", "Cloud Computing Infrastructure", "Content Automation", "ContentDistributionLayer", "CreativeAIApplicationsDomain", "Descript", "Dynamic Ad Insertion", "ElevenLabs", "EU AI Act Article 50" ] }, { "id": "automated-reasoning", "title": "Automated Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "Automated Reasoning is a subfield of artificial intelligence and computer science concerned with developing algorithms and systems that can perform logical inference, theorem proving, satisfiability checking, and knowledge-based deduction without continuous human guidance. It encompasses formal methods such as first-order logic, propositional calculus, and higher-order logics, implemented in systems ranging from SAT solvers and automated theorem provers to description logic reasoners underpinning knowledge graphs. Automated reasoning forms the rigorous backbone of formal verification, ontology inference, planning, and constraint satisfaction problems.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:automated-reasoning", "labels": [ "Automated Reasoning" ], "is_subclass_of": [ "Reasoning", "Symbolic Reasoning", "Symbolic AI" ], "wikilinks": [ "Symbolic Reasoning", "Formal Verification", "Knowledge Representation", "Logic Programming", "Expert Systems", "Knowledge Graph", "Automated Theorem Proving", "Constraint Satisfaction", "Deep Learning", "Symbolic AI", "Automated Planning", "Neuro Symbolic Ai", "Probabilistic Reasoning", "Description Logic", "SAT Solving", "Model Checking", "First-Order Logic", "Program Synthesis", "Inference Engine", "Formal Methods" ] }, { "id": "automated-summarization", "title": "Automated Summarization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Automated Summarization is an NLP task in which a system produces a condensed representation of one or more source documents that preserves salient information and is significantly shorter than the original. Extractive approaches select and concatenate verbatim sentences or passages from the source; abstractive approaches generate novel text that may paraphrase, fuse, or infer content not explicitly stated in the source. Modern large language model-based summarizers are predominantly abstractive, achieving strong performance on long-form documents, meeting transcripts, scientific papers, and news articles, but remain susceptible to hallucination, factual inconsistency, and salience mismatch relative to the reader's intent.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:automated-summarization", "labels": [ "Automated Summarization" ], "is_subclass_of": [ "AI Application", "Natural Language Processing" ], "wikilinks": [ "Artificial Intelligence", "Natural Language Processing", "Telecollaboration" ] }, { "id": "automated-testing", "title": "Automated Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Automated testing is the practice of using software tools to execute predefined test cases against a system and compare actual outcomes to expected results without manual intervention. It spans unit, integration, end-to-end, and regression tests, and is typically wired into build pipelines so defects are caught early and consistently. By making verification repeatable and fast, it enables continuous integration and reliable software delivery at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:automated-testing", "labels": [ "Automated Testing", "Automated Testing Engine" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "automated-theorem-proving", "title": "Automated Theorem Proving", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Automated theorem proving is the subfield of automated reasoning concerned with constructing formal proofs of mathematical or logical statements by machine. Given a set of axioms and a conjecture expressed in a formal logic, an automated theorem prover searches for a derivation that establishes the conjecture as a consequence of the axioms. Techniques span resolution, tableaux, term rewriting and decision procedures, and underpin formal verification, mathematics and AI reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:automated-theorem-proving", "labels": [ "Automated Theorem Proving" ], "is_subclass_of": [ "Automated Reasoning" ], "wikilinks": [] }, { "id": "automatic-differentiation", "title": "Automatic Differentiation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Automatic differentiation (AD) is a family of computational techniques for evaluating the derivative of a function specified by a computer program, by systematically applying the chain rule to elementary arithmetic operations rather than through symbolic algebra or finite-difference approximation. It operates in two principal modes: forward mode, which propagates tangent values alongside primal values, and reverse mode (backpropagation), which accumulates gradients in a backward pass over a recorded computation graph. AD produces machine-precision derivatives at a cost linear in the number of program operations, making it the computational backbone of modern deep learning frameworks.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:automatic-differentiation", "labels": [ "Automatic Differentiation" ], "is_subclass_of": [ "AI Technique", "Numerical Method", "Optimisation" ], "wikilinks": [ "Backpropagation", "Deep Learning", "Gradient Descent", "Stochastic Gradient Descent", "Neural Network", "Loss Function", "Optimisation", "JAX", "Differentiable Rendering", "Differentiable Architecture", "Just-In-Time Compilation", "Numerical Method", "Probabilistic Programming", "Physics Simulation", "Scientific Machine Learning", "Machine Learning Framework", "Forward Mode Differentiation", "Computation Graph", "Adam Optimiser", "Batch Normalisation" ] }, { "id": "automatic-prompt-optimisation", "title": "Automatic Prompt Optimisation", "domain": "ai", "domain_name": "Ai", "definition": "Automatic Prompt Optimisation (APO) is the family of algorithmic techniques that search for, refine, or generate the most effective natural-language or soft-token instructions for a large language model, replacing manual prompt crafting with automated search processes guided by an objective function. Methods span gradient-free discrete search (APE, OPRO, ProTeGi), evolutionary self-improvement loops (PromptBreeder, GEPA), LLM-as-optimiser meta-prompting, textual back-propagation (TextGrad), and compiler-based programmatic optimisation (DSPy), each scoring candidate prompts against a held-out evaluation metric and iteratively retaining the highest-performing variants. APO addresses the brittleness and labour cost of hand-tuned prompting and has become a core component of production LLM pipelines, enabling measurable, reproducible performance gains without requiring access to model weights.", "entityType": "Class", "qualityScore": 0.87, "maturity": "emerging", "iri": "urn:ngm:class:automatic-prompt-optimisation", "labels": [ "Automatic Prompt Optimisation" ], "is_subclass_of": [ "AI Technique", "Prompt Engineering", "Model Optimisation and Performance", "Machine Learning", "Natural Language Processing" ], "wikilinks": [ "AlgorithmLayer", "ApplicationLayer", "AI-GroundedDomain", "NLPDomain", "Prompt Engineering", "Large Language Models", "In-Context Learning", "Few-Shot Prompting", "Chain of Thought", "DSPy", "Reinforcement Learning", "Evaluation Benchmarks and Leaderboards", "Retrieval Augmented Generation", "Agents", "Natural Language Processing", "Machine Learning", "Transformer Architecture", "Training and Fine Tuning", "Instruction Following", "Model Optimisation and Performance" ] }, { "id": "automatic-speech-recognition", "title": "automatic speech recognition", "domain": "ai", "domain_name": "Ai", "definition": "Automatic Speech Recognition (ASR) is the technology that converts spoken language into written text by jointly modelling acoustic signals \u2014 frame-level spectral features derived from raw audio waveforms \u2014 and linguistic context, using neural sequence-to-sequence architectures trained on large corpora of paired audio and transcripts. Modern ASR systems built on transformer and conformer encoder-decoder designs achieve near-human word error rates on clean speech benchmarks (Whisper Large-v3 at 2.7% WER on LibriSpeech test-clean) and have been extended to multilingual and low-resource settings through large-scale self-supervised pre-training on unlabelled audio. ASR serves as a foundational component for voice assistants, real-time transcription services, accessibility tooling, and spoken language understanding pipelines, and underpins multimodal AI systems that must bridge the speech and text modalities.", "entityType": "Class", "qualityScore": 0.89, "maturity": "mature", "iri": "urn:ngm:class:automatic-speech-recognition", "labels": [ "Automatic Speech Recognition", "Speech Recogniser" ], "is_subclass_of": [ "Speech Processing", "Deep Learning", "Natural Language Processing", "Machine Learning", "Audio Signal Processing" ], "wikilinks": [ "Acoustic Model", "Language Model", "Feature Extraction", "Audio Signal Processing", "Training Data", "Connectionist Temporal Classification", "Speaker Diarisation", "Speaker Recognition", "Word Error Rate", "Keyword Spotting", "Spoken Language Understanding", "Real-Time Captioning", "Text-to-Speech", "Spatial Audio", "Distributed Collaboration", "Language Modeling", "AlgorithmLayer", "ApplicationLayer", "Whisper OpenAI", "wav2vec 2.0" ] }, { "id": "automatic1111", "title": "Automatic1111", "domain": "ai", "domain_name": "Ai", "definition": "An open-source, browser-based graphical user interface for Stable Diffusion and compatible diffusion models, providing extensive control over image-generation parameters, model loading, and an extensible plugin architecture. Automatic1111 (AUTOMATIC1111/stable-diffusion-webui on GitHub) became the dominant community-facing inference frontend for locally hosted image-generation models from 2022 onwards.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:automatic1111", "labels": [ "Automatic1111" ], "is_subclass_of": [ "Image Generation", "Generative AI", "Open Source Software" ], "wikilinks": [ "Diffusion Model", "Latent Diffusion", "Stable Diffusion", "Image Generation", "ComfyUI", "ControlNet", "Inpainting", "Outpainting", "GPU Acceleration", "Civitai", "Prompt Engineering", "Open Source Software", "LoRA", "Variational Autoencoder", "Classifier-Free Guidance", "SDXL", "Textual Inversion", "Generative AI", "Hypernetwork", "Image Editing" ] }, { "id": "automation-bias", "title": "Automation Bias", "domain": "governance", "domain_name": "Governance", "definition": "Automation bias is the human tendency to over-rely on automated systems, accepting their outputs uncritically and discounting contradictory information, including one's own judgement. It manifests as errors of commission, following an automated recommendation that is wrong, and errors of omission, failing to act because the system did not prompt. Driven by the perceived authority of computers and the cognitive ease of deferring, it grows with workload, trust, and system opacity. Automation bias is a central human-factors concern in the governance of decision-support systems and human-in-the-loop oversight, where it can erode the very safeguards that human supervision is meant to provide.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:automation-bias", "labels": [ "Automation Bias" ], "is_subclass_of": [ "Bias" ], "wikilinks": [] }, { "id": "automation", "title": "Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Automation is the application of technology, control systems, and software to execute processes, tasks, or workflows with minimal or no human intervention, transferring decision-making and execution from people to machines or algorithms. It encompasses industrial and manufacturing automation driven by programmable logic controllers and robotics, as well as software-driven process automation that replicates repetitive digital workflows. The primary goals are to increase throughput, reduce error rates, lower operational costs, and free human attention for non-routine judgement tasks. Automation exists on a spectrum from fixed mechanisation through programmable control, flexible automation, and fully autonomous adaptive systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:automation", "labels": [ "Automation", "Automation Engine", "Form Filling Automation", "Home Automation", "Rule-Based Automation" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Robotic Process Automation", "Robotics", "owl:Thing" ] }, { "id": "automerge", "title": "Automerge", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Automerge is a library and data format implementing a JSON-like CRDT that enables automatic merging of concurrent changes to shared documents without requiring a central server. It models document history as an append-only log of operations, allowing peers to exchange and apply changes in any order while converging to the same state. Automerge supports rich text, rich data structures, and is designed for local-first software where data lives on the user's device and syncs opportunistically.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:automerge", "labels": [ "Automerge" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "autonomous-ai-agents", "title": "Autonomous AI Agents", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autonomous AI Agents are intelligent software systems capable of independently perceiving their environment, reasoning over goals, planning multi-step action sequences, and executing tasks without continuous human intervention, typically powered by large language models or reinforcement learning. They differ from simpler AI tools by their ability to set sub-goals, call external tools, manage memory, and adapt dynamically to novel situations. Autonomous agents are increasingly deployed in enterprise workflows, research automation, customer service, and multi-agent collaborative systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-ai-agents", "labels": [ "Autonomous AI Agents", "Fully Autonomous AI" ], "is_subclass_of": [ "AI Application", "AI Agent System" ], "wikilinks": [ "Autonomous Robot", "MetaverseDomain" ] }, { "id": "autonomous-agent-framework-catalogue", "title": "Autonomous Agent Framework Catalogue", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A curated catalogue of open-source and commercial autonomous agent frameworks and projects, spanning general-purpose agents (AutoGPT, BabyAGI, CAMEL), coding agents (GPT Engineer, SWE-Agent), productivity agents, research agents, and DIY framework scaffolds (CrewAI, MetaGPT, AutoGen). The list provides practitioners with an entry point into the rapidly expanding agentic AI ecosystem.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-agent-framework-catalogue", "labels": [ "Autonomous Agent Framework Catalogue", "List Of Agent Projects" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "autonomous-agent", "title": "Autonomous Agent", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software entity capable of acting autonomously to achieve goals within a metaverse, exhibiting goal-directed behavior, decision-making, and adaptive responses without continuous human intervention.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:autonomous-agent", "labels": [ "Autonomous Agent", "Autonomous Agent Economy", "Autonomous Agents", "AutonomousAgent" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Action Executor", "AI System", "Autonomous Behavior", "Autonomous System", "Computational Resources", "Data Source", "Decision Engine", "Decision Support", "ETSI GR ARF 010", "Goal Specification", "Goal System", "Intelligent Environment", "Knowledge Base", "MSF Use Cases", "Perception Module", "Process Automation", "AI Framework", "Autonomous Robot", "ComputationAndIntelligenceDomain", "ComputeLayer" ] }, { "id": "autonomous-behavior", "title": "Autonomous Behavior", "domain": "ai", "domain_name": "Ai", "definition": "The capacity of an agent, robot, or AI system to select and execute actions in pursuit of goals without continuous human instruction, using internal models of the world, perception of environmental state, and learned or programmed decision policies. Autonomous behaviour spans a continuum from simple reactive reflexes to deliberative planning over extended time horizons.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-behavior", "labels": [ "Autonomous Behavior" ], "is_subclass_of": [ "Autonomous Operation", "Agentic AI", "Goal-Directed Behavior" ], "wikilinks": [ "Perception System", "Planning Module", "Autonomous Agent", "Autonomous Task Execution", "Agentic AI", "Behavioral Modeling", "Goal", "Feedback Loop", "Reinforcement Learning", "Multi-Agent Systems", "Cognitive Architecture", "Large Language Models", "AI Safety", "AI Alignment", "Human-in-the-Loop", "Reward Function", "Swarm Intelligence", "Agent-Based Modelling", "Task Planning", "Decision Making" ] }, { "id": "autonomous-business-operations", "title": "Autonomous Business Operations", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of artificial intelligence systems to independently manage, execute, and optimize core business functions and workflows with minimal human intervention.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:autonomous-business-operations", "labels": [ "Autonomous Business Operations" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "autonomous-coding", "title": "Autonomous Coding", "domain": "ai", "domain_name": "Ai", "definition": "Autonomous coding is the capability of an AI agent to write, modify, test and debug software with minimal human intervention, planning and executing multi-step coding tasks end to end. It builds on code generation but adds agentic control loops for tool use, self-correction and iterative verification against a goal. It is a core capability enabled by agent frameworks, orchestrators and execution sandboxes.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:autonomous-coding", "labels": [ "Autonomous Coding" ], "is_subclass_of": [ "Code Generation" ], "wikilinks": [] }, { "id": "autonomous-decision-making", "title": "Autonomous Decision Making", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autonomous decision making is the capacity of an artificial agent to select and commit to actions in pursuit of goals without requiring human authorisation for each choice. It combines perception of the environment, internal reasoning or learned policies, evaluation of expected outcomes, and action selection under uncertainty. Drawing on decision theory, planning, and reinforcement learning, it lets agents operate in dynamic settings where human supervision is impractical or too slow. Autonomous decision making is foundational to autonomous vehicles, robotic systems, and multi-agent coordination, and raises governance questions about accountability and human oversight.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-decision-making", "labels": [ "Autonomous Decision Making", "Autonomous Decision-Making" ], "is_subclass_of": [ "Decision Making" ], "wikilinks": [] }, { "id": "autonomous-delivery", "title": "Autonomous Delivery", "domain": "robotics", "domain_name": "Robotics", "definition": "Autonomous delivery is the use of self-navigating ground or aerial robots to transport goods from an origin to a destination without a human driver or operator. Delivery robots combine mapping, obstacle avoidance and route planning capabilities from autonomous mobile robotics with payload handling and secure drop-off mechanisms. It is deployed for last-mile parcel delivery, food delivery and campus or warehouse logistics, where it reduces labour cost and enables continuous operation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-delivery", "labels": [ "Autonomous Delivery" ], "is_subclass_of": [ "Autonomous Mobile Robots" ], "wikilinks": [] }, { "id": "autonomous-driving-perception", "title": "Autonomous Driving Perception", "domain": "ai", "domain_name": "Ai", "definition": "The subsystem of an autonomous vehicle responsible for interpreting sensor data to construct a structured understanding of the vehicle's immediate environment, including the detection, classification, and tracking of objects, lane geometry, road surfaces, traffic signage, and dynamic actors. Autonomous driving perception fuses inputs from cameras, LiDAR, radar, and ultrasonic sensors to produce a real-time scene representation sufficient for safe navigation decisions.", "entityType": "Class", "qualityScore": 0.85, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-driving-perception", "labels": [ "Autonomous Driving Perception", "Autonomous Vehicle Perception" ], "is_subclass_of": [ "Perception System", "Computer Vision" ], "wikilinks": [ "Sensor Fusion", "Computer Vision", "Lidar", "Deep Learning", "Autonomous Driving", "Object Detection", "Semantic Segmentation", "Point Cloud", "Depth Estimation", "Convolutional Neural Network", "Transformer Architecture", "Simultaneous Localisation and Mapping", "Object Tracking", "Bird's Eye View", "Reinforcement Learning", "Edge Computing", "Functional Safety", "Motion Planning", "HD Maps" ] }, { "id": "autonomous-driving", "title": "autonomous driving", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autonomous driving is the technical capability enabling a vehicle to perceive its environment, predict the behaviour of surrounding agents, plan a safe trajectory, and execute actuator commands without direct human intervention. The system architecture decomposes into perception (camera, Lidar, radar fusion), localisation (HD map matching, SLAM), prediction (probabilistic motion modelling), planning (route, behaviour, and motion planning layers), and control (longitudinal and lateral actuation). Safety assurance draws on formal verification, simulation, and real-world validation mileage, with regulatory oversight governed by frameworks such as SAE J3016, ISO 26262, and UN ECE WP.29. The field bridges robotics, machine learning, embedded systems, and transport infrastructure.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-driving", "labels": [ "Autonomous Driving" ], "is_subclass_of": [ "AI Application", "Robotics", "Safety-Critical Systems" ], "wikilinks": [ "Computer Vision", "Sensor Fusion", "Lidar", "Simultaneous Localisation and Mapping", "Object Detection", "Path Planning", "Deep Learning", "Reinforcement Learning", "Model Predictive Control", "Convolutional Neural Network", "Transformer Architecture", "Autonomous Vehicle", "Robotaxi", "Last-Mile Delivery", "Edge Computing", "V2X Communication", "GNSS", "SAE J3016", "ISO 26262", "Advanced Driver Assistance Systems" ] }, { "id": "autonomous-governance", "title": "Autonomous Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Autonomous governance is a model in which decision-making and rule enforcement are executed by code and protocol rather than by a central administrator, typically via smart contracts and on-chain voting. It enables decentralised organisations such as DAOs to allocate resources, upgrade parameters, and resolve disputes through transparent, programmatically binding processes. The approach trades human discretion for verifiable, tamper-resistant execution of collectively agreed rules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-governance", "labels": [ "Autonomous Governance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "autonomous-manipulation", "title": "Autonomous Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Autonomous manipulation is the capability of a robot to perceive, plan and physically interact with objects to achieve a task goal without step-by-step human teleoperation. It integrates perception, grasp and motion planning, force control and feedback to handle uncertainty in object pose, shape and contact dynamics. The aim is robust, closed-loop interaction in unstructured environments rather than the repetition of pre-programmed trajectories.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-manipulation", "labels": [ "Autonomous Manipulation" ], "is_subclass_of": [ "Robotic Manipulation" ], "wikilinks": [] }, { "id": "autonomous-manufacturing", "title": "Autonomous Manufacturing", "domain": "robotics", "domain_name": "Robotics", "definition": "Autonomous manufacturing is a production model in which robotic systems and automated machinery execute manufacturing tasks \u2014 assembly, inspection, material handling, and quality control \u2014 with minimal or no direct human intervention, coordinated by software that senses conditions and adapts in real time. It extends manufacturing automation by incorporating perception, planning, and decision-making capabilities that let equipment respond to variation on the production line rather than following fixed, pre-programmed sequences. Autonomous manufacturing is a core pillar of Industry 4.0 initiatives that seek to combine robotics, sensing, and data analytics across the factory floor.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-manufacturing", "labels": [ "Autonomous Manufacturing" ], "is_subclass_of": [ "Manufacturing Automation" ], "wikilinks": [] }, { "id": "autonomous-mobile-robots", "title": "Autonomous Mobile Robots", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotic platforms capable of self-directed locomotion through environments, navigating without fixed infrastructure guidance by using onboard sensing, mapping, localisation, and path-planning capabilities. Autonomous mobile robots (AMRs) operate in dynamic, human-shared spaces and are distinguished from automated guided vehicles (AGVs) by their ability to adapt routes in real time rather than following pre-defined paths.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:autonomous-mobile-robots", "labels": [ "Autonomous Mobile Robots", "Autonomous Mobile Robot" ], "is_subclass_of": [ "Mobile Robotics", "Mobile Robot" ], "wikilinks": [] }, { "id": "autonomous-navigation", "title": "Autonomous Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Autonomous Navigation encompasses the complete system capability for a robot or autonomous agent to move from one location to another without human guidance, integrating perception, localisation, mapping, path planning, obstacle avoidance, and control. Autonomous navigation systems employ SLAM, sensor fusion, and AI-based decision-making to operate safely in unknown or dynamic environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-navigation", "labels": [ "Autonomous Navigation", "Autonomous Navigation Support", "Autonomous Vehicle Navigation", "AutonomousNavigation" ], "is_subclass_of": [ "Autonomous Agent" ], "wikilinks": [ "Localisation", "Autonomous Robot", "MetaverseDomain", "Path Planning", "SLAM" ] }, { "id": "autonomous-operation", "title": "Autonomous Operation", "domain": "ai", "domain_name": "Ai", "definition": "The sustained execution of a system's core functions without ongoing human intervention, encompassing self-monitoring, self-configuration, self-healing, and self-optimisation capabilities that allow the system to maintain operational objectives across varying environmental conditions and failure states. Autonomous operation represents the highest level of system self-sufficiency on the automation continuum.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-operation", "labels": [ "Autonomous Operation" ], "is_subclass_of": [ "Autonomous System", "Workflow Automation", "Process Automation" ], "wikilinks": [ "Autonomous Behavior", "Feedback Control", "Autonomous Agent", "Autonomous Mobile Robots", "Agentic Workflow", "Agentic AI", "Functional Safety", "Monitoring System", "Autonomous System", "Autonomic Computing", "Self-Healing System", "Fault-Tolerant Control", "AI Safety", "Human-in-the-Loop", "EU AI Act", "Operational Design Domain", "Decision Engine", "Perception Module", "Reinforcement Learning", "Large Language Models" ] }, { "id": "autonomous-proof-generation", "title": "Autonomous Proof Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The capability of AI systems to independently formulate, generate, and verify mathematical proofs without human intervention.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:autonomous-proof-generation", "labels": [ "Autonomous Proof Generation" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "autonomous-robot", "title": "Autonomous Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotic system equipped with sensors, processing units, and actuators that operates independently to perform tasks without direct human control, using Artificial Intelligence and Autonomous Navigation to perceive, reason, and act in physical or virtual environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-robot", "labels": [ "Autonomous Robot", "AutonomousRobot" ], "is_subclass_of": [ "Robotic System" ], "wikilinks": [ "Actuators", "BlockchainIdentity", "dt:authenticatedBy", "dt:controlledBy", "dt:coordinatedBy", "dt:navigatesUsing", "dt:operatesIn", "hasSensor", "MultiAgentSystem", "navigatesIn", "Obstacle Detection", "performsTask", "Processing Units", "Real-time Control", "RobotPerception", "Sensors", "usesController", "AISystem", "Artificial Intelligence", "Autonomous Navigation" ] }, { "id": "autonomous-system", "title": "Autonomous System", "domain": "robotics", "domain_name": "Robotics", "definition": "An Autonomous System is a computational or physical entity capable of perceiving its environment through sensors, reasoning over that perception, and executing goal-directed actions without continuous human intervention, forming closed-loop sense-plan-act cycles. Architecturally, autonomous systems combine perception pipelines, world-model maintenance, deliberative or reactive planning, and actuator control, often augmented by machine learning to handle environmental uncertainty. They operate across a spectrum from fully automated (no human input required) to conditionally autonomous (human on the loop), and encompass platforms as diverse as self-driving vehicles, unmanned aerial systems, autonomous underwater vehicles, industrial robotic cells, and intelligent software agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:autonomous-system", "labels": [ "Autonomous System", "Autonomous Systems", "Autonomous Systems Deployment", "Autonomous Systems Platform" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "autonomous-task-execution", "title": "Autonomous Task Execution", "domain": "ai", "domain_name": "Ai", "definition": "Autonomous Task Execution refers to the capacity of AI agents or automated systems to independently carry out goal-directed actions \u2014 including planning, tool invocation, error recovery, and result validation \u2014 without continuous human supervision. It encompasses the full lifecycle from task decomposition through completion, spanning single-step tool calls and long-horizon multi-step workflows.", "entityType": "Class", "qualityScore": 0.92, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-task-execution", "labels": [ "Autonomous Task Execution", "Autonomous Task Completion" ], "is_subclass_of": [ "Agentic AI", "Autonomous Operation", "Workflow Automation", "Process Automation" ], "wikilinks": [ "Agentic AI", "Large Language Models", "Automated Planning", "Tool Use", "Working Memory", "Reinforcement Learning", "Agentic Workflow", "Multi-Agent Orchestration", "Process Automation", "Human-in-the-Loop", "Task Decomposition", "Error Recovery", "Sandboxed Execution", "Reasoning", "Context Management", "Robotic Process Automation", "LLM Agents", "Long-Horizon Planning", "ReAct Prompting", "Robotics" ] }, { "id": "autonomous-vehicle-control", "title": "Autonomous Vehicle Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Autonomous vehicle control is the set of algorithms and systems that compute steering, acceleration and braking commands for a self-driving vehicle in order to safely follow a planned trajectory. It draws on model-based control, which uses an explicit dynamics model to compute corrective actions, and on optimal control, which formulates control as minimising a cost function subject to the vehicle's dynamics and constraints. It must operate reliably under uncertainty from sensing, actuation and the surrounding environment.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:autonomous-vehicle-control", "labels": [ "Autonomous Vehicle Control" ], "is_subclass_of": [ "Autonomous Vehicle" ], "wikilinks": [] }, { "id": "autonomous-vehicle-testing", "title": "Autonomous Vehicle Testing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Autonomous vehicle testing is the discipline of validating the safety, reliability, and performance of self-driving systems through simulation, closed-course trials, and supervised public-road operation. It exercises the perception, planning, and control stack against a vast space of driving scenarios, including rare and hazardous edge cases that are impractical to encounter physically. Simulation-based testing has become central because it enables scalable, repeatable, and safe exploration of these scenarios before real-world deployment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-vehicle-testing", "labels": [ "Autonomous Vehicle Testing" ], "is_subclass_of": [ "Autonomous Vehicle" ], "wikilinks": [] }, { "id": "autonomous-vehicle", "title": "Autonomous Vehicle", "domain": "robotics", "domain_name": "Robotics", "definition": "An Autonomous Vehicle is a self-driving vehicle capable of navigating and operating without human intervention, employing artificial intelligence for perception, localisation, path planning, motion control, and decision-making. Autonomous vehicles integrate sensor fusion, computer vision, deep learning, and control algorithms to achieve SAE automation levels ranging from Level 1 (driver assistance) to Level 5 (full automation).", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:autonomous-vehicle", "labels": [ "Autonomous Vehicle", "Autonomous Vehicles" ], "is_subclass_of": [ "Autonomous Robot" ], "wikilinks": [ "ISO 21448", "ISO 26262", "ADAS", "Agents", "Autonomous Robot", "MetaverseDomain", "Path Planning", "Perception System", "Self Driving Car", "Sensor Fusion", "Social contract and jobs", "Some legacy Linked-JSON" ] }, { "id": "autonomy-level", "title": "Autonomy Level", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A graded classification (typically 0\u20135) quantifying the degree to which an agent perceives, decides, and acts without human intervention, ranging from fully manual operation through partial and conditional autonomy to full self-governance, with domain-specific scales for AI, robotics, blockchain DAOs, and multi-agent systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:autonomy-level", "labels": [ "Autonomy Level" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Agent Property", "Alignment", "ISO 21448", "Trust", "Agent", "AI Agent System", "BDI Model", "Blockchain", "EU AI Act", "Goal", "Human in the Loop", "Objective", "Safety" ] }, { "id": "autoregressive-decoding", "title": "Autoregressive Decoding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autoregressive decoding is the process by which a sequence model, such as a transformer language model, generates output one token at a time, conditioning each new token on all previously generated tokens. At every step the model produces a probability distribution over the vocabulary, a token is selected by a chosen strategy, and the token is appended and fed back as input for the next step. It is the dominant generation paradigm for large language models and is the primary target of inference optimisations such as caching and speculative methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:autoregressive-decoding", "labels": [ "Autoregressive Decoding" ], "is_subclass_of": [ "Large Language Model", "AI Technique" ], "wikilinks": [] }, { "id": "autoregressive-generation", "title": "Autoregressive Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Autoregressive generation is a sequence-modelling approach in which each output element is produced conditioned on all previously generated elements, factorising the joint probability of a sequence into a product of conditional next-element distributions. In language models it manifests as repeated next-token prediction, where the model samples or selects a token, appends it to the context, and repeats. This left-to-right dependency is the dominant decoding paradigm for large language models and underlies generative text, code, and other ordered outputs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:autoregressive-generation", "labels": [ "Autoregressive Generation" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "autoregressive-model", "title": "Autoregressive Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An autoregressive model generates sequences by predicting each element conditioned on all previously generated elements, thereby factorising the joint probability distribution of a sequence into an ordered product of conditional distributions via the chain rule of probability. This approach yields exact log-likelihoods and a straightforward maximum-likelihood training objective, making it the dominant paradigm for large language models, neural audio synthesis, and image generation. Inference is inherently sequential \u2014 each token must be sampled before the next can be computed \u2014 creating a fundamental latency trade-off relative to parallel decoding strategies. Modern architectures such as the Transformer exploit masked self-attention to parallelise training while preserving the strictly left-to-right conditional structure at inference time.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:autoregressive-model", "labels": [ "Autoregressive Model", "Autoregressive Language Model" ], "is_subclass_of": [ "Generative Model", "Deep Generative Model", "Probabilistic Model" ], "wikilinks": [ "Probabilistic Model", "Language Model", "Transformer", "Generative Model" ] }, { "id": "autoscaling", "title": "Autoscaling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Autoscaling is the automated adjustment of computing resources allocated to an application in response to observed demand, scaling capacity up under load and down when demand falls. It monitors metrics such as utilisation, request rate or queue depth and triggers provisioning or removal of compute instances or containers against defined policies. Autoscaling improves cost efficiency and availability by matching supply to demand without manual intervention.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:autoscaling", "labels": [ "Autoscaling" ], "is_subclass_of": [ "Resource Management" ], "wikilinks": [] }, { "id": "avalanche-effect", "title": "Avalanche Effect", "domain": "security", "domain_name": "Security", "definition": "The avalanche effect is a desirable property of cryptographic primitives whereby a tiny change in the input, such as flipping a single bit, produces an extensive, unpredictable change in the output, ideally altering about half of the output bits. In hash functions and ciphers it is the practical expression of diffusion, ensuring that outputs reveal no exploitable correlation with their inputs. A strong avalanche effect is essential for collision and preimage resistance and for resisting differential cryptanalysis.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:avalanche-effect", "labels": [ "Avalanche Effect" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "avalanche", "title": "Avalanche", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Avalanche is a high-throughput, low-latency layer-1 blockchain platform developed by Ava Labs that employs the Avalanche consensus family \u2014 a suite of leaderless, Byzantine-fault-tolerant protocols (Snowflake, Snowball, Avalanche) based on repeated random sub-sampled voting \u2014 to achieve probabilistic transaction finality in under two seconds. The platform is architecturally divided into three purpose-built chains: the Exchange Chain (X-Chain) for UTXO-model asset transfers, the Platform Chain (P-Chain) for validator and subnet management, and the Contract Chain (C-Chain) providing an EVM-compatible execution environment for Solidity smart contracts. Its subnet (now called L1) model allows developers to launch application-specific blockchains with custom virtual machines while optionally sharing primary-network validator sets, enabling both permissioned enterprise deployments and public decentralised applications within a unified interoperable ecosystem.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:avalanche", "labels": [ "Avalanche", "Avalanche Network" ], "is_subclass_of": [ "Blockchain Network" ], "wikilinks": [] }, { "id": "avatar-animation", "title": "Avatar Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar animation is the specialised subset of digital animation concerned with driving the motion and expression of user-controlled or AI-controlled avatar representations in real-time interactive environments. It encompasses full-body locomotion, facial expression synthesis, hand and gaze tracking, and upper-body gesture systems, all coordinated to produce social believability within virtual and extended reality spaces. Avatar animation systems must balance visual fidelity with low-latency responsiveness to maintain user embodiment and presence. Standards such as VRM and glTF define interchange formats that allow avatar animations to transfer across platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:avatar-animation", "labels": [ "Avatar Animation", "Avatar Animation System" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "avatar-behavior", "title": "Avatar Behavior", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The set of behavioural patterns, animation responses, and interactive capabilities that govern how avatars and NPCs act in virtual environments, driven by AI/ML inference, procedural animation, inverse kinematics, facial expression synthesis, and real-time response to user inputs and social context.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:avatar-behavior", "labels": [ "Avatar Behavior" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "DID Nostr Identity", "Metaverse" ] }, { "id": "avatar-creation", "title": "Avatar Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar Creation is the process of designing and generating 3D digital representations of users for metaverse environments, encompassing selfie-based AI generation, manual customization tools, full-body scanning, and procedural generation techniques that enable personalized virtual identities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:avatar-creation", "labels": [ "Avatar Creation" ], "is_subclass_of": [ "Content and Assets", "Interaction Technology", "Digital Identity Creation" ], "wikilinks": [ "3D Modeling Tools", "AI Generation Systems", "Customization Interfaces", "Digital Identity Creation", "Digital Persona Management", "Metaverse Participation", "Virtual Self-Expression", "DID Nostr Identity" ] }, { "id": "avatar-customization", "title": "Avatar Customization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar Customization is the practice of modifying and personalizing digital avatar attributes including physical features, clothing, accessories, animations, and expressions to create unique virtual representations that reflect user identity and preferences in metaverse and spatial-computing environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:avatar-customization", "labels": [ "Avatar Customization", "AvatarCustomization", "Customization Interfaces" ], "is_subclass_of": [ "Content and Assets", "Digital Personalization" ], "wikilinks": [ "Asset Libraries", "Customization Tools", "Digital Personalization", "Identity Representation", "Real-Time Preview", "Self-Expression", "Social Distinction", "DID Nostr Identity" ] }, { "id": "avatar-embodiment", "title": "Avatar Embodiment", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Avatar Embodiment is the psychological and technical phenomenon in which a user perceives a digital avatar as an extension or representation of their own body within a virtual or mixed reality environment. It involves mapping real-time motion capture, physiological signals, and expressive cues from the user onto the avatar to create a sense of ownership and presence. High-fidelity embodiment enhances social telepresence by allowing remote participants to express identity, emotion, and intent through their avatars.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:avatar-embodiment", "labels": [ "Avatar Embodiment" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "avatar-interoperability", "title": "Avatar Interoperability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Capability enabling an avatar's identity, appearance, and behaviors to function seamlessly across multiple metaverse platforms and virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:avatar-interoperability", "labels": [ "Avatar Interoperability", "Cross-Platform Avatar Interoperability" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "Behavior Mapping", "Cross-Platform Authentication", "Cross-Platform Presence", "Data Serialization", "glTF", "Identity Portability", "Identity Protocol", "MSF DG (Interoperable Avatars)", "Persistent Identity", "Platform API", "Seamless Migration", "Appearance Translation", "Avatar Standard", "DataLayer", "DID Nostr Identity", "HAnim Standard", "InteractionDomain", "MiddlewareLayer", "Universal Avatar", "VRM Format" ] }, { "id": "avatar-portability", "title": "Avatar Portability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar Portability is the capability to transfer digital avatar representations between different metaverse platforms, applications, and virtual worlds while maintaining visual fidelity, customization, and associated digital assets, enabled by standardized formats and interoperability frameworks.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:avatar-portability", "labels": [ "Avatar Portability" ], "is_subclass_of": [ "Standards and Interoperability", "Digital Asset Portability" ], "wikilinks": [ "Asset Continuity", "Digital Asset Portability", "glTF", "Khronos Group", "Metaverse Standards Forum", "Platform Integration", "Ready Player Me", "Seamless World Transitions", "Standardized Formats", "Translation Frameworks", "VRM Consortium", "Cross-Platform Identity", "DID Nostr Identity", "VRM Format" ] }, { "id": "avatar-standard", "title": "Avatar Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar Standard refers to technical specifications defining file formats, data structures, rigging conventions, and metadata schemas for 3D humanoid avatars, particularly the VRM format built on glTF 2.0 that enables cross-platform avatar interoperability in metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:avatar-standard", "labels": [ "Avatar Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standard" ], "wikilinks": [ "Format Compliance", "glTF 2.0", "Khronos Group", "Metadata Specification", "Metaverse Standards Forum", "Platform-Independent Identities", "Skeleton Configuration", "Standardized Rigging", "VRM 1.0", "VRM Consortium", "Avatar Interoperability", "Technical Standard" ] }, { "id": "avatar-system", "title": "Avatar System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Avatar System is the integrated technical architecture for creating, customizing, animating, and rendering digital representations of users in virtual environments, encompassing character models, animation systems, facial expression rigs, physics simulations, and real-time rendering pipelines.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:avatar-system", "labels": [ "Avatar System", "Avatar Systems" ], "is_subclass_of": [ "Content and Assets", "Virtual Character System" ], "wikilinks": [ "Animation Controller", "Embodied Presence", "Input Processing", "Social Interaction", "User Representation", "Virtual Character System", "3D Rendering Engine" ] }, { "id": "avatar-wearable", "title": "Avatar Wearable", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital clothing, accessories, and cosmetic items that can be equipped by avatars in virtual environments, often tokenised as NFTs to enable ownership, cross-platform portability, and secondary-market trading within metaverse economies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:avatar-wearable", "labels": [ "Avatar Wearable" ], "is_subclass_of": [ "Content and Assets", "Digital Asset" ], "wikilinks": [ "Digital Asset", "Metaverse", "Telecollaboration" ] }, { "id": "avatar", "title": "Avatar", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital representation of a person or autonomous agent used to perceive, act, and interact within a virtual or mixed-reality environment, embodying identity, appearance, and behavioural state.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:avatar", "labels": [ "Avatar", "Avatar Identity", "Avatar Pipeline", "Avatar Representation", "Photorealistic Avatar", "Virtual Avatar" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "ACM + Web3D HAnim", "Animation Rig", "User Embodiment", "UserExperienceLayer", "Visual Mesh", "3D Rendering Engine", "InteractionDomain", "Metaverse", "Social Presence" ] }, { "id": "avionics", "title": "Avionics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Avionics refers to the electronic systems used on aircraft and spacecraft for communication, navigation, flight control, monitoring, and mission management. These systems are safety-critical and must meet stringent real-time, reliability, and certification standards such as DO-178C and DO-254. Avionics is a canonical domain for hard real-time computing, where deterministic timing and fault tolerance are essential to flight safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:avionics", "labels": [ "Avionics", "Aerospace Avionics" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "awareness", "title": "Awareness", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Awareness in metaverse contexts refers to systems and mechanisms that provide users with perception of other participants, environmental changes, and relevant contextual information in shared virtual spaces, supporting presence, social interaction, and collaborative activities through visual, aud...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:awareness", "labels": [ "Awareness", "Interruptibility Awareness", "Mutual Awareness" ], "is_subclass_of": [ "Interaction Technology", "Social Computing" ], "wikilinks": [ "Collaborative Presence", "Notification Systems", "Situational Understanding", "Social Computing", "Social Interaction", "metaverse", "Presence Detection", "State Synchronization" ] }, { "id": "axelar", "title": "Axelar", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Axelar is a decentralised cross-chain communication network that enables general message passing and token transfers between heterogeneous blockchains via a proof-of-stake overlay network and a permissionless gateway smart contract model. It provides a Universal Message Passing (UMP) primitive that allows any contract on any connected chain to call any contract on any other connected chain as a single composable operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:axelar", "labels": [ "Axelar" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "axie-infinity", "title": "Axie Infinity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Axie Infinity is a blockchain-based game in which players collect, breed, and battle digital creatures called Axies that are represented as non-fungible tokens.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:axie-infinity", "labels": [ "Axie Infinity" ], "is_subclass_of": [ "NFT" ], "wikilinks": [ "NFT", "Smart Contract", "Ethereum" ] }, { "id": "aztec-network", "title": "Aztec Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The privacy-preserving Layer 2 network operated under the Aztec protocol, providing confidential transactions and private smart contracts settled on Ethereum.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:aztec-network", "labels": [ "Aztec Network" ], "is_subclass_of": [ "Aztec" ], "wikilinks": [ "Aztec Protocol", "Ethereum", "Privacy", "Layer 2 Networks", "Aztec" ] }, { "id": "aztec-protocol", "title": "Aztec Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The protocol underlying the Aztec network that combines zero-knowledge proofs with an encrypted note model to provide confidential transactions and private smart contracts on Ethereum, enabling programmable privacy through a UTXO-style note commitment scheme verified by recursive ZK rollup proofs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:aztec-protocol", "labels": [ "Aztec Protocol" ], "is_subclass_of": [ "Aztec" ], "wikilinks": [ "Zero-Knowledge Proof", "Ethereum", "Privacy", "Aztec Network", "Aztec" ] }, { "id": "aztec", "title": "Aztec", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A privacy-focused Layer 2 network for Ethereum that uses zero-knowledge proofs to enable confidential transactions and private smart contract execution, supporting encrypted notes, private state transitions, and programmable privacy at the application layer.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:aztec", "labels": [ "Aztec", "Aztec Connect" ], "is_subclass_of": [ "Layer 2 Networks" ], "wikilinks": [ "Zero-Knowledge Proof", "Ethereum", "Privacy", "Aztec Protocol", "Layer 2 Networks" ] }, { "id": "azure", "title": "Azure", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Azure is Microsoft's public cloud computing platform offering compute, storage, networking, identity and managed services.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:azure", "labels": [ "Azure", "Azure AI", "Azure Cloud", "Microsoft Azure", "Microsoft Azure Cognitive Services" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [ "Identity Management", "Web3 Infrastructure", "Microsoft Entra Verified ID", "Cloud Computing", "https://learn.microsoft.com/en-us/azure/", "https://azure.microsoft.com/" ] }, { "id": "b-tree-index", "title": "B-Tree Index", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A B-tree index is a self-balancing tree data structure used by database engines to maintain sorted data and support efficient logarithmic-time lookups, range scans, insertions, and deletions. Each node holds multiple sorted keys and child pointers, keeping the tree shallow and minimising disk reads relative to a binary tree. It is the default index structure in relational database systems such as PostgreSQL for primary keys and most secondary indexes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:b-tree-index", "labels": [ "B-Tree Index" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "bart", "title": "BART", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bidirectional and Auto-Regressive Transformers: a denoising sequence-to-sequence pre-training architecture developed by Meta AI that combines a bidirectional encoder (as in BERT) with an autoregressive decoder (as in GPT), achieving state-of-the-art results on abstractive summarisation, dialogue, and machine translation through flexible noise-based corruption objectives.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:bart", "labels": [ "BART", "BART Model" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "bbs-signature", "title": "BBS+ Signature", "domain": "security", "domain_name": "Security", "definition": "BBS+ Signature is a pairing-based digital signature scheme that signs a vector of messages simultaneously and supports the generation of zero-knowledge proofs that reveal only a chosen subset of those messages \u2014 a property known as selective disclosure \u2014 without revealing the full signed message set or enabling linkage of multiple presentations to the same credential. It is a cornerstone primitive for privacy-preserving verifiable credentials.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bbs-signature", "labels": [ "BBS+ Signature" ], "is_subclass_of": [ "Cryptographic Signature" ], "wikilinks": [] }, { "id": "bbs-signatures", "title": "BBS+ Signatures", "domain": "security", "domain_name": "Security", "definition": "BBS+ Signatures is the collective term for the class of pairing-based multi-message signature schemes derived from the BBS construction, encompassing both the base signature algorithm and the proof-of-knowledge protocols that enable selective disclosure and unlinkable presentation of signed credential attributes. As a scheme family, BBS+ Signatures represent the cryptographic foundation for privacy-preserving digital credential ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bbs-signatures", "labels": [ "BBS+ Signatures", "BBS Plus Signatures" ], "is_subclass_of": [ "Cryptographic Signature" ], "wikilinks": [] }, { "id": "bc-0120-consensus-mechanism", "title": "BC-0120-consensus-mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BC-0120-consensus-mechanism is an ontology catalogue entry representing the knowledge domain of blockchain consensus mechanisms \u2014 the family of distributed protocols by which a peer-to-peer network of nodes agrees on a canonical ordering of transactions and the current state of a shared ledger without a trusted central authority. It serves as an organising concept node within a blockchain knowledge graph taxonomy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-0120-consensus-mechanism", "labels": [ "BC-0120-consensus-mechanism" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "bc-0442-certification-and-compliance", "title": "BC-0442-certification-and-compliance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Certification and compliance in blockchain refers to the processes by which systems, service providers and tokens are assessed against technical standards and legal requirements, and formally attested as meeting them.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bc-0442-certification-and-compliance", "labels": [ "BC-0442-certification-and-compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "Regulatory Compliance", "Compliance" ] }, { "id": "bc-0456-virtual-asset-service-providers", "title": "BC-0456-virtual-asset-service-providers", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Virtual asset service providers are businesses that conduct activities such as exchange, transfer, custody or issuance of crypto-assets on behalf of others, and that are subject to financial regulation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bc-0456-virtual-asset-service-providers", "labels": [ "BC-0456-virtual-asset-service-providers", "Virtual Asset Service Provider" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "AML KYC Compliance", "Travel Rule", "Regulatory Compliance", "FATF" ] }, { "id": "bc-0480-kyc-requirements", "title": "BC-0480-kyc-requirements", "domain": "blockchain", "domain_name": "Blockchain", "definition": "KYC requirements are the obligations placed on regulated firms to identify and verify their customers, assess risk and monitor activity as part of anti-money-laundering and counter-terrorist-financing controls.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bc-0480-kyc-requirements", "labels": [ "BC-0480-kyc-requirements" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "Identity Verification", "Anti-Money Laundering", "Know Your Customer", "Regulatory Compliance" ] }, { "id": "bc-0482-eu-mica-regulation", "title": "BC-0482-eu-mica-regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The EU MiCA regulation is the European Union's Markets in Crypto-Assets framework, which establishes harmonised rules for the issuance, offering and provision of services relating to crypto-assets across member states.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bc-0482-eu-mica-regulation", "labels": [ "BC-0482-eu-mica-regulation", "MiCA Regulation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "Securities Regulation", "Regulatory Compliance", "MiCA Regulation" ] }, { "id": "bc-0484-markets-in-crypto-assets", "title": "BC-0484-markets-in-crypto-assets", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Markets in Crypto-Assets is the European Union regulatory framework that defines categories of crypto-assets and sets authorisation and conduct rules for their issuers and service providers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bc-0484-markets-in-crypto-assets", "labels": [ "BC-0484-markets-in-crypto-assets", "Markets in Crypto-Assets Regulation", "MiCA (Markets in Crypto-Assets)" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "Securities Regulation", "Regulatory Compliance", "MiCA Regulation" ] }, { "id": "bdi-model", "title": "BDI Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A deliberative agent architecture grounded in Bratman's theory of practical reasoning, structuring agent cognition around Beliefs (knowledge about the world), Desires (motivational goals), and Intentions (committed plans), with a reasoning cycle of belief revision, deliberation, means-end reasoning, and intention reconsideration.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:bdi-model", "labels": [ "BDI Model", "BDI Architecture" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Agent Architecture", "Deliberative Architecture", "Hybrid Architecture", "Multi-Agent System", "Plan", "Practical Reasoning", "Reactive Architecture", "Subsumption Architecture", "Agent", "Artificial Intelligence", "Autonomy Level", "Goal" ] }, { "id": "beir-benchmark", "title": "BEIR Benchmark", "domain": "ai", "domain_name": "Ai", "definition": "BEIR (Benchmarking IR) is a heterogeneous benchmark suite for evaluating the zero-shot generalisation of information retrieval models across eighteen diverse datasets spanning domains including biomedical, legal, financial, and Wikipedia text. It measures how well retrieval systems transfer across topic domains and retrieval task types without domain-specific fine-tuning, providing a standardised framework for comparing dense, sparse, and hybrid retrieval approaches.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:beir-benchmark", "labels": [ "BEIR Benchmark" ], "is_subclass_of": [ "Information Retrieval", "Evaluation benchmarks and leaderboards", "Evaluation Benchmark", "Natural Language Processing" ], "wikilinks": [ "Information Retrieval", "Embedding Model", "Semantic Search", "Dense Passage Retrieval", "BM25", "Retrieval-Augmented Generation", "Natural Language Processing", "Transformer" ] }, { "id": "bert", "title": "BERT", "domain": "ai", "domain_name": "Ai", "definition": "BERT (Bidirectional Encoder Representations from Transformers) is a pre-trained deep language representation model developed by Google AI Language, published in 2018, which applies a bidirectional Transformer encoder trained with masked language modelling (MLM) and next sentence prediction (NSP) objectives on large text corpora to produce contextualised word embeddings. Unlike prior unidirectional models, BERT conditions each token on its full left and right context simultaneously, enabling a richer semantic representation that generalises across diverse NLP tasks via fine-tuning. It established the pre-train-then-fine-tune paradigm for natural language processing, achieving state-of-the-art performance across eleven NLP benchmarks upon release, including GLUE and SQuAD. Its architecture directly underpins a large family of subsequent encoder models including RoBERTa, ALBERT, DistilBERT, and multilingual mBERT.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:bert", "labels": [ "BERT" ], "is_subclass_of": [ "Language Model", "AI Model Architecture", "Pre-Trained Language Model" ], "wikilinks": [] }, { "id": "bim-software", "title": "BIM Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specialized applications for Building Information Modeling that enable the generation, management, and collaboration on digital representations of physical and functional characteristics of buildings and infrastructure, supporting the entire asset lifecycle from design through construction to ope...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:bim-software", "labels": [ "BIM Software" ], "is_subclass_of": [ "Platform and Environment", "Design Software" ], "wikilinks": [ "CAD Capabilities", "Collaboration Tools", "Construction Coordination", "Facility Management", "3D Modeling", "Data Management", "Design Software", "metaverse" ] }, { "id": "bim-virtual-model", "title": "BIM Virtual Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive 3D digital representation of a building or infrastructure asset created through Building Information Modeling, containing geometric data, material specifications, and functional characteristics that enable visualization, simulation, quantity take-offs, and clash detection througho...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:bim-virtual-model", "labels": [ "BIM Virtual Model", "BIM", "BIM Model", "Building Information Modelling" ], "is_subclass_of": [ "Platform and Environment", "3D Model" ], "wikilinks": [ "Clash Detection", "Construction Simulation", "Design Data", "Design Visualization", "ISO (International Organization for Standardization)", "Modeling Standards", "3D Model", "BIM Software", "metaverse" ] }, { "id": "bip-process", "title": "BIP Process", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Bitcoin Improvement Proposal (BIP) process is the standardised social and technical workflow by which proposed changes to the Bitcoin protocol, best practices, and informational standards are drafted, reviewed, debated, and either accepted or rejected by the Bitcoin developer community. It defines the lifecycle of a proposal from Draft through Proposed, Final, Active, Withdrawn, or Rejected, providing a structured mechanism for decentralised protocol governance without a centralised authority.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bip-process", "labels": [ "BIP Process" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "bip-16", "title": "BIP-16", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-16 is the Bitcoin Improvement Proposal that introduced Pay-to-Script-Hash (P2SH), allowing funds to be sent to the hash of a redeem script rather than to a full script. The spender supplies the matching script and its satisfying inputs at redemption time, shifting the burden of specifying complex spending conditions from sender to recipient. P2SH made multi-signature and other complex scripts practical and is the standard mechanism behind multi-sig wallet addresses.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bip-16", "labels": [ "BIP-16", "BIP 16" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "bip-327", "title": "BIP-327", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-327 is a Bitcoin Improvement Proposal that specifies MuSig2, a two-round multi-party Schnorr signature protocol that enables a group of n signers to collaboratively produce a single aggregated Schnorr signature indistinguishable from a single-party signature. BIP-327 provides a cryptographically secure specification for key aggregation and signature combination compatible with BIP-340 Schnorr signatures and the Taproot upgrade, enabling private and efficient threshold custody and collaborative signing workflows on Bitcoin.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bip-327", "labels": [ "BIP-327" ], "is_subclass_of": [ "Bitcoin Improvement Proposals" ], "wikilinks": [] }, { "id": "bip-328", "title": "BIP-328", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-328 is a Bitcoin Improvement Proposal defining the Hierarchical Deterministic Keys (HD Keys) using the Silent Payments-related and descriptor key formats for derivation, standardising how extended keys and key origins are encoded. It supports the unambiguous representation of key paths and fingerprints used by descriptor-based and Taproot-aware wallets. By standardising key encoding, it improves interoperability between wallets handling Taproot assets and multi-party setups.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bip-328", "labels": [ "BIP-328" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "bip-329", "title": "BIP-329", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-329 is a Bitcoin Improvement Proposal that specifies a standard, wallet-agnostic format for exporting and importing wallet labels. It defines a JSONL structure associating human-readable labels and metadata with transactions, addresses, inputs, outputs, and extended public keys. The standard improves portability of bookkeeping and provenance information between wallets, which is valuable for asset tracking and compliance in Taproot-asset and multi-wallet workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bip-329", "labels": [ "BIP-329" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "bip-330", "title": "BIP-330", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-330 is a Bitcoin Improvement Proposal that introduces Erlay-style transaction reconciliation (the 'sendtxrcncl' protocol) to make mempool synchronisation between nodes more bandwidth-efficient. Instead of announcing every transaction to every peer, nodes use set-reconciliation sketches to identify only the transactions a peer is missing. This reduces redundant relay traffic, improving the scalability and privacy of the peer-to-peer network that supports applications like Taproot assets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:bip-330", "labels": [ "BIP-330" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "bip-340-schnorr-keypair", "title": "BIP-340 Schnorr Keypair", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A BIP-340 Schnorr Keypair is the 32-byte public key and associated private key pairing defined by Bitcoin Improvement Proposal 340, which introduced Schnorr signature support to Bitcoin via the Taproot upgrade activated in November 2021. Unlike the earlier ECDSA scheme used in Bitcoin, BIP-340 uses x-only public keys \u2014 just the x-coordinate of the secp256k1 curve point \u2014 reducing on-chain byte footprint and simplifying key aggregation. The scheme is provably secure under the discrete logarithm assumption and supports native key and signature aggregation via protocols such as MuSig2, enabling multi-party signing that is indistinguishable from single-party signing on-chain. BIP-340 keypairs are the cryptographic foundation for Taproot outputs, Tapscript, and the Nostr identity protocol.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bip-340-schnorr-keypair", "labels": [ "BIP-340 Schnorr Keypair" ], "is_subclass_of": [ "Cryptographic Keys" ], "wikilinks": [] }, { "id": "bip-340", "title": "BIP-340", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-340 is the Bitcoin Improvement Proposal that standardises Schnorr signatures for use on the Bitcoin network using the secp256k1 elliptic curve, defining a deterministic 64-byte signature scheme that is provably secure in the random oracle model. It introduces x-only 32-byte public key encoding, a tagged hash nonce-generation procedure, and rigorous test vectors to guarantee cross-implementation compatibility. BIP-340 is a strict prerequisite for Taproot (BIP-341) and Tapscript (BIP-342), enabling key aggregation via MuSig2 and linear signature composition that makes multisignature spends indistinguishable on-chain from single-key spends. The proposal represents the most significant cryptographic upgrade to Bitcoin since its inception, substantially improving privacy, efficiency, and the expressiveness of multi-party signing protocols.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:bip-340", "labels": [ "BIP-340", "BIP-340 Cryptography", "BIP-340 Pubkey" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "bip-341-taproot", "title": "BIP-341 Taproot", "domain": "standards", "domain_name": "Standards", "definition": "A Bitcoin Improvement Proposal defining Taproot, an upgrade introducing a new output type using Schnorr signatures and Merkle-branch script commitments. It improves privacy and flexibility of Bitcoin transactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bip-341-taproot", "labels": [ "BIP-341 Taproot" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Bitcoin", "Technical Standard" ] }, { "id": "bip-341", "title": "BIP-341", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-341 is the Bitcoin Improvement Proposal that specifies the Taproot upgrade, introducing Pay-to-Taproot (P2TR) as a native SegWit version 1 output type that unifies key-path and script-path spending under a single Schnorr-based commitment. It integrates Merklised Alternative Script Trees (MAST) so that only the executed spending branch is revealed on-chain, preserving the privacy of unexplored script conditions. BIP-341 activated on the Bitcoin mainnet at block 709,632 in November 2021 via the Speedy Trial soft-fork mechanism, requiring 90% miner signalling. It forms part of the Taproot bundle alongside BIP-340 (Schnorr signatures for Bitcoin) and BIP-342 (Tapscript), collectively the largest upgrade to Bitcoin's scripting system since SegWit.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:bip-341", "labels": [ "BIP-341" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "bip-342-tapscript", "title": "BIP-342 Tapscript", "domain": "standards", "domain_name": "Standards", "definition": "A Bitcoin Improvement Proposal defining Tapscript, the scripting semantics used within Taproot spends. It specifies the validation rules for scripts committed under the Taproot output type.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bip-342-tapscript", "labels": [ "BIP-342 Tapscript", "Tapscript" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Bitcoin", "Technical Standard" ] }, { "id": "bip-342", "title": "BIP-342", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-342 specifies Tapscript, the updated Bitcoin scripting language that replaces legacy script in the script-path spending branch of Taproot (BIP-341) outputs. Tapscript updates the signature opcodes to use BIP-340 Schnorr signatures, introduces the OP_CHECKSIGADD opcode to enable efficient threshold multisignature constructions, relaxes script resource limits, and establishes an extensible opcode upgrade mechanism via OP_SUCCESS opcodes for future soft forks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bip-342", "labels": [ "BIP-342" ], "is_subclass_of": [ "Bitcoin Improvement Proposals" ], "wikilinks": [] }, { "id": "bip39", "title": "BIP-39", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BIP-39 is a Bitcoin Improvement Proposal that defines how to encode wallet entropy as a human-readable mnemonic seed phrase and how to derive a binary seed from it. It maps random entropy plus a checksum onto words drawn from a fixed wordlist, then stretches the phrase and an optional passphrase into a seed via a key derivation function. The resulting seed feeds hierarchical deterministic wallets, making secrets easier to back up and transcribe.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bip39", "labels": [ "BIP-39", "BIP39" ], "is_subclass_of": [ "Seed Phrase" ], "wikilinks": [] }, { "id": "bis-innovation-hub", "title": "BIS Innovation Hub", "domain": "finance", "domain_name": "Finance", "definition": "The innovation arm of the Bank for International Settlements, established to develop public goods for central banks and explore financial technology including CBDCs, cross-border payments, tokenisation, and regulatory technology.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bis-innovation-hub", "labels": [ "BIS Innovation Hub" ], "is_subclass_of": [ "BIS" ], "wikilinks": [ "Central Bank Digital Currency", "Wholesale CBDC", "Central Bank", "Hong Kong", "BIS" ] }, { "id": "bis", "title": "BIS", "domain": "finance", "domain_name": "Finance", "definition": "The Bank for International Settlements (BIS) is an international financial institution owned by 63 member central banks, founded in 1930 and headquartered in Basel, Switzerland. It acts as banker to central banks, providing settlement, custody, and asset management services while fostering monetary and financial cooperation through research, policy forums, and standard-setting committees. The BIS hosts the Basel Committee on Banking Supervision, the Financial Stability Board secretariat, and the Committee on Payments and Market Infrastructures, which together produce globally binding prudential and operational standards such as the Basel Accords and CPMI principles. Through the BIS Innovation Hub it actively researches central bank digital currencies, tokenisation of financial assets, and cyber resilience for financial market infrastructures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:bis", "labels": [ "BIS" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "blake2", "title": "BLAKE2", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BLAKE2 is a high-performance cryptographic hash function designed as a faster, simpler alternative to SHA-2 and SHA-3 while preserving equivalent security guarantees. Introduced in 2012 by Jean-Philippe Aumasson, Samuel Neves, Zooko Wilcox-O'Hearn, and Christian Winnerlein, it produces digests of up to 512 bits (BLAKE2b) or 256 bits (BLAKE2s) and is widely adopted in blockchain systems, password-hashing schemes, and data integrity verification due to its speed advantage over SHA-256 on modern 64-bit processors. Its resistance to length-extension attacks and configurability\u2014including personalisation, salting, and variable digest length\u2014make it a versatile drop-in for many cryptographic protocol requirements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:blake2", "labels": [ "BLAKE2" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "DID Nostr Identity", "SecurityLayer" ] }, { "id": "blake3", "title": "BLAKE3", "domain": "security", "domain_name": "Security", "definition": "BLAKE3 is a cryptographic hash function released in 2020 that achieves exceptional speed through a tree-hashing construction enabling unlimited parallelism across SIMD lanes and CPU cores, while simultaneously functioning as a keyed hash, a key derivation function, and an extendable-output function (XOF). It is derived from the BLAKE2 family, inheriting its ARX (add-rotate-XOR) ChaCha-based compression function, and extends it with a Bao-style binary tree that allows verified streaming and incremental hashing. BLAKE3 produces digests of arbitrary length (defaulting to 256 bits), is formally specified under a Creative Commons public-domain dedication, and is designed to be faster than SHA-256 on modern hardware by factors of five to ten on multi-core systems. Its unified API replaces the need for separate HMAC, HKDF, or KDF constructions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blake3", "labels": [ "BLAKE3" ], "is_subclass_of": [ "Cryptographic Hash Function" ], "wikilinks": [] }, { "id": "bleu-score", "title": "BLEU Score", "domain": "ai", "domain_name": "Ai", "definition": "BLEU Score (Bilingual Evaluation Understudy) is an automatic evaluation metric for machine translation and text generation quality that measures the overlap of n-gram sequences between a candidate output and one or more human reference translations, applying a brevity penalty to discourage pathologically short outputs. Scores range from 0 to 1 (or 0 to 100 in percentage form), with higher values indicating closer correspondence to the reference. BLEU correlates moderately with human judgement at the corpus level but is known to be unreliable for single-sentence evaluation and insufficient alone for capturing semantic adequacy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bleu-score", "labels": [ "BLEU Score", "BLEU" ], "is_subclass_of": [ "Evaluation Metric", "AI Technique", "Performance Metrics" ], "wikilinks": [] }, { "id": "blip-2-captioner", "title": "BLIP-2 Captioner", "domain": "ai", "domain_name": "Ai", "definition": "BLIP-2 is a vision-language model that bridges a frozen image encoder and a frozen large language model with a lightweight Querying Transformer (Q-Former), and a BLIP-2 captioner uses this model to generate natural-language descriptions of images. It produces high-quality captions efficiently because only the Q-Former is trained, leaving the heavy backbones fixed. Such captioners are commonly used to auto-label image datasets for training diffusion models and fine-tuning pipelines.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:blip-2-captioner", "labels": [ "BLIP-2 Captioner" ], "is_subclass_of": [ "Computer Vision", "Multimodal AI Architecture", "Multimodal Learning" ], "wikilinks": [ "Computer Vision", "Multimodal AI Architecture", "Large Language Model", "Vision Transformer", "CLIP Encoder", "Diffusion Model", "Fine Tuning", "Parameter-Efficient Fine-Tuning", "KOHYA Dreambooth and similar", "Image Captioning", "Visual Question Answering", "Contrastive Learning", "LoRA Fine-Tuning", "Stable Diffusion", "CLIP", "LoRA DoRA etc", "Multimodal Learning", "Zero-Shot Learning", "Causal Language Modelling", "Attention Mechanism" ] }, { "id": "bls-signature", "title": "BLS Signature", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A BLS Signature (Boneh\u2013Lynn\u2013Shacham signature) is a cryptographic signature scheme built on bilinear pairings over elliptic curves that permits multiple individual signatures to be aggregated into a single constant-size signature, which can be verified against the aggregate of the corresponding public keys in a single pairing operation. This aggregation property drastically reduces bandwidth and verification cost when many parties must co-sign a message or attest to a block.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bls-signature", "labels": [ "BLS Signature", "BLS Signature Aggregation" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "bm25", "title": "BM25", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "BM25 (Best Match 25) is a probabilistic bag-of-words ranking function used in information retrieval to score documents against a query based on term frequency saturation, document length normalisation, and inverse document frequency weighting. Derived from the BM family of retrieval functions developed at the Robertson-Sp\u00e4rck Jones framework in the 1970s-1990s, BM25 extends TF-IDF by applying a saturation parameter (k1) that prevents very high term frequencies from dominating relevance scores and a length normalisation parameter (b) that adjusts for document verbosity. It remains the dominant sparse lexical retrieval baseline in modern information retrieval systems.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:bm25", "labels": [ "BM25" ], "is_subclass_of": [ "Search Algorithm", "Information Retrieval", "Probabilistic Relevance Model", "Keyword Search" ], "wikilinks": [ "Information Retrieval", "Search Algorithm", "Inverted Index", "Semantic Search", "Dense Retrieval", "Hybrid Retrieval", "Retrieval Augmented Generation - RAG", "Enterprise Search", "Natural Language Processing", "Keyword Search", "Reciprocal Rank Fusion", "TF-IDF", "Tokenisation", "Embeddings", "Large Language Model", "Vector Database", "Knowledge Retrieval", "Cross-Encoder Reranking", "Document Retrieval", "Agentic RAG" ] }, { "id": "bnb-chain", "title": "BNB Chain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain compatible with the Ethereum Virtual Machine, operated with a delegated Proof-of-Staked-Authority validator set and used widely for decentralised trading, lending, token issuance, and other on-chain applications within the Binance ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bnb-chain", "labels": [ "BNB Chain" ], "is_subclass_of": [ "Blockchain", "Blockchain Network" ], "wikilinks": [ "EVM", "Proof of Stake", "Binance", "Smart Contract", "Blockchain" ] }, { "id": "bolt-specifications", "title": "BOLT Specifications", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The BOLT (Basis of Lightning Technology) specifications are the open, community-maintained set of documents defining the Lightning Network protocol, covering peer messaging, channel establishment and closure, commitment transactions, onion-routed payments, and invoice formats. By specifying the wire protocol and behaviour precisely, BOLTs ensure that independently developed Lightning implementations interoperate on a single payment network layered atop Bitcoin. They are the normative reference that makes the Lightning Network a multi-implementation open standard rather than a single product.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bolt-specifications", "labels": [ "BOLT Specifications", "BOLT Protocol", "BOLT Specification", "Lightning Network BOLT Specifications" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "bolt", "title": "BOLT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BOLT, the Basis of Lightning Technology, is the set of specifications that define how Lightning Network implementations interoperate. The documents cover channels, routing, and messaging.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bolt", "labels": [ "BOLT" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning", "BOLT11", "BOLT12", "Payment Channel", "Lightning Network", "https://github.com/lightning/bolts", "https://github.com/lightning/bolts/blob/master/00-introduction.md" ] }, { "id": "bolt-11", "title": "BOLT11", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BOLT11 is the Lightning Network specification for the invoice format used to request a single payment. It encodes payment amount, payment hash, description, expiry, and optional routing hints into a bech32-encoded string that the payer scans or pastes to initiate an off-chain payment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bolt-11", "labels": [ "BOLT11" ], "is_subclass_of": [ "BOLT" ], "wikilinks": [ "BOLT", "Lightning", "BOLT12", "https://github.com/lightning/bolts/blob/master/11-payment-encoding.md", "https://www.bolt11.org" ] }, { "id": "bolt12-offers", "title": "BOLT12 Offers", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BOLT12 Offers is a Lightning Network specification defining reusable, static payment requests called offers, improving on single-use BOLT11 invoices. An offer is a long-lived, shareable code from which payers fetch a fresh invoice on demand via onion messages, enabling recurring payments, refunds, and donation links without a server issuing each invoice. It also improves privacy through blinded paths that hide the recipient's node identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bolt12-offers", "labels": [ "BOLT12 Offers", "BOLT12" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [] }, { "id": "bolt-12", "title": "BOLT12", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BOLT12 is a Lightning Network specification that defines offers, a reusable payment request format that improves on single-use invoices. It supports recurring and static payment requests, onion messaging for invoice fetching, and improved privacy through blinded paths.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bolt-12", "labels": [ "BOLT12" ], "is_subclass_of": [ "BOLT" ], "wikilinks": [ "BOLT", "Phoenix", "BOLT11", "https://github.com/lightning/bolts/blob/master/12-offer-encoding.md", "https://bolt12.org" ] }, { "id": "brc-20", "title": "BRC-20", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BRC-20 is an experimental fungible token standard on the Bitcoin blockchain that encodes deploy, mint, and transfer operations as JSON-formatted Ordinals inscriptions written onto individual satoshis. Token balances are not enforced by Bitcoin consensus rules but are instead computed off-chain by indexers that parse inscription data in sequential ordinal order. Proposed by the pseudonymous developer @domo in March 2023, BRC-20 was the first widely adopted approach to creating transferable fungible tokens natively on Bitcoin without requiring a sidechain or layer-2 network. Its simplicity and permissionless nature drove rapid experimentation but also exposed limitations around scalability, indexer consensus, and on-chain fee pressure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:brc-20", "labels": [ "BRC-20" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [ "Ordinals", "Fungible Token", "Bitcoin", "Token Standard" ] }, { "id": "brdf", "title": "BRDF", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A Bidirectional Reflectance Distribution Function (BRDF) is a function that defines how light is reflected at an opaque surface, giving the ratio of reflected radiance to incident irradiance for each pair of incoming and outgoing directions. It encodes material appearance such as diffuse, glossy, and specular behaviour and is the core of physically based shading. Renderers evaluate the BRDF at surface intersections to compute realistic light transport.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:brdf", "labels": [ "BRDF", "BRDF Function" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "bsi-germany", "title": "BSI Germany", "domain": "security", "domain_name": "Security", "definition": "The Bundesamt f\u00fcr Sicherheit in der Informationstechnik (BSI), or Federal Office for Information Security, is Germany's national authority for cybersecurity, responsible for protecting digital infrastructure, certifying IT products, and developing security standards. It functions as a central advisory, regulatory, and technical body for both public administration and the private sector across Germany and the European Union.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bsi-germany", "labels": [ "BSI Germany", "BSI" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "btc-layer-3", "title": "BTC Layer 3", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Layer 3 (Bitcoin Layer 3) designates the application-layer stratum of the Bitcoin protocol stack \u2014 a collection of protocols, virtual machines, asset-issuance systems, and programmability frameworks constructed atop Layer 2 scaling networks (principally the Lightning Network and emerging", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:btc-layer-3", "labels": [ "BTC Layer 3" ], "is_subclass_of": [ "Protocol and Consensus", "Network Component", "Bitcoin Protocol Stack", "Application Layer Protocol", "Blockchain Scalability Solution", "Smart Contract Platform", "Distributed Ledger Technology" ], "wikilinks": [ "AI agent", "AI Agent Smart Contracts", "Alpen", "AluVM", "Application Layer Protocol", "Aptos", "Asigna", "Atomic Swaps", "Babylon", "Babylon Protocol", "Back 2014 Enabling Blockchain Innovations Pegged Sidechains", "Bifrost", "BIP-341 Taproot", "BIP-342 Tapscript", "BitFlow", "BitVM", "BitVM Alliance 2025 BitVM2 Bridging Bitcoin to Second Layers", "BitVM Bridge", "BitVM2", "Bitcoin DeFi" ] }, { "id": "btc", "title": "BTC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ticker for bitcoin, the native unit of the Bitcoin network and the first widely adopted decentralised digital currency.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:btc", "labels": [ "BTC" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [ "Bitcoin Network", "Bitcoin", "Store of Value", "Cryptocurrency" ] }, { "id": "bvh-acceleration-structure", "title": "BVH Acceleration Structure", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A bounding volume hierarchy (BVH) is a tree-based spatial data structure that recursively partitions scene geometry into nested bounding volumes to accelerate ray-geometry intersection queries. By testing rays against coarse parent volumes before descending into child nodes, a BVH reduces intersection complexity from linear in primitive count to roughly logarithmic. It is the dominant acceleration structure in modern ray tracing because it supports fast rebuilds for dynamic scenes and maps efficiently to GPU hardware.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bvh-acceleration-structure", "labels": [ "BVH Acceleration Structure" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "babylon-js", "title": "Babylon Js", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Babylon.js is an open-source, JavaScript and TypeScript real-time 3D rendering engine for the web that drives interactive graphics, games, and immersive experiences through WebGL and WebGPU backends. It provides a scene graph, physically based rendering, a node-based material system, glTF asset loading, an animation system, physics integration, and first-class WebXR support for virtual and augmented reality. As a spatial-computing framework it is a common foundation for browser-delivered metaverse and 3D applications, abstracting low-level GPU programming behind a high-level scene and entity API.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:babylon-js", "labels": [ "Babylon Js", "Babylon.js" ], "is_subclass_of": [ "Rendering Engine" ], "wikilinks": [] }, { "id": "back-translation", "title": "Back-Translation", "domain": "ai", "domain_name": "Ai", "definition": "Back-translation is a data augmentation and quality-assurance technique in machine translation where text translated into a target language is translated back into the source language. As an augmentation method, monolingual target-language data is translated into the source language to create synthetic parallel pairs that improve translation models. As a QA method, the round-trip output is compared with the original to detect meaning drift.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:back-translation", "labels": [ "Back-Translation" ], "is_subclass_of": [ "Natural Language Processing", "Data Augmentation" ], "wikilinks": [ "Machine Translation", "Neural Machine Translation", "Natural Language Processing", "Data Augmentation", "Transformer Architecture", "Low-Resource Language", "Parallel Corpus", "Monolingual Data", "BLEU Score", "Sequence-to-Sequence Model", "Encoder Decoder Architecture", "Attention Mechanism", "Large Language Models", "Quality Estimation", "Semi-Supervised Learning", "Synthetic Data", "Transfer Learning", "Language Model", "Domain Adaptation", "Text Generation" ] }, { "id": "backdoor-attack", "title": "Backdoor Attack", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A training-time adversarial attack that embeds a hidden trigger pattern into an AI model via data poisoning, causing the model to behave normally on standard inputs but produce attacker-chosen outputs when the trigger is present, creating a covert vulnerability that survives fine-tuning and is exploitable post-deployment.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:backdoor-attack", "labels": [ "Backdoor Attack" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "backdrivability", "title": "Backdrivability", "domain": "robotics", "domain_name": "Robotics", "definition": "Backdrivability is the property of a mechanical actuator or transmission that allows external forces applied at the output to drive the input in reverse, so motion and force pass freely in both directions. A highly backdrivable joint exhibits low reflected inertia and friction, enabling it to comply with and sense external contact. It is a key enabler of safe, compliant physical human-robot interaction and force-controlled robots.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:backdrivability", "labels": [ "Backdrivability" ], "is_subclass_of": [ "Actuator" ], "wikilinks": [] }, { "id": "backlash", "title": "Backlash", "domain": "robotics", "domain_name": "Robotics", "definition": "Backlash - The non-linear loss of motion in a mechanical transmission system caused by gaps, tolerances, or wear between gears, joints, or actuators, compromising Precision, Repeatability, and Accuracy in robotic manipulation.", "entityType": "Class", "qualityScore": 0.48, "maturity": "draft", "iri": "urn:ngm:class:backlash", "labels": [ "Backlash" ], "is_subclass_of": [ "Actuation and Control", "Robot Dynamics", "Robotics" ], "wikilinks": [ "Backlash Measurement", "Control Algorithm Tuning", "GDPR", "ISO 8373:2021", "Joint Mechanics", "Mechanical Compensation", "Performance Degradation Detection", "Repeatability", "Accuracy", "Precision", "Robot Dynamics", "Robotics", "RoboticsDomain" ] }, { "id": "backpropagation-through-time", "title": "Backpropagation Through Time", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Backpropagation through time (BPTT) is the extension of the backpropagation algorithm used to train recurrent neural networks, in which the recurrent network is unrolled across time steps into an equivalent feedforward network so that gradients of the loss can be computed and propagated backward through every step of the sequence. Because gradients must flow through many time steps, BPTT is prone to vanishing and exploding gradients on long sequences, which motivated architectures such as LSTM that mitigate the problem through gating. Truncated BPTT, which limits the number of steps backpropagated, is commonly used in practice to control computational cost on long sequences.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:backpropagation-through-time", "labels": [ "Backpropagation Through Time" ], "is_subclass_of": [ "Backpropagation" ], "wikilinks": [] }, { "id": "backpropagation", "title": "Backpropagation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Backpropagation is the algorithm for computing gradients of the loss function with respect to each weight in a neural network by applying the chain rule of calculus in reverse through the computation graph. It enables efficient gradient calculation across all layers in a single backward pass, making large-scale neural network training computationally feasible.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:backpropagation", "labels": [ "Backpropagation", "Backpropagation Update", "Neural Network Backpropagation" ], "is_subclass_of": [ "Optimization Algorithm" ], "wikilinks": [ "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "NIST (National Institute of Standards and Technology)", "ArtificialIntelligenceDomain" ] }, { "id": "backtracking-search", "title": "Backtracking Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A depth-first search strategy for combinatorial problems that incrementally extends a partial assignment one variable at a time, checks it against the problem's constraints, and on any violation abandons the current branch by undoing the most recent choice and trying an alternative. Because it prunes every extension of an inconsistent partial assignment, it explores a small fraction of the full assignment space, and with constraint propagation and intelligent variable ordering it forms the core complete algorithm of constraint solvers and SAT solvers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:backtracking-search", "labels": [ "Backtracking Search" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [ "Search Algorithm", "Constraint Solver", "Constraint Propagation", "Depth-First Search", "Constraint Satisfaction" ] }, { "id": "backup-and-recovery", "title": "Backup and Recovery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Backup and Recovery is the set of processes, technologies, and policies that ensure data and system state can be copied to a secondary store and restored to a known-good condition following data loss, corruption, or infrastructure failure. It encompasses full, incremental, and differential backup strategies alongside recovery time objectives (RTO) and recovery point objectives (RPO) that define acceptable loss and restoration windows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:backup-and-recovery", "labels": [ "Backup and Recovery", "Backup Strategy", "Backup Systems", "Data Backup" ], "is_subclass_of": [ "Disaster Recovery" ], "wikilinks": [] }, { "id": "backward-chaining", "title": "Backward Chaining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A goal-driven inference strategy that starts from a hypothesis to be proved and works backwards through the rule base, decomposing each goal into the subgoals given by the antecedents of rules whose consequents match it, recursing until every subgoal is grounded in known facts or fails. It is the query-answering counterpart to forward chaining, the evaluation strategy underlying Prolog's SLD resolution, and the diagnostic engine of classic expert systems such as MYCIN.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:backward-chaining", "labels": [ "Backward Chaining" ], "is_subclass_of": [ "Inference" ], "wikilinks": [ "Inference", "Rule-Based Systems", "Forward Chaining", "Expert Systems", "Inference Engine" ] }, { "id": "backward-compatibility", "title": "Backward Compatibility", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Backward compatibility is the property of a system, interface, or data format that allows newer versions to continue working correctly with inputs, clients, or data produced for older versions. It ensures that existing consumers do not break when a producer is upgraded, preserving established contracts while permitting evolution. Maintaining backward compatibility typically requires additive, non-breaking changes and careful deprecation policies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:backward-compatibility", "labels": [ "Backward Compatibility" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "bagging", "title": "Bagging", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bagging is a artificial intelligence concept and a type of Ensemble Mods. that enables Parallel Training.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:bagging", "labels": [ "Bagging" ], "is_subclass_of": [ "AI Technique", "Ensemble Methods" ], "wikilinks": [ "Parallel Training", "Ensemble Methods" ] }, { "id": "balance-of-payments", "title": "Balance of Payments", "domain": "finance", "domain_name": "Finance", "definition": "The balance of payments is a systematic record of all economic transactions between a country's residents and the rest of the world over a given period, comprising the current account (trade in goods, services, and income) and the capital and financial accounts (investment and asset flows). It is a core macroeconomic accounting framework used to assess a country's external economic position and currency stability. Historically it was central to the gold standard's adjustment mechanism, where payment imbalances triggered gold flows that altered domestic money supply.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:balance-of-payments", "labels": [ "Balance of Payments" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "balancer", "title": "Balancer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Balancer is a decentralised automated market maker (AMM) protocol on Ethereum and compatible EVM chains that generalises the constant-product AMM model to support weighted multi-asset pools, where pool weights can be set arbitrarily (e.g., 80/20 or 60/20/20 distributions) rather than the 50/50 split of Uniswap. It functions simultaneously as a self-rebalancing portfolio manager, a liquidity provider, and a price sensor.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:balancer", "labels": [ "Balancer" ], "is_subclass_of": [ "Automated Market Maker" ], "wikilinks": [] }, { "id": "ballistic-coefficient", "title": "Ballistic Coefficient", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ballistic-coefficient", "labels": [ "Ballistic Coefficient" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bandwidth-adaptation", "title": "Bandwidth Adaptation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Bandwidth adaptation is the real-time, algorithmic process by which a media transmission system continuously measures available network capacity and dynamically adjusts encoding parameters \u2014 bitrate, resolution, frame rate, codec profile, and layer selection \u2014 to maintain the highest achievable [...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bandwidth-adaptation", "labels": [ "Bandwidth Adaptation" ], "is_subclass_of": [ "Protocol and Infrastructure", "Network Quality-of-Service", "Real-time Transport Optimisation", "Adaptive Bitrate Streaming", "Congestion Control", "Quality of Experience" ], "wikilinks": [ "3GPP 5G Standards", "5G Networks", "ABR Algorithm", "Adaptive Bitrate Streaming", "Apple HLS RFC 8216", "AR VR Streaming", "Bandwidth Estimator", "BBR Congestion Control", "Bitrate Ladder", "BOLA Algorithm", "Buffer Manager", "Buffering-Only Adaptation", "CDN Infrastructure", "Circuit Switching", "Cloud Gaming", "Codec Ecosystem", "CodecLayer", "Codec Parameter Controller", "Congestion Control", "Congestion Signal" ] }, { "id": "bandwidth-optimization", "title": "Bandwidth Optimization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Bandwidth Optimisation is the set of techniques and algorithms that maximise the effective throughput of network links by reducing unnecessary data transmission, prioritising critical traffic, and intelligently managing congestion. It encompasses data compression, deduplication, caching, traffic shaping, protocol selection, and adaptive bitrate strategies applied across local, wide-area, and wireless network infrastructures.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bandwidth-optimization", "labels": [ "Bandwidth Optimization" ], "is_subclass_of": [ "Quality Of Service" ], "wikilinks": [] }, { "id": "bandwidth", "title": "Bandwidth", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The maximum rate at which data can be transferred over a network path or communication channel, commonly measured in bits per second. In signal processing it also refers to the range of frequencies a channel can carry.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bandwidth", "labels": [ "Bandwidth", "Bandwidth Constraints", "Bandwidth Management" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Video Streaming", "Latency", "Network Protocol" ] }, { "id": "bank-secrecy-act", "title": "Bank Secrecy Act", "domain": "governance", "domain_name": "Governance", "definition": "The Bank Secrecy Act (BSA), enacted in the United States in 1970, is the primary federal anti-money-laundering statute that requires financial institutions to assist US government agencies in detecting and preventing money laundering and financial crime. It mandates customer identification, suspicious activity reporting (SARs), currency transaction reporting (CTRs), and record-keeping \u2014 establishing the foundational framework for AML compliance in the US financial system.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:bank-secrecy-act", "labels": [ "Bank Secrecy Act", "Bank Secrecy Act 1970" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "bank-for-international-settlements", "title": "Bank for International Settlements", "domain": "finance", "domain_name": "Finance", "definition": "The Bank for International Settlements (BIS) is an international financial institution, established in 1930 and headquartered in Basel, that serves as a bank for central banks and fosters international monetary and financial cooperation. It provides banking services to its central bank members, hosts standard-setting bodies such as the Basel Committee on Banking Supervision, and conducts economic research on global financial stability. Through its Innovation Hub it also coordinates experimentation with central bank digital currencies and the modernisation of cross-border payments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:bank-for-international-settlements", "labels": [ "Bank for International Settlements" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [] }, { "id": "bank-of-england", "title": "Bank of England", "domain": "finance", "domain_name": "Finance", "definition": "The Bank of England (BoE) is the United Kingdom's central bank, founded in 1694 by Royal Charter to finance the Crown, and nationalised in 1946. It is responsible for monetary policy via the Monetary Policy Committee (MPC), which sets Bank Rate to maintain a 2% CPI inflation target; macroprudential financial stability policy via the Financial Policy Committee (FPC); and microprudential supervision of deposit-taking institutions, insurers, and systemically important financial market infrastructures via the Prudential Regulation Authority (PRA). The Bank also operates the UK's Real-Time Gross Settlement (RTGS) payment infrastructure and issues sterling banknotes, acting as the operational backbone of the UK financial system.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bank-of-england", "labels": [ "Bank of England" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [] }, { "id": "banking-regulation", "title": "Banking Regulation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Banking Regulation encompasses the body of laws, rules, and supervisory frameworks that govern the formation, operation, and dissolution of banks and other deposit-taking institutions. Its primary purposes are to ensure systemic financial stability, protect depositors, prevent money laundering, and enforce prudential capital and liquidity requirements. Key regulatory regimes include Basel III, the EU's CRD/CRR framework, and the US Dodd-Frank Act.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:banking-regulation", "labels": [ "Banking Regulation" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "banking-system", "title": "Banking System", "domain": "finance", "domain_name": "Finance", "definition": "An organised network of financial institutions, regulatory bodies, payment infrastructure, and legal frameworks that collectively facilitate the creation, custody, transfer, and lending of money within an economy. A banking system includes central banks, commercial banks, investment banks, and clearing houses, all operating under a body of prudential regulation. It underpins economic activity by providing mechanisms for capital allocation, risk distribution, and monetary policy transmission. Modern banking systems increasingly integrate digital and real-time payment rails.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:banking-system", "labels": [ "Banking System" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "banking", "title": "Banking", "domain": "finance", "domain_name": "Finance", "definition": "Banking is the system of financial intermediation through which licensed institutions accept deposits, extend credit, and provide payment and related services to individuals, businesses, and governments. Banks transform short-term deposits into longer-term loans, manage liquidity and risk, and operate the payment rails that move money through the economy. As a regulated, systemically important sector, banking sits at the centre of monetary policy transmission and financial stability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:banking", "labels": [ "Banking" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "barcode", "title": "Barcode", "domain": "standards", "domain_name": "Standards", "definition": "A barcode is a machine-readable optical representation of data that encodes information in the widths and spacings of parallel lines (one-dimensional) or in patterns of dots, squares and other geometric shapes (two-dimensional). Barcodes are scanned by optical readers or cameras to retrieve an identifier that links a physical item to a record in an information system. They underpin automatic identification and data capture in retail, logistics and inventory control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:barcode", "labels": [ "Barcode" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "barter-system", "title": "Barter System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A decentralized exchange mechanism enabling peer-to-peer trading of goods, services, or digital assets without monetary intermediaries, enhanced in digital contexts through blockchain-based matching algorithms and smart contracts that solve the traditional \"double coincidence of wants\" problem.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:barter-system", "labels": [ "Barter System" ], "is_subclass_of": [ "AI Governance and Ethics", "Exchange Mechanism" ], "wikilinks": [ "Asset Exchange", "Matching Algorithm", "Peer-to-Peer Trading", "Trust Mechanism", "Exchange Mechanism", "metaverse", "Smart Contracts", "Virtual Economy" ] }, { "id": "barycentric-dynamical-time", "title": "Barycentric Dynamical Time", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:barycentric-dynamical-time", "labels": [ "Barycentric Dynamical Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "base-fee", "title": "Base Fee", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Base Fee is the algorithmically determined minimum fee per unit of gas that every transaction must pay to be included in an Ethereum block, introduced by EIP-1559. It adjusts automatically each block based on whether the previous block was above or below its gas target, increasing when demand is high and decreasing when low. Unlike miner tips, the base fee is burned rather than paid to validators, creating a deflationary pressure on ETH supply and making gas price prediction more reliable for users.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:base-fee", "labels": [ "Base Fee" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Economic Mechanism", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain", "Virtual Economy" ] }, { "id": "base-learner", "title": "Base Learner", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A base learner is an individual model trained as one member of an ensemble, whose predictions are combined with those of other base learners to produce a final output. Base learners are typically simple or weak relative to the ensemble as a whole - for example shallow decision trees in a random forest or boosting sequence - so that their errors are diverse and can partially cancel when aggregated. The choice of base learner architecture and training procedure directly affects the diversity and accuracy of the resulting ensemble.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:base-learner", "labels": [ "Base Learner" ], "is_subclass_of": [ "Machine Learning Model" ], "wikilinks": [] }, { "id": "base", "title": "Base", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum Layer 2 network built by Coinbase using the Optimism OP Stack, providing low-cost, EVM-compatible transactions settled on and secured by the Ethereum mainnet.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:base", "labels": [ "Base", "Base Chain", "Base Network" ], "is_subclass_of": [ "Layer 2 Scaling" ], "wikilinks": [ "Ethereum", "Optimism", "DeFi", "Coinbase", "Layer 2 Scaling" ] }, { "id": "base64url-encoding", "title": "Base64URL Encoding", "domain": "security", "domain_name": "Security", "definition": "Base64URL encoding is a variant of Base64 that replaces the characters plus and slash with hyphen and underscore and typically omits padding, so that encoded data can be safely embedded in URLs and filenames without escaping. It is the encoding used for the header, payload and signature segments of a JSON Web Token, and for the compact serialisation used by SD-JWT. Its safety for use in URL query parameters and path segments is the reason it was chosen over standard Base64 for these token formats.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:base64url-encoding", "labels": [ "Base64URL Encoding", "Base64url Encoding" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "basel-accords", "title": "Basel Accords", "domain": "finance", "domain_name": "Finance", "definition": "The Basel Accords are a series of international banking-supervision agreements developed by the Basel Committee on Banking Supervision under the Bank for International Settlements, establishing minimum standards for bank capital, leverage, and liquidity. Successive accords progressively strengthened requirements for risk-weighted capital, introduced leverage and liquidity ratios, and refined the treatment of credit, market, and operational risk. They are non-binding recommendations that member jurisdictions transpose into national law to enhance the resilience and stability of the global financial system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:basel-accords", "labels": [ "Basel Accords" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "basel-committee-on-banking-supervision", "title": "Basel Committee on Banking Supervision", "domain": "finance", "domain_name": "Finance", "definition": "The Basel Committee on Banking Supervision (BCBS) is the primary global standard-setter for the prudential regulation of banks, operating under the auspices of the Bank for International Settlements in Basel, Switzerland. It develops minimum capital, liquidity, and leverage requirements \u2014 collectively the Basel Accords (Basel I, II, III, and the finalised Basel III framework completed in 2017) \u2014 that are adopted into national law by its 45-member jurisdictions. The Committee does not possess formal supranational authority; compliance is achieved through voluntary adoption by member central banks and supervisory agencies.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:basel-committee-on-banking-supervision", "labels": [ "Basel Committee on Banking Supervision", "Banking Supervision" ], "is_subclass_of": [ "Regulatory Authority" ], "wikilinks": [] }, { "id": "basel-committee", "title": "Basel Committee", "domain": "finance", "domain_name": "Finance", "definition": "The Basel Committee on Banking Supervision (BCBS) is an international standard-setting body hosted by the Bank for International Settlements (BIS) in Basel, Switzerland, that develops global regulatory standards for the prudential supervision of banks. Its accords \u2014 Basel I (1988), Basel II (2004), and Basel III (2010, finalised 2017) \u2014 set internationally agreed minimum capital adequacy, liquidity, and leverage requirements for commercial banks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:basel-committee", "labels": [ "Basel Committee" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "basel-ii", "title": "Basel II", "domain": "finance", "domain_name": "Finance", "definition": "Basel II is the second of the Basel Accords, issued by the Basel Committee on Banking Supervision in 2004 to refine the international framework for bank capital adequacy. It organised prudential regulation around three pillars: minimum capital requirements made more risk-sensitive through standardised and internal-ratings-based approaches to credit risk plus explicit treatment of operational risk; supervisory review of institutions' own capital adequacy assessments; and market discipline through enhanced disclosure. Distinct from both its predecessor Basel I and its successor Basel III, Basel II is a specific accord version; the 2007-2009 financial crisis exposed its underweighting of liquidity and leverage, prompting the tighter Basel III reforms that superseded and contrast with it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:basel-ii", "labels": [ "Basel II" ], "is_subclass_of": [ "Basel Accords" ], "wikilinks": [ "Basel Accords", "Basel III", "Basel Committee" ] }, { "id": "basel-iii", "title": "Basel III", "domain": "finance", "domain_name": "Finance", "definition": "Basel III is an international set of banking regulatory standards on capital adequacy, stress-tested liquidity buffers, and leverage limits, developed by the Basel Committee on Banking Supervision in response to the 2007-2009 global financial crisis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:basel-iii", "labels": [ "Basel III", "BIS Basel III CRE20", "Basel III Compliance", "Basel III Crypto Exposure Rules", "Basel III Standards" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Regulatory Frameworks", "Financial Stability", "Risk Management", "Financial Regulation", "https://www.bis.org/bcbs/basel3.htm", "https://www.bis.org/bcbs/" ] }, { "id": "batch-image-processing", "title": "Batch Image Processing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Batch image processing is the automated application of a fixed sequence of image operations, such as resizing, format conversion, colour correction, or generative transformation, across a large set of images without per-image manual intervention. It is exposed by image-generation platforms and pipeline tools as queued or scripted workflows that process many inputs sequentially or in parallel under a shared configuration. Batch processing trades per-image customisation for throughput, making it suited to dataset preparation and bulk content production.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:batch-image-processing", "labels": [ "Batch Image Processing" ], "is_subclass_of": [ "Image Processing" ], "wikilinks": [] }, { "id": "batch-inference", "title": "Batch Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The process of applying a trained machine learning model to a collection of inputs simultaneously rather than processing each input individually in real time. Batch inference amortises per-request overhead by grouping inputs into tensors that saturate GPU or accelerator memory bandwidth, significantly reducing per-sample latency at scale. It is the dominant serving pattern for offline analytics, embedding generation, and large-scale data enrichment pipelines where latency deadlines are relaxed. Contrast with online inference, which prioritises low single-request latency over throughput.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:batch-inference", "labels": [ "Batch Inference" ], "is_subclass_of": [ "Machine Learning", "Model Serving", "AI Infrastructure" ], "wikilinks": [ "Machine Learning", "Model Serving", "Neural Network", "GPU Computing", "Tensor Processing", "Embedding Generation", "Data Pipeline", "MLOps", "Online Inference", "Streaming Inference", "Throughput Optimisation", "Model Quantisation", "Distributed Training", "Machine Learning Infrastructure", "Scalability", "Inference Pipeline", "Deep Learning Framework", "Distributed Inference", "Flash Attention", "Large Language Models" ] }, { "id": "batch-normalisation", "title": "Batch Normalisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A technique that normalises layer inputs within a mini-batch to zero mean and unit variance, stabilising training dynamics, enabling higher learning rates, and acting as a form of regularisation in deep neural networks. Introduced by Ioffe and Szegedy (2015), it reduces internal covariate shift and has become a standard component in convolutional and other deep learning architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:batch-normalisation", "labels": [ "Batch Normalisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ISO (International Organization for Standardization)", "NIST (National Institute of Standards and Technology)", "ArtificialIntelligenceDomain" ] }, { "id": "batch-processing", "title": "Batch Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A computational paradigm in which jobs are accumulated and executed as a group rather than individually in real-time. Batch processing optimises throughput by amortising fixed overhead across many records, enabling efficient ETL pipelines, model training over large datasets, report generation, and vulnerability scanning. Scheduling may be time-based, event-triggered, or dependency-driven.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:batch-processing", "labels": [ "Batch Processing", "Batch Data Processing", "Batch Request" ], "is_subclass_of": [ "Data Pipeline" ], "wikilinks": [ "Core Technology", "Processing Model" ] }, { "id": "batch-size", "title": "Batch Size", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Batch Size is the number of training examples processed together in a single forward and backward pass before model parameters are updated. It is a critical hyperparameter governing the trade-off between training speed, memory usage, gradient noise, and convergence stability\u2014small batches introduce regularising noise via stochastic gradient estimates, whilst large batches enable faster hardware utilisation but may generalise less well.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:batch-size", "labels": [ "Batch Size" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ISO (International Organization for Standardization)", "NIST (National Institute of Standards and Technology)", "ArtificialIntelligenceDomain" ] }, { "id": "batch-verification", "title": "Batch Verification", "domain": "security", "domain_name": "Security", "definition": "Batch verification is a cryptographic technique that checks the validity of many signatures or proofs together using a single, combined computation, rather than verifying each one independently. It exploits algebraic structure in schemes such as Schnorr signatures and Bulletproofs to amortise the cost of verification across a batch, giving large speed-ups when many proofs must be checked, for example during blockchain block validation. The trade-off is that a failing batch identifies only that some element is invalid, not which one, without additional bisection.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:batch-verification", "labels": [ "Batch Verification" ], "is_subclass_of": [ "Cryptographic Verification" ], "wikilinks": [] }, { "id": "batching", "title": "Batching", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Batching is the practice of grouping multiple discrete operations, requests or data items so that they are processed together in a single pass rather than individually. It amortises fixed per-operation overheads such as draw calls, network round trips or kernel launches across many items, improving throughput at the cost of added latency for the items that wait to be grouped. Batching appears throughout computing, from GPU rendering (combining draw calls to reduce state changes) to machine learning inference (grouping requests to maximise accelerator utilisation).", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:batching", "labels": [ "Batching" ], "is_subclass_of": [ "Batch Processing" ], "wikilinks": [] }, { "id": "bathymetry", "title": "Bathymetry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bathymetry", "labels": [ "Bathymetry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "battery-management-system", "title": "Battery Management System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Battery Management System (BMS) is an embedded electronic system that monitors, protects, and optimises the operation of a rechargeable battery pack by continuously measuring cell voltages, currents, and temperatures, then enforcing safety limits and balancing cell state-of-charge to extend pack lifetime. It provides the digital intelligence layer between raw electrochemical energy storage and the broader power electronics or vehicle system, communicating pack status via standardised interfaces such as CAN bus. Accurate state-of-charge and state-of-health estimation algorithms are the defining computational challenge of modern BMS design.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:battery-management-system", "labels": [ "Battery Management System", "Battery Management" ], "is_subclass_of": [ "Embedded Systems" ], "wikilinks": [] }, { "id": "bayes-filter", "title": "Bayes Filter", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "BayesFilter denotes the canonical recursive probabilistic framework for sequential state estimation in partially-observable stochastic dynamical systems, computing at each timestep the posterior belief bel(x_t) = p(x_t | z_{1:t}, u_{1:t}) over the latent system state x_t conditioned on the full h...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bayes-filter", "labels": [ "Bayes Filter" ], "is_subclass_of": [ "AI Technique", "Recursive Estimator", "Bayesian Inference", "State Estimation", "Probabilistic Filter", "Sequential Monte Carlo" ], "wikilinks": [ "Active Perception", "Aerospace Guidance Navigation Control", "AlgorithmLayer", "Arasaratnam Haykin 2009 Cubature Kalman Filters", "Arulampalam Maskell Gordon Clapp 2002 Particle Filter Tutorial", "Autonomous Driving", "Bar-Shalom Li Kirubarajan 2001 Estimation with Tracking and Navigation", "Batch Optimisation", "Bayes Rule", "Bayesian Inference", "Belief-Space Planning", "Belief State", "Campos Elvira Gomez Rodriguez Montiel Tardos 2021 ORB-SLAM3", "Chapman-Kolmogorov Equation", "ControlAndEstimationDomain", "Control Inputs", "Davison Reid Molton Stasse 2007 MonoSLAM", "Dellaert Fox Burgard Thrun 1999 Monte Carlo Localization", "Deterministic Observer", "Doucet de Freitas Gordon 2001 Sequential Monte Carlo Methods" ] }, { "id": "bayes-theorem", "title": "Bayes Theorem", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bayes' theorem is a fundamental result in probability theory that describes how to update the probability of a hypothesis in light of new evidence. It expresses the posterior probability as proportional to the product of the prior probability and the likelihood of the evidence, normalised by the total probability of the evidence. The theorem is the mathematical foundation of Bayesian inference and probabilistic reasoning in artificial intelligence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:bayes-theorem", "labels": [ "Bayes Theorem", "Bayes' Theorem" ], "is_subclass_of": [ "Probability Theory", "Statistical Inference" ], "wikilinks": [ "Probability Theory", "Bayesian Inference", "Statistical Inference", "Probabilistic Model", "Prior Distribution", "Posterior Distribution", "Likelihood Function", "Conditional Probability", "Machine Learning", "Naive Bayes Classifier", "Bayesian Network", "Gaussian Process", "Markov Chain Monte Carlo", "Variational Inference", "Uncertainty Quantification", "Decision Theory", "Bayesian Decision Theory", "Bayesian Optimisation", "Bayesian Deep Learning", "Natural Language Processing" ] }, { "id": "bayesian-decision-theory", "title": "Bayesian Decision Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Bayesian decision theory is a framework for making decisions under uncertainty by combining probabilities of outcomes with a loss or utility function to choose actions that minimise expected loss. It uses Bayesian updating to incorporate evidence.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bayesian-decision-theory", "labels": [ "Bayesian Decision Theory" ], "is_subclass_of": [ "Statistics", "Decision Theory", "Statistical Decision Theory" ], "wikilinks": [ "Probability Theory", "Statistics", "Machine Learning", "Bayes Theorem", "Bayesian Inference", "Decision Theory", "Loss Function", "Utility Function", "Posterior Distribution", "Prior Distribution", "Risk Minimisation", "Optimal Classifier", "Reinforcement Learning", "Markov Decision Process", "Active Learning", "Bayesian Optimisation", "Uncertainty Quantification", "Statistical Decision Theory", "Game Theory", "Operations Research" ] }, { "id": "bayesian-deep-learning", "title": "Bayesian Deep Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Bayesian deep learning combines neural networks with Bayesian inference to represent uncertainty over model parameters and predictions. It treats network weights as random variables with prior and posterior distributions rather than fixed point estimates, enabling calibrated uncertainty quantification over both model parameters and outputs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bayesian-deep-learning", "labels": [ "Bayesian Deep Learning" ], "is_subclass_of": [ "Deep Learning", "Bayesian Inference", "Probabilistic Machine Learning" ], "wikilinks": [ "Deep Learning", "Bayesian Inference", "Uncertainty Quantification", "Variational Inference", "Probabilistic Inference" ] }, { "id": "bayesian-inference", "title": "Bayesian Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bayesian Inference is a principled statistical framework that updates the probability of a hypothesis as new evidence is observed, using Bayes' theorem to combine a prior distribution over parameters with a likelihood function derived from data to yield a posterior distribution. Unlike frequentist methods, it treats probability as a degree of belief and propagates uncertainty through every step of reasoning, enabling calibrated predictions and natural model comparison via marginal likelihoods. In machine learning, it underpins probabilistic graphical models, Gaussian processes, Bayesian neural networks, variational inference, and Markov Chain Monte Carlo methods. Its capacity to incorporate domain knowledge through priors and to quantify epistemic uncertainty makes it foundational for safety-critical AI, active learning, and sequential decision-making.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bayesian-inference", "labels": [ "Bayesian Inference", "Sequential Bayesian Updating" ], "is_subclass_of": [ "AI Technique", "Probabilistic Inference", "Statistical Inference" ], "wikilinks": [] }, { "id": "bayesian-knowledge-tracing", "title": "Bayesian Knowledge Tracing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bayesian Knowledge Tracing (BKT) is a probabilistic modelling technique that estimates a learner's mastery of a skill over time by treating knowledge as a latent binary state inferred from a sequence of correct and incorrect responses. Using a hidden Markov model with parameters for prior knowledge, learning, guessing, and slipping, BKT updates the probability that a student has mastered each skill after every interaction. It is a cornerstone of intelligent tutoring systems and adaptive learning, enabling personalised pacing and content selection.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:bayesian-knowledge-tracing", "labels": [ "Bayesian Knowledge Tracing" ], "is_subclass_of": [ "Adaptive Learning", "Probabilistic Model", "Learner Model", "Machine Learning", "Bayesian Inference" ], "wikilinks": [ "Adaptive Learning", "Intelligent Tutoring System", "Machine Learning", "Personalised Learning", "Cognitive Science", "Hidden Markov Model", "Probabilistic Inference", "Bayesian Inference", "Educational Technology", "Item Response Theory", "Spaced Repetition", "Deep Knowledge Tracing", "Knowledge Component Model", "Mastery Learning", "Formative Assessment", "Learner Model", "Expectation-Maximisation", "Reinforcement Learning", "Neural Network", "Computerised Adaptive Testing" ] }, { "id": "bayesian-optimisation", "title": "Bayesian Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Bayesian Optimisation is a sequential, sample-efficient strategy for optimising expensive black-box objective functions by constructing a probabilistic surrogate model \u2014 most commonly a Gaussian Process \u2014 over the function's input space, then using an acquisition function to select the next evaluation point by trading off exploration of uncertain regions against exploitation of known optima. The method accumulates knowledge about the objective between evaluations, making it uniquely valuable when each function evaluation is computationally or financially costly. It maintains a posterior distribution over the objective that quantifies uncertainty and guides convergence to a global optimum with far fewer evaluations than grid search or random search. Applications span hyperparameter tuning, neural architecture search, drug discovery, materials science, and robotics controller design.", "entityType": "Class", "qualityScore": 0.93, "maturity": "established", "iri": "urn:ngm:class:bayesian-optimisation", "labels": [ "Bayesian Optimisation" ], "is_subclass_of": [ "Optimisation", "Probabilistic Model", "Machine Learning", "Bayesian Inference" ], "wikilinks": [ "Gaussian Process", "Acquisition Function", "Surrogate Model", "Bayesian Inference", "Probabilistic Model", "Kernel Function", "Hyperparameter Tuning", "Neural Architecture Search", "AutoML", "Automated Experiment Design", "Multi-Objective Optimisation", "Probabilistic Inference", "Machine Learning", "Deep Learning", "Drug Discovery", "Materials Science", "Grid Search", "Random Search", "Evolutionary Algorithm", "Gradient Descent" ] }, { "id": "beacon-chain", "title": "Beacon Chain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Beacon Chain is the proof-of-stake consensus backbone introduced to Ethereum that coordinates validators, manages staking, and finalises blocks. It organises time into slots and epochs, assigns block-proposal and attestation duties, and applies a finality gadget that locks in the canonical chain. The Beacon Chain decouples consensus from execution, providing the security and randomness that the broader sharded and execution layers build upon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:beacon-chain", "labels": [ "Beacon Chain" ], "is_subclass_of": [ "Consensus Layer" ], "wikilinks": [] }, { "id": "beam-footprint", "title": "Beam Footprint", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:beam-footprint", "labels": [ "Beam Footprint" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "beam-handover", "title": "Beam Handover", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:beam-handover", "labels": [ "Beam Handover" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "beam-search-decoding", "title": "Beam Search Decoding", "domain": "ai", "domain_name": "Ai", "definition": "Beam search decoding is the application of the beam search algorithm specifically to the inference phase of neural sequence-to-sequence and autoregressive language models, where token-by-token predictions are generated by retaining the k highest-scoring partial sequences at each step. It serves as the primary decoding strategy for tasks requiring high-fidelity, deterministic outputs such as translation, summarisation, and structured text generation.", "entityType": "Class", "qualityScore": 0.95, "maturity": "mature", "iri": "urn:ngm:class:beam-search-decoding", "labels": [ "Beam Search Decoding" ], "is_subclass_of": [ "Beam Search", "Natural Language Processing", "Sequence-to-Sequence Learning", "Autoregressive Model" ], "wikilinks": [ "Beam Search", "Large Language Models", "Encoder Decoder Architecture", "Machine Translation", "Text Generation", "Sampling", "Natural Language Processing", "Transformer Architecture", "Attention Mechanism", "Autoregressive Model", "Greedy Decoding", "Nucleus Sampling", "Text Summarisation", "Code Generation", "Sequence-to-Sequence Learning", "Speculative Decoding", "Constrained Decoding", "Language Model", "Natural Language Generation", "Log-Probability" ] }, { "id": "beam-search", "title": "Beam Search", "domain": "ai", "domain_name": "Ai", "definition": "Beam search is a heuristic search algorithm that explores a graph by expanding the most promising nodes within a fixed-width frontier, called the beam, at each step. In sequence generation tasks it retains the top-k candidate sequences at each decoding step rather than pursuing a single greedy choice, balancing exploration against computational cost.", "entityType": "Class", "qualityScore": 0.95, "maturity": "mature", "iri": "urn:ngm:class:beam-search", "labels": [ "Beam Search" ], "is_subclass_of": [ "Search Algorithm", "Heuristic Methods", "Natural Language Processing", "Sequence-to-Sequence Learning" ], "wikilinks": [ "Search Algorithm", "Greedy Decoding", "Machine Translation", "Text Generation", "Speech Recognition", "Natural Language Processing", "Large Language Models", "Encoder Decoder Architecture", "Sequence-to-Sequence Learning", "Language Model", "Heuristic Methods", "Probability Distribution", "Sequence Model", "Nucleus Sampling", "Monte Carlo Tree Search", "Breadth-First Search", "Beam Search Decoding", "Constrained Decoding", "Speculative Decoding", "Robotics" ] }, { "id": "beamforming", "title": "Beamforming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Beamforming is a signal processing technique that combines the outputs of an antenna or sensor array with calculated phase and amplitude weightings to steer transmission or reception toward a specific spatial direction. It increases signal gain toward intended targets while suppressing interference from other directions, without physically moving the array. Beamforming is central to modern wireless communication standards and to radar systems that must resolve direction as well as range.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:beamforming", "labels": [ "Beamforming" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "bearer-token", "title": "Bearer Token", "domain": "security", "domain_name": "Security", "definition": "A bearer token is a security credential that grants access to a protected resource to any party that presents it, without requiring the holder to prove possession of an associated cryptographic key. Commonly issued by authorisation servers and transmitted in an HTTP Authorization header, it is simple to use but must be protected in transit and at rest because anyone who obtains it can use it. Bearer tokens are central to modern API authorisation flows such as OAuth.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bearer-token", "labels": [ "Bearer Token" ], "is_subclass_of": [ "Token" ], "wikilinks": [] }, { "id": "behavior-analysis", "title": "Behavior Analysis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Behavior Analysis is an AI and machine learning technique concerned with the systematic observation, modelling, and interpretation of patterns in entity actions \u2014 whether human users, software agents, or autonomous systems \u2014 in order to detect anomalies, predict future actions, or classify intent. It combines statistical modelling, sequence analysis, and supervised or unsupervised learning to extract actionable intelligence from behavioural streams. Typical applications include cybersecurity threat detection, user experience optimisation, game AI design, and fraud prevention.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:behavior-analysis", "labels": [ "Behavior Analysis", "Behavioral Analysis", "User Behavior Analysis" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning" ] }, { "id": "behavioral-learning", "title": "Behavioral Learning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An AI-driven approach that analyzes and predicts user behavior patterns from digital interactions including clicks, browsing patterns, movement trajectories, and gaze tracking, enabling automated decision-making and personalized experiences through machine learning and pattern recognition.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:behavioral-learning", "labels": [ "Behavioral Learning" ], "is_subclass_of": [ "Content and Assets", "Machine Learning Discipline" ], "wikilinks": [ "Data Collection", "Personalization", "User Experience Optimization", "Machine Learning", "Machine Learning Models", "metaverse", "Pattern Recognition", "Predictive Analytics" ] }, { "id": "behavioral-modeling", "title": "Behavioral Modeling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Behavioral Modeling is the application of machine learning and statistical methods to represent, predict, and simulate how agents (humans, robots, or software entities) act under varying conditions. It underpins applications ranging from user-behaviour analytics and fraud detection to reinforcement learning policies and autonomous agent control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:behavioral-modeling", "labels": [ "Behavioral Modeling", "Behavioural Modelling" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique", "Predictive Analytics" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning" ] }, { "id": "behaviour-tree", "title": "Behaviour Tree", "domain": "ai", "domain_name": "Ai", "definition": "A Behaviour Tree (BT) is a hierarchical, directed acyclic graph structure used to model the decision-making logic of autonomous agents, non-player characters (NPCs), and robots. Internal nodes represent control-flow composites \u2014 Sequence, Selector, Parallel, and Decorator \u2014 while leaf nodes represent atomic Actions or Conditions; execution propagates through the tree and each node returns Success, Failure, or Running to its parent. Behaviour Trees superseded Finite State Machines (FSMs) in many game and robotics contexts because they offer superior modularity, reusability, and comprehensibility: sub-trees encapsulate coherent behaviours that can be composed without explicit inter-state transition wiring. First popularised in the game-development community circa 2005 and later formalised in robotics frameworks such as BehaviorTree.CPP, BTs are now a standard control architecture in both real-time interactive media and autonomous robotic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:behaviour-tree", "labels": [ "Behaviour Tree", "Behavior Tree", "Behaviour Tree Control", "Behaviour Tree Execution" ], "is_subclass_of": [ "Automated Planning", "Navigation and Planning", "Reactive System", "Control Architecture", "Agent Architecture", "Decision Making" ], "wikilinks": [] }, { "id": "behaviour-trees", "title": "Behaviour Trees", "domain": "robotics", "domain_name": "Robotics", "definition": "Behaviour trees are a model for organising the decision logic of autonomous agents and robots into a tree of tasks and control nodes. They are used in robotics and game artificial intelligence.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:behaviour-trees", "labels": [ "Behaviour Trees" ], "is_subclass_of": [ "Robot Control" ], "wikilinks": [ "Task and Motion Planning", "Behaviour Tree", "Robot Control", "https://en.wikipedia.org/wiki/Behavior_tree_(artificial_intelligence,_robotics_and_control)", "https://www.behaviortree.dev" ] }, { "id": "behavioural-analytics", "title": "Behavioural Analytics", "domain": "security", "domain_name": "Security", "definition": "Behavioural analytics is the security and analytics discipline that establishes baseline patterns of user, device and entity activity and detects meaningful deviations that may indicate threats, fraud or compromise. It applies statistical modelling and machine learning to telemetry such as login times, access sequences and transaction behaviour to surface anomalies that signature-based controls miss. The approach underpins user and entity behaviour analytics (UEBA) within modern security operations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:behavioural-analytics", "labels": [ "Behavioural Analytics", "Behavioral Analytics", "Behavioral Tracking" ], "is_subclass_of": [ "Security", "AI Technique", "Machine Learning Discipline" ], "wikilinks": [] }, { "id": "behavioural-economics", "title": "Behavioural Economics", "domain": "finance", "domain_name": "Finance", "definition": "Behavioural economics is an interdisciplinary field that integrates insights from cognitive psychology and social science into economic modelling, documenting systematic ways in which human decision-making deviates from the predictions of classical rational-choice theory. It explains phenomena such as loss aversion, anchoring, present bias, and herd behaviour that standard utility maximisation models cannot account for.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:behavioural-economics", "labels": [ "Behavioural Economics", "Behavioral Economics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "behavioural-feedback-loop", "title": "Behavioural Feedback Loop", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Recurring cycle where user actions influence environment responses which in turn modify subsequent user behavior through adaptive learning and reinforcement mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:behavioural-feedback-loop", "labels": [ "Behavioural Feedback Loop" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Action Detection", "Adaptive Experience", "Adaptive Reinforcement", "Affective Computing", "AI Model", "Dynamic Adjustment", "Environment Response", "IEEE Affective Systems", "Personalized Interaction", "State Management", "User Modeling", "User Tracking", "Behavior Analysis", "Behavioral Learning", "ComputationAndIntelligenceDomain", "ComputeLayer", "Machine Learning", "NetworkLayer", "Real-time Processing" ] }, { "id": "behaviourism", "title": "Behaviourism", "domain": "ai", "domain_name": "Ai", "definition": "A school of psychology and associated philosophy of mind holding that behaviour should be explained through observable stimulus\u2013response relationships and learning histories rather than by appeal to inner mental states. Dominant from Watson's 1913 manifesto through Skinner's radical behaviourism of the mid-twentieth century, it produced the experimental analysis of classical and operant conditioning, was displaced as a general paradigm by the cognitive revolution, and survives today in applied behaviour analysis and in reinforcement learning's core concepts of reward-driven learning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:behaviourism", "labels": [ "Behaviourism" ], "is_subclass_of": [ "Psychology" ], "wikilinks": [ "Psychology", "Cognitive Psychology", "Cognitive Science", "Philosophy of Mind" ] }, { "id": "belief-desire-intention", "title": "Belief Desire Intention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Belief-Desire-Intention (BDI) model is a framework for designing rational software agents in which mental states are represented as beliefs (what the agent knows about the world), desires (goals the agent wishes to achieve), and intentions (committed plans of action). Originating from Bratman's philosophical work on practical reasoning, BDI has been formalised into agent programming languages such as AgentSpeak and frameworks such as JADE. It underpins many autonomous and multi-agent systems requiring deliberative reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:belief-desire-intention", "labels": [ "Belief Desire Intention", "Belief-Desire-Intention" ], "is_subclass_of": [ "Artificial Intelligence", "Multi-Agent System", "Cognitive Architecture", "Agent Architecture", "Deliberative System", "Rational Agent Model" ], "wikilinks": [] }, { "id": "belief-propagation", "title": "Belief Propagation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Belief propagation is a message-passing algorithm for performing inference on graphical models by iteratively exchanging local messages between nodes representing variables and the factors that relate them. On tree-structured graphs it computes exact marginal distributions; on graphs with cycles, loopy belief propagation provides an approximate inference scheme that often works well in practice. It is the basis of efficient decoding for modern error-correcting codes and of probabilistic reasoning over structured domains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:belief-propagation", "labels": [ "Belief Propagation" ], "is_subclass_of": [ "Graphical Model", "Probabilistic Inference" ], "wikilinks": [] }, { "id": "bellman-equation", "title": "Bellman Equation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Bellman equation is a recursive relationship that expresses the value of a state as the immediate reward plus the discounted value of successor states under a given policy. It is the mathematical foundation of dynamic programming and reinforcement learning, characterising optimal behaviour through the principle of optimality. Solving or approximating the Bellman equation yields value functions and optimal policies for sequential decision problems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:bellman-equation", "labels": [ "Bellman Equation" ], "is_subclass_of": [ "Dynamic Programming" ], "wikilinks": [] }, { "id": "benchmark-dataset", "title": "Benchmark Dataset", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A benchmark dataset is a standardised, publicly shared collection of data with established splits and evaluation protocols, used to compare the performance of machine-learning models on a common task. By fixing the data, metrics, and procedure, benchmarks enable fair, reproducible comparison and track progress over time. Well-known benchmarks have driven advances in computer vision, natural language processing, and many other fields.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:benchmark-dataset", "labels": [ "Benchmark Dataset" ], "is_subclass_of": [ "Dataset", "Labelled Dataset" ], "wikilinks": [ "Dataset", "Model Evaluation", "Benchmarking", "Ground Truth Labels", "Reproducibility", "Data Annotation", "Data Curation", "Leaderboard", "Evaluation Metric", "Training Data", "Benchmark Evaluation", "Evaluation Harness", "Performance Benchmarks", "Image Classification", "Natural Language Processing", "Supervised Learning", "Transfer Learning", "Large Language Models", "Computer Vision", "Data Quality" ] }, { "id": "benchmark-evaluation", "title": "Benchmark Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "Benchmark evaluation is the systematic measurement of a model or system against a standardised dataset and scoring protocol so that results are comparable across systems and over time. A benchmark specifies the task, the data splits, the permitted inputs, and the metrics used to rank performance. In machine learning, benchmark evaluation drives the field's empirical progress, but it is subject to well-known failure modes including train-test contamination, overfitting to leaderboards, and construct validity gaps between the benchmark and the real-world capability it purports to measure.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:benchmark-evaluation", "labels": [ "Benchmark Evaluation", "Cityscapes Benchmark", "Evaluation Benchmark", "MLPerf Benchmark", "WMT Benchmark" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation", "Benchmarking" ], "wikilinks": [ "Model Evaluation", "Benchmark Dataset", "Evaluation Metric", "Evaluation Harness", "Performance Benchmarks", "Leaderboard", "Reproducibility", "Benchmarking", "Model Evaluation Results", "Benchmark Standard", "AI Benchmark Epistemological Critique", "Frontier Model Evaluation", "Capability Evaluation", "Human Evaluation", "Large Language Models", "Ground Truth Labels", "Natural Language Processing", "Data Annotation", "Overfitting", "Bias" ] }, { "id": "benchmark-standard", "title": "Benchmark Standard", "domain": "standards", "domain_name": "Standards", "definition": "A vendor-agnostic reference specification for evaluating and comparing system performance, establishing standardized metrics and methodologies that enable fair, repeatable, and meaningful comparisons across computing systems, processes, or technologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:benchmark-standard", "labels": [ "Benchmark Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Measurement Methodology", "Performance Comparison", "Reference Implementation", "System Evaluation", "metaverse", "Quality Assurance", "Technical Standard", "Validation Process" ] }, { "id": "benchmarking", "title": "Benchmarking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Benchmarking is the disciplined practice of measuring the performance, accuracy, efficiency, or quality of a system, component, or model against a standardised workload and a set of comparable baselines. It produces reproducible quantitative metrics \u2014 such as latency, throughput, resource utilisation, or task accuracy \u2014 that allow engineers to compare alternatives, detect regressions, and guide optimisation. Robust benchmarking requires controlled environments, representative workloads, statistically sound measurement, and transparent reporting to avoid misleading or non-generalisable results.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:benchmarking", "labels": [ "Benchmarking" ], "is_subclass_of": [ "Performance Optimization" ], "wikilinks": [] }, { "id": "benchmarks", "title": "Benchmarks", "domain": "ai", "domain_name": "Ai", "definition": "Benchmarks are standardised tasks, datasets, or workloads used to measure and compare the capabilities or performance of systems, models, or components under controlled, reproducible conditions. In artificial intelligence they encompass curated evaluation suites that probe language understanding, mathematical reasoning, coding, and multimodal perception to produce comparable scores across model generations and research groups. In computing, robotics, and hardware engineering, benchmarks quantify throughput, latency, accuracy, and energy efficiency against fixed reference workloads. Benchmark results are published via leaderboards and model cards to support reproducible science, informed procurement, and regulatory accountability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:benchmarks", "labels": [ "Benchmarks", "Benchmark", "SPEC Benchmarks" ], "is_subclass_of": [ "Evaluation benchmarks and leaderboards", "Performance Metrics", "Reproducibility" ], "wikilinks": [] }, { "id": "beneficial-ownership-disclosure", "title": "Beneficial Ownership Disclosure", "domain": "governance", "domain_name": "Governance", "definition": "Beneficial ownership disclosure is the regulatory requirement for legal entities such as companies, trusts, and partnerships to identify and register the natural persons who ultimately own or control them above a defined threshold, typically 25% of shares or voting rights. The primary purpose is to prevent the misuse of corporate structures for money laundering, tax evasion, corruption, and sanctions circumvention by making true ownership transparent to regulators and, increasingly, the public.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:beneficial-ownership-disclosure", "labels": [ "Beneficial Ownership Disclosure", "Beneficial Ownership Reporting" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "beneficial-ownership", "title": "Beneficial Ownership", "domain": "finance", "domain_name": "Finance", "definition": "Beneficial ownership identifies the natural persons who ultimately own or control a legal entity or arrangement, or on whose behalf a transaction is conducted, regardless of the formal legal title. It distinguishes the real human beneficiaries from nominee shareholders, trustees and layered corporate structures. Establishing beneficial ownership is a core obligation in anti-financial-crime regimes, enabling regulators and institutions to pierce opaque structures and attribute responsibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:beneficial-ownership", "labels": [ "Beneficial Ownership" ], "is_subclass_of": [ "Corporate Governance" ], "wikilinks": [] }, { "id": "benign-overfitting", "title": "Benign Overfitting", "domain": "ai", "domain_name": "Ai", "definition": "Benign overfitting is the phenomenon, observed in heavily overparameterised models such as deep neural networks, where a model fits its training data exactly, including noise, yet still generalises well to unseen data. This contradicts the classical bias-variance trade-off, which predicts that interpolating noise should harm generalisation. It is closely associated with the double-descent risk curve and is a central puzzle in modern statistical learning theory.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:benign-overfitting", "labels": [ "Benign Overfitting" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Research Area", "Statistical Learning Theory", "Generalisation Theory" ], "wikilinks": [] }, { "id": "bent-pipe-transponder", "title": "Bent-pipe Transponder", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bent-pipe-transponder", "labels": [ "Bent-pipe Transponder" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "best-practice", "title": "Best Practice", "domain": "standards", "domain_name": "Standards", "definition": "A method, technique, or configuration that a professional community has found through accumulated experience to reliably produce superior results, codified so that others can adopt it without rediscovering it. Best practices sit between raw craft knowledge and formal standards: they are advisory rather than mandatory, evolve faster than standardisation processes allow, and are the substance that bodies of knowledge, reference architectures, and accessibility or security guidelines distil and disseminate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:best-practice", "labels": [ "Best Practice" ], "is_subclass_of": [ "Knowledge Management" ], "wikilinks": [ "Knowledge Management", "Technical Standard", "Reference Architecture", "Governance Framework", "Interoperability" ] }, { "id": "beyond-low-earth-orbit-spacecraft", "title": "Beyond Low Earth Orbit Spacecraft", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:beyond-low-earth-orbit-spacecraft", "labels": [ "Beyond Low Earth Orbit Spacecraft" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "beyond-low-earth-orbit-spaceflight", "title": "Beyond Low Earth Orbit Spaceflight", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:beyond-low-earth-orbit-spaceflight", "labels": [ "Beyond Low Earth Orbit Spaceflight" ], "is_subclass_of": [ "Spaceflight" ], "wikilinks": [] }, { "id": "bfloat16", "title": "Bfloat16", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bfloat16 (brain floating point) is a 16-bit floating-point format that keeps the same 8-bit exponent as 32-bit IEEE float but truncates the mantissa to 7 bits. It preserves the dynamic range of single precision while halving memory and bandwidth, making it well suited to deep-learning training and inference. Because it trades precision for range, it avoids the overflow and underflow problems that affect narrower formats during gradient computation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bfloat16", "labels": [ "Bfloat16", "BFLOAT16", "BFloat16" ], "is_subclass_of": [ "Model Compression", "Mixed Precision Training" ], "wikilinks": [] }, { "id": "bi-elliptic-transfer", "title": "Bi-elliptic Transfer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bi-elliptic-transfer", "labels": [ "Bi-elliptic Transfer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bias-and-fairness", "title": "Bias And Fairness", "domain": "ai", "domain_name": "Ai", "definition": "Bias and Fairness is the area of responsible AI concerned with detecting, measuring, and mitigating systematic disparities in how machine learning systems treat individuals and groups. It addresses biases that enter through training data, model design, and deployment context, and it formalises competing notions of fairness such as demographic parity, equalised odds, and individual fairness. The goal is to ensure that automated decisions do not unjustifiably disadvantage protected populations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bias-and-fairness", "labels": [ "Bias And Fairness", "Bias and Fairness" ], "is_subclass_of": [ "Responsible AI", "AI Ethics", "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "bias-detection-methods", "title": "Bias Detection Methods", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bias Detection Methods are systematic analytical techniques for identifying algorithmic bias in AI systems through statistical hypothesis testing, fairness audits, counterfactual analysis, intersectional evaluation, and causal inference. These methods examine model predictions across protected demographic groups to detect disparate impacts, unequal error rates, and discriminatory patterns, producing bias audit reports that document severity, affected populations, and regulatory compliance status.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:bias-detection-methods", "labels": [ "Bias Detection Methods", "Bias Detection", "Bias Detection Metrics", "Bias Detection Protocol" ], "is_subclass_of": [ "AI Governance and Ethics", "Algorithmic Auditing", "Fairness Assessment" ], "wikilinks": [ "IEEE P7003-2021", "ISO/IEC TR 24027", "NIST SP 1270", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "bias-mitigation-techniques", "title": "Bias Mitigation Techniques", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Methods and interventions designed to reduce algorithmic bias and improve fairness in AI systems through modifications at pre-processing (data reweighting, resampling), in-processing (fairness constraints, adversarial debiasing), and post-processing (threshold optimisation) stages of the machine learning pipeline. Each approach involves tradeoffs between fairness improvement and predictive accuracy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bias-mitigation-techniques", "labels": [ "Bias Mitigation Techniques", "Bias Mitigation", "Bias Reduction" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics", "Responsible AI" ], "wikilinks": [ "AIF360", "Fairlearn", "IEEE P7003-2021", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "bias-in-ai", "title": "Bias in AI", "domain": "governance", "domain_name": "Governance", "definition": "The systematic and unfair skewing of artificial intelligence outputs against particular groups or outcomes, arising from unrepresentative or historically prejudiced training data, flawed problem framing, proxy variables, feedback loops, and deployment context. It manifests as measurable performance and treatment disparities \u2014 in face recognition, hiring, credit, healthcare and content moderation \u2014 and is addressed through bias auditing, fairness metrics, data curation, and governance obligations now codified in regulation such as the EU AI Act.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bias-in-ai", "labels": [ "Bias in AI" ], "is_subclass_of": [ "AI Ethics" ], "wikilinks": [ "AI Ethics", "Algorithmic Bias", "Responsible AI", "Synthetic Data", "Healthcare AI", "Content Moderation" ] }, { "id": "bias-in-large-language-models", "title": "Bias in Large Language Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Bias in Large Language Models is the systematic skew in the outputs, representations, and decisions of transformer-based foundation models (GPT-4o/4.5, Claude 3.5/4 Sonnet/Opus, Gemini 1.5/2.0 Pro, Llama 3/4, Mistral Large, Qwen 2.5, DeepSeek-V3) toward particular social groups, viewpoints, langu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bias-in-large-language-models", "labels": [ "Bias in Large Language Models" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Safety", "Sociotechnical Harm", "Algorithmic Bias", "Harmful Bias", "AI Risks", "Representational Harm" ], "wikilinks": [ "Abid et al. 2021 Anti-Muslim Bias", "ACL", "Age Bias", "AIES", "Allocational Bias", "Bai et al. 2022 Constitutional AI", "Barocas et al. 2017 Allocative vs Representational Harm", "BBQ Benchmark", "Bender et al. 2021 Stochastic Parrots", "Bias Auditing", "Blodgett et al. 2020 Language Technology is Power", "BOLD Benchmark", "Bolukbasi et al. 2016 Debiasing Word Embeddings", "Caliskan et al. 2017 WEAT Science", "Cao et al. 2023 Cross-Cultural Alignment ChatGPT", "Casper et al. 2023 Open Problems RLHF", "Common Crawl", "Confirmation Bias", "CrowS-Pairs", "Cultural Homogenization" ] }, { "id": "bias-variance-tradeoff", "title": "Bias-Variance Tradeoff", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The bias-variance tradeoff is the central principle in supervised learning describing how a model's expected generalisation error decomposes into bias, variance, and irreducible noise, and how reducing one component tends to increase the other. Bias is error from overly simplistic assumptions that cause systematic underfitting, while variance is error from excessive sensitivity to the training sample that causes overfitting. Effective model selection seeks a complexity sweet spot that minimises total expected error on unseen data, balancing these competing sources rather than eliminating either alone.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bias-variance-tradeoff", "labels": [ "Bias-Variance Tradeoff", "Bias Variance Tradeoff" ], "is_subclass_of": [ "Statistical Learning Theory", "Hyperparameter Tuning", "Machine Learning", "Model Evaluation" ], "wikilinks": [ "Supervised Learning", "Overfitting", "Underfitting", "Model Complexity", "Regularization", "Cross-Validation", "Ensemble Methods", "Model Evaluation", "Hyperparameter Tuning", "Statistical Learning Theory", "Random Forest", "Gradient Boosting", "Neural Network", "Deep Learning", "Double Descent", "Loss Function", "Generalisation Error", "Model Selection", "Learning Curve", "Bagging" ] }, { "id": "bias", "title": "Bias", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Systematic deviation from fairness, objectivity, or expected outcomes in an AI system that leads to prejudiced results favouring or disfavouring particular groups, individuals, or outcomes, arising from data, algorithms, or deployment contexts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:bias", "labels": [ "Bias", "Bias Amplification", "Bias Component" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "confusion matrices", "GDPR (General Data Protection Regulation)", "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "NIST (National Institute of Standards and Technology)", "AI Risks", "ArtificialIntelligenceDomain", "Bias in Large Language Models", "Deep Learning", "EU AI Act", "Large Language Models", "Machine Learning", "Safety and alignment", "Update Cycle" ] }, { "id": "bid-ask-spread", "title": "Bid Ask Spread", "domain": "finance", "domain_name": "Finance", "definition": "The bid-ask spread is the difference between the highest price a buyer is willing to pay (the bid) and the lowest price a seller will accept (the ask) for an asset at a given moment. It is a core measure of market liquidity and an implicit transaction cost borne by traders who cross the spread. Narrow spreads indicate liquid, competitive markets, while wide spreads reflect thin liquidity, volatility, or elevated risk for liquidity providers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bid-ask-spread", "labels": [ "Bid Ask Spread", "Bid-Ask Spread" ], "is_subclass_of": [ "Market Microstructure" ], "wikilinks": [] }, { "id": "bidirectional-reflectance-distribution-function", "title": "Bidirectional Reflectance Distribution Function", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bidirectional-reflectance-distribution-function", "labels": [ "Bidirectional Reflectance Distribution Function" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bifrost-protocol", "title": "Bifrost Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bifrost is a Bitcoin-focused protocol stack associated with RGB and client-side validation, providing transport and interoperability for off-chain smart-contract state. In the RGB ecosystem it handles the peer-to-peer exchange of consignments and validation data so that asset and contract state can move between parties without publishing it on-chain. By keeping contract data client-side, it preserves Bitcoin's scalability and privacy while enabling rich asset issuance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:bifrost-protocol", "labels": [ "Bifrost Protocol" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "big-data", "title": "Big Data", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Big data denotes datasets whose volume, velocity and variety exceed the capacity of conventional single-machine tools, demanding distributed storage and parallel computation. It is characterised by horizontally scalable architectures, schema-flexible stores, and batch or streaming processing frameworks that move computation to where data resides. The term also names the discipline of extracting value from such datasets through analytics, mining and machine learning at scale.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:big-data", "labels": [ "Big Data" ], "is_subclass_of": [ "Data Processing" ], "wikilinks": [] }, { "id": "bilinear-pairing", "title": "Bilinear Pairing", "domain": "security", "domain_name": "Security", "definition": "A bilinear pairing is a mathematical map between two groups to a third group that is linear in each argument separately, commonly realised as the Weil or Tate pairing on elliptic curves over finite fields. Bilinear pairings enable advanced cryptographic constructions including identity-based encryption, short signature schemes (BLS), and zero-knowledge proof systems. They form the algebraic foundation of pairing-based cryptography, which underpins threshold signatures and SNARKs used in modern blockchain protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bilinear-pairing", "labels": [ "Bilinear Pairing" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "bilingual-rag-hallucination-mitigation-framework", "title": "Bilingual RAG Hallucination Mitigation Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Jing is a multi-agent AI framework designed to minimise hallucination in Chinese-English bilingual retrieval-augmented generation (RAG) pipelines, specifically applied to emotional-support chatbots for end-of-life counselling and palliative care. The system combines multiple large language models with open-source text-to-speech engines and real-time 3D avatars to deliver culturally sensitive, affectively aware conversational companions, initially targeting Japanese VR screen deployment contexts. Its multi-agent architecture distributes validation, retrieval, and generation tasks across specialised model instances to improve factual grounding and reduce hallucinated responses in high-stakes care settings.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:bilingual-rag-hallucination-mitigation-framework", "labels": [ "Bilingual RAG Hallucination Mitigation Framework", "jing" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "binance", "title": "Binance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Binance is a centralised cryptocurrency exchange founded in 2017 by Changpeng Zhao (CZ) and Yi He, which rapidly grew to become one of the highest-volume digital-asset trading venues globally. It provides spot trading, perpetual and futures derivatives, margin lending, staking, savings products, and a token launchpad across hundreds of supported assets. Binance created the BNB token as a native utility and fee-discount instrument, and subsequently developed BNB Chain, an EVM-compatible Layer-1 blockchain that hosts a substantial decentralised-finance and Web3 application ecosystem. The exchange has faced extensive regulatory scrutiny across multiple jurisdictions, including a landmark 2023 settlement with the United States Department of Justice and FinCEN, resulting in substantial financial penalties and a change in executive leadership.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:binance", "labels": [ "Binance" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [ "Coinbase", "Digital Asset Domain" ] }, { "id": "binary-buffer", "title": "Binary Buffer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A contiguous region of memory used to temporarily store binary data during transfer, processing, or rendering operations, serving as an intermediary between data sources and destinations in graphics pipelines, network communications, and computational workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:binary-buffer", "labels": [ "Binary Buffer" ], "is_subclass_of": [ "Standards and Interoperability", "Data Structure" ], "wikilinks": [ "Buffer Management", "Data Formatting", "Data Transfer", "Graphics Rendering", "Memory Allocation", "Stream Processing", "Data Structure", "metaverse" ] }, { "id": "binary-data-mask", "title": "Binary Data Mask", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:binary-data-mask", "labels": [ "Binary Data Mask" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "binary-encoding", "title": "Binary Encoding", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A data representation format that converts information into sequences of binary digits (0s and 1s) for efficient storage, transmission, and processing, offering compact machine-friendly representations that optimize bandwidth, reduce latency, and enable cross-platform data exchange.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:binary-encoding", "labels": [ "Binary Encoding", "Binary Encoding Standard", "Binary Serialisation" ], "is_subclass_of": [ "Standards and Interoperability", "Data Format" ], "wikilinks": [ "Data Serialization", "Efficient Storage", "Encoding Rules", "Network Transmission", "Parser Implementation", "Schema Definition", "Data Format", "metaverse" ] }, { "id": "binaural-audio", "title": "Binaural Audio", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Binaural audio is two-channel sound designed to reproduce the directional and spatial cues a listener would experience naturally, delivering an immersive three-dimensional image over headphones. It can be captured with a dummy head or in-ear microphones, or synthesised by convolving sources with head-related transfer functions. By preserving interaural time and level differences and pinna spectral cues, binaural audio conveys not just left-right panning but full surrounding placement including elevation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:binaural-audio", "labels": [ "Binaural Audio" ], "is_subclass_of": [ "Spatial Audio" ], "wikilinks": [] }, { "id": "binaural-rendering", "title": "Binaural Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Binaural rendering is the audio signal processing technique that synthesises a three-dimensional sound field deliverable over standard headphones by convolving audio sources with head-related transfer functions (HRTFs) that model the acoustic filtering imposed by the human head, pinnae, and torso. The result is the perceptual illusion of sounds emanating from specific spatial locations outside the headphones, enabling immersive audio experiences in virtual and augmented reality environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:binaural-rendering", "labels": [ "Binaural Rendering" ], "is_subclass_of": [ "Spatial Audio" ], "wikilinks": [] }, { "id": "binding-corporate-rules", "title": "Binding Corporate Rules", "domain": "governance", "domain_name": "Governance", "definition": "Binding Corporate Rules (BCRs) are an intra-group data transfer mechanism approved by EU data protection authorities that allow multinational corporate groups to transfer personal data from the European Economic Area to group entities in third countries lacking an EU adequacy decision, provided the group adopts and enforces a comprehensive, legally binding internal data protection code. BCRs are assessed and approved by a lead supervisory authority under the GDPR framework.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:binding-corporate-rules", "labels": [ "Binding Corporate Rules" ], "is_subclass_of": [ "Cross-Border Data Transfer Rule" ], "wikilinks": [] }, { "id": "binned-data-state", "title": "Binned Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:binned-data-state", "labels": [ "Binned Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bio-terror", "title": "Bio Terror", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Bioterrorism (biological terrorism) is the deliberate use, threat, or weaponisation of pathogenic microorganisms (bacteria, viruses, fungi, prions), biological toxins, or genetically engineered agents by state or non-state actors to cause mass casualties, generate fear, coerce governments, or des...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bio-terror", "labels": [ "Bio Terror" ], "is_subclass_of": [ "Legal and Regulatory", "AI Risks", "Biosecurity", "Dual-Use Research", "Weapons of Mass Destruction", "Global Catastrophic Risk", "Pandemic Preparedness" ], "wikilinks": [ "Aerosol Dispersal Mechanism", "Agroterrorism", "AI Protein Design", "Amerithrax 2001", "Aum Shinrikyo Biological Programme", "Australia Group", "Australia Group Export Controls", "Automated Biofoundry", "BARDA Medical Countermeasures", "Biodefence Research", "Biodefence Stockpiling", "Biological Toxin", "Biological Weapons Convention", "Biological Weapons Convention BWC 1972", "Biosafety", "Biosafety Level BSL-1-4 Framework", "Biosafety Level Laboratory", "Biosecurity", "BiosecurityDomain", "Bioweapon Delivery System" ] }, { "id": "biodiversity-conservation", "title": "Biodiversity Conservation", "domain": "governance", "domain_name": "Governance", "definition": "Biodiversity conservation is the protection, restoration and sustainable management of the variety of life on Earth \u2014 genes, species, habitats and ecosystems \u2014 to maintain ecological function and resilience. It combines protected areas, species recovery, habitat restoration, sustainable use and policy instruments, increasingly informed by monitoring data and remote sensing. As a pillar of environmental sustainability, it is closely linked to climate change mitigation, ecosystem services and emerging biodiversity finance markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:biodiversity-conservation", "labels": [ "Biodiversity Conservation" ], "is_subclass_of": [ "Environmental Sustainability" ], "wikilinks": [] }, { "id": "biodiversity-monitoring", "title": "Biodiversity Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:biodiversity-monitoring", "labels": [ "Biodiversity Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bioinformatics", "title": "Bioinformatics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Bioinformatics is the interdisciplinary field that develops computational methods and software to acquire, store, analyse, and interpret biological data, especially molecular sequences and structures. It combines biology, computer science, statistics, and increasingly machine learning to address problems such as genome assembly, sequence alignment, protein structure prediction, and the inference of biological networks. As high-throughput sequencing and imaging generate vast datasets, bioinformatics provides the algorithms and pipelines that turn raw measurements into testable biological insight.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bioinformatics", "labels": [ "Bioinformatics" ], "is_subclass_of": [ "Computational Biology", "Data Science", "Scientific Computing" ], "wikilinks": [] }, { "id": "biomarker-discovery", "title": "Biomarker Discovery", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Biomarker discovery is the process of identifying measurable biological indicators, such as gene expression patterns, proteins or metabolites, that correlate with a disease state, treatment response or physiological process. Machine learning methods applied to genomic, proteomic and clinical datasets accelerate this process by detecting patterns across high-dimensional biological data that would be impractical to find manually. Validated biomarkers underpin precision medicine and inform drug discovery by identifying targets and patient stratification criteria.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:biomarker-discovery", "labels": [ "Biomarker Discovery" ], "is_subclass_of": [ "Bioinformatics" ], "wikilinks": [] }, { "id": "biomechanics", "title": "Biomechanics", "domain": "robotics", "domain_name": "Robotics", "definition": "Biomechanics is the study of the mechanical principles governing biological systems, analysing forces, motion and structure in living organisms. In robotics it informs the design of actuators, limbs and gaits that emulate or assist biological movement, bridging physiology and mechanical engineering. It supplies the models of kinematics and dynamics used to make legged, humanoid and wearable machines move efficiently and safely.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:biomechanics", "labels": [ "Biomechanics" ], "is_subclass_of": [ "Robotics", "Actuation and Control" ], "wikilinks": [] }, { "id": "biomedical-engineering", "title": "Biomedical Engineering", "domain": "robotics", "domain_name": "Robotics", "definition": "Biomedical engineering is the discipline that applies engineering principles, materials, and computational methods to medicine and biology in order to design devices, systems, and processes that diagnose, monitor, treat, or restore human function. It spans medical instrumentation, biomaterials, prosthetics and exoskeletons, rehabilitation robotics, and biosignal processing, bridging mechanical, electrical, and control engineering with the life sciences and clinical practice.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:biomedical-engineering", "labels": [ "Biomedical Engineering" ], "is_subclass_of": [ "Robotics", "Medical Robot" ], "wikilinks": [] }, { "id": "biometric-authentication", "title": "Biometric Authentication", "domain": "security", "domain_name": "Security", "definition": "Biometric authentication is an identity verification method that uses measurable, unique physiological or behavioural characteristics of an individual \u2014 such as fingerprints, facial geometry, iris patterns, voice, or behavioural signals like keystroke dynamics \u2014 to confirm the claimed identity of a subject. Unlike password-based or token-based mechanisms, biometric credentials are inherent to the individual and cannot be forgotten, shared, or easily replicated, though they introduce irreversibility concerns because compromised biometric data cannot be reset. A complete system comprises enrolment, secure template storage (typically on-device in a secure enclave), a matching engine evaluated by false acceptance rate and false rejection rate, and presentation attack detection to counter spoofing. Biometric authentication underpins modern identity assurance frameworks such as FIDO2/WebAuthn, eIDAS 2.0 digital identity wallets, and biometric border control systems worldwide.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:biometric-authentication", "labels": [ "Biometric Authentication", "Biometric Authentication Subsystem", "BiometricAuthentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "biometric-binding-mechanism", "title": "Biometric Binding Mechanism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical hardware device that captures and verifies biometric traits (fingerprints, iris patterns, facial geometry) to bind digital identities to authenticated users.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:biometric-binding-mechanism", "labels": [ "Biometric Binding Mechanism" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Anti-Spoofing Module", "Authentication Server", "Biometric Template Database", "EdgeLayer", "Encrypted Storage", "Facial Recognition Sensor", "FIDO Alliance", "Fingerprint Scanner", "Iris Recognition Camera", "ISO 19794", "Multi-Factor Authentication", "Secure Element Chip", "Template Matching Processor", "Zero Trust Security Framework", "Access Control", "Identity Management System", "Identity Verification", "Infrared Illuminator", "Non-Repudiation", "Optical Sensor Array" ] }, { "id": "biometric-data", "title": "Biometric Data", "domain": "governance", "domain_name": "Governance", "definition": "Biometric data is a category of personal data comprising unique physiological or behavioural characteristics of an individual \u2014 such as fingerprints, facial geometry, iris patterns, voice prints, gait, and keystroke dynamics \u2014 that can be used to identify or authenticate that person. Under data protection frameworks such as GDPR, biometric data processed for identification purposes is classified as a special category of personal data subject to heightened protection. Its irrevocable nature (an individual cannot change their biometrics) makes breaches particularly consequential.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:biometric-data", "labels": [ "Biometric Data" ], "is_subclass_of": [ "Personal Data" ], "wikilinks": [] }, { "id": "biometric-identification", "title": "Biometric Identification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Biometric identification is the recognition of an individual by measuring and matching distinctive physiological or behavioural characteristics, such as fingerprints, facial geometry, iris patterns, voice or gait. It captures a sample, extracts a feature template and compares it against enrolled templates to verify a claimed identity (one-to-one) or identify an unknown subject (one-to-many). It underpins access control, border management and device authentication, and raises significant privacy considerations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:biometric-identification", "labels": [ "Biometric Identification" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [] }, { "id": "biometric-verification", "title": "Biometric Verification", "domain": "security", "domain_name": "Security", "definition": "Biometric verification is the one-to-one matching process in which a live biometric sample presented by a claimant is compared against a single pre-enrolled template associated with the claimed identity, producing an accept or reject decision based on a similarity threshold. It is distinct from biometric identification, which performs a one-to-many search across an entire database, making verification faster and more privacy-preserving for authentication use cases.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:biometric-verification", "labels": [ "Biometric Verification" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [] }, { "id": "biosecurity", "title": "Biosecurity", "domain": "governance", "domain_name": "Governance", "definition": "Biosecurity is the set of policies, practices, and controls aimed at preventing the accidental or deliberate misuse of biological agents, including pathogens, toxins, and biotechnology. It spans laboratory containment, pathogen access controls, dual-use research oversight, and increasingly the governance of AI tools that could lower barriers to engineering dangerous organisms. As a field, it is central to mitigating biological existential and catastrophic risks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:biosecurity", "labels": [ "Biosecurity" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "biosensing-interface", "title": "Biosensing Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "Physical sensor hardware system that detects physiological signals such as heart rate, electroencephalography (EEG), galvanic skin response (GSR), and electromyography (EMG) to enable real-time adaptation of virtual interaction and user experience.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:biosensing-interface", "labels": [ "Biosensing Interface" ], "is_subclass_of": [ "Human-Robot Interaction", "Sensor System" ], "wikilinks": [ "Affective Computing", "Affective Computing Framework", "Biofeedback Systems", "Cloud Analytics Service", "Communication Protocols", "Control Systems", "EdgeLayer", "Electrocardiogram Sensor", "Electroencephalography Sensor", "Emotional State Detection", "Galvanic Skin Response Sensor", "IEEE P2733", "ISO 9241-960", "Neurotechnology", "Physiological Computing System", "Pulse Oximeter", "Rehabilitation Robots", "Sensor System", "Signal Processing Unit", "Skin Contact Electrodes" ] }, { "id": "biosignature", "title": "Biosignature", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:biosignature", "labels": [ "Biosignature" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "biospecimen", "title": "Biospecimen", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:biospecimen", "labels": [ "Biospecimen" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bipedal-balance", "title": "Bipedal Balance", "domain": "robotics", "domain_name": "Robotics", "definition": "Bipedal Balance is the set of sensing, computation, and actuation mechanisms that enable a two-legged robot or agent to maintain postural stability during standing, walking, and dynamic manoeuvres. It encompasses whole-body control strategies, inertial sensing, and real-time torque control to keep the centre of mass within a supportable base of support. The field integrates mechanics, control theory, and machine learning to achieve robust locomotion across uneven terrain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bipedal-balance", "labels": [ "Bipedal Balance" ], "is_subclass_of": [ "Legged Locomotion" ], "wikilinks": [] }, { "id": "bipropellant-propulsion", "title": "Bipropellant Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bipropellant-propulsion", "labels": [ "Bipropellant Propulsion" ], "is_subclass_of": [ "Chemical Propulsion" ], "wikilinks": [] }, { "id": "bit-error-rate", "title": "Bit Error Rate", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bit-error-rate", "labels": [ "Bit Error Rate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bit-go", "title": "BitGo", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BitGo is a company that provides digital asset custody, wallet, and security services, including multi-signature wallet technology, for institutional clients.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bit-go", "labels": [ "BitGo", "BitGo Wallet" ], "is_subclass_of": [ "Institutional Custody" ], "wikilinks": [ "Multisignature", "Institutional Custody", "Key Management" ] }, { "id": "bit-vm", "title": "BitVM", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BitVM is a proposed approach for expressing complex computations that can be verified on Bitcoin without changing its consensus rules. It uses fraud proofs and challenge-response interactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bit-vm", "labels": [ "BitVM" ], "is_subclass_of": [ "Bitcoin Script" ], "wikilinks": [ "Taproot", "Fraud Proof", "Layer 2 Scaling", "Smart Contract", "Bitcoin Script", "https://bitvm.org", "https://github.com/BitVM/BitVM" ] }, { "id": "bitcoin-as-money", "title": "Bitcoin As Money", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin As Money is the analytical framework within monetary economics and philosophy that interrogates wher Bitcoin \u2014 the fixed-supply, decentralised, proof-of-work-secured digital commodity created by Satoshi Nakamoto \u2014 satisfies the classical three functions of money: medium of exchang...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-as-money", "labels": [ "Bitcoin As Money" ], "is_subclass_of": [ "DeFi and Economics", "Token and Asset", "Monetary Theory", "Sound Money", "Digital Asset Economics", "Austrian Economics", "Cryptocurrency Economics" ], "wikilinks": [ "Africa", "AI Agent Payments", "ainvest 2025 Lightning Network 1.17 Billion Monthly Volume", "Alden 2025 What is Money", "Ammous 2018 The Bitcoin Standard", "Andrew M. Bailey", "arXiv 2025 Bitcoin Structural Shortcomings as Money", "ARK Invest", "Ark Protocol", "Austrian Economics", "AustrianEconomicsDomain", "Austrian Hard Money Theory", "BaseLayerBitcoin", "BIP32 HD Wallets", "Bitcoin Beach", "Bitcoin Beach FBCE 2025 Circular Economies Summit", "Bitcoin Network", "Bitcoin Standard", "Bitcoin Treasury Strategy", "Blinded Paths" ] }, { "id": "bitcoin-cash", "title": "Bitcoin Cash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cryptocurrency and blockchain created in 2017 as a hard fork of Bitcoin, with a larger block size intended to increase on-chain transaction throughput. It uses the same proof-of-work consensus model as Bitcoin.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-cash", "labels": [ "Bitcoin Cash" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [ "Mining", "Hash Function", "Payment Channel", "Bitcoin", "Bitcoin Network", "Cryptocurrency", "https://bitcoincash.org/" ] }, { "id": "bitcoin-centralisation-risks", "title": "Bitcoin Centralisation Risks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Centralisation Risks refers to the ensemble of systemic vulnerabilities arising from the progressive concentration of economic power, computational control, physical infrastructure, custody arrangements, development authority, and payment routing within Bitcoin's nominally decentralised e...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-centralisation-risks", "labels": [ "Bitcoin Centralisation Risks" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain", "Systemic Risk", "Blockchain Governance Risk", "Decentralisation", "Network Security", "Concentration Risk" ], "wikilinks": [ "ACINQ", "ACM TOSEM 2024 Bitcoin Governance Empirical Study", "Anchorage Digital", "AntPool", "ArXiv 2025 Concentration Within Distribution Bitcoin Network Science", "Argo Blockchain", "ASIC", "ASIC Manufacturer Duopoly", "Biais Bisiere Bouvard Casamatta 2019 Blockchain Folk Theorem", "Binance", "BIS 2022 Prudential Treatment Cryptoassets", "BIS Crypto Risk Assessment Framework", "Bitcoin Cash", "Bitcoin Core", "Bitcoin Network", "Bitcoin Security Budget", "Bitcoin Whitepaper", "Bitmain", "BlackRock", "BlackRock IBIT 2025 Annual Filing" ] }, { "id": "bitcoin-core", "title": "Bitcoin Core", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Core is the canonical, open-source reference implementation of the Bitcoin protocol, written primarily in C++, which full nodes run to independently validate transactions, enforce consensus rules, and maintain a complete local copy of the blockchain. Originally released by Satoshi Nakamoto in January 2009 and subsequently maintained by a global community of contributors, it establishes the authoritative behaviour of the Bitcoin network through its codebase. The software bundles a peer-to-peer networking layer, a UTXO-based scripting engine, a deterministic wallet, a mining interface, and a JSON-RPC API for integration with higher-level applications. Protocol changes are introduced exclusively via Bitcoin Improvement Proposals, requiring broad community review and backward-compatible soft-fork activation to preserve network cohesion.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-core", "labels": [ "Bitcoin Core", "Bitcoin Core Development", "Bitcoin Core Reference Implementation" ], "is_subclass_of": [ "Blockchain Protocol" ], "wikilinks": [] }, { "id": "bitcoin-custody", "title": "Bitcoin Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin custody is the practice and infrastructure for securely holding the private keys that control bitcoin, spanning self-custody, collaborative custody, and qualified institutional custodians. Approaches differ in how they distribute key control and recovery, using techniques such as multi-signature, hardware security modules, and multi-party computation to balance security against operational availability. Custody design is foundational to wallets, federations, and regulated products such as exchange-traded funds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-custody", "labels": [ "Bitcoin Custody" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "bitcoin-de-fi", "title": "Bitcoin DeFi", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin DeFi is the emerging set of decentralised finance applications, such as lending, swaps, and stablecoins, built on or anchored to the Bitcoin network. Because Bitcoin's base layer has limited scripting, these applications typically rely on layers and sidechains, including Lightning, Taproot assets, RGB, and Layer-3 protocols, to add programmability. The goal is to bring DeFi functionality to Bitcoin's liquidity and security while preserving its conservative base-layer design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-de-fi", "labels": [ "Bitcoin DeFi" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "bitcoin-distribution", "title": "Bitcoin Distribution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Distribution is the sub-field of Blockchain economics that measures, models, and interprets the allocation of Bitcoin's circulating supply across identifiable holder categories \u2014 individuals, corporations, governments, miners, exchanges, custodians, and the permanently lost or provabl...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-distribution", "labels": [ "Bitcoin Distribution" ], "is_subclass_of": [ "DeFi and Economics", "Bitcoin Proof-of-Work Protocol", "Bitcoin Technical Overview", "Blockchain", "Bitcoin Value Proposition", "Bitcoin As Money", "Bitcoin Environmental Issues" ], "wikilinks": [ "AnalyticsLayer", "BitInfoCharts", "Bitcoin Halving", "Bitcoin Halving", "Bitcoin Scarcity", "BlackRock", "Blockchain Analysis", "Blockchain Analysis", "Blockchain Analysis", "Blockchain Analysis", "Blockchain Analysis", "Chainalysis", "CryptoQuant", "CryptoeconomicsDomain", "Cryptographic Addresses", "Dormant Supply", "FinancialEconomicsDomain", "GINI Coefficient", "Glassnode", "HODL Waves" ] }, { "id": "bitcoin-etf-custody", "title": "Bitcoin ETF Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin ETF custody is the institutional safekeeping of the bitcoin backing a spot exchange-traded fund, performed by qualified custodians under regulatory and audit requirements. It typically uses cold storage with multi-signature or multi-party-computation key management, segregation of client assets, insurance, and proof-of-reserves attestation. This custody model bridges traditional regulated finance and on-chain asset control, and is central to the integrity of spot Bitcoin ETFs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-etf-custody", "labels": [ "Bitcoin ETF Custody" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "bitcoin-etf", "title": "Bitcoin ETF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bitcoin ETF (Bitcoin Exchange-Traded Fund) is a regulated, exchange-listed pooled investment vehicle providing investors fungible share-based exposure to the spot price of Bitcoin (or, in the case of futures-based variants, to rolling CME Bitcoin futures contracts) through traditional broke...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-etf", "labels": [ "Bitcoin ETF" ], "is_subclass_of": [ "Governance and Regulation", "Token and Asset", "Exchange-Traded Fund", "Pooled Investment Vehicle", "Regulated Investment Company", "Bitcoin Exposure Product", "Passive Index Product" ], "wikilinks": [ "Anchorage Digital", "Aramonte Doerr Huang Schrimpf 2024 BIS DeFi and Spot Bitcoin ETFs", "Authorized Participant", "Authorized Participant Network", "Bank of England Financial Stability Reports 2023-2024", "Ben-David Franzoni Moussawi 2017 Do ETFs Increase Volatility", "Bianchi Babiak Dickerson 2024 Impact of Spot Bitcoin ETF Approval", "BitGo", "Bitcoin Custody", "Bitcoin Exposure Product", "Bitcoin Inclusion in Model Portfolio", "Bitcoin Lightning Wallet", "Bitcoin Liquidity", "Bitcoin Mining Equity", "Bitcoin Network", "Bitcoin Options", "Bitcoin Reference Rate", "BITO", "BlackRock", "BlackRock iShares Bitcoin Trust Form S-1 2023" ] }, { "id": "bitcoin-environmental-issues", "title": "Bitcoin Environmental Issues", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Environmental Issues refers to the multi-dimensional ecological footprint generated by Bitcoin's Proof of Work (PoW) consensus mechanism and the extensive ASIC mining hardware infrastructure that sustains it.", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-environmental-issues", "labels": [ "Bitcoin Environmental Issues" ], "is_subclass_of": [ "Governance and Regulation", "Environmental Impact Assessment", "Energy Economics", "Blockchain Sustainability", "Digital Asset Ecology", "Climate Finance" ], "wikilinks": [ "ACS Sustainable Chemistry Engineering 2023 From Mining to Mitigation", "ACS Sustainable Chemistry Engineering 2023 Renewable Energy Facilitated by Bitcoin", "Ammous 2018 The Bitcoin Standard", "ASIC", "ASIC E-Waste", "ASIC Hardware", "Bastian-Pinto 2021 Hedging Renewable Energy with Bitcoin Mining", "Blandin 2020 Global Cryptoasset Benchmarking Study Cambridge", "Cambridge Bitcoin Electricity Consumption Index", "Cambridge Centre for Alternative Finance", "Cambridge Centre for Alternative Finance 2025 Digital Mining Industry Report", "Carbon Credits", "Carbon Footprint", "Carbon Neutrality Goals", "CBECI", "CBECI Cambridge Bitcoin Electricity Consumption Index", "CleanSpark", "Climate Finance", "ClimateFinanceDomain", "Climate Policy" ] }, { "id": "bitcoin-fungible-token-protocols", "title": "Bitcoin Fungible Token Protocols", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Runes and Glyphs refers to two complementary Bitcoin token protocols: Runes, which encodes fungible token balances directly in Bitcoin transaction outputs using OP_RETURN data, and Glyphs, a related protocol developed by Melvin Carvalho that encodes token metadata as on-chain inscriptions compatible with the Ordinals framework. Together they extend Bitcoin's base layer with native fungible asset issuance without requiring a separate blockchain, enabling DAOs, project tokens, and programmable digital objects on Bitcoin testnet and mainnet.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-fungible-token-protocols", "labels": [ "Bitcoin Fungible Token Protocols", "Runes and Glyphs" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Digital Objects", "Melvin Carvalho", "Runes and Glyphs", "Testnet" ] }, { "id": "bitcoin-halving", "title": "Bitcoin Halving", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Bitcoin halving is a protocol-enforced event, occurring approximately every 210,000 blocks (roughly four years), at which the block subsidy paid to miners is cut in half. It implements Bitcoin's disinflationary monetary policy, capping total supply at 21 million coins and progressively reducing new issuance until it reaches zero around the year 2140. Halvings have historically preceded major shifts in mining economics and market price, and they are central to Bitcoin's narrative as a scarce, predictably issued digital asset.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-halving", "labels": [ "Bitcoin Halving" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "bitcoin-improvement-proposals", "title": "Bitcoin Improvement Proposals", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Improvement Proposals (BIPs) are the formal design documents through which changes, enhancements, and new features are proposed and ratified for the Bitcoin protocol and its broader ecosystem. First introduced by Amir Taaki in 2011, modelled on Python's PEP and Python Enhancement Proposal process, BIPs serve as the primary coordination mechanism for a decentralised developer community that lacks any central authority. Each BIP progresses through a defined lifecycle \u2014 draft, proposed, final, and optionally superseded \u2014 and must achieve rough consensus among economic nodes, miners, and users before activation. BIPs are organised into three tracks: Standards Track (protocol-level changes requiring network-wide adoption), Informational (guidelines and best practices), and Process (procedural rules governing the BIP process itself).", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-improvement-proposals", "labels": [ "Bitcoin Improvement Proposals", "BIP-340", "Bitcoin Improvement Proposal" ], "is_subclass_of": [ "Blockchain Governance" ], "wikilinks": [] }, { "id": "bitcoin-layer-2", "title": "Bitcoin Layer 2", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Layer 2 refers to protocols built on top of the Bitcoin base chain that increase transaction throughput, reduce fees, or add functionality while inheriting Bitcoin's security. Examples include the Lightning Network for fast payments and federated systems such as Fedimint and Cashu for custodial and ecash-style scaling. These layers settle to the base chain periodically, trading some on-chain finality for speed and cost efficiency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-layer-2", "labels": [ "Bitcoin Layer 2", "Bitcoin Layer-2 Ecosystem" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "bitcoin-lightning-network", "title": "Bitcoin Lightning Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Bitcoin Lightning Network is a layer-two payment protocol built on top of Bitcoin that enables fast, low-cost transactions through bidirectional payment channels. Two parties lock funds in a multisignature channel and exchange signed balance updates off-chain, settling the final state on the Bitcoin blockchain only when the channel closes. Payments can be routed across a network of connected channels, allowing transfers between parties that do not share a direct channel.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-lightning-network", "labels": [ "Bitcoin Lightning Network" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Domain" ], "wikilinks": [ "Payment Channel", "HTLC", "Bitcoin", "Multisignature", "Micropayments", "Instant Settlement", "Payment Systems Domain", "Liquid Network", "Blockchain Domain" ] }, { "id": "bitcoin-mining", "title": "Bitcoin Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Mining is the proof-of-work Nakamoto consensus process by which the Bitcoin blockchain is extended, secured, and monetised through competitive computation of SHA-256 double-hash preimages whose output, interpreted as a 256-bit unsigned integer, falls below a network-ad...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-mining", "labels": [ "Bitcoin Mining", "Bitcoin Mining Hardware" ], "is_subclass_of": [ "Protocol and Consensus", "Token and Asset", "Proof-of-Work", "Consensus Mechanism", "Cryptographic Mining", "Distributed Computing", "Industrial Process" ], "wikilinks": [ "2PIC1Pump", "3nm", "4-Coincident Peak", "5nm", "Adam Back", "Aggelos Kiayias", "AI HPC", "AI HPC Pivot", "Alex de Vries", "Alex de Vries (Digiconomist)", "Andrew Webber", "AntPool", "Application-Specific Integrated Circuit", "Argo Blockchain", "Argo Blockchain plc", "ARPU per terahash", "ArtForz", "ASIC", "ASIC Hardware", "ASIC Manufacturer" ] }, { "id": "bitcoin-network", "title": "Bitcoin Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Bitcoin Network is the global, permissionless peer-to-peer infrastructure underpinning the Bitcoin cryptocurrency, comprising full nodes, mining nodes, and the gossip protocol that propagates signed transactions and mined blocks across the internet without central coordination. It employs the Nakamoto consensus mechanism\u2014proof-of-work mining on the SHA-256 hash function\u2014to achieve Byzantine-fault-tolerant agreement on a single shared transaction ledger among mutually distrusting participants. The network enforces deterministic monetary policy through its protocol rules, automatically adjusting mining difficulty every 2,016 blocks to target a ten-minute inter-block interval and capping total issuance at 21 million BTC via a geometric halving schedule. Launched in January 2009, it is the longest-continuously-operated public blockchain and the foundational reference implementation for decentralised digital value transfer.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-network", "labels": [ "Bitcoin Network", "Bitcoin P2P Network" ], "is_subclass_of": [ "Peer-to-Peer Network" ], "wikilinks": [] }, { "id": "bitcoin-ordinals", "title": "Bitcoin Ordinals", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Ordinals is a numbering scheme and inscription protocol that assigns a unique serial number to every individual satoshi on the Bitcoin network based on the order of its mining, enabling arbitrary content \u2014 text, images, code \u2014 to be embedded directly into Bitcoin transactions via the witness data introduced by SegWit and Taproot. Each inscription is permanently stored on-chain without requiring a separate token contract, making it a native non-fungible artefact within the Bitcoin base layer. The protocol was introduced by Casey Rodarmor in January 2023.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-ordinals", "labels": [ "Bitcoin Ordinals", "Bitcoin-Ordinals" ], "is_subclass_of": [ "Bitcoin Network" ], "wikilinks": [] }, { "id": "bitcoin-proof-of-work-protocol", "title": "Bitcoin Proof-of-Work Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin is the first decentralised peer-to-peer electronic cash system, introduced by the pseudonymous Satoshi Nakamoto in a 2008 white paper and launched in January 2009. It operates on a public, permissionless blockchain secured by proof-of-work consensus, with a fixed supply cap of 21 million BTC enforced by its protocol. Bitcoin is simultaneously a payment network, a monetary asset, and the foundational reference implementation for the broader cryptocurrency and blockchain industry.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol", "labels": [ "Bitcoin Proof-of-Work Protocol", "Bitcoin BIPs", "Bitcoin Ecosystem", "Bitcoin Governance", "Bitcoin NFTs", "Bitcoin Proof-of-Work", "Bitcoin Scarcity" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [] }, { "id": "bitcoin-protocol", "title": "Bitcoin Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The set of rules governing how Bitcoin transactions are formed, validated, and ordered into blocks by proof-of-work consensus, including the UTXO model, scripting system, block reward schedule, and difficulty adjustment mechanism.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bitcoin-protocol", "labels": [ "Bitcoin Protocol" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Consensus Protocol", "UTXO", "Bitcoin Script", "Block Reward", "Bitcoin" ] }, { "id": "bitcoin-script", "title": "Bitcoin Script", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Script is a non-Turing-complete, stack-based scripting language embedded in every Bitcoin transaction that defines the conditions under which unspent transaction outputs (UTXOs) may be spent. It consists of a constrained set of opcodes operating on a last-in-first-out (LIFO) stack, deliberately designed without loops or unbounded recursion to ensure guaranteed termination and predictable resource consumption. Script programs are expressed as paired locking scripts (scriptPubKey) and unlocking scripts (scriptSig or SegWit witness data) that encode spending conditions including digital signature verification, multisignature requirements, hash pre-image revelation, and time locks. Standard output templates \u2014 P2PKH, P2SH, P2WPKH, P2WSH, and P2TR \u2014 formalise the most common spending patterns used across the Bitcoin network.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bitcoin-script", "labels": [ "Bitcoin Script" ], "is_subclass_of": [ "Scripting Language" ], "wikilinks": [] }, { "id": "bitcoin-standard", "title": "bitcoin standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Bitcoin Standard is an economic and monetary thesis proposing that Bitcoin's algorithmically enforced fixed supply of 21 million units, its decentralised Proof of Work consensus mechanism, and its censorship-resistant disinflationary issuance schedule collectively make it a superior store of value and the basis for a new international monetary order analogous to the historical gold standard. Drawing heavily on Austrian economics \u2014 particularly Ludwig von Mises's concept of sound money and Saifedean Ammous's eponymous 2018 treatise \u2014 the thesis holds that Bitcoin's predictable, politically immutable monetary policy immunises it against the purchasing-power erosion endemic to fiat currencies managed by central banks. In practice the standard informs corporate treasury reserve strategies, nation-state legal-tender experiments, and the design of Bitcoin ETF and custody infrastructure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-standard", "labels": [ "Bitcoin Standard", "Saifedean Ammous Bitcoin Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "bitcoin-technical-overview", "title": "Bitcoin Technical Overview", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Technical Overview is the canonical ontological concept encapsulating the complete technical architecture of the Bitcoin protocol \u2014 the first decentralised, permissionless, proof-of-work secured, UTXO-based peer-to-peer electronic cash system described by Satoshi Nakamoto in the 2008 whit...", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-technical-overview", "labels": [ "Bitcoin Technical Overview" ], "is_subclass_of": [ "Protocol and Consensus", "Peer-to-Peer Network", "Blockchain", "Distributed Ledger Technology", "Consensus Mechanism", "Cryptographic Protocol" ], "wikilinks": [ "Antonopoulos 2023 Mastering Bitcoin", "BIP-141 Segregated Witness", "BIP-32 Hierarchical Deterministic Wallets", "BIP-327 MuSig2", "BIP-340", "BIP-340 Schnorr Signatures", "BIP-341", "BIP-341 Taproot", "BIP-342", "BIP-342 Tapscript", "BIP-347 OP CAT", "BIP-360", "BIP-360 P2TSH Quantum Resistance", "BIP-39 Mnemonic Code", "BIP Process", "Bitcoin Core", "Bitcoin Optech", "Bitcoin Script", "Block Structure", "blockchain analytics" ] }, { "id": "bitcoin-transaction", "title": "Bitcoin Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bitcoin transaction is a signed data structure that transfers value on the Bitcoin network by consuming unspent transaction outputs (UTXOs) and creating new ones. Each transaction references prior outputs, provides cryptographic signatures satisfying their spending conditions, and specifies new outputs locked to recipient scripts. Validated transactions are broadcast to the mempool and ultimately confirmed when included in a mined block.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bitcoin-transaction", "labels": [ "Bitcoin Transaction" ], "is_subclass_of": [ "Blockchain Transaction", "Bitcoin" ], "wikilinks": [] }, { "id": "bitcoin-value-proposition", "title": "Bitcoin Value Proposition", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Value Proposition is the aggregate set of economic, monetary, and philosophical arguments that justify holding, using, or building on Bitcoin as a distinct and superior form of money, property, or network infrastructure.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-value-proposition", "labels": [ "Bitcoin Value Proposition" ], "is_subclass_of": [ "DeFi and Economics", "Store of Value", "Bitcoin Proof-of-Work Protocol", "Sound Money", "Monetary Theory", "Digital Asset", "Economic Philosophy" ], "wikilinks": [ "21 Million Supply Cap", "AI agents", "AI Economy", "AI Economy Payments", "Ammous 2018 The Bitcoin Standard", "Argentina", "arxiv 2024 Bitcoin Structural Shortcomings 2512-07840", "ASIC Mining Hardware", "Austrian Economics", "AustrianEconomicsLayer", "Batten 2024 Bitcoin Sustainable Energy BEEST", "Ben Hunt", "BITCOIN Act of 2025", "Bitcoin Improvement Proposals", "Bitcoin Miners", "Bitcoin Network", "Bitcoin Standard Adoption", "Bitcoin Whitepaper", "BlackRock", "BlackRock Bitcoin ETF" ] }, { "id": "bitcoin-whitepaper", "title": "Bitcoin Whitepaper", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The October 2008 technical paper by pseudonymous author Satoshi Nakamoto, titled 'Bitcoin: A Peer-to-Peer Electronic Cash System', which introduced the design of a decentralised digital currency that eliminates reliance on trusted third parties. It proposed a chain of cryptographically linked blocks secured by proof-of-work consensus to prevent double-spending without a central authority. The paper synthesised prior work on digital cash, cryptographic hash functions, and distributed timestamps into a coherent, deployable protocol that was subsequently realised in the January 2009 Bitcoin genesis block launch.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-whitepaper", "labels": [ "Bitcoin Whitepaper" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "Hashcash", "Bitcoin Protocol", "Satoshi Nakamoto", "Bitcoin" ] }, { "id": "bitcoin-related-links", "title": "Bitcoin related links", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin Related Links is a curated reference collection aggregating news, technical resources, regulatory developments, and community analysis pertaining to the Bitcoin blockchain ecosystem. It encompasses coverage of Bitcoin as money, mining economics, environmental debates, ETF instruments, and the broader cryptocurrency landscape.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:bitcoin-proof-of-work-protocol-related-links", "labels": [ "Bitcoin related links" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Address Distribution", "Africa", "African Bitcoin", "Alex Gladstein", "Argentina", "ASIC Mining", "Asset Class", "Asset Recovery", "BIP85", "Bitcoin Adoption", "Bitcoin Bonds", "Bitcoin Culture", "Bitcoin Legal Status", "Bitcoin Price Prediction", "Bitcoin Whitepaper", "Blockstream", "CBDC", "CBDC Competition", "Central Banks", "Charles Edwards" ] }, { "id": "bitcoin", "title": "Bitcoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin is the first decentralised, permissionless, peer-to-peer electronic cash system, introduced by the pseudonymous Satoshi Nakamoto in a 2008 whitepaper and launched as open-source software in January 2009. It maintains a globally shared, tamper-evident ledger \u2014 the blockchain \u2014 through a proof-of-work consensus mechanism called Nakamoto Consensus, in which miners compete to extend the chain by finding nonces satisfying a difficulty-adjusted SHA-256 hash target, earning new bitcoin and transaction fees as reward. The monetary supply is strictly bounded by a 21 million coin cap enforced deterministically via a halving schedule that reduces the block subsidy approximately every four years, conferring programmatic scarcity and censorship-resistant value transfer without reliance on any trusted intermediary. Bitcoin's UTXO model, Script-based transaction authorisation, and secp256k1 elliptic-curve cryptography together form the foundational substrate upon which Lightning Network payment channels, Taproot smart contracting, and a growing ecosystem of Layer 2 protocols are built.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bitcoin", "labels": [ "Bitcoin", "Bitcoin Blockchain", "Bitcoin Covenant" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [] }, { "id": "bitfinex", "title": "Bitfinex", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitfinex is a cryptocurrency exchange that allows trading of digital assets and is associated with the issuer of the Tether stablecoin. It is operated by iFinex.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bitfinex", "labels": [ "Bitfinex" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [ "Bitcoin", "Tether", "Cryptocurrency", "https://www.bitfinex.com", "https://docs.bitfinex.com" ] }, { "id": "bitmain", "title": "Bitmain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitmain is a Chinese company that designs and manufactures application-specific integrated circuit mining hardware, best known for its Antminer product line. It also operates mining services.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bitmain", "labels": [ "Bitmain" ], "is_subclass_of": [ "ASIC" ], "wikilinks": [ "Bitcoin Mining", "AntPool", "Hardware", "ASIC", "https://www.bitmain.com", "https://shop.bitmain.com" ] }, { "id": "bitrate", "title": "Bitrate", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Bitrate is the rate at which bits are transmitted, encoded or processed per unit of time, typically expressed in kilobits or megabits per second. It determines the bandwidth a signal consumes and, for audio and video, the trade-off between file size or transmission cost and perceptual quality. Bitrate is a fundamental parameter in broadcast systems and adaptive streaming, where it is dynamically adjusted, and in video compression, where techniques such as motion estimation influence how efficiently a target bitrate is achieved.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:bitrate", "labels": [ "Bitrate" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "bitstring-status-list", "title": "Bitstring Status List", "domain": "governance", "domain_name": "Governance", "definition": "Bitstring Status List is a W3C specification for expressing the revocation or suspension status of Verifiable Credentials through a compressed, publicly-hosted bitstring in which each credential is assigned a position index; setting that bit to 1 indicates revocation without disclosing which specific credential holder triggered the change. The mechanism is privacy-preserving because verifiers observe only a large compressed list, not individual credential identifiers. It supersedes the earlier Credential Status List 2021 specification and is designed to be bandwidth-efficient and herd-privacy-compatible.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bitstring-status-list", "labels": [ "Bitstring Status List" ], "is_subclass_of": [ "Revocation Mechanism" ], "wikilinks": [] }, { "id": "bittensor", "title": "Bittensor", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A decentralised network that uses blockchain-based incentives to coordinate and reward the contribution of machine learning models and compute by independent participants.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:bittensor", "labels": [ "Bittensor" ], "is_subclass_of": [ "Artificial Intelligence", "Distributed Training", "Decentralised System", "Peer-to-Peer Network" ], "wikilinks": [ "Blockchain", "Token Economics", "Distributed Computing", "Peer-to-Peer Network", "Artificial Intelligence" ] }, { "id": "bittorrent", "title": "Bittorrent", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "BitTorrent is a peer-to-peer protocol for distributing files by splitting them into pieces that participants download from and upload to one another rather than from a single server. Each participant who holds the complete file can seed it, while downloaders simultaneously share the pieces they already have, so aggregate capacity grows with demand. The protocol uses content hashing to verify pieces and a tracker or distributed hash table to help peers discover one another.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bittorrent", "labels": [ "Bittorrent", "BitTorrent" ], "is_subclass_of": [ "Peer To Peer Network" ], "wikilinks": [] }, { "id": "black-hole", "title": "Black Hole", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:black-hole", "labels": [ "Black Hole" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "black-box-model", "title": "Black-Box Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A black-box model is a predictive system whose internal decision logic is opaque or too complex for a human to inspect directly, so it can only be understood through its inputs and outputs. Many high-performing machine-learning systems, such as deep neural networks and large ensembles, are black-box in nature, trading interpretability for accuracy. This opacity motivates post-hoc explanation techniques and contrasts with inherently interpretable white-box models where the reasoning is transparent.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:black-box-model", "labels": [ "Black-Box Model", "Black Box Model" ], "is_subclass_of": [ "Model Interpretability", "Machine Learning Model", "Predictive Model" ], "wikilinks": [] }, { "id": "black-rock-buidl", "title": "BlackRock BUIDL", "domain": "finance", "domain_name": "Finance", "definition": "BlackRock BUIDL is a tokenised money market fund issued by BlackRock on public blockchains that invests in cash, US Treasury bills, and repurchase agreements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:black-rock-buidl", "labels": [ "BlackRock BUIDL" ], "is_subclass_of": [ "Asset Tokenisation" ], "wikilinks": [ "Asset Tokenisation", "Tokenisation", "Institutional Adoption" ] }, { "id": "black-rock-bitcoin-etf", "title": "BlackRock Bitcoin ETF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The BlackRock Bitcoin ETF (iShares Bitcoin Trust, ticker IBIT) is a spot Bitcoin exchange-traded fund issued by BlackRock and approved by the US SEC in January 2024. It holds bitcoin in qualified custody and tracks the spot price, giving traditional investors regulated, brokerage-accessible exposure to bitcoin without managing keys. As one of the largest and fastest-growing spot Bitcoin ETFs, it is frequently cited as evidence of institutional adoption.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:black-rock-bitcoin-etf", "labels": [ "BlackRock Bitcoin ETF" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "black-rock", "title": "BlackRock", "domain": "finance", "domain_name": "Finance", "definition": "BlackRock is the world's largest asset management firm, providing investment management, risk advisory, and financial technology services to institutional investors, sovereign wealth funds, pension funds, and retail clients globally. The firm operates Aladdin, a proprietary risk management and operating system platform that processes risk analytics for trillions of dollars in assets across thousands of financial institutions. BlackRock has expanded beyond traditional asset management into digital assets, sustainable investing, and multi-asset solutions, including launching spot Bitcoin and Ethereum ETFs that attracted significant institutional capital. Its scale and systemic importance across public equities, fixed income, alternatives, and infrastructure positions it as a central node in global capital markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:black-rock", "labels": [ "BlackRock" ], "is_subclass_of": [ "Asset Management" ], "wikilinks": [ "Risk Management", "Investment Management", "Bitcoin ETF", "Asset Management" ] }, { "id": "blackboard-pattern", "title": "Blackboard Pattern", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A coordination architecture in which multiple specialised agents collaborate not by messaging each other directly but by reading from and writing to a shared, structured workspace \u2014 the blackboard. Each agent watches the blackboard for a state it can act on, contributes its partial result back to the shared space, and lets other agents build on that contribution in turn, so a solution accretes incrementally through many opportunistic updates rather than through a fixed pipeline. The pattern decouples the collaborators from one another: an agent needs to understand only the blackboard's contents, not the identity, order, or availability of its peers.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:blackboard-pattern", "labels": [ "Blackboard Pattern" ], "is_subclass_of": [ "Multi-Agent Orchestration", "MultiAgentOrchestration" ], "wikilinks": [ "MultiAgentOrchestration", "MemoryBank", "MultiAgentSystem", "SupervisorWorkerPattern" ] }, { "id": "blend-shape", "title": "Blend Shape", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A blend shape, also called a morph target, is a stored deformation of a 3D mesh that is blended with a base shape by a weighted interpolation of vertex positions. By combining multiple blend shapes at varying weights, animators produce smooth transitions between expressions and poses without altering the mesh topology. Blend shapes are central to facial animation, where subtle muscle movements are captured as named targets driven by animation rigs or performance capture.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:blend-shape", "labels": [ "Blend Shape" ], "is_subclass_of": [ "Facial Animation" ], "wikilinks": [] }, { "id": "blend-tree", "title": "Blend Tree", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A blend tree is a structure within an animation system that smoothly blends multiple animation clips according to one or more continuous parameters, producing a single output pose. It is commonly used for locomotion, where clips such as idle, walk and run are interpolated by speed and direction to avoid abrupt transitions. Blend trees are typically composed within an animation controller alongside state machines that govern when each tree is active.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:blend-tree", "labels": [ "Blend Tree" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "blended-finance", "title": "Blended Finance", "domain": "finance", "domain_name": "Finance", "definition": "Blended finance is the strategic use of concessional public or philanthropic capital to mobilise additional private investment toward development and sustainability goals. By absorbing early-stage or disproportionate risk, the concessional tranche improves the risk-adjusted return of a project enough to attract commercial investors who would not otherwise participate. It is widely applied to fund infrastructure, climate, and social projects in markets perceived as too risky for purely commercial capital.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:blended-finance", "labels": [ "Blended Finance" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "blender-3d-creation-suite", "title": "Blender 3D Creation Suite", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Blender is a free and open-source 3D creation suite supporting modelling, animation, rendering, and related workflows.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:blender-3d-creation-suite", "labels": [ "Blender 3D Creation Suite", "Blender", "Blender Geometry Nodes" ], "is_subclass_of": [ "Open Source Software" ], "wikilinks": [ "Open Source Software", "Animation", "Game Engine" ] }, { "id": "bletchley-declaration", "title": "bletchley declaration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Bletchley Declaration is a multilateral political agreement signed in November 2023 by 28 nations and the European Union at the inaugural AI Safety Summit at Bletchley Park, UK, establishing the first intergovernmental consensus on the risks of frontier AI models. Signatories committed to international cooperation on safety evaluation, information sharing among national AI safety institutes, and the development of governance frameworks to address potentially catastrophic risks from highly capable foundation models. The declaration catalysed the formation of AI safety institutes in the UK, United States, and other signatory nations, and set the precedent for government-coordinated pre-deployment evaluations of frontier AI systems. It was followed by the Seoul AI Safety Summit in 2024, which extended commitments toward operational safety testing requirements for frontier model developers.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:bletchley-declaration", "labels": [ "Bletchley Declaration" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance", "International Agreement", "Policy Instrument" ], "wikilinks": [] }, { "id": "blind-signatures", "title": "Blind Signatures", "domain": "security", "domain_name": "Security", "definition": "Blind Signatures are a cryptographic primitive invented by David Chaum in 1982 that allow a signer to sign a message without being able to see its content, enabling the message author to later unblind the signature and present a valid signature from the signer without the signer being able to link the signing event to the subsequent presentation. The scheme preserves the unlinkability property \u2014 the signer cannot correlate a signing request with a later use of that signature \u2014 making it foundational for privacy-preserving payment systems and anonymous credential issuance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blind-signatures", "labels": [ "Blind Signatures", "Blind Signature", "Blind Signature Scheme", "BlindSignatures", "Chaumian Blind Signature Scheme" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "blob-transaction", "title": "Blob Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blob transaction is an Ethereum transaction type introduced by EIP-4844 (Proto-Danksharding) that carries a large binary payload of 'blob' data alongside the standard transaction fields, priced separately from calldata gas. Blobs are stored by consensus nodes for a limited retention window rather than persisted in perpetuity by the EVM, which keeps the cost of posting rollup data far lower than embedding it as ordinary calldata. Blob transactions are the mechanism by which layer-2 rollups post the data needed for fraud or validity proofs back to layer 1.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:blob-transaction", "labels": [ "Blob Transaction" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "block-cipher", "title": "Block Cipher", "domain": "security", "domain_name": "Security", "definition": "A block cipher is a deterministic symmetric-key encryption algorithm that transforms a fixed-size block of plaintext, such as 128 bits, into a ciphertext block of the same size under a shared secret key, with the same key correctly reversing the transformation. Algorithms such as AES operate this way, and block ciphers are combined with a mode of operation to encrypt messages longer than a single block. As a primitive, a block cipher underlies symmetric encryption schemes and is a common building block for constructing message authentication codes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:block-cipher", "labels": [ "Block Cipher" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "block-explorer", "title": "Block Explorer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Block Explorer is a web-based or API-accessible tool that indexes and presents the contents of a blockchain in human-readable form, enabling users to search, inspect, and verify blocks, transactions, addresses, smart contracts, and network statistics without operating a full node locally. Block explorers ingest raw node data, parse it according to the chain's consensus rules, and expose it through search interfaces and REST or GraphQL APIs. They are the primary transparency and audit interface for blockchain networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:block-explorer", "labels": [ "Block Explorer" ], "is_subclass_of": [ "Blockchain Analytics" ], "wikilinks": [] }, { "id": "block-header", "title": "Block Header", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The fixed-size metadata section of a blockchain block that encodes the previous block hash, Merkle root of transactions, timestamp, difficulty target, and nonce. The block header is the unit that miners and validators hash during proof-of-work or verify during proof-of-stake, and it commits to the complete block body via the Merkle root without including raw transactions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:block-header", "labels": [ "Block Header" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "block-height", "title": "Block Height", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Height is a monotonically increasing integer representing the position of a specific block in a blockchain, defined as the count of confirmed blocks preceding it in the canonical chain (the genesis block has height 0). Block height serves as the primary temporal reference for blockchain state: smart contracts use it for time-locked operations, miners use it to calculate block rewards and halving events, and consensus rules use it to enforce protocol upgrade activation thresholds. In the presence of forks, two competing chains may share block height values; the canonical chain is determined by the fork choice rule applied at the point of divergence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-height", "labels": [ "Block Height", "Block Number Reference", "block-height" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "block-production", "title": "Block Production", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block production is the process by which a blockchain network assembles pending transactions into a new block, orders them, and proposes that block for inclusion in the canonical chain. In proof-of-stake and similar systems, validators are selected to produce blocks for given slots or epochs, executing the consensus protocol that determines who may extend the chain and when. Block production is the heartbeat of a blockchain, governing throughput, latency and the fair ordering of transactions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:block-production", "labels": [ "Block Production" ], "is_subclass_of": [ "Blockchain Consensus" ], "wikilinks": [] }, { "id": "block-propagation-time", "title": "Block Propagation Time", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Propagation Time is the latency metric measuring how long it takes for a newly mined or validated block to be disseminated to all (or a target percentage of) nodes in a blockchain peer-to-peer network after its initial announcement. It is a critical determinant of blockchain security and throughput: long propagation times increase the probability of temporary forks (stale/orphan blocks), waste miner effort, and reduce effective network throughput. Bitcoin historically achieved median propagation times of approximately 1\u20132 seconds to reach 50% of nodes, while Ethereum's uncle mechanism tolerated higher latency by including competing blocks in the canonical chain.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-propagation-time", "labels": [ "Block Propagation Time" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Network Component", "NetworkComponent" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "block-propagation", "title": "Block Propagation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Propagation is the process by which a newly mined or validated block is broadcast across a blockchain peer-to-peer network so that all full nodes can update their local copy of the chain. Propagation latency directly influences the orphan/stale block rate, security against selfish mining, and the degree of centralisation pressure towards large, well-connected mining pools. Techniques such as Compact Block Relay (Bitcoin BIP 152) and Graphene reduce bandwidth requirements by sending block sketches rather than full transaction lists, exploiting the fact that recipient nodes already hold most transactions in their mempools.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-propagation", "labels": [ "Block Propagation" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "block-proposal", "title": "Block Proposal", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block proposal is the step in a blockchain consensus protocol where a designated participant assembles a candidate block of ordered transactions and broadcasts it to the network for validation and agreement. The proposer selects transactions from the mempool, constructs the block header referencing the prior block, and signs the proposal so peers can verify its authorship. In proof-of-stake systems the proposer is chosen by a leader-election procedure for each slot, after which validators attest to the proposed block.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:block-proposal", "labels": [ "Block Proposal" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "block-reward-halving", "title": "Block Reward Halving", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The block reward halving is the protocol rule in Bitcoin and similar proof-of-work chains that cuts the newly minted coin subsidy paid to miners in half at fixed block intervals, roughly every four years for Bitcoin. This geometric reduction enforces a predictable, disinflationary issuance schedule that converges on a fixed total supply of 21 million bitcoin. Halvings are central to Bitcoin's scarcity narrative and have historically been focal points for market and security-budget analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:block-reward-halving", "labels": [ "Block Reward Halving" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "block-reward", "title": "Block Reward", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Block Reward is the cryptoeconomic incentive paid to the producer of a valid block\u2014comprising a protocol-specified subsidy (newly minted tokens) plus the aggregate transaction fees included in that block\u2014which compensates validators or miners for expending resources to extend the canonical chain and maintain network security. Block reward schedules are a core parameter of a blockchain's monetary policy, directly governing token inflation, miner revenue, and the long-run security budget of the network.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-reward", "labels": [ "Block Reward" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure", "Virtual Economy" ] }, { "id": "block-size", "title": "Block Size", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Size is the maximum data capacity\u2014measured in bytes or weight units\u2014that a single block in a blockchain may contain, governing how many transactions can be confirmed per block and therefore setting a fundamental upper bound on network throughput. Larger blocks increase per-block transaction capacity but raise propagation latency, orphan-block rates, and hardware requirements for full nodes, while smaller blocks favour decentralisation and faster propagation at the cost of throughput and higher fee markets when demand exceeds capacity. The Bitcoin block-size debate (resulting in the SegWit upgrade and eventual Bitcoin Cash fork) made block size one of the most consequential and politically contentious protocol parameters in blockchain history.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-size", "labels": [ "Block Size", "Block Size Increase" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "block-snapshot", "title": "Block Snapshot", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A block snapshot is a record of on-chain state, typically token balances or governance weights, captured at a specific block height. In decentralised governance it freezes voting power at a chosen block so that token holdings used to weight votes cannot be manipulated by transfers during the voting window. Snapshots underpin gasless off-chain voting systems and fair airdrop distributions by providing a tamper-evident, reproducible baseline state.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:block-snapshot", "labels": [ "Block Snapshot" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "block-storage", "title": "Block Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Block Storage is an infrastructure storage paradigm that presents raw fixed-size blocks of data to a host operating system or hypervisor, which then manages formatting, file system placement, and I/O scheduling directly. Unlike object storage or file storage, block storage exposes a low-level disk abstraction enabling high-performance, low-latency random read/write operations suitable for databases, virtual machine boot disks, and transactional workloads. Cloud providers implement block storage as network-attached volumes (e.g., AWS EBS, GCP Persistent Disk) that can be dynamically provisioned and attached to compute instances.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:block-storage", "labels": [ "Block Storage" ], "is_subclass_of": [ "Storage Systems" ], "wikilinks": [] }, { "id": "block-structure", "title": "Block Structure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block structure is the internal data layout of a block in a blockchain or distributed ledger, specifying the fields, encoding formats, and cryptographic commitments that constitute a valid unit of the chain. A block consists of a header containing metadata \u2014 including the parent block hash, timestamp, nonce, and Merkle root of the transaction set \u2014 and a body containing the ordered list of transactions or state transitions. The specific fields, size limits, and serialisation rules of the block structure are defined by the network's consensus protocol and directly govern the chain's security properties, throughput, and upgrade path.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:block-structure", "labels": [ "Block Structure" ], "is_subclass_of": [ "Block" ], "wikilinks": [] }, { "id": "block-time", "title": "Block Time", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Time is the average elapsed time between the creation of consecutive blocks on a blockchain, governed by the network's consensus mechanism and difficulty adjustment algorithm. Shorter block times increase transaction throughput and reduce confirmation latency but raise the risk of forks due to block propagation delays, creating a fundamental tradeoff between speed and chain security.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:block-time", "labels": [ "Block Time" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "block-timestamp", "title": "Block Timestamp", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A block timestamp is the time value that a block producer embeds in a blockchain block header, recording when the block was created according to that producer's clock. Because it is set by the block producer rather than a trusted external clock, it is only approximately accurate and is bounded by consensus rules to prevent large manipulation; difficulty adjustment algorithms and timelocks both depend on it. It provides the ordering and timing primitive on which many on-chain time-dependent mechanisms are built.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:block-timestamp", "labels": [ "Block Timestamp" ], "is_subclass_of": [ "Block" ], "wikilinks": [] }, { "id": "block-trails", "title": "Block Trails", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Trails is a minimal Bitcoin-native state-anchoring primitive that records an evolving sequence of states as a cryptographic chain of key tweaks mirroring a Bitcoin spend chain. Each state transition derives a tweak t_i = SHA256(state_i) mod n which is scalar-added to the previous key (d_i = d_(i-1) + t_i), producing a fresh pay-to-Taproot (P2TR) key-path output; spending that output to create the next commitment makes every UTXO a single-use seal. The state bytes themselves live off-chain on IPFS, Git, or Nostr relays, while Bitcoin's UTXO model supplies ordering and double-spend protection so that exactly one valid history can exist. Because trails use full secp256k1 keys, existing Nostr identities can own and advance a trail without key conversion. Application semantics are defined by Profiles such as MRC20 (a fungible-token ledger) and Git-mark (anchoring Git commits).", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:block-trails", "labels": [ "Block Trails", "Block Trails Protocol", "Blocktrails" ], "is_subclass_of": [ "Client-Side Validation" ], "wikilinks": [] }, { "id": "block-validation", "title": "Block Validation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Block Validation is the process by which a blockchain node verifies that a candidate block satisfies all protocol rules before accepting it into the chain. It checks the proof-of-work or consensus proof, the block header structure, the Merkle root, and the validity of every contained transaction including signatures, double-spend constraints, and balances. Successful validation is the precondition for extending the canonical chain and is central to trustless consensus.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:block-validation", "labels": [ "Block Validation" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "block", "title": "Block", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The fundamental unit of a blockchain: a cryptographically linked, immutable data container that batches a set of transactions together with a block header containing the Merkle root of those transactions, a timestamp, the previous block's hash, and consensus-specific fields such as a nonce or validator signature. Sequential blocks form the chain that provides an append-only audit trail.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:block", "labels": [ "Block" ], "is_subclass_of": [ "Distributed Data Structure", "Blockchain Entity", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure", "Telecollaboration" ] }, { "id": "blockchain-analysis", "title": "Blockchain Analysis", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain analysis is the systematic examination of distributed ledger transaction data to identify patterns, trace fund flows, attribute addresses to entities, and detect illicit activity. It applies graph analytics, heuristic clustering, and machine-learning techniques to publicly available on-chain records to support compliance, law enforcement, and financial intelligence functions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blockchain-analysis", "labels": [ "Blockchain Analysis", "Chain Analysis" ], "is_subclass_of": [ "Digital Forensics" ], "wikilinks": [] }, { "id": "blockchain-analytics-platform", "title": "Blockchain Analytics Platform", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain analytics platform is software that ingests, clusters, and analyses on-chain transaction data to trace fund flows, attribute addresses to entities, and assess risk. Platforms such as Chainalysis and Elliptic apply heuristics, machine learning, and labelled datasets to support anti-money-laundering, sanctions screening, and investigations. They are central tooling for exchange compliance teams, regulators, and law-enforcement agencies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-analytics-platform", "labels": [ "Blockchain Analytics Platform" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "blockchain-analytics", "title": "Blockchain Analytics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Analytics is the application of data-science, graph-analysis, and heuristic-clustering techniques to the publicly visible transaction records of distributed ledgers, with the aim of tracing fund flows, identifying entities, detecting illicit activity, and producing compliance intelligence for regulators and financial institutions. It encompasses on-chain transaction monitoring, address clustering, cross-chain tracing, DeFi flow analysis, and risk-scoring of wallets and counterparties. Leading commercial providers include Chainalysis, Elliptic, and TRM Labs, whose tooling has become integral to AML compliance programmes worldwide.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blockchain-analytics", "labels": [ "Blockchain Analytics" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "blockchain-anchoring", "title": "Blockchain Anchoring", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Anchoring is a technique for binding external data or documents to a blockchain by embedding a cryptographic hash of that data in a blockchain transaction, thereby creating a tamper-evident, timestamped proof of existence and integrity that can be independently verified by any party with access to the document and the chain. The blockchain's immutability and distributed consensus guarantee that the anchoring transaction cannot be altered retroactively, providing a trust anchor without requiring the document itself to be stored on-chain. Applications span document notarisation, supply chain provenance, audit logs, and verifiable credential revocation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blockchain-anchoring", "labels": [ "Blockchain Anchoring" ], "is_subclass_of": [ "Blockchain Infrastructure", "Cryptographic Hash Function" ], "wikilinks": [] }, { "id": "blockchain-application", "title": "Blockchain Application", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralized application (dApp) that runs on a blockchain or peer-to-peer network rather than centralized servers, combining smart contract backend logic with frontend interfaces to provide enhanced security, transparency, censorship resistance, and zero downtime through distributed execution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-application", "labels": [ "Blockchain Application" ], "is_subclass_of": [ "Network Component", "Blockchain", "Distributed Application" ], "wikilinks": [ "Aave", "Axie Infinity", "Blur", "Chainlink", "Compound", "Cryptocurrency Wallet", "Decentraland", "Distributed Application", "ENS", "MakerDAO", "OpenSea", "Rarible", "The Graph", "The Sandbox", "Uniswap", "AI Agent System", "Blockchain", "Blockchain Infrastructure", "Consensus Mechanism", "Decentralized Finance (DeFi)" ] }, { "id": "blockchain-as-a-service", "title": "Blockchain As A Service", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain as a Service (BaaS) is a managed cloud delivery model in which a third-party provider provisions, operates, and maintains the distributed ledger infrastructure, consensus-node orchestration, cryptographic identity services, smart contract deployment pipelines, and monitoring tooling re...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:blockchain-as-a-service", "labels": [ "Blockchain As A Service", "BC-0434-blockchain-as-a-service" ], "is_subclass_of": [ "Network Component", "Cloud Service Model", "Managed Service", "Distributed Ledger Technology", "Enterprise Software Platform", "Platform as a Service" ], "wikilinks": [ "Alibaba Cloud BaaS", "Amazon Managed Blockchain", "Androulaki et al. 2018 Hyperledger Fabric EuroSys", "API Management", "Asset Tokenisation", "AWS IAM", "AWS Lambda", "Azure Active Directory", "Azure Blockchain Service", "Bank of England 2023 Digital Pound Consultation CP7/23", "Baseline Protocol", "Bessani Sousa Alchieri 2014 BFT-SMART DSN", "Blockchain API Gateway", "Brown et al. 2016 Corda Introduction R3", "Buterin 2013 Ethereum White Paper", "Casino Dasaklis Patsakis 2019 Systematic Literature Review Blockchain TI", "Castro Liskov 1999 PBFT OSDI", "CBDC", "CBDC Infrastructure", "Central Bank Digital Currencies" ] }, { "id": "blockchain-collaboration", "title": "Blockchain Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The use of blockchain distributed ledger technology, smart contracts, and cryptocurrency systems to coordinate, govern, and compensate geographically distributed teams, enabling trustless collaboration through cryptographic verification, automated enforcement of agreements, and transparent recor...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-collaboration", "labels": [ "Blockchain Collaboration", "TELE-250-blockchain-collaboration" ], "is_subclass_of": [ "Protocol and Infrastructure", "Telecollaboration", "TELE-002-telecollaboration" ], "wikilinks": [ "DecentralisedAutonomousOrganisation", "TELE-002-telecollaboration", "TELE-251-smart-contract-coordination", "TELE-252-dao-governance-telecollaboration", "TELE-253-cryptocurrency-remuneration", "ConsensusProtocol", "SmartContracts", "TransparentDecisionMaking" ] }, { "id": "blockchain-compliance", "title": "Blockchain Compliance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The framework of regulatory requirements, technical controls, and operational procedures ensuring blockchain systems and crypto-asset service providers comply with applicable laws including AML/KYC, GDPR, securities regulations, and jurisdiction-specific mandates while maintaining transparency and security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-compliance", "labels": [ "Blockchain Compliance" ], "is_subclass_of": [ "Governance and Regulation", "Regulatory Compliance" ], "wikilinks": [ "6AMLD (Directive 2015/849)", "AML (Anti-Money Laundering)", "AMLA (EU AML Authority)", "Anti-Money Laundering", "Blockchain Analytics", "CCPA (California Consumer Privacy Act)", "CTA (Corporate Transparency Act)", "CTF (Counter-Terrorist Financing)", "Decentralized Identity", "FinCEN (Financial Crimes Enforcement Network)", "Financial Regulation", "GDPR (General Data Protection Regulation)", "GENIUS Act (2025)", "Institutional Adoption", "Know Your Customer", "KYC (Know Your Customer)", "MiCA (Markets in Crypto-Assets)", "SEC (Securities and Exchange Commission)", "Blockchain", "Identity Verification" ] }, { "id": "blockchain-consensus", "title": "Blockchain Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Consensus refers to the family of distributed agreement protocols by which independent, potentially adversarial nodes in a blockchain network reach agreement on a single canonical version of the transaction ledger, including the ordering, validity, and finality of all blocks. These protocols must tolerate Byzantine faults \u2014 nodes that may behave maliciously or arbitrarily \u2014 and operate across open, permissionless networks without a trusted coordinator. The choice of consensus mechanism fundamentally shapes a blockchain's security model, throughput, energy consumption, and degree of decentralisation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:blockchain-consensus", "labels": [ "Blockchain Consensus", "Blockchain Consensus Protocol", "BlockchainConsensus" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "blockchain-core", "title": "Blockchain Core", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain core denotes the foundational base-layer components of a blockchain system: the consensus protocol, the append-only ledger data structure, the peer-to-peer network, and the validation rules that together produce a shared, tamper-resistant state. It is the substrate on which higher-level constructs such as tokens, wallets, and smart contracts are built. As the trust anchor of a chain, its design dictates the network's security, decentralisation, and throughput.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-core", "labels": [ "Blockchain Core" ], "is_subclass_of": [ "Blockchain", "Blockchain Protocol" ], "wikilinks": [] }, { "id": "blockchain-data", "title": "Blockchain Data", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Data refers to the structured information stored on a distributed ledger, comprising transaction records, state data, smart contract bytecode, event logs, and cryptographic proofs organised into immutable, hash-linked blocks. It is characterised by append-only semantics, cryptographic integrity, and public verifiability. The data model differs fundamentally from traditional databases in that history cannot be altered without recomputing the entire subsequent chain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:blockchain-data", "labels": [ "Blockchain Data" ], "is_subclass_of": [ "Blockchain", "Network Component (Blockchain)" ], "wikilinks": [] }, { "id": "blockchain-document-trail-infrastructure", "title": "Blockchain Document Trail Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A concept node representing resource trails related to blockchain exploration tools and Nostr-based document sharing infrastructure. In the NarrativeGoldmine context, blocktrails links to split-screen markdown editing and decentralised document storage via NosDAV and the Nostr protocol.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-document-trail-infrastructure", "labels": [ "Blockchain Document Trail Infrastructure", "blocktrails" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "blockchain-economics", "title": "Blockchain Economics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The study and design of economic mechanisms governing blockchain networks, including incentive structures for validators, fee markets, token supply schedules, governance models, and the emergent macro-economic properties of decentralised systems. Blockchain economics integrates mechanism design, game theory, and monetary theory to sustain network security and participant alignment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-economics", "labels": [ "Blockchain Economics" ], "is_subclass_of": [ "Token Economics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "blockchain-energy-consumption", "title": "Blockchain Energy Consumption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The electrical power and computational resources consumed by blockchain networks during transaction validation, block creation, and network security operations, varying significantly across consensus mechanisms from energy-intensive Proof-of-Work (Bitcoin: ~140 TWh/year, 0.65% of global electricity) to energy-efficient Proof-of-Stake (Ethereum post-Merge: 99.95% reduction), with ongoing research into sustainable consensus algorithms, renewable energy mining, carbon credit tokenization, and environmental impact measurement frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-energy-consumption", "labels": [ "Blockchain Energy Consumption" ], "is_subclass_of": [ "Governance and Regulation", "EnvironmentalImpact", "BlockchainPerformance", "Sustainability" ], "wikilinks": [ "AI Energy Optimisation", "Bitcoin Mining Council", "BlockchainPerformance", "Cambridge Centre for Alternative Finance", "CarbonEmissions", "ComputationalPower", "ElectricityConsumption", "EnergySector", "EnergySource", "EnvironmentalDomain", "EnvironmentalImpact", "Ethereum Foundation", "HashRate", "IEEE Blockchain Standards", "International Energy Agency (IEA)", "MiningHardware", "NetworkSize", "BlockchainDomain", "ConsensusMechanism", "Sustainability" ] }, { "id": "blockchain-entity", "title": "Blockchain Entity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Blockchain Entity is the foundational abstract conceptual class representing any distinct component, construct, or participant within a blockchain ecosystem.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:blockchain-entity", "labels": [ "Blockchain Entity" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Autonomous Governance", "Cryptographic Key", "Decentralised Coordination", "Distributed Trust", "ISO/IEC 23257:2021", "Network Participation", "NIST Blockchain Technology Overview", "Transparent Audit Trails", "W3C Blockchain Community Group", "AI Agent System", "Block", "Blockchain", "BlockchainDomain", "ConceptualLayer", "Consensus Mechanism", "Cryptographic Security", "Cryptographic System", "Digital Twin", "Distributed Data Structure", "Distributed Ledger" ] }, { "id": "blockchain-environmental-impact-assessment", "title": "Blockchain Environmental Impact Assessment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A systematic emodology for measuring, analyzing, and reporting the environmental consequences of blockchain network operations, encompassing energy consumption measurement (electricity usage per transaction, annual network consumption), carbon emissions calculation (CO\u2082e from electricity generati...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:blockchain-environmental-impact-assessment", "labels": [ "Blockchain Environmental Impact Assessment" ], "is_subclass_of": [ "Governance and Regulation", "EnvironmentalAssessment", "SustainabilityMetric", "BlockchainGovernance", "LifeCycleAssessment" ], "wikilinks": [ "AI Energy Optimisation", "Bitcoin Mining Council", "Cambridge Centre for Alternative Finance (CCAF)", "CarbonEmissionsCalculation", "DataCenterMetrics", "EnergyConsumptionMeasurement", "EnergyMix", "EnvironmentalDomain", "Ethereum Foundation", "EWasteAnalysis", "HardwareInventory", "International Energy Agency (IEA)", "InvestorDisclosure", "ISO 14040 Life Cycle Assessment", "LifeCycleAssessment", "ScopeDefinition", "SustainabilityMetric", "SustainabilityReporting", "WaterUsageTracking", "BlockchainDomain" ] }, { "id": "blockchain-gaming", "title": "Blockchain Gaming", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A gaming paradigm in which in-game assets, economies, and ownership records are managed on distributed ledgers, enabling players to hold verifiable ownership of digital items as NFTs, participate in play-to-earn economies, and trade assets across compatible platforms without relying on centralised game servers for provenance or scarcity guarantees.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-gaming", "labels": [ "Blockchain Gaming", "Gaming Asset Ownership", "Gaming Assets", "Gaming NFTs" ], "is_subclass_of": [ "Decentralised Application" ], "wikilinks": [ "owl:Thing" ] }, { "id": "blockchain-governance", "title": "Blockchain Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Governance refers to the mechanisms, processes, and structures through which decisions are made about the development, operation, and evolution of blockchain networks and protocols, encompassing on-chain voting systems, decentralised autonomous organisations, and community-driven decision-making frameworks.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:blockchain-governance", "labels": [ "Blockchain Governance", "Blockchain Governance Framework", "BlockchainGovernance" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Entity", "Decentralized Organization" ], "wikilinks": [ "Blockchain Technology", "Collective Intelligence", "Community Alignment", "DAO Structure", "Decentralized Decision-Making", "Decentralized Organization", "Protocol Evolution", "Token-Based Representation", "Voting Mechanism", "Blockchain Entity", "On-Chain Voting", "Proposal System", "Smart Contract", "Smart Contracts", "Telecollaboration", "Token" ] }, { "id": "blockchain-identity", "title": "Blockchain Identity", "domain": "security", "domain_name": "Security", "definition": "Blockchain Identity is an approach to digital identity in which identifiers, credentials and attestations are anchored to a distributed ledger rather than a single central authority. It gives users cryptographic control over their identity through key pairs, enables verifiable claims that any party can check against on-chain or anchored data, and underpins self-sovereign identity models. By decentralising the registry of identifiers, it reduces reliance on intermediaries and supports portable, tamper-evident identity across services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-identity", "labels": [ "Blockchain Identity" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [] }, { "id": "blockchain-infrastructure", "title": "Blockchain Infrastructure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The foundational technical components enabling blockchain networks to operate, including nodes, networking protocols, consensus mechanisms, storage systems, and cryptographic primitives that together provide the physical and logical substrate for distributed ledger operation and decentralized application execution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:blockchain-infrastructure", "labels": [ "Blockchain Infrastructure" ], "is_subclass_of": [ "Blockchain", "Network Component", "Distributed Systems Infrastructure" ], "wikilinks": [ "AI Energy Optimisation", "Blockchain Node", "Distributed Systems Infrastructure", "Layer 2 Scaling", "Web3 Infrastructure", "Blockchain", "Blockchain Application", "Consensus Mechanism", "Cross-Chain Bridge", "Cryptography", "Peer-to-Peer Network", "Smart Contracts" ] }, { "id": "blockchain-interoperability", "title": "Blockchain Interoperability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Interoperability is the technical and economic capability for distinct, sovereign blockchain networks to securely communicate state, transfer assets, invoke smart contracts, and propagate consensus across heterogeneous trust domains without relinquishing each chain's independent finali...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:blockchain-interoperability", "labels": [ "Blockchain Interoperability", "BC-0440-blockchain-interoperability" ], "is_subclass_of": [ "Network Component", "Cross-Chain Communication", "Distributed Systems Protocol", "Cryptographic Protocol", "Network Interoperability", "Blockchain Infrastructure" ], "wikilinks": [ "Across Protocol", "Al-Bassam Sonnino Buterin 2018 Fraud Data Availability Proofs", "AML (Anti-Money Laundering)", "Arbitrum Orbit", "Axelar", "Axelar Cross-Chain Communication Whitepaper", "Belchior et al 2021 Blockchain Interoperability Survey", "Bhuptani 2021 Interoperability Trilemma", "Blockchain Consensus", "Bridge Contract", "Buchman 2016 Tendermint BFT MSc Thesis", "Burn-and-Mint Bridge", "Buterin 2022 Endgame of Cross-Chain Bridges", "CBDC Cross-Border Settlement", "Celestia", "Centralised Exchange Bridging", "Chainalysis 2024 Crypto Crime Report Bridges", "Chainlink CCIP", "Chainlink CCIP Architecture", "Consensus Finality" ] }, { "id": "blockchain-ledger", "title": "Blockchain Ledger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Blockchain Ledger is the authoritative, append-only register of all validated transactions and state changes maintained collectively by participants in a blockchain network, representing the shared ground truth from which account balances, ownership records, and contract states are derived. Unlike a traditional centralised ledger maintained by a single institution, a blockchain ledger is replicated across potentially thousands of nodes, with consensus rules ensuring that all honest participants converge on the same view of history. Its integrity derives from cryptographic chaining rather than institutional trust.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blockchain-ledger", "labels": [ "Blockchain Ledger", "BlockchainLedger" ], "is_subclass_of": [ "Distributed Ledger" ], "wikilinks": [] }, { "id": "blockchain-network", "title": "Blockchain Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Blockchain Network is a permissionless or permissioned peer-to-peer (P2P) overlay network through which participating nodes collectively maintain, validate, and propagate a shared append-only ledger (the blockchain) without relying on any central coordinator, achieving agreement through a deter...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:blockchain-network", "labels": [ "Blockchain Network", "Blockchain Network Data", "BlockchainNetwork" ], "is_subclass_of": [ "Network Component", "Peer to Peer Network", "Distributed System", "Fault Tolerant System", "Overlay Network", "Byzantine Fault Tolerant System" ], "wikilinks": [ "Apostolaki et al 2017 BGP Hijacking Bitcoin", "Archive Node", "Bagaria et al 2019 Prism", "Bitcoin Improvement Proposals", "Bitnodes 2025 Bitcoin Node Statistics", "Bittensor", "Blockdaemon 2025 Ethereum Exit Queue", "Buterin Griffith 2017 Casper FFG", "Byzantine Fault Tolerant System", "Cambridge CCAF 2025 Cryptoasset Benchmarking", "Casper FFG", "Castro Liskov 1999 PBFT", "Centralised Database", "Client-Server Architecture", "ConsensusLayer", "Cosmos", "Cryptographic Hash Function", "Decker Wattenhofer 2013 Bitcoin Propagation", "devP2P", "discv5" ] }, { "id": "blockchain-node", "title": "Blockchain Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain node is a computer running the protocol software of a blockchain network that maintains a copy of the ledger, validates transactions and blocks, and relays them to peers. Nodes collectively enforce the consensus rules and provide the decentralisation and redundancy that make the network trustworthy. They range from full nodes that store and verify the complete chain to light clients that verify selectively, and validator nodes that additionally participate in block production.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:blockchain-node", "labels": [ "Blockchain Node" ], "is_subclass_of": [ "Blockchain", "Network Component" ], "wikilinks": [] }, { "id": "blockchain-oracle", "title": "Blockchain Oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralized service that bridges blockchain smart contracts with external real-world data, enabling smart contracts to access off-chain information, execute based on real-world events, and interact with traditional systems while maintaining trustless verification and data integrity.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:blockchain-oracle", "labels": [ "Blockchain Oracle", "Blockchain Oracle Integration" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Aave", "Aavegotchi", "AI", "AI-Enhanced Oracles", "API3", "API3 DAO", "API3 token", "Arbitrum", "Atomic Finance", "Augur", "Avalanche", "AWS IoT", "Axie Infinity", "Band Protocol", "BAND token", "Binance Research", "binary options", "Bitcoin Oracle Expansion", "blockchains", "BNB Chain" ] }, { "id": "blockchain-process", "title": "Blockchain Process", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Process encompasses the operational workflows and procedural mechanisms that govern blockchain network functioning, including consensus execution, block validation, transaction verification, mempool management, state transitions, and protocol upgrade procedures. These processes collectively maintain network integrity and enable trustless distributed agreement.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-process", "labels": [ "Blockchain Process" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Blockchain" ] }, { "id": "blockchain-protocol", "title": "Blockchain Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Foundational rule set governing distributed ledger system operation comprising consensus mechanisms (algorithmic procedures enabling network-wide agreement on canonical state without centralized authority including Proof of Work PoW SHA-256 hash puzzles requiring computational expenditure Bitcoin...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:blockchain-protocol", "labels": [ "Blockchain Protocol", "Blockchain Protocol Foundation" ], "is_subclass_of": [ "Protocol and Consensus", "Peerto Peer Protocol", "Distributed System Protocol", "Consensus Algorithm", "Peer-to-Peer Protocol", "Cryptographic Protocol", "Network Protocol", "State Machine Replication" ], "wikilinks": [ "BFT Consensus", "Bitcoin Improvement Proposals BIP", "Bitcoin Whitepaper Nakamoto 2008", "Block Structure", "Block Validation", "Byzantine Generals Problem Lamport 1982", "CAP Theorem Brewer", "Cosmos IBC Protocol", "Cross-Chain Communication", "Cross-Chain Interoperability", "Cryptographic Primitives", "DeFi Applications", "DeFi Protocol Standards", "Decentralized Consensus", "Digital Signature Scheme", "Double-Spend Prevention", "Economic Incentives", "Economic Model", "Enterprise Ethereum Alliance", "Ethereum Improvement Proposals EIP" ] }, { "id": "blockchain-provenance", "title": "Blockchain Provenance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Provenance is the application of distributed ledger technology to record and verify the complete historical lineage of an asset, document, or data artefact, such that its origin, custody chain, and transformations are cryptographically authenticated and tamper-resistant. Each provenance event is anchored as an immutable transaction on-chain, enabling any party to independently reconstruct and audit the full lifecycle of the item without relying on a central authority.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-provenance", "labels": [ "Blockchain Provenance", "Blockchain Provenance Tracking" ], "is_subclass_of": [ "Blockchain Technology" ], "wikilinks": [] }, { "id": "blockchain-scalability", "title": "Blockchain Scalability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Scalability encompasses the technical solutions and architectural approaches designed to increase the transaction throughput, reduce latency, and improve the efficiency of blockchain networks while maintaining security and decentralisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:blockchain-scalability", "labels": [ "Blockchain Scalability", "Blockchain Scalability Trilemma", "BlockchainScalability" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "System Performance", "Protocol Architecture" ], "wikilinks": [ "AI Energy Optimisation", "Blockchain Technology", "High Throughput", "Layer 2 Solution", "Low Latency", "Mass Adoption", "Protocol Architecture", "Rollup", "Rollups", "Sharding", "System Performance", "Blockchain Entity", "Consensus Mechanism", "Data Compression", "Network Infrastructure", "Smart Contract", "State Channel" ] }, { "id": "blockchain-security", "title": "Blockchain Security", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain security is the multidisciplinary field concerned with protecting distributed ledger systems against attacks on their consensus mechanisms, smart contract logic, cryptographic primitives, and network topology, while preserving the properties of immutability, censorship resistance, and trustless operation. It encompasses threat modelling, formal verification of on-chain code, cryptographic auditing, and economic game-theory analysis to prevent incentive manipulation. The discipline extends across permissionless and permissioned blockchain architectures, addressing layer-specific attack surfaces from peer-to-peer networking through execution environments to cross-chain interoperability bridges.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:blockchain-security", "labels": [ "Blockchain Security", "Blockchain Security Audit", "Blockchain Security Model", "Blockchain-Security", "Chain Security" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "blockchain-sustainability", "title": "Blockchain Sustainability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The practice of designing, operating, and evolving blockchain networks to minimize environmental impact through energy-efficient consensus mechanisms, renewable energy integration, and carbon offset strategies, encompassing metrics, governance, and cross-sector environmental accountability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-sustainability", "labels": [ "Blockchain Sustainability" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain" ], "wikilinks": [ "AI Energy Optimisation", "Carbon Credit", "ESG Compliance", "Green Finance", "Blockchain", "Proof-of-Stake", "Proof-of-Work" ] }, { "id": "blockchain-technology-laboratory", "title": "Blockchain Technology Laboratory", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An academic or institutional research unit dedicated to the study, prototyping and evaluation of distributed ledger systems and their applications. Such laboratories produce experimental implementations, measurements and analyses rather than production services.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:blockchain-technology-laboratory", "labels": [ "Blockchain Technology Laboratory" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus", "Blockchain Domain" ], "wikilinks": [ "Distributed Ledger Technology", "Consensus Mechanism", "Blockchain", "Scalability", "Blockchain Domain", "https://www.bitcoin.com/" ] }, { "id": "blockchain-technology", "title": "Blockchain Technology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain Technology is a class of distributed ledger systems in which validated transactions are grouped into blocks that are cryptographically linked in an append-only sequence, maintained by a peer-to-peer network through a consensus mechanism. Each block header includes the cryptographic hash of its predecessor, a timestamp, and a Merkle root of its transaction set, ensuring that altering any historical record requires re-computing all subsequent proofs, which is computationally or economically prohibitive. The design eliminates the need for a trusted central authority by replacing it with algorithmic agreement, enabling trustless settlement, programmable value transfer via smart contracts, and tamper-evident audit trails across diverse application domains.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:blockchain-technology", "labels": [ "Blockchain Technology", "BlockchainTechnology" ], "is_subclass_of": [ "Distributed Ledger Technology" ], "wikilinks": [] }, { "id": "blockchain-transaction", "title": "Blockchain Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Blockchain Transaction is an atomic, cryptographically signed instruction that encodes a state-change on a distributed ledger \u2014 such as a transfer of digital assets, invocation of a smart contract function, or mutation of on-chain data. Transactions are broadcast to a peer-to-peer network, validated against protocol rules and the active consensus mechanism, and permanently recorded in an ordered block once accepted. The transaction model differs fundamentally between UTXO-based chains (e.g. Bitcoin) and account-based chains (e.g. Ethereum), affecting parallelism, privacy, and composability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:blockchain-transaction", "labels": [ "Blockchain Transaction", "Low-Cost Blockchain Transaction" ], "is_subclass_of": [ "Blockchain", "Network Component (Blockchain)" ], "wikilinks": [ "IEEE (Institute of Electrical and Electronics Engineers)", "Blockchain" ] }, { "id": "blockchain-wallet", "title": "Blockchain Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain wallet is software or hardware that manages the cryptographic keys used to control assets and identities on a blockchain. Rather than storing assets directly, a wallet stores private keys, derives public addresses, and signs transactions that authorise the movement of on-chain assets. Wallets range from custodial services that hold keys on a user's behalf to self-custody solutions where the user retains exclusive control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:blockchain-wallet", "labels": [ "Blockchain Wallet" ], "is_subclass_of": [ "Key Management" ], "wikilinks": [] }, { "id": "blockchain", "title": "Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain is a distributed, cryptographically-secured data structure consisting of an ordered chain of blocks, where each block contains a cryptographic hash of the previous block, a timestamp, and transaction data, maintained through a consensus mechanism across a peer-to-peer network without...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:blockchain", "labels": [ "Blockchain", "BC-0001-blockchain", "Blockchain Integration", "Blockchain Recording", "Programmable Blockchain", "Transparent Blockchain" ], "is_subclass_of": [ "Thing", "Distributed Data Structure", "Distributed Ledger", "Cryptographic System" ], "wikilinks": [ "AI Energy Optimisation", "Cryptographic Hash Function", "Decentralised Finance", "Distributed Consensus", "ISO/IEC 23257:2021", "ISO/IEC 23455:2019", "ISO/IEC TR", "ITU-T Y.4460", "Supply Chain Transparency", "AI Agent System", "Block", "BlockchainDomain", "ConceptualLayer", "Consensus Mechanism", "Cryptocurrency", "Cryptographic System", "Digital Asset", "Digital Twin", "Distributed Data Structure", "Distributed Ledger" ] }, { "id": "blockstream", "title": "Blockstream", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockstream is a Bitcoin-focused technology company founded in 2014 that develops open-source infrastructure and commercial products for the Bitcoin ecosystem, including the Liquid Network federated sidechain, c-lightning (now Core Lightning), and satellite-based blockchain distribution via Blockstream Satellite. The company is led by co-founders including Adam Back (inventor of Hashcash) and is a major contributor to Bitcoin Core protocol development. Blockstream's work spans cryptographic research, Layer 2 payment channels, hardware security modules, and institutional-grade Bitcoin financial products such as Blockstream AMP (Asset Management Platform).", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:blockstream", "labels": [ "Blockstream" ], "is_subclass_of": [ "Blockchain Infrastructure", "Network Component (Blockchain)" ], "wikilinks": [ "Bitcoin", "Lightning Network", "Layer 2 Scaling" ] }, { "id": "bloom-filter", "title": "Bloom Filter", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Bloom Filter is a space-efficient probabilistic data structure that tests whether an element is a member of a set, accepting a controllable false-positive rate while guaranteeing zero false negatives. Invented by Burton Howard Bloom in 1970, the structure uses multiple hash functions to map elements to bit positions within a fixed-size bit array. Membership queries are answered in constant time regardless of set size, making Bloom Filters indispensable in high-throughput systems where exact lookup is prohibitively expensive. They are widely deployed in databases, networking, distributed caches, and blockchain nodes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bloom-filter", "labels": [ "Bloom Filter" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "bluetooth-le", "title": "Bluetooth LE", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Bluetooth LE (Low Energy), also known as Bluetooth Smart, is a short-range wireless communication protocol introduced in Bluetooth 4.0 (2010) and standardised by the Bluetooth SIG, designed to transmit small amounts of data at low duty cycles while consuming a fraction of the power of Classic Bluetooth. It operates in the 2.4 GHz ISM band with 40 channels, using frequency-hopping spread spectrum, and is optimised for battery-constrained devices that need to remain active for months or years on a coin cell.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bluetooth-le", "labels": [ "Bluetooth LE" ], "is_subclass_of": [ "Bluetooth Low Energy" ], "wikilinks": [] }, { "id": "bluetooth-low-energy", "title": "Bluetooth Low Energy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Bluetooth Low Energy (BLE) is the formal designation of the low-power wireless communication mode introduced in the Bluetooth 4.0 Core Specification (2010), standardised by the Bluetooth SIG, enabling battery-constrained devices to communicate over short ranges (typically 10\u2013100 m) using duty-cycled radio bursts with peak currents measured in milliamps. It defines a complete protocol stack from physical layer through application profiles, and has become the dominant short-range radio for wearable, medical, industrial, and IoT devices.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:bluetooth-low-energy", "labels": [ "Bluetooth Low Energy" ], "is_subclass_of": [ "Internet of Things" ], "wikilinks": [] }, { "id": "bluetooth-sig", "title": "Bluetooth SIG", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Bluetooth Special Interest Group (Bluetooth SIG) is the non-profit trade association responsible for developing, maintaining, and licensing the Bluetooth wireless communication standard. Founded in 1998 by Ericsson, IBM, Intel, Nokia, and Toshiba, it governs the Bluetooth Core Specification and the library of Bluetooth profiles, managing qualification and branding programmes that ensure interoperability among the over 50,000 member companies worldwide.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bluetooth-sig", "labels": [ "Bluetooth SIG" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "bluetooth", "title": "Bluetooth", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A short-range wireless communication standard for exchanging data between devices over radio frequencies in the 2.4 gigahertz band. It is widely used for peripherals, audio, and low-power sensors.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:bluetooth", "labels": [ "Bluetooth" ], "is_subclass_of": [ "Network and Communication", "owl:Thing" ], "wikilinks": [ "Bluetooth Low Energy", "Internet of Things", "owl:Thing" ] }, { "id": "board-level-oversight", "title": "Board-Level Oversight", "domain": "governance", "domain_name": "Governance", "definition": "Board-Level Oversight refers to the governance responsibility of a corporate or institutional board of directors to monitor, evaluate, and guide the organisation's strategic direction, risk appetite, executive conduct, and compliance posture. It encompasses the board's duty to act as an informed check on executive management, reviewing material risks\u2014including operational, financial, legal, reputational, and increasingly technology risks such as AI adoption\u2014on behalf of shareholders and other stakeholders.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:board-level-oversight", "labels": [ "Board-Level Oversight" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "boardroom", "title": "Boardroom", "domain": "governance", "domain_name": "Governance", "definition": "The Boardroom is the physical or virtual environment in which a company's board of directors convenes to exercise its governance responsibilities, encompassing both the formal meeting space and the institutional setting\u2014including procedures, information flows, and power dynamics\u2014that shape board decision-making. As a governance construct, it represents the apex deliberative forum where strategy, risk, and oversight responsibilities are discharged.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:boardroom", "labels": [ "Boardroom" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "bode-plot", "title": "Bode Plot", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Bode Plot is a pair of frequency-domain graphs used in control engineering and signal analysis to characterise the frequency response of a linear time-invariant system: one graph plots the magnitude of the system's transfer function in decibels against log-frequency, and the second plots the phase angle in degrees against log-frequency. Together they reveal gain margins, phase margins, bandwidth, and resonant behaviour, making Bode plots the primary graphical tool for assessing open-loop stability and designing compensators.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bode-plot", "labels": [ "Bode Plot" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "boosting", "title": "Boosting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A sequential ensemble learning technique that combines multiple weak learners into a strong predictor by iteratively training each new model to correct the errors of its predecessors. Instance weights are adjusted after each round so that misclassified examples receive more attention; the final prediction is a weighted vote across all weak learners. Key algorithms include AdaBoost, Gradient Boosting, XGBoost, LightGBM, and CatBoost.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:boosting", "labels": [ "Boosting", "Gradient Boosting" ], "is_subclass_of": [ "AI Technique", "Ensemble Methods" ], "wikilinks": [ "CatBoost", "LightGBM", "XGBoost", "Accuracy", "Bias", "Ensemble Methods" ] }, { "id": "bootstrap-node", "title": "Bootstrap Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bootstrap Node is a well-known, stable network entry point that newly joining peers contact to obtain their initial list of active participants in a peer-to-peer blockchain network. By providing a curated, long-lived set of peer addresses, bootstrap nodes solve the cold-start problem: without them a new client would have no means of discovering the network. They do not typically validate or store blocks themselves; their function is purely topological\u2014seeding new participants into the gossip network so that subsequent peer discovery can proceed organically.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:bootstrap-node", "labels": [ "Bootstrap Node" ], "is_subclass_of": [ "Blockchain Entity" ], "wikilinks": [ "AI Energy Optimisation", "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "boston-dynamics-spot", "title": "Boston Dynamics Spot", "domain": "robotics", "domain_name": "Robotics", "definition": "Spot is a quadruped mobile robot developed by Boston Dynamics, designed for inspection and data collection across varied terrain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:boston-dynamics-spot", "labels": [ "Boston Dynamics Spot" ], "is_subclass_of": [ "Mobile Robotics", "Robot Type" ], "wikilinks": [ "Legged Locomotion", "SLAM", "Robot Perception", "Sensors", "Mobile Robotics", "https://bostondynamics.com/products/spot/", "https://dev.bostondynamics.com/" ] }, { "id": "bounded-rationality", "title": "Bounded Rationality", "domain": "ai", "domain_name": "Ai", "definition": "Bounded rationality is the principle that decision-makers are constrained by limited information, finite cognitive resources and time pressure, and therefore seek satisfactory rather than optimal outcomes. It contrasts with the idealised fully rational agent of classical economics and informs models of human and artificial decision behaviour. The concept grounds heuristics, satisficing and resource-rational approaches to reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bounded-rationality", "labels": [ "Bounded Rationality" ], "is_subclass_of": [ "Decision Making", "Behavioural Economics", "Cognitive Science" ], "wikilinks": [] }, { "id": "bounding-box-regression", "title": "Bounding Box Regression", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Bounding Box Regression is the supervised learning task within object detection of predicting the precise coordinates of axis-aligned rectangular boxes that tightly enclose detected objects. A model outputs four continuous values \u2014 typically centre x, centre y, width, and height relative to an anchor \u2014 and is trained using smooth-L1 or IoU-based loss functions that penalise deviation from ground-truth boxes. It is jointly trained with an object classification head and combined with [[Non Maximum Suppression]] to produce the final detections. Accurate regression is critical for downstream tasks such as instance segmentation and 3-D pose estimation that consume the predicted boxes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bounding-box-regression", "labels": [ "Bounding Box Regression" ], "is_subclass_of": [ "Object Detection", "Regression", "Computer Vision Task", "Supervised Learning" ], "wikilinks": [] }, { "id": "bounding-volume-hierarchy", "title": "Bounding Volume Hierarchy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A tree-structured acceleration data structure that recursively organizes geometric objects within nested bounding volumes, enabling efficient spatial queries, collision detection, and ray-scene intersection testing by rapidly culling large portions of geometry that cannot intersect with a query.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:bounding-volume-hierarchy", "labels": [ "Bounding Volume Hierarchy" ], "is_subclass_of": [ "Display and Rendering", "Spatial Data Structure" ], "wikilinks": [ "Bounding Volume", "Frustum Culling", "Traversal Algorithm", "Tree Construction", "Collision Detection", "metaverse", "Ray Tracing", "Spatial Data Structure" ] }, { "id": "bounding-volume", "title": "Bounding Volume", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Bounding Volume is a simple geometric shape\u2014typically a sphere, axis-aligned bounding box (AABB), oriented bounding box (OBB), or convex hull\u2014that encloses a more complex geometric object or set of objects. By testing intersections or containment against the bounding volume rather than the full geometry, real-time rendering engines, physics simulators, and spatial query systems achieve orders-of-magnitude speedups during broad-phase culling and collision detection.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bounding-volume", "labels": [ "Bounding Volume", "Axis-Aligned Bounding Box" ], "is_subclass_of": [ "Spatial Data Structure" ], "wikilinks": [] }, { "id": "bow-shock", "title": "Bow Shock", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:bow-shock", "labels": [ "Bow Shock" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "bradley-terry-model", "title": "Bradley Terry Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Bradley-Terry model is a probabilistic model for paired comparisons that estimates a latent strength or quality score for each item and predicts the probability that one item beats another via the logistic of their score difference. Fitted by maximum-likelihood from observed comparison outcomes, it produces a global ranking from local pairwise data. In machine learning it underpins preference modelling and reward learning from human comparisons.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:bradley-terry-model", "labels": [ "Bradley Terry Model", "Bradley-Terry Model" ], "is_subclass_of": [ "Machine Learning", "Reward Modelling", "Probabilistic Model", "Statistical Model" ], "wikilinks": [] }, { "id": "brain-computer-interfaces", "title": "Brain Computer Interfaces", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Brain-Computer Interfaces (BCIs), also termed brain-machine interfaces (BMIs) or direct neural interfaces, are biomedical and neurotechnological systems establishing a direct, real-time bidirectional or unidirectional communication channel between the central nervous system (predominantly the cer...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:brain-computer-interfaces", "labels": [ "Brain Computer Interfaces", "Brain-Computer Interface", "Brain-Computer Interfaces" ], "is_subclass_of": [ "Computing and Cloud", "Neural Interface", "Biomedical Device", "Human-Computer Interface", "Assistive Technology", "Neuroprosthetic System" ], "wikilinks": [ "AccessibilityDomain", "Action Potential Recording", "Aflalo et al. 2015 Posterior Parietal Cortex BCI", "ALS Communication", "Assistive Technology", "Battery or Inductive Power Link", "BCI Society", "Biocompatible Encapsulation", "Biocompatible Engineering", "Biocompatible Materials", "Biomedical Device", "BiomedicalEngineeringDomain", "Bouton et al. 2016 NeuroLife Restoring Cortical Control", "Card et al. 2024 BrainGate Self-Paced NEJM", "ClinicalLayer", "Clinical Trial Infrastructure", "Closed-Loop Neuromodulation", "Cognitive Augmentation", "Cognitive Enhancement", "Collinger et al. 2013 Pittsburgh Neuroprosthetic Lancet" ] }, { "id": "branch-and-bound", "title": "Branch and Bound", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An exact algorithmic paradigm for combinatorial optimisation that systematically explores a tree of candidate solution subsets, using bounds from relaxations of the problem to prune any branch that provably cannot contain a solution better than the best one found so far. Introduced by Land and Doig in 1960 for integer programming, branch and bound guarantees optimality while often avoiding exhaustive enumeration, and remains the backbone of modern mixed-integer programming and constraint solvers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:branch-and-bound", "labels": [ "Branch and Bound" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [ "Search Algorithm", "Combinatorial Optimisation", "Constraint Solver", "Constraint Satisfaction", "Dynamic Programming" ] }, { "id": "brand-identity", "title": "Brand Identity", "domain": "data", "domain_name": "Data", "definition": "Brand identity is the collection of visible and conceptual elements through which an organisation presents itself and is recognised, including its name, logo, colour palette, typography, voice, and the values it communicates. It is the deliberately constructed expression of a brand that aims to shape how audiences perceive and remember it, distinguishing it from competitors. A coherent brand identity aligns visual design, messaging, and behaviour so that experiences across channels reinforce a consistent and trusted impression.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:brand-identity", "labels": [ "Brand Identity" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [] }, { "id": "breadth-first-search", "title": "Breadth-First Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Breadth-first search (BFS) is a graph and tree traversal algorithm that explores all neighbours of a node before moving to nodes at the next depth level, expanding the search frontier in order of increasing distance from the source. It is implemented with a first-in-first-out queue and, on unweighted graphs, finds the shortest path in terms of edge count from the start vertex to every reachable vertex. BFS runs in time linear in the number of vertices and edges and is foundational to pathfinding, connectivity analysis and level-order processing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:breadth-first-search", "labels": [ "Breadth-First Search" ], "is_subclass_of": [ "Graph Algorithms", "Search Algorithm", "Graph Search", "Uninformed Search" ], "wikilinks": [] }, { "id": "breakout-room", "title": "Breakout Room", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Breakout rooms are temporary, parallel sub-session spaces embedded within a primary virtual meeting or conferencing environment that partition a large group of participants into smaller, purpose-bounded clusters for focused collaborative work, deliberation, or learning activities, enabling concur...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:breakout-room", "labels": [ "Breakout Room", "Sub-room, Small Group Session, Syndicate Room, Parallel Session, Group Discussion Space, Zoom Room, Teams Breakout, Workshop Room, Virtual Syndicate" ], "is_subclass_of": [ "Communication Technology", "Virtual Meeting Structure", "Group Facilitation Tools", "Synchronous Communication", "Collaborative Learning Infrastructure", "Remote Work Technology" ], "wikilinks": [ "AI Group Formation", "Asynchronous Discussion Forum", "Automatic Grouping", "Broadcast Webinar", "Chat Messaging", "Collaborative Learning Infrastructure", "Conference Facilitation", "Cooperative Learning", "Corporate Training", "Cross-functional Team Collaboration", "DistributedCollaborationDomain", "Distributed Team Collaboration", "EducationalTechnologyDomain", "Fishbowl Dialogue", "Flipped Classroom", "Focused Sub-team Discussions", "Group Facilitation Tools", "Hackathon Events", "Host Broadcast Channel", "Hosting Privileges" ] }, { "id": "breez", "title": "Breez", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Breez is a company and open-source platform providing a non-custodial [[Lightning Network]] wallet and a [[Software Development Kit]] (SDK) that enables developers to embed Bitcoin Lightning payments into applications without managing node infrastructure. Breez operates as a Lightning Service Provider (LSP), handling inbound liquidity and channel management on behalf of users while users retain sole custody of their private keys. The Breez SDK abstracts away the complexity of [[Payment Channel]] lifecycle management, splicing, and [[Liquidity]] provisioning, exposing a clean API for instant, low-fee Bitcoin payments. Its architecture is grounded in the [[LNURL]] protocol suite and [[Greenlight]] node-as-a-service infrastructure from Blockstream.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:breez", "labels": [ "Breez" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Bitcoin", "Payment Channel", "Self-Custody", "Lightning Network" ] }, { "id": "bridge-contract", "title": "Bridge Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bridge Contract is a smart contract deployed on one or more blockchain networks that facilitates the transfer of tokens, messages, or state between two separate chains by locking or burning assets on the source chain and minting or releasing equivalent assets on the destination chain. Bridge contracts are the on-chain components of cross-chain bridge protocols, enforcing custody, verifying cryptographic proofs of source-chain events, and coordinating with off-chain relayers or decentralised oracle networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:bridge-contract", "labels": [ "Bridge Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "bridge", "title": "Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A protocol that transfers assets or messages between separate ledgers by locking, burning, or attesting value on one chain and reproducing a representation on another.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bridge", "labels": [ "Bridge", "Blockchain Bridge", "Intent-Based Bridge" ], "is_subclass_of": [ "Blockchain Interoperability" ], "wikilinks": [ "Smart Contract", "Cross Chain Asset Transfer", "Cross-Chain Bridge", "Cross-Chain Messaging", "Blockchain Interoperability" ] }, { "id": "bright-id", "title": "BrightID", "domain": "blockchain", "domain_name": "Blockchain", "definition": "BrightID is a decentralised social identity network that establishes the uniqueness of human participants by analysing the topology of a peer-to-peer social connection graph, without collecting personally identifying information. Each user creates an account identified solely by a public key, then builds mutual verification links with trusted contacts; graph-analysis algorithms infer whether an account corresponds to a distinct human rather than a bot or duplicate. The system provides applications with a binary or graded uniqueness signal usable for Sybil-resistant resource allocation, quadratic voting, and universal basic income distribution without requiring biometric data or a central identity authority.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:bright-id", "labels": [ "BrightID" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [ "Sybil Resistance", "Reputation System", "Digital Identity", "Sybil Attack" ] }, { "id": "brightness-temperature", "title": "Brightness Temperature", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:brightness-temperature", "labels": [ "Brightness Temperature" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "broadband-connectivity", "title": "Broadband Connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Broadband Connectivity is high-capacity, always-on access to the internet delivered over fixed or wireless infrastructure such as fibre, cable, DSL, or cellular networks, characterised by data rates substantially higher than legacy dial-up. It provides the underlying transport for digital services, remote work, streaming, cloud applications, and the Internet of Things. Universal broadband is treated as essential infrastructure, and disparities in its availability constitute the digital divide. Its delivery depends on telecommunications infrastructure, network protocols, and last-mile access technologies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:broadband-connectivity", "labels": [ "Broadband Connectivity" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "broadcast-production", "title": "Broadcast Production", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Broadcast Production is the end-to-end process of creating, processing, and distributing audio-visual content for transmission to audiences over television, radio, or internet broadcast channels, encompassing pre-production planning, live or recorded acquisition, signal processing, editing, graphics integration, encoding, and playout to distribution networks. It operates under stringent real-time constraints and technical quality standards defined by regulatory and industry bodies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:broadcast-production", "labels": [ "Broadcast Production" ], "is_subclass_of": [ "Production Pipeline" ], "wikilinks": [] }, { "id": "broadcast-television", "title": "Broadcast Television", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Broadcast television is the distribution of moving-image and audio content to a general audience over terrestrial, satellite, or cable channels, in which a single signal is transmitted simultaneously to many passive receivers. It evolved from analogue standards such as PAL and NTSC to digital systems like DVB and ATSC that carry compressed video over the same spectrum. As a one-to-many medium with scheduled programming, it contrasts with on-demand internet streaming.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:broadcast-television", "labels": [ "Broadcast Television" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "broadcasting", "title": "Broadcasting", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The one-to-many distribution of a signal or message from a single source to all receivers within reach, in contrast to unicast (one-to-one) and multicast (one-to-selected-group) delivery. In telecommunications it denotes terrestrial and satellite radio and television transmission over allocated spectrum, and by extension the broadcast primitives of computer networks, where a frame or packet is addressed to every node on a segment. The same word names a distinct but analogous mechanism in array computing, where NumPy-style broadcasting stretches arrays of differing shapes to a common shape so that element-wise operations apply one value across many without copying data.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:broadcasting", "labels": [ "Broadcasting" ], "is_subclass_of": [ "Telecommunications" ], "wikilinks": [ "Telecommunications", "Spectrum Allocation", "Satellite Communication", "NumPy" ] }, { "id": "brown-dwarf", "title": "Brown Dwarf", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:brown-dwarf", "labels": [ "Brown Dwarf" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "browser-automation", "title": "Browser Automation", "domain": "ai", "domain_name": "Ai", "definition": "Browser Automation is the programmatic control of web browsers to execute tasks\u2014navigation, form submission, data extraction, and UI interaction\u2014without direct human input. It spans a spectrum from low-level Chrome DevTools Protocol (CDP) scripting and WebDriver-based frameworks to high-level tools such as Playwright and Puppeteer, as well as AI-driven computer-use agents that perceive and act upon rendered page content. Browser automation underpins modern software testing pipelines, robotic process automation, and agentic AI workflows that must interact with web-based services. The field has evolved rapidly with the emergence of large language model agents capable of interpreting visual page state and generating action sequences autonomously.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:browser-automation", "labels": [ "Browser Automation", "Browser Automation by LLMs", "Browser Use" ], "is_subclass_of": [ "Workflow Automation", "Process Automation", "Software Engineering" ], "wikilinks": [] }, { "id": "browser-engine", "title": "Browser Engine", "domain": "standards", "domain_name": "Standards", "definition": "A browser engine is the software component that parses and renders web content, implementing the HTML, CSS and JavaScript execution semantics defined by web standards to turn markup and script into an interactive display. Examples include Blink, Gecko and WebKit, each of which independently implements the same standards with varying levels of conformance and performance. Browser engines form the runtime substrate on which web applications, and automated agents that operate a browser, ultimately depend.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:browser-engine", "labels": [ "Browser Engine" ], "is_subclass_of": [ "Runtime Environment" ], "wikilinks": [] }, { "id": "browser-based-screen-capture", "title": "Browser-Based Screen Capture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Browser-based screen capture is the in-browser acquisition of screen, window, or tab video using web APIs such as the Screen Capture API (getDisplayMedia), without installing native software. The captured MediaStream can be recorded locally, encoded, or streamed in real time over WebRTC, making it the foundation for web conferencing, asynchronous video messaging, and screen recording tools. It runs under explicit user permission and within the browser's security sandbox.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:browser-based-screen-capture", "labels": [ "Browser-Based Screen Capture" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "building-automation", "title": "Building Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Building Automation refers to the centralised monitoring and control of a building's mechanical, electrical, and environmental systems \u2014 including heating, ventilation, air conditioning, lighting, access control, and fire safety \u2014 through networked sensor-actuator architectures and programmable control logic. It aims to optimise occupant comfort, reduce energy consumption, and enable remote facility management via integrated software platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:building-automation", "labels": [ "Building Automation", "Building Automation System" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "building-information-modelling", "title": "Building Information Modelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Building Information Modelling (BIM) is a collaborative process for creating and managing a shared digital representation of the physical and functional characteristics of a built asset across its lifecycle. A BIM model is an object-oriented, parametric 3D database in which geometry is enriched with semantic data such as materials, costs, schedules and performance properties. It supports coordinated design, clash detection, quantity take-off and facility management by allowing architects, engineers and contractors to work against a single federated source of truth. BIM underpins the convergence of construction practice with digital twin and spatial computing technologies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:building-information-modelling", "labels": [ "Building Information Modelling", "Building Information Modeling" ], "is_subclass_of": [ "Spatial Computing", "Digital Twin" ], "wikilinks": [] }, { "id": "bullet-physics", "title": "Bullet Physics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Bullet Physics is an open-source, real-time physics simulation library that provides collision detection, rigid body dynamics, and soft body simulation for use in games, robotics, visual effects, and scientific computing. Originally developed by Erwin Coumans, it implements discrete and continuous collision detection alongside a constraint solver, enabling physically plausible interactions between complex 3D geometries at interactive frame rates.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:bullet-physics", "labels": [ "Bullet Physics" ], "is_subclass_of": [ "Physics Engine" ], "wikilinks": [] }, { "id": "bulletproofs", "title": "Bulletproofs", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bulletproofs are a class of short, non-interactive zero-knowledge proof system that enables efficient range proofs and arbitrary arithmetic circuit satisfiability without a trusted setup. Based on the discrete logarithm assumption over elliptic curves, they produce logarithmically-sized proofs that can be aggregated and batched, making them particularly well-suited to confidential transaction systems in public blockchains where proof size and verification cost are critical constraints.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bulletproofs", "labels": [ "Bulletproofs", "Bulletproofs Range Proof", "Bulletproofs+" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [] }, { "id": "bundle-adjustment", "title": "bundle adjustment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Bundle adjustment is a non-linear least-squares optimisation technique in computer vision and photogrammetry that simultaneously refines estimates of 3-D structure (point cloud positions) and camera parameters (extrinsics and intrinsics) so that reprojected 3-D points best match their observed 2-D image locations across all views. The name derives from the bundle of light rays connecting each camera centre to visible scene points; optimising the geometry tightens these bundles. It is the gold-standard global refinement step in structure-from-motion pipelines, visual SLAM, and aerial photogrammetry, producing metrically accurate 3-D reconstructions from overlapping images or video frames. Computationally it exploits the sparse block structure of the Jacobian via Schur complement elimination and is typically solved with the Levenberg-Marquardt algorithm.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:bundle-adjustment", "labels": [ "Bundle Adjustment" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "buoy", "title": "Buoy", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:buoy", "labels": [ "Buoy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "burn-and-mint-bridge", "title": "Burn-and-Mint Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Burn-and-Mint Bridge is a cross-chain asset transfer mechanism in which tokens are irreversibly destroyed (burned) on the source blockchain and an equivalent quantity of canonical tokens is newly created (minted) on the destination blockchain, ensuring that the total circulating supply across chains remains constant. This contrasts with lock-and-mint approaches by eliminating custodial reserves, instead relying on protocol-level guarantees and verifiable burn proofs to authorise minting on the receiving chain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:burn-and-mint-bridge", "labels": [ "Burn-and-Mint Bridge" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "burned-area-mapping", "title": "Burned Area Mapping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:burned-area-mapping", "labels": [ "Burned Area Mapping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "burning-mechanism", "title": "Burning Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Burning Mechanism is an economic design pattern in tokenised blockchain systems whereby tokens are permanently removed from circulating supply by sending them to an unspendable address (a null or black-hole address) or by protocol-enforced destruction, permanently contracting the total token supply. Token burning is deployed as a deflationary monetary policy tool to counteract inflationary issuance, to create token scarcity as a value-accrual mechanism, to implement fee markets (as in Ethereum's EIP-1559 base fee burn), and to regulate supply in algorithmic stablecoins and tokenomics models. The economic effect depends critically on the rate and predictability of burning relative to issuance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:burning-mechanism", "labels": [ "Burning Mechanism" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain", "Virtual Economy" ] }, { "id": "business-continuity", "title": "Business Continuity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Business continuity is the discipline of ensuring that an organisation's critical functions can continue, or be rapidly restored, during and after a disruptive event such as an outage, disaster, or cyber-incident. It combines planning, redundancy, recovery procedures, and testing to limit downtime and data loss against defined objectives. In technical infrastructure it is realised through high availability, disaster recovery, fault tolerance, and resilient architecture, and is closely coupled with risk management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:business-continuity", "labels": [ "Business Continuity" ], "is_subclass_of": [ "Resilience" ], "wikilinks": [] }, { "id": "business-intelligence", "title": "Business Intelligence", "domain": "data", "domain_name": "Data", "definition": "Business intelligence is the set of technologies, processes, and practices for collecting, integrating, analysing, and presenting business data to support managerial decision-making. It encompasses data warehousing, reporting, dashboards, online analytical processing (OLAP), and ad-hoc querying that turn raw operational records into actionable insight. BI matters because it converts dispersed enterprise data into a coherent, queryable foundation for performance monitoring and strategic planning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:business-intelligence", "labels": [ "Business Intelligence" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "business-logic-layer", "title": "Business Logic Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Business Logic Layer (BLL) is the architectural tier in a multi-tier application that encapsulates domain rules, workflows, and computations specific to the problem domain, sitting between the presentation layer and the data access layer. It is responsible for validating inputs, enforcing business constraints, orchestrating data transformations, and coordinating service calls, ensuring that domain invariants are maintained independently of user interface or persistence concerns.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:business-logic-layer", "labels": [ "Business Logic Layer" ], "is_subclass_of": [ "Service Layer" ], "wikilinks": [] }, { "id": "business-process-automation", "title": "Business Process Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "The use of technology to automate complex, multi-step business processes beyond individual tasks, integrating systems and people to streamline operations. It targets end-to-end workflows rather than isolated actions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:business-process-automation", "labels": [ "Business Process Automation" ], "is_subclass_of": [ "Process Automation" ], "wikilinks": [ "Workflow Automation", "Automation", "Robotic Process Automation", "Process Automation" ] }, { "id": "business-process-management", "title": "Business Process Management", "domain": "governance", "domain_name": "Governance", "definition": "Business Process Management (BPM) is a discipline that uses methods, techniques, and software tools to discover, model, analyse, measure, improve, and automate business processes \u2014 end-to-end workflows that produce value for internal stakeholders or external customers. It combines organisational management practice with process automation technology to make organisational operations more efficient, agile, and aligned with strategic goals.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:business-process-management", "labels": [ "Business Process Management", "Business Process Model" ], "is_subclass_of": [ "Enterprise Architecture" ], "wikilinks": [] }, { "id": "business-rules-engine", "title": "Business Rules Engine", "domain": "governance", "domain_name": "Governance", "definition": "A Business Rules Engine (BRE) is a software component that externalises and executes business logic as declarative rules separate from application code. Rules are authored by domain experts in natural-language-like syntax and evaluated against facts at runtime using algorithms such as the Rete network. BREs enable non-technical stakeholders to modify organisational policies without redeploying software, making them a key enabler of agile governance and compliance automation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:business-rules-engine", "labels": [ "Business Rules Engine" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "byte-pair-encoding", "title": "Byte Pair Encoding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A subword tokenisation algorithm that iteratively merges the most frequent pairs of characters or character sequences to build a vocabulary, originally developed for data compression and widely adopted in neural language models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:byte-pair-encoding", "labels": [ "Byte Pair Encoding", "Byte-Pair Encoding" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "bytecode", "title": "Bytecode", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bytecode is a compact, platform-independent instruction set produced by compiling source code, designed to be executed by a virtual machine rather than directly by hardware. In blockchain systems, smart contracts written in high-level languages such as Solidity are compiled to bytecode that the Ethereum Virtual Machine deterministically executes across every node. Bytecode strikes a balance between the portability of source code and the efficiency of native machine code.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:bytecode", "labels": [ "Bytecode" ], "is_subclass_of": [ "Smart Contract Execution" ], "wikilinks": [] }, { "id": "byzantine-agreement", "title": "Byzantine Agreement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Byzantine Agreement is a class of distributed consensus protocols that guarantee correct operation even when a subset of participating nodes behave arbitrarily \u2014 sending conflicting, malicious, or unpredictable messages. Originating from the Byzantine Generals Problem formalised by Lamport, Shostak, and Pease in 1982, these protocols ensure that all honest nodes reach the same decision provided the number of faulty nodes does not exceed one-third of the total. Byzantine Agreement is foundational to blockchain consensus mechanisms and safety-critical distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:byzantine-agreement", "labels": [ "Byzantine Agreement" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "byzantine-fault-tolerance", "title": "Byzantine Fault Tolerance", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The feature of a distributed network to reach consensus on the same value even when some nodes fail to respond or respond with incorrect information. Enables networks to function correctly despite malicious or faulty nodes comprising less than one-third of the network. Derived from the Byzantine Generals Problem; the fundamental theorem states consensus is achievable if and only if more than two-thirds of participants are honest.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:byzantine-fault-tolerance", "labels": [ "Byzantine Fault Tolerance", "Byzantine Fault Tolerant", "Byzantine Fault-Tolerant Consensus" ], "is_subclass_of": [ "Consensus Mechanism", "Distributed Agreement" ], "wikilinks": [ "AI Energy Optimisation", "Consensus Safety", "Cryptographic Signature", "Distributed Agreement", "Finality Gadget", "Leader Election", "Malicious Node Handling", "Message Authentication", "Quorum", "Voting Round", "Blockchain", "Consensus Mechanism", "Distributed Systems", "Fault Tolerance" ] }, { "id": "byzantine-fault-tolerant-system", "title": "Byzantine Fault Tolerant System", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A Byzantine Fault Tolerant System is a distributed computing system designed to continue operating correctly even when a fraction of its nodes exhibit arbitrary, potentially malicious failures \u2014 including sending conflicting, incorrect, or no messages to different peers. Such systems implement consensus protocols that guarantee safety (agreement) and liveness (progress) provided strictly fewer than one-third of participating nodes are faulty, a bound established by the foundational Byzantine Generals Problem. BFT systems underpin the security of permissioned and permissionless blockchain networks, replicated state machines, and safety-critical distributed infrastructure where adversarial behaviour must be tolerated without compromising overall correctness.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:byzantine-fault-tolerant-system", "labels": [ "Byzantine Fault Tolerant System" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "byzantine-generals-problem", "title": "Byzantine Generals Problem", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The Byzantine Generals Problem is a foundational thought experiment in distributed computing describing how loyal participants can reach agreement when some participants are traitorous and may send arbitrary or conflicting messages. It formalises the difficulty of achieving consensus in the presence of arbitrary (Byzantine) faults, where faulty nodes behave maliciously rather than merely crashing. The problem establishes that agreement is solvable only when fewer than one-third of participants are faulty.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:byzantine-generals-problem", "labels": [ "Byzantine Generals Problem" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "c2-pa-content-credentials", "title": "c2pa content credentials", "domain": "security", "domain_name": "Security", "definition": "C2PA Content Credentials are cryptographically signed metadata manifests, defined by the Coalition for Content Provenance and Authenticity (C2PA) technical specification, that are embedded in or bound to digital media assets to record their origin, capture conditions, AI generation provenance, and editing history in a tamper-evident chain. Each Content Credential is a JUMBF-structured (JPEG Universal Metadata Box Format) assertion set signed using the COSE (CBOR Object Signing and Encryption) standard, anchored to the asset via a cryptographic hash binding that detects post-signing modifications. Verifiers \u2014 including browser extensions, social media platforms, AI disclosure tools, and editing software \u2014 can retrieve and display the full credential chain, enabling transparent and auditable provenance for photographs, video, audio, documents, and AI-generated synthetic content. The mechanism serves as the primary interoperability layer between hardware capture devices, editing software, AI generation systems, and distribution platforms participating in the broader content authenticity ecosystem.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:c2-pa-content-credentials", "labels": [ "C2PA Content Credentials", "Content Credentials" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "c2-pa-standard", "title": "C2PA Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An open technical standard developed by the Coalition for Content Provenance and Authenticity that establishes cryptographically-signed Content Credentials to certify the origin, history, and authenticity of digital media, functioning like a tamper-evident nutrition label for digital content.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:c2-pa-standard", "labels": [ "C2PA Standard", "C2PA Audio Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standard" ], "wikilinks": [ "Authenticity Certification", "Content Provenance", "Cryptographic Signing", "ISO (International Organization for Standardization)", "Media Verification", "Metadata Framework", "Trust Model", "W3C (World Wide Web Consortium)", "metaverse", "Technical Standard" ] }, { "id": "c2-pa", "title": "c2pa", "domain": "standards", "domain_name": "Standards", "definition": "C2PA (Coalition for Content Provenance and Authenticity) is a cross-industry open technical standards body and specification that cryptographically binds signed provenance metadata \u2014 covering capture origin, editing history, and AI generation disclosure \u2014 to digital media assets including images, video, audio, and documents. Founded by Adobe, Arm, BBC, Intel, Microsoft, and Truepic under the Joint Development Foundation, the C2PA specification defines the manifest container format (JUMBF/ISO 19566-5), assertion vocabularies, COSE-based signing mechanism, X.509 PKI trust model, and the resulting Content Credentials artefact. C2PA consolidates two prior initiatives \u2014 Adobe's Content Authenticity Initiative (CAI) and the BBC-led Project Origin \u2014 into a single interoperable standard for end-to-end media provenance across capture, editing, distribution, and verification stages.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:c2-pa", "labels": [ "C2PA", "C2PA Content Provenance Standards", "C2pa" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "c4-model", "title": "C4 Model", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The C4 Model is a hierarchical software architecture notation system developed by Simon Brown that uses four abstraction levels \u2014 Context, Containers, Components, and Code \u2014 to describe and communicate the structure of a software system at progressively increasing levels of technical detail. Each level targets a different audience, from non-technical stakeholders at the Context level to developers at the Code level, using a minimal, tool-agnostic set of diagram elements that are easy to reason about and keep up to date.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:c4-model", "labels": [ "C4 Model", "C4 Model Convention" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "cad-software", "title": "CAD Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computer-Aided Design (CAD) software is a category of application that enables engineers, architects, and designers to create, modify, analyse, and optimise two-dimensional drawings and three-dimensional geometric models of physical artefacts and structures. CAD tools encode geometry using boundary representation (B-rep) or constructive solid geometry (CSG) methods, support parametric constraint-based modelling, and produce outputs usable in manufacturing, simulation, and visualisation pipelines.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:cad-software", "labels": [ "CAD Software", "CAD", "CAD Tool" ], "is_subclass_of": [ "Design Software" ], "wikilinks": [] }, { "id": "can-bus", "title": "CAN Bus", "domain": "robotics", "domain_name": "Robotics", "definition": "CAN Bus (Controller Area Network) is a reliable serial communication standard that lets microcontrollers and devices exchange messages over a shared two-wire bus without a host computer, widely used in vehicles and robotics.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:can-bus", "labels": [ "CAN Bus" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "Embedded Systems", "Real-Time Control", "Motor Driver", "micro-ROS", "Power Electronics", "Communication Protocol" ] }, { "id": "cap-theorem", "title": "CAP Theorem", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The CAP Theorem states that a distributed data store cannot simultaneously guarantee all three of consistency, availability and partition tolerance. When a network partition occurs and messages between nodes are lost or delayed, a system must choose between remaining available, by serving possibly stale data, and remaining consistent, by refusing requests it cannot safely satisfy. Formulated by Eric Brewer and later proved formally by Gilbert and Lynch, it frames a fundamental trade-off in distributed systems design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cap-theorem", "labels": [ "CAP Theorem" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Distributed Database Design", "Distributed Systems Domain", "Consensus", "Eventual Consistency", "owl:Thing", "Gilbert and Lynch (2002)" ] }, { "id": "cbdc-cross-border-settlement", "title": "CBDC Cross-Border Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "CBDC cross-border settlement refers to the use of central bank digital currencies to settle payments between parties in different jurisdictions. It aims to reduce cost and delay in international transfers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cbdc-cross-border-settlement", "labels": [ "CBDC Cross-Border Settlement" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [ "Settlement", "Payment System", "Digital Pound", "Central Bank Digital Currency", "https://www.bis.org/about/bisih/topics/cbdc.htm", "https://www.bis.org/publ/othp59.htm" ] }, { "id": "cbdc-frameworks", "title": "CBDC Frameworks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "CBDC Frameworks are the integrated bodies of technical specifications, regulatory governance structures, design principles, and interoperability standards that central banks and international monetary institutions use to research, pilot, and deploy Central Bank Digital Currencies \u2014 sovereign digi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cbdc-frameworks", "labels": [ "CBDC Frameworks" ], "is_subclass_of": [ "Governance and Regulation", "Monetary Policy Framework", "Payment System Standard", "Financial Regulation", "Digital Money", "Central Bank Infrastructure" ], "wikilinks": [ "AML CFT Compliance Layer", "AML CFT Framework", "AML CFT Regulation", "Andolfatto 2021 Economic Journal CBDC Banking Competition", "API Standards", "Atlantic Council CBDC Tracker July 2025", "Atomic Delivery versus Payment", "Bank of England", "Bank of England Digital Pound Progress Update 2025", "Bank of England Digital Pound Update October 2025", "BIS", "BIS Annual Economic Report 2025 Unified Ledger Blueprint", "BIS CBDC Design Principles 2020", "BIS CBDC Principles", "BIS CBDC System Design 2025 Others 88", "BIS CBDC System Design and Interoperability 2022", "BIS Central Bank Digital Currencies Foundational Principles 2020", "BIS mBridge MVP 2024", "BIS Papers 159 Advancing in Tandem 2024 CBDC Survey", "BIS Project Nexus Blueprint July 2024" ] }, { "id": "cbdc-infrastructure", "title": "CBDC Infrastructure", "domain": "finance", "domain_name": "Finance", "definition": "CBDC infrastructure refers to the technical systems, protocols, and institutional arrangements that underpin the issuance, distribution, settlement, and management of central bank digital currencies. It encompasses the ledger technology (centralised, distributed, or hybrid), API layers enabling interoperability with commercial banks and payment service providers, offline payment capability, and the privacy-preserving and security mechanisms required for sovereign digital money at scale. CBDC infrastructure must satisfy central bank requirements for finality, programmability, resilience, and monetary policy control that differ fundamentally from commercial digital payment systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:cbdc-infrastructure", "labels": [ "CBDC Infrastructure" ], "is_subclass_of": [ "Payment Infrastructure" ], "wikilinks": [] }, { "id": "cbdcs", "title": "CBDCs", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Central Bank Digital Currencies (CBDCs) are sovereign digital money issued directly by a nation's central bank, representing a liability of the issuing central bank rather than a commercial bank or private institution, denominated in the official unit of account and designed to function as legal ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cbdcs", "labels": [ "CBDCs", "CBDC" ], "is_subclass_of": [ "Governance and Regulation", "Token and Asset", "Digital Currency", "Legal Tender", "Central Bank Money", "Sovereign Money", "Payment Instrument" ], "wikilinks": [ "Agur Ari DellAriccia 2022 Designing CBDCs JFE", "AI Agent", "Alipay", "Allen 2020 CBDC Design Choices Penn Law Review", "Anti Money Laundering", "API Standards", "Auer Cornelli Frost 2022 Rise of CBDC BIS Working Paper 880", "Bahamas Sand Dollar", "Bank of Canada 2025 Privacy Enhancing Technologies CBDC Staff Discussion Paper 2025-1", "Bank of England", "Bank of England 2025 Digital Pound Progress Update October", "Bank of England HMT 2023 Digital Pound Consultation Paper February", "Bank of Ghana", "Basel Committee", "Bindseil 2020 Tiered CBDC ECB Working Paper 2351", "BIS 2020 CBDC Foundational Principles Report", "BIS 2024 CBDC Survey Papers 159 Advancing in Tandem", "BIS 2024 mBridge MVP Documentation othp59", "BIS CPMI Standards", "BIS MAS 2024 Project Nexus Blueprint Phase 3" ] }, { "id": "cbeci-methodology", "title": "CBECI Methodology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The CBECI methodology is the estimation framework underlying the Cambridge Bitcoin Electricity Consumption Index, used to model the Bitcoin network's real-time electricity demand. It derives a plausible consumption range by combining network hashrate, a profitability-weighted mix of mining hardware efficiencies, data-centre overheads, and assumptions about miner economics. The methodology matters because it provides the most widely cited, transparent academic estimate of Bitcoin's energy use and the basis for its carbon analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cbeci-methodology", "labels": [ "CBECI Methodology" ], "is_subclass_of": [ "Evaluation Metric", "Blockchain Environmental Impact Assessment" ], "wikilinks": [] }, { "id": "cbeci", "title": "CBECI", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Cambridge Bitcoin Electricity Consumption Index (CBECI) is a real-time model produced by the Cambridge Centre for Alternative Finance that estimates the annualised electricity consumption and carbon footprint of the Bitcoin proof-of-work mining network. It aggregates hardware efficiency data from known ASIC models, network hash rate, and electricity price assumptions to produce lower-, central-, and upper-bound estimates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cbeci", "labels": [ "CBECI" ], "is_subclass_of": [ "Benchmark Evaluation", "Electricity Consumption" ], "wikilinks": [] }, { "id": "cbor", "title": "CBOR", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Concise Binary Object Representation (CBOR) is a binary data serialisation format specified in RFC 7049 (superseded by RFC 8949) designed to enable extremely compact encoding of structured data with a data model that is a superset of JSON. CBOR encodes values using a type-length-value scheme, eliminating the overhead of textual delimiters and key quotation, which makes it particularly well-suited for constrained environments such as IoT devices, embedded systems, and low-bandwidth protocols where minimising message size and parsing complexity is critical.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cbor", "labels": [ "CBOR" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [] }, { "id": "ccpa", "title": "CCPA", "domain": "governance", "domain_name": "Governance", "definition": "The California Consumer Privacy Act (CCPA), a California state statute enacted in 2018 that grants California residents rights over the personal information that businesses collect about them, including the right to know, the right to delete, the right to opt out of sale, and the right to non-discrimination.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ccpa", "labels": [ "CCPA", "CCPA Compliance" ], "is_subclass_of": [ "Data Protection Law" ], "wikilinks": [ "Data Protection", "Privacy", "Consumer Protection", "GDPR", "Data Privacy", "Data Protection Law" ] }, { "id": "cdla", "title": "CDLA", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Community Data License Agreement (CDLA) is a family of open data licences published by the Linux Foundation to govern the sharing and reuse of curated datasets. Its principal variants are CDLA-Permissive, which allows broad reuse with attribution, and CDLA-Sharing, which adds a copyleft-style obligation to share derived datasets under the same terms. CDLA matters for AI because it provides clear, machine-readable licensing for training-data corpora, reducing legal uncertainty over data provenance and downstream use.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cdla", "labels": [ "CDLA" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "cdn", "title": "CDN", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Content Delivery Network (CDN) is a geographically distributed network of proxy servers and data centres that caches and serves web content from locations physically close to end users, reducing latency, relieving origin server load, and improving resilience against traffic spikes and denial-of-service attacks. CDNs operate through anycast routing, edge caching policies, and real-time traffic steering algorithms to deliver optimal user experience across diverse network conditions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:cdn", "labels": [ "CDN", "Content Delivery Network" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ce-marking", "title": "CE Marking", "domain": "robotics", "domain_name": "Robotics", "definition": "CE marking is a mandatory conformity mark indicating that a product placed on the European Economic Area market meets the applicable EU health, safety, and environmental protection requirements. For machinery and robots it signals compliance with directives such as the Machinery Regulation, the EMC Directive, and relevant harmonised standards, often supported by a declaration of conformity and technical file. It matters because it is a legal precondition for market access and embeds robot safety standards into commercial deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ce-marking", "labels": [ "CE Marking" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "ceo-led-ai-strategy", "title": "CEO-led AI Strategy", "domain": "governance", "domain_name": "Governance", "definition": "An organizational approach where the Chief Executive Officer provides direct, high-level oversight and strategic direction for artificial intelligence initiatives to ensure alignment with business goals.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ceo-led-ai-strategy", "labels": [ "CEO-led AI Strategy" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "cftc", "title": "CFTC", "domain": "governance", "domain_name": "Governance", "definition": "The Commodity Futures Trading Commission, an independent agency of the United States government that regulates the derivatives markets, including futures, swaps and certain options.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cftc", "labels": [ "CFTC" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Financial Regulation", "Consumer Protection", "Financial Stability", "Securities Regulation" ] }, { "id": "ci-cd", "title": "CI/CD", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "CI/CD (continuous integration and continuous delivery/deployment) is the software engineering practice of automatically building, testing and releasing every code change through a versioned pipeline. Continuous integration merges work into a shared trunk many times a day behind automated test gates; continuous delivery keeps the mainline permanently releasable; continuous deployment pushes each passing change to production automatically. Together they shrink feedback loops, reduce integration risk and enable progressive release strategies such as canary and zero-downtime deployments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:ci-cd", "labels": [ "CI/CD" ], "is_subclass_of": [ "DevOps" ], "wikilinks": [ "DevOps", "Continuous Integration", "Continuous Deployment", "Canary Deployment" ] }, { "id": "ci-cd-automation", "title": "CI-CD Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Continuous integration and continuous delivery (CI/CD) automation is the practice of automatically building, testing, and deploying software changes through a defined pipeline triggered by source-control events. Continuous integration merges and validates changes frequently to detect defects early; continuous delivery extends this to produce always-deployable artefacts, while continuous deployment automates release to production. CI/CD automation reduces integration risk, shortens feedback loops, and makes deployments repeatable and auditable. It is implemented through pipeline-as-code definitions executed by orchestrators such as GitHub Actions, GitLab CI, and Jenkins.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:ci-cd-automation", "labels": [ "CI-CD Automation", "CI/CD", "CI/CD Pipeline" ], "is_subclass_of": [ "Automation" ], "wikilinks": [] }, { "id": "cid", "title": "CID", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Content Identifier (CID) is a self-describing, cryptographic label used in the InterPlanetary File System (IPFS) and related protocols to uniquely and permanently address a piece of content based on its cryptographic hash rather than its location. CIDs encode the hash function used, the hash digest, and the codec describing the data format, enabling content-addressed storage where identical content always produces the same identifier regardless of where it is stored. CIDs are the primary addressing primitive of the IPLD (InterPlanetary Linked Data) ecosystem.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cid", "labels": [ "CID" ], "is_subclass_of": [ "Content Identifier" ], "wikilinks": [] }, { "id": "cli-multi-agent-systems", "title": "CLI Multi-Agent Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "CLI multi-agent systems are software architectures in which networks of autonomous AI agents operate natively within command-line and terminal environments, coordinating to decompose, plan, and execute complex long-horizon tasks through tool invocation, sandboxed code execution, inter-agent messa...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cli-multi-agent-systems", "labels": [ "CLI Multi-Agent Systems", "CLI multi agent systems" ], "is_subclass_of": [ "AI Application", "Agentic Workflow", "Multi Agent Systems", "Autonomous Agents", "LLM Agents", "Tool Use" ], "wikilinks": [ "Agent Orchestrator", "AgentSystemsDomain", "Agent2Agent Protocol", "Agentic Workflow", "AutoGen", "Automated Code Review", "Automated Testing", "Autonomous Agents", "Autonomous Debugging", "Bash", "CI/CD Automation", "CodeAct", "Code Execution", "CrewAI", "DevSecOps", "Firecracker", "Git", "Inter-Agent Communication", "LangGraph", "Large Language Model" ] }, { "id": "clip-encoder", "title": "CLIP Encoder", "domain": "ai", "domain_name": "Ai", "definition": "A CLIP Encoder is either of the two paired neural network encoders \u2014 an image encoder and a text encoder \u2014 within the Contrastive Language-Image Pre-training (CLIP) framework developed by OpenAI. Each encoder maps its respective modality into a shared high-dimensional embedding space where semantically related image-text pairs are placed in proximity, enabling zero-shot image classification, cross-modal retrieval, and semantic image search without task-specific fine-tuning. CLIP Encoders serve as foundational components in multimodal AI pipelines, diffusion model conditioning, and vision-language models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:clip-encoder", "labels": [ "CLIP Encoder", "CLIP Image Encoder" ], "is_subclass_of": [ "Encoder", "Neural Network", "Embedding Model", "Foundation Model Component", "Deep Learning Model" ], "wikilinks": [] }, { "id": "clip", "title": "clip", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "CLIP (Contrastive Language-Image Pre-training) is a dual-encoder neural network architecture developed by OpenAI in which a vision encoder (Vision Transformer or CNN) and a text encoder (Transformer) are jointly trained on large-scale internet-sourced image-text pairs using an InfoNCE contrastive objective. The training maximises cosine similarity between embeddings of matched image-text pairs and minimises it for unmatched pairs within a batch, yielding a shared multimodal embedding space where semantically related images and text are geometrically proximate. This shared space enables zero-shot image classification by comparing image embeddings to natural language class descriptions without task-specific fine-tuning, and has become a foundational component in text-to-image diffusion models, open-vocabulary object detection, and cross-modal retrieval systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:clip", "labels": [ "CLIP", "CLIP Skip" ], "is_subclass_of": [ "AI Model Architecture", "Foundation Model", "Multimodal AI" ], "wikilinks": [] }, { "id": "cnc-machining", "title": "CNC Machining", "domain": "robotics", "domain_name": "Robotics", "definition": "CNC (computer numerical control) machining is a subtractive manufacturing process in which programmed instructions drive automated cutting tools to remove material from a workpiece and produce precise parts. Toolpaths derived from a digital model control multi-axis motion of mills, lathes and routers to achieve tight tolerances and repeatable results. It is a cornerstone of precision manufacturing across metals, plastics and composites.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cnc-machining", "labels": [ "CNC Machining" ], "is_subclass_of": [ "Precision Manufacturing" ], "wikilinks": [] }, { "id": "cncf", "title": "CNCF", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Cloud Native Computing Foundation (CNCF) is a vendor-neutral, open-source foundation established in 2015 under the Linux Foundation to steward cloud-native software projects including Kubernetes, Prometheus, Envoy, and over 150 additional hosted projects. CNCF defines cloud-native computing as building and running scalable applications in modern, dynamic environments such as public, private, and hybrid clouds using techniques including containers, microservices, and declarative APIs. The Foundation operates a Technical Oversight Committee (TOC) that governs project acceptance through sandbox, incubating, and graduated maturity stages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cncf", "labels": [ "CNCF", "Cloud Native Computing Foundation" ], "is_subclass_of": [ "Linux Foundation" ], "wikilinks": [] }, { "id": "coco-dataset", "title": "COCO Dataset", "domain": "ai", "domain_name": "Ai", "definition": "The COCO Dataset (Common Objects in Context) is a large-scale benchmark dataset for computer vision research comprising over 330,000 images with dense per-instance annotations covering object detection, instance segmentation, panoptic segmentation, keypoint estimation, and image captioning across 80 object categories in natural everyday scenes. Released in 2014 by Microsoft Research, it became the de-facto standard evaluation corpus for detection and segmentation models because its annotations capture objects in realistic, cluttered contexts rather than artificially isolated settings. COCO metrics\u2014specifically Average Precision (AP) averaged across IoU thresholds\u2014are the primary performance currency for reporting state-of-the-art results in visual perception research.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:coco-dataset", "labels": [ "COCO Dataset", "COCO Benchmark" ], "is_subclass_of": [ "Benchmarks" ], "wikilinks": [] }, { "id": "coco-wholebody", "title": "COCO WholeBody", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "COCO WholeBody is an extension of the COCO dataset that adds dense whole-body keypoint annotations, covering the body, hands, feet, and face with 133 keypoints per person rather than the 17 body-only keypoints of the original COCO annotations. It is used to train and benchmark whole-body pose estimation models, including hand and facial landmark detectors used by tools such as DWPose. The extended annotation set enables downstream applications such as gesture recognition, sign-language interpretation, and fine-grained human motion capture from images.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:coco-wholebody", "labels": [ "COCO WholeBody" ], "is_subclass_of": [ "COCO Dataset" ], "wikilinks": [] }, { "id": "colmap", "title": "COLMAP", "domain": "metaverse", "domain_name": "Metaverse", "definition": "COLMAP is an open-source general-purpose Structure-from-Motion and multi-view stereo pipeline used to reconstruct three-dimensional scenes from unordered collections of photographs. It performs feature extraction, image matching, incremental camera pose estimation, triangulation, and dense point-cloud generation. COLMAP is widely used as a preprocessing step for neural radiance fields and 3D Gaussian Splatting pipelines. Its modular architecture supports both GPU-accelerated and CPU-only execution, making it accessible across a range of hardware configurations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:colmap", "labels": [ "COLMAP" ], "is_subclass_of": [ "Structure-from-Motion" ], "wikilinks": [] }, { "id": "comet-metric", "title": "COMET Metric", "domain": "ai", "domain_name": "Ai", "definition": "A family of learned, reference-based and reference-free evaluation metrics for machine translation that use pre-trained multilingual language model representations to predict human-quality assessments of translation output. COMET models correlate more strongly with human judgements than n-gram-overlap metrics such as BLEU by capturing semantic and contextual similarity rather than surface string matching.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:comet-metric", "labels": [ "COMET Metric", "COMET Benchmark" ], "is_subclass_of": [ "Evaluation Metric", "AI Technique", "Natural Language Processing" ], "wikilinks": [] }, { "id": "complete-ontology-index", "title": "COMPLETE ONTOLOGY INDEX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive reference catalogue documenting and interconnecting all ontology terms, semantic classifications, and knowledge relationships across multiple domains and knowledge graphs. This index enables discovery, navigation, and semantic reasoning across the entire knowledge base, serving as the master pointer for all class hierarchies and inter-domain relations.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:complete-ontology-index", "labels": [ "COMPLETE ONTOLOGY INDEX" ], "is_subclass_of": [ "Standards and Interoperability", "Metaverse Ontology" ], "wikilinks": [ "IEEE (Institute of Electrical and Electronics Engineers)", "IETF (Internet Engineering Task Force)", "ISO (International Organization for Standardization)", "Knowledge Base", "Knowledge Relationships", "NIST (National Institute of Standards and Technology)", "Ontology", "RFC (Request for Comments)", "Semantic Classifications", "MetaverseDomain" ] }, { "id": "complete-hri-terms-reference", "title": "COMPLETE_HRI_TERMS_REFERENCE", "domain": "robotics", "domain_name": "Robotics", "definition": "A structured terminology reference documenting standardised vocabulary for Human-Robot Interaction (HRI), encompassing communication protocols, behavioural patterns, and engagement modalities. It enables consistent annotation and semantic understanding of interaction contexts across robotic and autonomous systems, aligned with IEEE and ISO standards.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:complete-hri-terms-reference", "labels": [ "COMPLETE_HRI_TERMS_REFERENCE" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [ "Behavioural Patterns", "CCPA (California Consumer Privacy Act)", "Communication Protocols", "Engagement Modalities", "GDPR (General Data Protection Regulation)", "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "EU AI Act", "Human-Robot Interaction", "MetaverseDomain" ] }, { "id": "completion-report", "title": "COMPLETION_REPORT", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Completion Report is a formal documentation artefact summarising the outcomes, achievements, and validation status of a completed project or work package. It captures deliverables, quality metrics, sign-off records, and governance evidence required for stakeholder communication and project closure in regulated or standards-aligned environments.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:completion-report", "labels": [ "COMPLETION_REPORT" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Deliverables", "IEEE (Institute of Electrical and Electronics Engineers)", "Project", "Project Governance", "Quality Metrics", "Sign-off Records", "Stakeholder Communication", "W3C (World Wide Web Consortium)", "Work Package", "MetaverseDomain" ] }, { "id": "cpmi-iosco-pfmi", "title": "CPMI-IOSCO PFMI", "domain": "finance", "domain_name": "Finance", "definition": "The Principles for Financial Market Infrastructures are international standards published by the Committee on Payments and Market Infrastructures and the International Organization of Securities Commissions. They set requirements for the safety and efficiency of payment, clearing and settlement systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cpmi-iosco-pfmi", "labels": [ "CPMI-IOSCO PFMI" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [ "Payment System", "SWIFT Messaging", "Central Bank" ] }, { "id": "cpu-architecture", "title": "CPU Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "CPU architecture is the design of general-purpose central processors optimised for low-latency execution of sequential and branch-heavy code: a small number of powerful cores with deep pipelines, aggressive out-of-order execution, sophisticated branch prediction, and large multi-level caches. It is the deliberate opposite pole to GPU architecture, which trades single-thread latency for massive throughput; CPUs devote silicon to making one instruction stream fast, GPUs to running tens of thousands of streams concurrently.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:cpu-architecture", "labels": [ "CPU Architecture" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [ "Computer Hardware", "GPU Architecture", "Instruction Set Architecture", "Parallel Computing" ] }, { "id": "cpu-computing", "title": "CPU Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "CPU computing is the execution of computational workloads on a general-purpose central processing unit, which performs instructions sequentially across a small number of high-clock cores optimised for low latency and complex control flow. It excels at branch-heavy, irregular, and serial tasks but contrasts with the massively parallel throughput model of GPU computing. CPUs follow a stored-program architecture with deep cache hierarchies, out-of-order execution, and rich instruction sets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cpu-computing", "labels": [ "CPU Computing" ], "is_subclass_of": [ "Parallel Computing" ], "wikilinks": [] }, { "id": "cpu", "title": "CPU", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A central processing unit (CPU) is the primary general-purpose processor of a computer, responsible for fetching, decoding, and executing the instructions of programs. It performs arithmetic, logic, control, and input/output operations dictated by software, coordinating the activity of the whole machine. As a versatile sequential and lightly parallel engine, the CPU contrasts with specialised accelerators such as GPUs that favour massive data parallelism.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cpu", "labels": [ "CPU" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "cqrs", "title": "CQRS", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Command Query Responsibility Segregation (CQRS) is a software architecture pattern that separates the model used to update state (commands) from the model used to read state (queries). By using distinct write and read paths, systems can independently optimise, scale, and secure each side, often pairing the write model with event sourcing. It is commonly applied in high-throughput, complex-domain systems where read and write workloads diverge.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cqrs", "labels": [ "CQRS" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "crdt", "title": "CRDT", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Conflict-free Replicated Data Types (CRDTs) are a family of data structures with mathematically proven convergence guarantees that allow multiple distributed replicas to be independently modified without requiring coordination, locks, or a consensus protocol, and which automatically merge to a si...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:crdt", "labels": [ "CRDT", "CRDT Synchronisation" ], "is_subclass_of": [ "Protocol and Infrastructure", "Distributed Data Synchronisation", "Eventually Consistent Data Structure", "Replicated Data Structure", "Mergeable Data Type", "Commutative Data Structure", "Monotone Data Structure" ], "wikilinks": [ "Associativity", "Automerge", "CAP Theorem", "Causal Broadcast", "Causal Broadcast Protocol", "Causal Consistency", "Causal Delivery", "Causal History", "Causal History Graph", "Causal Ordering", "Collaborative IDEs", "Commutative Data Structure", "Commutativity", "Coordination-free Computing", "CQRS", "DatabasesDomain", "Delta State", "Delta Synchronisation", "DistributedCollaborationDomain", "Distributed Data Synchronisation" ] }, { "id": "crm-integration", "title": "CRM Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "CRM integration is the connection of customer-relationship-management software with other business systems\u2014such as telephony, email, marketing, and AI assistants\u2014so that customer data and interactions flow automatically across tools. It typically uses APIs, webhooks, and event syncs to keep contact records, activity logs, and pipelines consistent. It is essential for unified customer views and automated sales and support workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:crm-integration", "labels": [ "CRM Integration" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [] }, { "id": "crm-systems", "title": "CRM Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "CRM systems are software platforms that centralise the management of an organisation's interactions with current and prospective customers across sales, marketing, and service. They store contact records, track pipelines and tickets, automate outreach, and increasingly embed AI for summarisation and recommendation. They are core operational systems for revenue and customer-experience teams.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:crm-systems", "labels": [ "CRM Systems" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "crm", "title": "CRM", "domain": "data", "domain_name": "Data", "definition": "Customer Relationship Management (CRM) is a category of enterprise software and associated business processes that centralises the storage, tracking, and analysis of all interactions between an organisation and its customers, prospects, and partners across the full sales, marketing, and service lifecycle. A CRM system provides a shared record of each contact and account, enabling sales pipelines, service-case management, marketing campaign tracking, and analytical reporting from a single platform.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:crm", "labels": [ "CRM" ], "is_subclass_of": [ "Enterprise Software Platform" ], "wikilinks": [] }, { "id": "csrd-compliance", "title": "CSRD Compliance", "domain": "governance", "domain_name": "Governance", "definition": "CSRD compliance is the practice of meeting the reporting obligations of the European Union's Corporate Sustainability Reporting Directive, which mandates standardised disclosure of environmental, social and governance impacts. It requires in-scope companies to report under the European Sustainability Reporting Standards (ESRS), including double-materiality assessments and audited climate and carbon metrics. Compliance underpins demonstrable carbon accounting for sectors such as energy-intensive computing and blockchain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:csrd-compliance", "labels": [ "CSRD Compliance", "CSRD" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "cuda", "title": "cuda", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "CUDA (Compute Unified Device Architecture) is NVIDIA's proprietary parallel computing platform and programming model that exposes GPU hardware through a C/C++ extension, enabling thousands of threads to execute concurrently across thousands of GPU cores via a three-level hierarchy of grids, blocks, and warps. It provides a unified memory model spanning host (CPU) and device (GPU) address spaces, complemented by curated libraries such as cuBLAS, cuDNN, and NCCL that deliver optimised primitives for linear algebra, deep neural network operations, and multi-GPU communication. Originally released in 2006, CUDA has become the de facto execution substrate for production deep learning training and inference, underpinning frameworks such as PyTorch and TensorFlow, and is the primary determinant of GPU vendor lock-in in the AI infrastructure stack.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:cuda", "labels": [ "CUDA", "CUDA Kernel", "CUDA Runtime", "CUDA Toolkit", "NVIDIA CUDA" ], "is_subclass_of": [ "Parallel Computing" ], "wikilinks": [] }, { "id": "cvpr", "title": "CVPR", "domain": "ai", "domain_name": "Ai", "definition": "CVPR (Conference on Computer Vision and Pattern Recognition) is the premier annual peer-reviewed academic conference for computer vision and pattern recognition research, co-sponsored by IEEE and the Computer Vision Foundation (CVF). It is consistently ranked among the most impactful venues in artificial intelligence, serving as the primary publication channel for breakthroughs in image recognition, object detection, semantic segmentation, generative modelling, video understanding, and 3D vision. The conference has been held annually since 1983 and its open-access proceedings, published through the CVF, represent a foundational evidence base for both academic research and applied AI development.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cvpr", "labels": [ "CVPR", "IEEE CVPR" ], "is_subclass_of": [ "AI Research Area", "Academic Conference", "Scientific Publication Venue" ], "wikilinks": [] }, { "id": "ca-browser-forum", "title": "Ca Browser Forum", "domain": "security", "domain_name": "Security", "definition": "The CA/Browser Forum (CA/B Forum) is a voluntary industry consortium of Certification Authorities (CAs), web browser vendors, and other relying parties that collaboratively develops and enforces minimum standards for the issuance and management of X.509 digital certificates used in TLS/HTTPS, code signing, and S/MIME email. Its Baseline Requirements documents define mandatory technical and procedural controls that CAs must meet to remain trusted by member browsers such as Chrome, Firefox, Safari, and Edge. Compliance is a prerequisite for inclusion in browser root stores, giving the Forum significant de facto regulatory power over internet PKI.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ca-browser-forum", "labels": [ "Ca Browser Forum", "CA/Browser Forum" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "cable-drive", "title": "Cable Drive", "domain": "robotics", "domain_name": "Robotics", "definition": "A cable drive (cable-driven transmission) is a mechanical actuation mechanism that transmits force and motion through tensioned cables, tendons, or belts routed over pulleys, rather than through rigid gear trains. It allows actuators to be located remotely from the joint they drive, reducing distal mass and inertia in robotic limbs and surgical instruments. Cable drives matter in robotics because they enable lightweight, compliant, backdrivable joints with low backlash.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cable-drive", "labels": [ "Cable Drive" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "cache-layer", "title": "Cache Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cache layer is an intermediate storage tier interposed between a data consumer and an authoritative data source, holding frequently or recently accessed data in faster storage to reduce latency, decrease load on origin systems, and improve overall system throughput. Cache layers operate at multiple levels of a computing stack\u2014CPU caches, in-process memory caches, distributed shared caches, and edge CDN caches\u2014each exploiting temporal or spatial locality of access patterns. Correctness requires coherence protocols or explicit invalidation strategies to prevent stale data from being served.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cache-layer", "labels": [ "Cache Layer", "Cache", "Cache System" ], "is_subclass_of": [ "Storage Layer" ], "wikilinks": [] }, { "id": "caching", "title": "Caching", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Caching is the technique of storing copies of frequently accessed data or computed results in a faster, closer storage tier so that subsequent requests can be served without repeating the expensive original operation. It trades additional memory or storage for reduced latency and lower load on backend systems, governed by policies for placement, expiry and invalidation. Caches appear at every layer of a system, from CPU registers and operating systems to content delivery networks and application-level stores.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:caching", "labels": [ "Caching" ], "is_subclass_of": [ "Performance Optimization" ], "wikilinks": [] }, { "id": "calculation-parameters", "title": "Calculation Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Configurable variables and settings that control the behavior, accuracy, and performance of computational simulations and models, including solver options, convergence criteria, time steps, and optimization constraints that fine-tune simulation outputs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:calculation-parameters", "labels": [ "Calculation Parameters" ], "is_subclass_of": [ "Content and Assets", "Configuration Setting" ], "wikilinks": [ "Model Calibration", "Parameter Definition", "Performance Tuning", "Sensitivity Analysis", "Simulation Control", "Validation Rules", "Configuration Setting", "metaverse" ] }, { "id": "calculus", "title": "Calculus", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Calculus is the branch of mathematics concerned with continuous change, comprising differential calculus (rates of change and slopes via derivatives) and integral calculus (accumulation and areas via integrals). In machine learning it provides the foundational machinery for optimisation: gradients computed through differentiation drive parameter updates, while integration underpins probability, expectation and continuous-time models. It is indispensable for understanding how learning algorithms adjust models to minimise loss.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:calculus", "labels": [ "Calculus" ], "is_subclass_of": [ "Machine Learning", "Mathematical Science", "Mathematics", "Applied Mathematics", "Continuous Mathematics" ], "wikilinks": [] }, { "id": "calibrated-data-state", "title": "Calibrated Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:calibrated-data-state", "labels": [ "Calibrated Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "calibration-equipment", "title": "Calibration Equipment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical instruments and reference artefacts used to measure, adjust, and verify the accuracy of sensors, displays, and tracking systems in spatial computing deployments. Calibration equipment includes colour reference targets, IMU calibration boards, structured-light patterns, and photometric probes, all of which underpin the geometric and radiometric correctness of XR pipelines.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:calibration-equipment", "labels": [ "Calibration Equipment" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "calibration-standards", "title": "Calibration Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Documented specifications and reference materials that establish measurement accuracy requirements, traceability to national standards, and systematic procedures for aligning sensors, displays, and imaging systems to known reference values in XR and computer vision applications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:calibration-standards", "labels": [ "Calibration Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standard" ], "wikilinks": [ "Device Alignment", "Measurement Accuracy", "Reference Materials", "Traceability Chain", "Validation Procedures", "metaverse", "Quality Assurance", "Technical Standard" ] }, { "id": "calibration-system", "title": "Calibration System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Calibration System is an integrated assembly of hardware targets, measurement instruments, algorithms, and software workflows used to determine and correct systematic errors in sensors, cameras, and measurement devices, establishing a known and traceable relationship between a sensor's raw outputs and the physical quantities they represent. Calibration systems are essential preconditions for accurate spatial measurement, computer vision, robotics, and extended reality applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:calibration-system", "labels": [ "Calibration System" ], "is_subclass_of": [ "Calibration Equipment" ], "wikilinks": [] }, { "id": "calibration-target", "title": "Calibration Target", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A physical or displayed reference pattern with precisely known geometric, photometric, or colorimetric properties used to determine camera intrinsic and extrinsic parameters, enabling accurate lens distortion correction, spatial measurement, and color reproduction in imaging systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:calibration-target", "labels": [ "Calibration Target" ], "is_subclass_of": [ "Standards and Interoperability", "Reference Standard" ], "wikilinks": [ "Camera Calibration", "Color Accuracy", "Flatness Control", "Lens Distortion Correction", "Pattern Definition", "Precision Manufacturing", "metaverse", "Reference Standard" ] }, { "id": "calibration-tools", "title": "Calibration Tools", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software and hardware utilities used to align, tune, and verify the accuracy of XR sensors, displays, and tracking systems. Calibration establishes ground-truth references for IMUs, cameras, optical tracking arrays, and display geometry, ensuring that virtual content is correctly registered to the physical world and that tracking data is within acceptable error tolerances.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:calibration-tools", "labels": [ "Calibration Tools" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "calibration-traceability", "title": "Calibration Traceability", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:calibration-traceability", "labels": [ "Calibration Traceability" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "calibration", "title": "calibration", "domain": "robotics", "domain_name": "Robotics", "definition": "Calibration is the systematic process of establishing, verifying, and correcting the quantitative relationship between a measurement instrument's or computational model's output and a known reference standard, encompassing intrinsic parameter estimation (gain, offset, nonlinearity, bias), extrinsic parameter determination (spatial pose and orientation relative to a reference frame), and inter-device consistency alignment. In physical systems it removes systematic error between raw sensor readings and true physical quantities; in machine learning it aligns predicted probability distributions to empirical frequencies. Calibration is a prerequisite for metrically accurate perception, reliable closed-loop control, trustworthy probabilistic inference, and coherent multi-modal data fusion across robotics, spatial computing, and AI systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:calibration", "labels": [ "Calibration", "Calibration Dataset", "Dual Quaternion Hand-Eye Calibration", "Intrinsic Calibration", "Photometric Calibration", "Robot Calibration", "Self-Calibration" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "california-ai-bill", "title": "California AI bill", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "California AI bill is the collective designation for a sequence of California state legislative instruments enacted or proposed between 2023 and 2026 to regulate the development, deployment, and disclosure obligations of large-scale Frontier Models \u2014 most prominently Senate Bill 1047 (SB 1047...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:california-ai-bill", "labels": [ "California AI bill" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Regulation", "Technology Policy", "Legislative Instrument", "State Law", "Frontier Model Governance" ], "wikilinks": [ "AB 1008", "AB 2013", "AB 489", "AI Alliance", "AI Regulation", "AI Safety Transparency", "AIGovernanceDomain", "Anthropic", "Attorney General Enforcement", "C2PA Content Credentials", "California Attorney General", "California Department of Justice", "California Effect", "California Legislature", "California Office of Emergency Services", "Catastrophic Risk Assessment", "CCPA", "Center for AI Safety", "Compliance Statement", "Compute Threshold" ] }, { "id": "call-centres", "title": "Call Centres", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Call Centres (contact centres \u2014 the operationally preferred modern term reflecting multichannel reality) are coordinated operational environments in which human agents and increasingly autonomous AI Agents handle inbound and outbound customer communications across voice, chat, email, social m...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:call-centres", "labels": [ "Call Centres" ], "is_subclass_of": [ "AI Application", "Customer Service Automation", "Conversational AI", "Enterprise AI", "Human-in-the-Loop Learning", "Operational AI" ], "wikilinks": [ "24/7 Customer Service", "Agent Assist", "Agent Productivity Augmentation", "Agent Wellbeing", "AI Agents", "Amazon Bedrock", "Amazon Connect", "Automated Quality Monitoring", "Automatic Call Distribution", "Automatic Speech Recognition", "Average Handle Time Reduction", "Biometric Authentication", "Contact Centre as a Service", "CRM", "CRM Integration", "CSAT Improvement", "Customer Experience", "CustomerExperienceDomain", "Customer Identity Verification", "Customer Service Automation" ] }, { "id": "calldata-compression", "title": "Calldata Compression", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Calldata compression is the practice of encoding transaction data more compactly before posting it as calldata to a base-layer blockchain, reducing the number of bytes that must be paid for and stored. Rollups such as Arbitrum and Base apply techniques including signature aggregation, zero-byte-heavy encoding and custom serialisation formats to shrink batch size. Because calldata gas cost dominates rollup transaction fees, compression directly reduces the cost of settling Layer 2 state on the base chain.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:calldata-compression", "labels": [ "Calldata Compression" ], "is_subclass_of": [ "Calldata" ], "wikilinks": [] }, { "id": "calldata", "title": "Calldata", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Calldata is the read-only, immutable byte array supplied with a transaction or message call on the Ethereum Virtual Machine, carrying the function selector and ABI-encoded arguments that tell a smart contract what to execute. Because it lives outside contract storage and is comparatively cheap to include, calldata is the primary channel through which external inputs reach contracts. It is especially significant for rollups, which post compressed transaction batches as calldata to the base layer to inherit its data availability and security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:calldata", "labels": [ "Calldata" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "cambridge-bitcoin-electricity-consumption-index", "title": "Cambridge Bitcoin Electricity Consumption Index", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An index published by the University of Cambridge that estimates the electricity consumption of the Bitcoin network from mining hardware efficiency and network hashrate. It provides a transparent methodology and range of estimates rather than a single fixed figure.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cambridge-bitcoin-electricity-consumption-index", "labels": [ "Cambridge Bitcoin Electricity Consumption Index" ], "is_subclass_of": [ "Bitcoin Mining" ], "wikilinks": [ "Mining", "Carbon Accounting", "Bitcoin Network", "Sustainability", "Bitcoin Mining", "https://ccaf.io/cbnsi/cbeci" ] }, { "id": "cambridge-centre-for-alternative-finance", "title": "Cambridge Centre for Alternative Finance", "domain": "finance", "domain_name": "Finance", "definition": "The Cambridge Centre for Alternative Finance (CCAF) is an interdisciplinary research centre at the University of Cambridge Judge Business School that conducts empirical studies on financial instruments and channels operating outside conventional banking systems. It is globally recognised for producing landmark indices and benchmark datasets on cryptoasset markets, distributed ledger technology adoption, digital payments, and the energy consumption of proof-of-work networks. CCAF research informs regulatory policy, industry strategy, and academic discourse through rigorous survey-based methodologies and open data publications. Its work bridges fintech innovation, environmental sustainability analysis, and international financial governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cambridge-centre-for-alternative-finance", "labels": [ "Cambridge Centre for Alternative Finance" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [ "Bitcoin", "Cryptocurrency", "Bitcoin Environmental Issues", "Financial Infrastructure Domain" ] }, { "id": "cambridge-judge-business-school", "title": "Cambridge Judge Business School", "domain": "finance", "domain_name": "Finance", "definition": "Cambridge Judge Business School is the business school of the University of Cambridge, located in Cambridge, England. It conducts research and teaching in management, finance and alternative finance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cambridge-judge-business-school", "labels": [ "Cambridge Judge Business School" ], "is_subclass_of": [ "University of Cambridge" ], "wikilinks": [ "Economics", "University of Cambridge" ] }, { "id": "camera-calibration", "title": "camera calibration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Camera calibration is the process of estimating a camera's intrinsic parameters \u2014 focal length, principal point, pixel skew, and radial and tangential distortion coefficients \u2014 and, when multiple sensors are involved, extrinsic parameters describing the rigid-body transformation (rotation and translation) between coordinate frames. Accurate calibration is a prerequisite for metrically correct 3D reconstruction, stereo depth estimation, augmented-reality overlay registration, and robot visual servoing. The Zhang method, which solves for camera parameters from multiple images of a planar chessboard target via non-linear reprojection-error minimisation, is the dominant practical technique and is standardised in OpenCV and MATLAB's Computer Vision Toolbox. Calibration quality must be maintained over a device's operational lifetime, driving development of online self-calibration methods that exploit ego-motion and scene structure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:camera-calibration", "labels": [ "Camera Calibration", "Camera Calibration Workflow", "Camera Intrinsic Calibration", "CameraCalibration" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "camera-intrinsics", "title": "Camera Intrinsics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Camera Intrinsics are the internal optical and geometric parameters of a camera that define the mathematical relationship between 3D points in the camera's coordinate frame and their 2D projections onto the image sensor. The intrinsic parameter matrix encodes focal length in pixel units along each image axis, the principal point (optical axis intersection with the sensor), and skew, whilst associated distortion coefficients correct for lens aberrations that cause deviations from the ideal pinhole projection model.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:camera-intrinsics", "labels": [ "Camera Intrinsics" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "camera-model", "title": "Camera Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A camera model is a mathematical description that maps three-dimensional scene points to two-dimensional image coordinates, capturing the geometry of how a camera projects the world. It encodes intrinsic parameters such as focal length and principal point alongside extrinsic parameters describing pose, and may model lens distortion. Camera models underpin calibration, reconstruction and pose estimation in computer vision and spatial computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:camera-model", "labels": [ "Camera Model" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "camera-parameters", "title": "Camera Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The intrinsic and extrinsic mathematical values that define a camera's optical characteristics and spatial positioning, comprising focal length, optical center, distortion coefficients (intrinsic) and rotation/translation relative to world coordinates (extrinsic), essential for 3D reconstruction ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:camera-parameters", "labels": [ "Camera Parameters" ], "is_subclass_of": [ "Display and Rendering", "Imaging Parameters" ], "wikilinks": [ "Camera Calibration", "Object Measurement", "Scene Localization", "3D Reconstruction", "Calibration Target", "Computer Vision", "Imaging Parameters", "metaverse", "Optimization Algorithm" ] }, { "id": "camera-sensor", "title": "Camera Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A camera sensor is the photosensitive component, typically a CMOS or CCD array, that converts incident light focused by the optics into an electrical signal forming an image. Its characteristics, including resolution, pixel size, dynamic range, noise and frame rate, determine the quality of the captured imagery and shutter behaviour. It is the primary exteroceptive vision input for robots, autonomous vehicles and immersive devices, and is frequently fused with complementary sensors such as lidar.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:camera-sensor", "labels": [ "Camera Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "camera-tracking-system", "title": "Camera Tracking System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A camera tracking system is a hardware and software assembly that continuously records the position, orientation, and lens parameters of a physical camera in three-dimensional space, transmitting this data in real time to rendering or compositing engines so that computer-generated imagery can be seamlessly integrated with live-action footage. Such systems underpin virtual production stages, augmented reality overlays, broadcast graphics, and visual effects pipelines. Tracking technologies include optical encoder arrays, inertial measurement units, infrared LED constellations, machine vision fiducial markers, and LiDAR-based spatial mapping.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:camera-tracking-system", "labels": [ "Camera Tracking System" ], "is_subclass_of": [ "Camera Tracking" ], "wikilinks": [] }, { "id": "camera-tracking", "title": "Camera Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Camera Tracking is the process of continuously estimating the position and orientation (pose) of a camera in 3D space relative to a fixed reference frame or scene, typically using image feature analysis, optical flow, or fiducial marker detection. It underpins augmented reality, visual effects compositing, robotic navigation, and autonomous vehicle perception by enabling virtual or computed elements to be correctly registered to the physical world as the camera moves.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:camera-tracking", "labels": [ "Camera Tracking" ], "is_subclass_of": [ "Motion Tracking" ], "wikilinks": [] }, { "id": "camera", "title": "Camera", "domain": "robotics", "domain_name": "Robotics", "definition": "An imaging sensor device that captures visual information in robotics and spatial-computing systems, enabling computer vision applications including object detection, SLAM, 3D reconstruction, visual servoing, and semantic scene understanding through various modalities (RGB, depth, thermal, event-based) and sensor technologies (CCD, CMOS, ToF, structured light).", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:camera", "labels": [ "Camera", "Camera System", "Device Camera", "Event Camera" ], "is_subclass_of": [ "Sensor", "VisualPerception" ], "wikilinks": [ "CameraCalibration", "ComputerVisionDomain", "Computer Vision Standards", "IEEE", "ImageSensor", "ISO/IEC", "Lens", "ProcessingUnit", "Robotics Research", "VisualPerception", "VisualServoing", "3D Reconstruction", "ObjectDetection", "RoboticsDomain", "SemanticSegmentation", "Sensor", "SLAM" ] }, { "id": "canaan", "title": "Canaan", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Canaan is a Chinese company that designs and produces Bitcoin mining hardware under the Avalon brand. It is a publicly listed manufacturer of ASIC machines.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:canaan", "labels": [ "Canaan" ], "is_subclass_of": [ "ASIC" ], "wikilinks": [ "Bitcoin Mining", "Transaction Validation", "Hardware", "ASIC", "https://www.canaan.io", "https://www.canaan.io/products" ] }, { "id": "canary-deployment", "title": "Canary Deployment", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Canary deployment is a progressive release strategy in which a new version of a service or model is exposed to a small subset of traffic before being rolled out more widely. Operators monitor health, performance and quality metrics on the canary cohort, promoting the release only if it behaves acceptably and otherwise rolling back. It limits the blast radius of a faulty change and is widely used in continuous delivery pipelines and in serving machine-learning models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:canary-deployment", "labels": [ "Canary Deployment" ], "is_subclass_of": [ "Continuous Deployment", "Progressive Delivery", "Release Engineering", "Deployment Strategy" ], "wikilinks": [] }, { "id": "canonical-json", "title": "Canonical JSON", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Canonical JSON is a normalised serialisation form of JSON that specifies a unique, deterministic byte-level representation for any given JSON value, enabling reliable hashing, digital signatures, and binary equality comparisons over JSON documents. Because RFC 8259 JSON permits equivalent representations that differ in whitespace, key ordering, and number formatting, canonical forms such as JCS (JSON Canonicalization Scheme, RFC 8785) and earlier proposals impose rules including lexicographic key sorting, no insignificant whitespace, and IEEE 754 double-precision serialisation for numbers. This determinism is essential in cryptographic contexts where the same logical object must produce the same digest regardless of which implementation serialised it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:canonical-json", "labels": [ "Canonical JSON" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "capability-advertisement", "title": "Capability Advertisement", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Capability Advertisement is the mechanism by which an autonomous agent publishes machine-readable metadata describing the skills, tools and interfaces it exposes, allowing other agents or an orchestrator to discover what it can do without prior hard-coded knowledge. It typically takes the form of a structured manifest, such as an agent card, exchanged during a discovery or handshake phase. It underpins dynamic task routing in multi-agent systems built on protocols such as the Agent Communication Protocol and Agent-to-Agent Protocol.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:capability-advertisement", "labels": [ "Capability Advertisement" ], "is_subclass_of": [ "Service Discovery" ], "wikilinks": [] }, { "id": "capability-elicitation", "title": "Capability Elicitation", "domain": "ai", "domain_name": "Ai", "definition": "Capability elicitation is the systematic process of designing prompts, scaffolding, and evaluation protocols to uncover the true maximum performance of an AI model on a given task or domain, distinguishing what a model is genuinely capable of from what it demonstrates under default conditions. It is a central concern in AI safety research and frontier model evaluation because models may possess latent capabilities \u2014 such as the ability to reason about dangerous knowledge, produce deceptive outputs, or autonomously pursue goals \u2014 that are not revealed by standard benchmarks but can be surfaced through carefully constructed elicitation methods including chain-of-thought prompting, multi-step scaffolding, and adversarial jailbreaking.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:capability-elicitation", "labels": [ "Capability Elicitation" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation", "Safety Evaluation", "AI Safety Research", "Benchmark Evaluation" ], "wikilinks": [] }, { "id": "capability-evaluation", "title": "Capability Evaluation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Capability evaluation is the systematic measurement of what an AI model can do, especially the elicitation and assessment of potentially dangerous capabilities such as autonomous replication, cyber-offence, or assistance with weapons. It combines benchmarks, structured tasks, and adversarial elicitation (including red-teaming) to establish upper bounds on model behaviour under best-effort prompting and tooling. Results feed safety cases and trigger the thresholds defined in responsible scaling and preparedness frameworks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:capability-evaluation", "labels": [ "Capability Evaluation", "Capability Evaluations" ], "is_subclass_of": [ "AI Evaluation" ], "wikilinks": [] }, { "id": "capability-forecasting", "title": "Capability Forecasting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Capability forecasting is the practice of predicting the future capabilities of AI systems before they are built or deployed, typically by extrapolating from scaling laws, benchmark trends, and historical progress. It aims to anticipate when models will reach particular performance thresholds so that safety, governance, and deployment decisions can be made proactively. Forecasts are inherently uncertain because of emergent behaviour and discontinuous jumps in capability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:capability-forecasting", "labels": [ "Capability Forecasting" ], "is_subclass_of": [ "AI Governance", "Risk Assessment", "Anticipatory Governance" ], "wikilinks": [] }, { "id": "capability-overhang", "title": "Capability Overhang", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The disparity between the technical capability of an AI system to perform a task and the actual rate of adoption or deployment of that capability in practice.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:capability-overhang", "labels": [ "Capability Overhang" ], "is_subclass_of": [ "Enterprise AI" ], "wikilinks": [] }, { "id": "capacity-planning", "title": "Capacity Planning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Capacity planning is the process of determining the production, infrastructure, or service capacity required by an organisation to meet changing demand over a defined time horizon, balancing the cost of over-provisioned resources against the risk of under-provisioned systems that cannot meet service-level objectives. In technology contexts, capacity planning encompasses compute, storage, network bandwidth, and human resources, employing demand forecasting models, utilisation metrics, and growth projections to derive procurement and scaling roadmaps. In manufacturing, the same principles apply to machine-hours, floor space, and workforce shifts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:capacity-planning", "labels": [ "Capacity Planning" ], "is_subclass_of": [ "Cloud Infrastructure" ], "wikilinks": [] }, { "id": "capital-adequacy", "title": "Capital Adequacy", "domain": "finance", "domain_name": "Finance", "definition": "Capital adequacy is the regulatory and risk management requirement that financial institutions maintain a minimum level of capital relative to their risk-weighted assets, ensuring they can absorb losses without becoming insolvent and thereby protecting depositors, counterparties, and the broader financial system. The Basel Committee on Banking Supervision (BCBS) has codified capital adequacy requirements through successive accords \u2014 Basel I (1988), Basel II (2004), Basel III (2010, phased through 2028) \u2014 specifying the composition of qualifying capital (Common Equity Tier 1, Additional Tier 1, Tier 2) and the risk-weighting methodologies for credit, market, and operational risk exposures.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:capital-adequacy", "labels": [ "Capital Adequacy", "Capital Adequacy Framework", "Capital Adequacy Ratio", "Minimum Capital Requirements" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "capital-allocation", "title": "Capital Allocation", "domain": "finance", "domain_name": "Finance", "definition": "Capital allocation is the process by which firms, investors and markets distribute scarce financial resources across competing uses to maximise risk-adjusted return or strategic value. It spans corporate decisions about investing, returning capital and funding, as well as investor decisions about deploying funds across assets and opportunities. Effective allocation channels capital towards its most productive uses, making it central to both firm performance and the efficiency of the wider financial system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:capital-allocation", "labels": [ "Capital Allocation" ], "is_subclass_of": [ "Investment Management" ], "wikilinks": [] }, { "id": "capital-efficiency", "title": "Capital Efficiency", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Capital efficiency is the degree to which deployed capital is put to productive use to generate returns or provide a service, maximising output per unit of locked or committed funds. In decentralised finance it describes designs such as concentrated liquidity and over-collateralisation tuning that let liquidity providers and protocols achieve more depth or yield from less capital. Higher capital efficiency reduces idle assets and lowers the cost of providing liquidity and credit.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:capital-efficiency", "labels": [ "Capital Efficiency" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "capital-formation", "title": "Capital Formation", "domain": "governance", "domain_name": "Governance", "definition": "Capital formation is the economic and financial process by which savings are channelled into productive investment, raising the stock of capital goods and funding enterprises. In capital markets it refers specifically to the issuance and sale of securities \u2014 equity and debt \u2014 that allow firms to raise funds from investors. Efficient capital formation underpins economic growth and is a central policy objective of securities regulation, which balances investor protection against ease of raising capital.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:capital-formation", "labels": [ "Capital Formation", "Capital Accumulation" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "capital-markets", "title": "Capital Markets", "domain": "finance", "domain_name": "Finance", "definition": "Capital markets are the financial markets in which long-term debt and equity instruments are issued and traded, channelling savings from investors to issuers such as companies and governments. They comprise primary markets, where new securities are issued, and secondary markets, where existing securities trade among investors. They are central to capital formation, price discovery and liquidity, and are increasingly intersecting with tokenisation and blockchain-based settlement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:capital-markets", "labels": [ "Capital Markets" ], "is_subclass_of": [ "Traditional Finance" ], "wikilinks": [] }, { "id": "carbon-accounting-software", "title": "Carbon Accounting Software", "domain": "governance", "domain_name": "Governance", "definition": "Carbon accounting software is a category of enterprise application that automates the collection, calculation, and reporting of an organisation's greenhouse gas (GHG) emissions across Scope 1 (direct), Scope 2 (purchased energy), and Scope 3 (value chain) categories in accordance with established standards such as the GHG Protocol Corporate Standard and ISO 14064. These platforms ingest activity data from financial systems, energy bills, logistics records, and supplier databases; apply emission factors; and produce auditable emissions inventories aligned with regulatory disclosure frameworks including the EU Corporate Sustainability Reporting Directive (CSRD) and the SEC Climate Disclosure Rule.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:carbon-accounting-software", "labels": [ "Carbon Accounting Software" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "carbon-accounting", "title": "Carbon Accounting", "domain": "governance", "domain_name": "Governance", "definition": "The systematic process of measuring, recording, and reporting an organization's greenhouse gas emissions across direct operations (Scope 1), purchased energy (Scope 2), and value chain activities (Scope 3), expressed in CO2 equivalent units to quantify climate impact and enable reduction strategies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:carbon-accounting", "labels": [ "Carbon Accounting", "Carbon Accounting Standard", "Carbon Project Accounting", "CarbonAccounting", "CarbonEmissionsCalculation", "Internal Carbon Accounting" ], "is_subclass_of": [ "Environmental Accounting" ], "wikilinks": [ "Climate Action Planning", "Data Collection", "Emission Factors", "Emissions Reporting", "ISO (International Organization for Standardization)", "SEC (Securities and Exchange Commission)", "Verification Process", "Blockchain", "Environmental Accounting", "metaverse", "Regulatory Compliance" ] }, { "id": "carbon-calculator", "title": "Carbon Calculator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital tool or software application that estimates greenhouse gas emissions from activities such as energy consumption, transportation, and production by applying standardized emission factors to user-provided data, enabling individuals and organizations to quantify and understand their carbon...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:carbon-calculator", "labels": [ "Carbon Calculator" ], "is_subclass_of": [ "Governance and Safety", "Sustainability Tool" ], "wikilinks": [ "Activity Data", "Calculation Methodology", "Emission Factors", "Emissions Estimation", "Footprint Awareness", "Reduction Planning", "Blockchain", "metaverse", "Sustainability Tool" ] }, { "id": "carbon-credit-retirement", "title": "Carbon Credit Retirement", "domain": "governance", "domain_name": "Governance", "definition": "Carbon credit retirement is the permanent cancellation of a carbon credit in a registry so that it can no longer be sold, transferred, or double-counted, formally claiming the underlying emission reduction or removal. Retirement is the terminal step that converts a tradable credit into a fulfilled offset claim against a specific buyer's emissions. It is fundamental to carbon market integrity because it guarantees that each tonne of avoided or removed CO2 is claimed only once.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:carbon-credit-retirement", "labels": [ "Carbon Credit Retirement" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "carbon-credit-token", "title": "Carbon Credit Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital token representing verified carbon-offset value tradeable across platforms for emissions reduction and environmental sustainability.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:carbon-credit-token", "labels": [ "Carbon Credit Token", "CarbonCreditToken" ], "is_subclass_of": [ "Token and Asset", "Virtual Economy" ], "wikilinks": [ "Blockchain Record", "Emissions Trading", "Environmental Compliance", "ISO 14065", "Siemens + OMA3", "SmartContractLayer", "Sustainability Framework", "Sustainability Reporting", "Token Smart Contract", "Trading Platform", "UNFCCC", "Verification Authority", "Verification Metadata", "Blockchain", "BlockchainDomain", "Carbon Offset Certificate", "Carbon Offset Trading", "Carbon Registry", "DataLayer", "Digital Wallet" ] }, { "id": "carbon-credit-tracking", "title": "Carbon Credit Tracking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Carbon Credit Tracking is the application of Blockchain and distributed ledger technology to record, verify, transfer, and permanently retire carbon emission reduction credits across their full lifecycle \u2014 from project origination through third-party verification to final environmental claim ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:carbon-credit-tracking", "labels": [ "Carbon Credit Tracking" ], "is_subclass_of": [ "Network Component", "Real World Asset Tokenisation", "Environmental Finance", "Distributed Ledger Application", "Carbon Markets", "Green Finance" ], "wikilinks": [ "AI", "Article 6 Compliance", "Automated Carbon Accounting", "Biodiversity Credits", "Carbon Credits", "Carbon Markets", "CarbonPlan 2022 Zombies on the Blockchain", "Carbon Price Discovery", "Carbon Registries", "Carbon Registry Bridge", "Carbon Standard Certification", "Carbonplace 2024 Ledger Insights Live Pilot Seed Round", "Centralised Carbon Exchange", "Chainlink", "Chainlink Oracles", "Climate Action Reserve", "Climate Tech", "ClimateTechDomain", "Columbia SIPA CGEP 2024 Operationalise Article 6", "Compliance Cap-and-Trade Registry" ] }, { "id": "carbon-credits", "title": "carbon credits", "domain": "finance", "domain_name": "Finance", "definition": "Carbon credits are tradeable instruments, each representing the verified reduction or removal of one tonne of carbon dioxide equivalent (tCO\u2082e) from the atmosphere, used in both compliance and voluntary markets to incentivise greenhouse gas mitigation. In compliance cap-and-trade systems, regulators issue a capped total of allowances and require regulated emitters to surrender one allowance per tonne emitted, creating a price signal for abatement. Voluntary carbon markets allow organisations to purchase credits from validated offset projects \u2014 such as reforestation, renewable energy deployment, or methane capture \u2014 to offset residual emissions. Market integrity depends on independently certified additionality, permanence, and measurability, overseen by standards bodies such as Verra, Gold Standard, and the American Carbon Registry.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:carbon-credits", "labels": [ "Carbon Credits", "Carbon Credit", "Tokenized Carbon Credits" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "carbon-cycle-monitoring", "title": "Carbon Cycle Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:carbon-cycle-monitoring", "labels": [ "Carbon Cycle Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "carbon-footprint-assessment", "title": "Carbon Footprint Assessment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive emodology that quantifies total greenhouse gas emissions associated with a product, service, or organization across its entire lifecycle using Life Cycle Assessment principles, expressed in CO2 equivalent units to identify emission hotspots and reduction opportunities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:carbon-footprint-assessment", "labels": [ "Carbon Footprint Assessment" ], "is_subclass_of": [ "Governance and Safety", "Environmental Assessment" ], "wikilinks": [ "Boundary Definition", "Emission Factors", "Hotspot Identification", "Impact Quantification", "ISO (International Organization for Standardization)", "Lifecycle Data", "Reduction Strategy", "Blockchain", "Environmental Assessment", "metaverse" ] }, { "id": "carbon-footprint-indicator", "title": "Carbon Footprint Indicator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A quantitative environmental performance metric that measures and tracks an organization's greenhouse gas emissions in CO2 equivalent units, serving as a key sustainability KPI for monitoring progress toward emission reduction targets and enabling comparative analysis across operations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:carbon-footprint-indicator", "labels": [ "Carbon Footprint Indicator" ], "is_subclass_of": [ "Governance and Safety", "Environmental K P I", "Environmental KPI" ], "wikilinks": [ "Benchmarking", "Emissions Data", "Environmental KPI", "ISO (International Organization for Standardization)", "Measurement Methodology", "Performance Monitoring", "Reporting Framework", "Target Tracking", "Blockchain", "metaverse" ] }, { "id": "carbon-footprint-measurement", "title": "Carbon Footprint Measurement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systematic quantitative modology for calculating the total greenhouse gas (GHG) emissions attributable to an organisation, product, service, or activity across its full operational and value-chain scope, expressed in tonnes of carbon dioxide equivalent (tCO\u2082e), applying IPCC Sixth Assessment Repo...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:carbon-footprint-measurement", "labels": [ "Carbon Footprint Measurement" ], "is_subclass_of": [ "Legal and Regulatory", "Security and Identity", "Sustainability Reporting", "Environmental Accounting", "Measurement Frameworks", "Corporate Disclosure", "Climate Risk Management" ], "wikilinks": [ "Activity Data", "AI", "Boundary Setting", "Carbon Accounting Software", "Carbon Credits", "Carbon Offsetting", "Climate Risk Management", "Climate Tech", "Corporate Disclosure", "Corporate Sustainability Reporting Directive", "CSRD Compliance", "Decarbonisation Strategy", "Digital Product Passport", "Emission Factor Databases", "Emission Factors", "Emission Inventory", "ERP Systems", "ESG Investing", "ESRS E1", "GHG Protocol" ] }, { "id": "carbon-footprint", "title": "Carbon Footprint", "domain": "governance", "domain_name": "Governance", "definition": "A carbon footprint is the total quantity of greenhouse gases, expressed as carbon dioxide equivalent (CO2e), emitted directly and indirectly by an individual, organisation, product, or activity over a defined period. It aggregates scope 1, 2, and 3 emissions to provide a single comparable measure of climate impact. Carbon footprints matter because they are the baseline metric for emissions reduction targets, disclosure, and offsetting strategies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:carbon-footprint", "labels": [ "Carbon Footprint", "Carbon Footprint of Bitcoin", "Carbon Footprinting", "Corporate Carbon Footprint", "Product Carbon Footprint" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "carbon-markets", "title": "Carbon Markets", "domain": "finance", "domain_name": "Finance", "definition": "Carbon markets are regulated or voluntary trading systems in which carbon credits or allowances \u2014 each representing one metric tonne of CO\u2082 equivalent avoided, reduced, or removed \u2014 are issued, bought, sold, and retired to place a price on greenhouse gas emissions and channel investment towards least-cost mitigation activities. Compliance markets operate under mandatory emissions-trading schemes (ETS) such as the EU ETS, California Cap-and-Trade, and the Article 6 mechanisms of the Paris Agreement, where regulated emitters must surrender allowances matching verified output. Voluntary carbon markets (VCMs) allow organisations and individuals outside regulatory caps to purchase independently verified credits from projects ranging from avoided deforestation to direct air capture, enabling net-zero and carbon-neutrality claims. Market integrity depends on robust project standards, third-party verification, transparent registry infrastructure, and governance frameworks to prevent double counting and ensure additionality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:carbon-markets", "labels": [ "Carbon Markets", "Carbon Market", "CarbonMarkets", "On-Chain Carbon Markets", "Tokenised Carbon Markets" ], "is_subclass_of": [ "Environmental Asset Market" ], "wikilinks": [] }, { "id": "carbon-neutral-blockchain", "title": "Carbon Neutral Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Carbon Neutral Blockchain is any distributed ledger infrastructure that achieves net-zero or carbon-negative lifecycle greenhouse gas emissions through one or more of three principal mechanisms: (1) Consensus mechanism transition: Adoption of low-energy Consensus Mechanisms \u2014 pre-eminently", "entityType": "Class", "qualityScore": 0.53, "maturity": "established", "iri": "urn:ngm:class:carbon-neutral-blockchain", "labels": [ "Carbon Neutral Blockchain" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain", "Distributed Ledger Technology", "Sustainable Technology", "Green Infrastructure", "Climate-Aligned Finance" ], "wikilinks": [ "Algorand", "Algorand Foundation 2022-2025 Sustainability Carbon-Negative Certification", "ArXiv 2024 Blockchain Carbon Markets Standards Overview", "Cambridge Blockchain Network Sustainability Index CBECI", "Cambridge CCAF 2025 Digital Mining Industry Report", "CarbonAccountingLayer", "Carbon Credit", "Carbon Credits", "Carbon Markets", "Carbon Neutrality", "Carbon Offset", "Carbon Offset Programme", "Carbonmark 2025 State of Voluntary Carbon Market", "Centralised Carbon Market", "Climate-Aligned Finance", "Climate Commitments", "Climate Disclosure", "ClimateTrade", "ClimateTrade 2021 Algorand Carbon-Negative Announcement", "ConsensusLayer" ] }, { "id": "carbon-neutrality-planning", "title": "Carbon Neutrality Planning", "domain": "governance", "domain_name": "Governance", "definition": "The strategic process of developing a comprehensive roadmap to achieve net-zero greenhouse gas emissions, encompassing emission measurement, science-based reduction targets, interim milestones, decarbonization actions, and residual emission neutralization through permanent carbon removal.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:carbon-neutrality-planning", "labels": [ "Carbon Neutrality Planning" ], "is_subclass_of": [ "Governance and Safety", "Strategic Planning" ], "wikilinks": [ "Climate Action", "Decarbonization", "GHG Baseline", "Implementation Strategy", "Net Zero Achievement", "Reduction Targets", "Blockchain", "metaverse", "Strategic Planning" ] }, { "id": "carbon-neutrality-verification", "title": "Carbon Neutrality Verification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The independent third-party audit process that validates an organization's carbon neutrality claims by verifying emission calculations, reduction measures, and offset quality against established standards such as ISO 14068-1 and PAS 2060.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:carbon-neutrality-verification", "labels": [ "Carbon Neutrality Verification", "Carbon Verification" ], "is_subclass_of": [ "Governance and Safety", "Environmental Verification" ], "wikilinks": [ "Credibility Assurance", "Documentation Package", "Greenwashing Prevention", "ISO (International Organization for Standardization)", "Stakeholder Trust", "Blockchain", "Environmental Verification", "metaverse", "Third Party Auditor", "Verification Standard" ] }, { "id": "carbon-neutrality", "title": "Carbon Neutrality", "domain": "governance", "domain_name": "Governance", "definition": "Carbon neutrality is a state in which an entity \u2014 an organisation, product, event, or country \u2014 achieves a net-zero balance between its greenhouse gas emissions and the carbon dioxide it removes or offsets, such that its net contribution to atmospheric CO\u2082 concentration is zero over a defined accounting period. It is typically achieved through a combination of direct emissions reductions, procurement of renewable energy, and the purchase and retirement of independently verified carbon credits representing emissions avoided or removed elsewhere. Carbon neutrality is distinct from net-zero emissions, which generally requires deeper reductions and limits the role of offsets, and from climate positivity, which requires net negative emissions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:carbon-neutrality", "labels": [ "Carbon Neutrality", "Carbon Neutrality Goals", "Corporate Carbon Neutrality", "PAS 2060 Carbon Neutrality" ], "is_subclass_of": [ "Climate Policy" ], "wikilinks": [] }, { "id": "carbon-offset-certificate", "title": "Carbon Offset Certificate", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A verified document representing the reduction or removal of one metric ton of CO2 equivalent emissions, issued by recognized standards bodies after independent validation that confirms the underlying project's additionality, permanence, and measurability.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:carbon-offset-certificate", "labels": [ "Carbon Offset Certificate" ], "is_subclass_of": [ "Governance and Safety", "Environmental Certificate" ], "wikilinks": [ "Carbon Neutrality", "Climate Investment", "Emissions Compensation", "Project Verification", "Registry Issuance", "Third Party Validation", "Blockchain", "Environmental Certificate", "metaverse" ] }, { "id": "carbon-offset-programme", "title": "Carbon Offset Programme", "domain": "governance", "domain_name": "Governance", "definition": "A carbon offset programme is a structured scheme that generates, certifies, issues, and retires carbon credits representing greenhouse gas emissions reductions or removals achieved by specific projects or activities, providing a mechanism by which organisations or individuals can compensate for their own emissions by financing equivalent climate action elsewhere. Programmes operate under standards bodies \u2014 including the Verified Carbon Standard (Verra), Gold Standard, American Carbon Registry (ACR), and Climate Action Reserve \u2014 that define methodologies for quantifying emissions impacts, require independent verification, and maintain public registries to prevent double-counting of credits. Projects include avoided deforestation (REDD+), renewable energy, methane capture, reforestation, and engineered carbon removal.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:carbon-offset-programme", "labels": [ "Carbon Offset Programme", "Carbon Offset Program" ], "is_subclass_of": [ "Carbon Offset Trading" ], "wikilinks": [] }, { "id": "carbon-offset-trading", "title": "Carbon Offset Trading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The buying and selling of verified carbon credits on voluntary and compliance markets through exchanges and over-the-counter transactions, enabling price discovery, liquidity provision, and efficient allocation of climate finance to emission reduction projects.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:carbon-offset-trading", "labels": [ "Carbon Offset Trading", "Carbon Trading" ], "is_subclass_of": [ "Governance and Safety", "Environmental Trading" ], "wikilinks": [ "Carbon Price Discovery", "Climate Finance", "Market Infrastructure", "Offset Liquidity", "Trading Platform", "Blockchain", "Carbon Registry", "Environmental Trading", "metaverse" ] }, { "id": "carbon-offset", "title": "Carbon Offset", "domain": "governance", "domain_name": "Governance", "definition": "A carbon offset is a tradable instrument representing a measurable reduction, avoidance, or removal of one tonne of CO2-equivalent emissions, purchased to compensate for emissions occurring elsewhere. Offsets are generated by certified projects such as reforestation, renewable energy, or direct air capture and verified against recognised methodologies. They matter as a market mechanism for funding mitigation, though their credibility hinges on additionality, permanence, and avoidance of double-counting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:carbon-offset", "labels": [ "Carbon Offset", "Carbon Offset Procurement", "Offset Project" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "carbon-offsetting", "title": "Carbon Offsetting", "domain": "governance", "domain_name": "Governance", "definition": "Carbon offsetting is the practice whereby an entity compensates for its greenhouse gas emissions by financing or undertaking activities that achieve an equivalent reduction or removal of carbon dioxide equivalent (CO2e) elsewhere in the global economy. Offsets are quantified in standardised units \u2014 typically one tonne of CO2e per credit \u2014 and must satisfy criteria of additionality, permanence, measurability, and independent verification to be considered credible. The mechanism operates across compliance markets (e.g. the EU Emissions Trading System) and voluntary markets (e.g. Gold Standard, Verra VCS), and spans project types including afforestation, improved cookstoves, methane capture, and direct air capture. Controversy around offset quality, permanence risks, and potential for greenwashing has driven increasing scrutiny from regulators, standards bodies, and civil society.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:carbon-offsetting", "labels": [ "Carbon Offsetting", "Emissions Offsetting" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Carbon Accounting", "Sustainability", "Carbon Credits", "Voluntary Carbon Market" ] }, { "id": "carbon-price-discovery", "title": "Carbon Price Discovery", "domain": "finance", "domain_name": "Finance", "definition": "Carbon price discovery is the process by which market participants collectively determine the equilibrium price of carbon emission allowances or credits through the interaction of supply and demand in regulated compliance markets and voluntary carbon markets. It encompasses the mechanisms, venues, and information flows that translate abatement costs, regulatory constraints, and economic activity into observable carbon prices. Transparent price signals are essential for directing investment towards emissions reduction and for evaluating the economic cost of decarbonisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:carbon-price-discovery", "labels": [ "Carbon Price Discovery" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [] }, { "id": "carbon-registry", "title": "Carbon Registry", "domain": "governance", "domain_name": "Governance", "definition": "A centralized database system that tracks the issuance, ownership, transfer, and retirement of carbon credits, assigning unique serial numbers to each credit for full lifecycle traceability and preventing double-counting across voluntary and compliance carbon markets.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:carbon-registry", "labels": [ "Carbon Registry" ], "is_subclass_of": [ "Environmental Registry" ], "wikilinks": [ "Credit Tracking", "Database Infrastructure", "Double Counting Prevention", "Market Transparency", "Unique Identifiers", "Verification Process", "Blockchain", "Environmental Registry", "metaverse" ] }, { "id": "carbon-standard-certification", "title": "Carbon Standard Certification", "domain": "governance", "domain_name": "Governance", "definition": "Carbon standard certification is the process by which carbon-reduction projects and the credits they generate are independently validated and verified against a recognised standard such as Verra VCS, Gold Standard, or the ICVCM framework. Certification confirms that claimed emission reductions are real, additional, permanent, and properly quantified before credits are issued into a registry. It is the trust anchor of voluntary and compliance carbon markets, distinguishing high-integrity credits from low-quality ones.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:carbon-standard-certification", "labels": [ "Carbon Standard Certification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "carbon-tax", "title": "Carbon Tax", "domain": "governance", "domain_name": "Governance", "definition": "A carbon tax is a government levy charged per tonne of carbon dioxide (or CO2-equivalent) emitted, putting an explicit, predictable price on greenhouse-gas pollution so that emitters internalise the climate damage their activities cause. Unlike an emissions trading scheme, which caps the quantity of emissions and lets the price float, a carbon tax fixes the price and lets emitted quantities adjust; examples include Sweden's levy of over 100 euros per tonne and carbon taxes in Canada, Singapore and South Africa.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:carbon-tax", "labels": [ "Carbon Tax" ], "is_subclass_of": [ "Climate Policy" ], "wikilinks": [ "Climate Policy", "Emissions Trading Scheme", "Carbon Credits", "Carbon Markets" ] }, { "id": "carbon-aware-computing", "title": "Carbon-Aware Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Carbon-aware computing is the practice of scheduling and placing computational workloads to minimise their associated greenhouse-gas emissions by responding to the time-varying and location-varying carbon intensity of electricity. Rather than only reducing energy use, it shifts flexible work to periods and regions where the grid is cleaner. The approach combines real-time grid carbon-intensity signals with workload orchestration to lower the carbon footprint of data centres and cloud services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:carbon-aware-computing", "labels": [ "Carbon-Aware Computing" ], "is_subclass_of": [ "Green Computing" ], "wikilinks": [] }, { "id": "cardano", "title": "Cardano", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cardano is a public proof-of-stake blockchain platform launched in 2017 by IOHK, with Charles Hoskinson, a co-founder of Ethereum, among its founders. It is distinctive for its emphasis on peer-reviewed academic research and formal methods, and its consensus protocol Ouroboros was the first proof-of-stake protocol with published security proofs. The platform separates a settlement layer for its ADA cryptocurrency from a computation layer for smart contracts, which were enabled through the Alonzo upgrade in 2021. Cardano uses an extended UTXO accounting model and the functional language Plutus for on-chain logic.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cardano", "labels": [ "Cardano" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus", "Blockchain Domain" ], "wikilinks": [ "Proof of Stake", "UTXO", "Smart Contract", "Ethereum", "Blockchain Domain", "Kiayias et al. 2017, Ouroboros: A Provably Secure Proof-of-Stake Blockchain Protocol" ] }, { "id": "career-strategy", "title": "Career Strategy", "domain": "governance", "domain_name": "Governance", "definition": "The deliberate planning and adaptation of individual professional skills, roles, and trajectories in response to technological and market shifts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:career-strategy", "labels": [ "Career Strategy" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "carrier-to-noise-ratio", "title": "Carrier-to-noise Ratio", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:carrier-to-noise-ratio", "labels": [ "Carrier-to-noise Ratio" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cartographic-generalisation", "title": "Cartographic Generalisation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cartographic-generalisation", "labels": [ "Cartographic Generalisation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cascading-water", "title": "Cascading Water", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cascading-water", "labels": [ "Cascading Water" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "case-management-system", "title": "Case Management System", "domain": "governance", "domain_name": "Governance", "definition": "A software platform that tracks discrete units of investigative or remedial work\u2014cases\u2014from intake through triage, assignment, evidence gathering, escalation, and resolution, providing structured workflows, deadline management, and a complete audit trail. In compliance and legal operations, a case management system is the destination for alerts raised by monitoring controls, ensuring each potential breach is documented, investigated by accountable owners, and closed with a defensible record for regulators.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:case-management-system", "labels": [ "Case Management System" ], "is_subclass_of": [ "Enterprise Software" ], "wikilinks": [ "Compliance Monitoring", "Workflow Automation", "Audit Trail" ] }, { "id": "case-management", "title": "Case Management", "domain": "governance", "domain_name": "Governance", "definition": "A business-process discipline for handling unstructured, knowledge-intensive work items \u2014 'cases' \u2014 whose resolution path cannot be fully predetermined, such as compliance investigations, insurance claims, legal matters, and customer disputes. Each case aggregates documents, tasks, participants, deadlines, and an audit trail around a single subject, and progresses through investigation, decision, and resolution stages guided by worker judgement rather than a rigid predefined flow.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:case-management", "labels": [ "Case Management" ], "is_subclass_of": [ "Business Process Management" ], "wikilinks": [ "Business Process Management", "Transaction Monitoring", "Case Management System", "Regulatory Compliance" ] }, { "id": "cash-app", "title": "Cash App", "domain": "finance", "domain_name": "Finance", "definition": "Cash App is a mobile payment service operated by Block, Inc. that lets users send money, hold balances, and buy and sell bitcoin and stocks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cash-app", "labels": [ "Cash App" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [ "Payment System", "Bitcoin", "Block" ] }, { "id": "cashu", "title": "Cashu", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cashu is a free and open-source Chaumian ecash protocol designed for Bitcoin and the Lightning Network, first released in October 2022 by the pseudonymous developer Calle (alias @calle), a PhD physicist and Bitcoin developer.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cashu", "labels": [ "Cashu" ], "is_subclass_of": [ "Token and Asset", "Bitcoin Layer 2", "ChaumianEcash", "BitcoinPrivacyTools", "LightningApplications", "Chaumian Ecash", "Bearer Instrument", "Privacy Protocol", "Digital Cash" ], "wikilinks": [ "AI Agent Payments", "Anonymous Payments", "Ark Protocol", "b-money", "Bearer Instrument", "Bearer Token Model", "Bearer Token Transfer", "Bearer Tokens", "BIS 2023 Project Tourbillon", "BIS Project Tourbillon", "Bitcoin Design", "Bitcoin Improvement Proposals", "Bitcoin Layer 2", "Bitcoin Magazine 2024 Cashu Vision", "BitcoinPrivacyTools", "Bitcoin UTXO", "Bitfinex Blog 2024 Cashu Ecash", "Blind Signature Scheme", "BlindSignatures", "Blind Signatures" ] }, { "id": "casper-ffg", "title": "Casper Ffg", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Casper FFG (the Friendly Finality Gadget) is a proof-of-stake finality mechanism that overlays a checkpoint-based voting protocol on an underlying block proposal chain. Validators stake deposits and vote in two rounds to justify and then finalise checkpoints, after which reverting them would require destroying at least one third of the total stake. It introduces economic finality with slashing penalties for equivocation, providing strong accountability without requiring a full consensus overhaul.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:casper-ffg", "labels": [ "Casper Ffg", "Casper FFG" ], "is_subclass_of": [ "Finality Gadget" ], "wikilinks": [] }, { "id": "catastrophic-risk-assessment", "title": "Catastrophic Risk Assessment", "domain": "ai", "domain_name": "Ai", "definition": "Catastrophic risk assessment is the systematic evaluation of low-probability, high-severity hazards that could cause widespread harm, including the danger that advanced AI systems may enable mass casualties or societal-scale disruption. In AI governance it involves measuring a model's potential to contribute to chemical, biological, cyber, or autonomous-weapon threats before deployment. It matters because frontier-AI regulation, such as proposed state safety bills, conditions release on credible assessment and mitigation of these tail risks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:catastrophic-risk-assessment", "labels": [ "Catastrophic Risk Assessment" ], "is_subclass_of": [ "AI Safety", "Risk Assessment", "Probabilistic Risk Assessment" ], "wikilinks": [] }, { "id": "catastrophic-risk-reduction", "title": "Catastrophic Risk Reduction", "domain": "ai", "domain_name": "Ai", "definition": "Catastrophic risk reduction is the set of interventions, controls, and governance measures aimed at lowering the probability or severity of large-scale harms, particularly those posed by advanced AI systems. It spans technical safeguards such as alignment and capability control, organisational measures such as staged deployment and incident response, and policy measures such as compute governance. It is a core goal of AI safety and alignment work because it targets the tail risks that could threaten societal stability or human survival.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:catastrophic-risk-reduction", "labels": [ "Catastrophic Risk Reduction" ], "is_subclass_of": [ "AI Safety", "Global Catastrophic Risk", "Risk Management" ], "wikilinks": [] }, { "id": "category-hierarchies", "title": "Category Hierarchies", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured classification system that organizes concepts, objects, or information into parent-child relationships forming a tree-like taxonomy, enabling logical navigation from broad categories to specific subcategories for knowledge organization and content management.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:category-hierarchies", "labels": [ "Category Hierarchies", "Hierarchical Category" ], "is_subclass_of": [ "Standards and Interoperability", "Knowledge Organization System" ], "wikilinks": [ "Classification Rules", "Content Classification", "Navigation Structure", "Semantic Organization", "Taxonomy Design", "Term Relationships", "Computer Vision", "Knowledge Organization System", "metaverse" ] }, { "id": "causal-attention", "title": "Causal Attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Causal Attention (also called masked self-attention) is an attention mechanism where each token position attends only to itself and earlier positions in the sequence, enforced via an upper-triangular mask applied before softmax. This unidirectional constraint is essential for autoregressive language model training and inference, ensuring predictions at position i depend only on positions less than i.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:causal-attention", "labels": [ "Causal Attention" ], "is_subclass_of": [ "AI Model Architecture", "Attention Mechanism" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "causal-inference", "title": "Causal Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Causal inference is the scientific and statistical discipline concerned with drawing conclusions about cause-and-effect relationships from data, distinguishing genuine causal mechanisms from mere statistical association. It employs frameworks such as potential outcomes (Rubin causal model), structural causal models (Pearl's do-calculus), and graphical models (directed acyclic graphs) to formalise interventions and reason about counterfactuals. Applications span medicine, economics, social science, and AI alignment, wherever understanding the effect of an action \u2014 not merely its correlation with outcomes \u2014 is required.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:causal-inference", "labels": [ "Causal Inference" ], "is_subclass_of": [ "Bayesian Inference", "Statistical Learning", "Mathematical Reasoning", "Inference" ], "wikilinks": [] }, { "id": "causal-language-modelling", "title": "Causal Language Modelling", "domain": "ai", "domain_name": "Ai", "definition": "Causal language modelling is a self-supervised pre-training objective in which a neural network learns to predict the next token in a sequence given all preceding tokens, modelling the joint probability of text as an autoregressive product of conditional distributions. The term 'causal' refers to the unidirectional (left-to-right) attention mask that enforces the temporal ordering of tokens, preventing the model from attending to future context. This objective is the foundation of decoder-only transformer architectures such as GPT, LLaMA, and Claude, which power the majority of state-of-the-art large language models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:causal-language-modelling", "labels": [ "Causal Language Modelling", "Autoregressive Language Modeling", "Causal Language Model" ], "is_subclass_of": [ "Large Language Model Training", "Self-Supervised Learning", "Language Modelling", "Pre-Training" ], "wikilinks": [] }, { "id": "causal-loop-diagram", "title": "Causal Loop Diagram", "domain": "data", "domain_name": "Data", "definition": "A causal loop diagram (CLD) is a system-dynamics visualisation tool that maps variables as nodes connected by directed causal links, each marked with a positive or negative polarity. Closed chains of links form reinforcing or balancing feedback loops that reveal the structural drivers of a system's behaviour over time. CLDs matter because they expose feedback structure and delays that linear cause-effect thinking misses, supporting qualitative analysis before quantitative simulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:causal-loop-diagram", "labels": [ "Causal Loop Diagram" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [] }, { "id": "celestia", "title": "Celestia", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Celestia is a modular blockchain network that decouples data availability from execution and consensus, functioning as a dedicated data availability layer that rollups and sovereign chains can use to publish and order transaction data without requiring a monolithic execution environment. It employs data availability sampling (DAS) via erasure coding so that light nodes can probabilistically verify that block data has been published without downloading it in full, enabling Celestia nodes to scale with the number of light clients rather than validators. Celestia introduced the concept of sovereign rollups, in which chains publish data to Celestia for ordering and availability whilst handling their own execution and settlement independently.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:celestia", "labels": [ "Celestia", "Celestia Specifications" ], "is_subclass_of": [ "Data Availability" ], "wikilinks": [] }, { "id": "celestial-mechanics", "title": "Celestial Mechanics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:celestial-mechanics", "labels": [ "Celestial Mechanics" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "celestial-reference-frame", "title": "Celestial Reference Frame", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:celestial-reference-frame", "labels": [ "Celestial Reference Frame" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cellular-automata", "title": "Cellular Automata", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A cellular automaton is a discrete computational model consisting of a regular grid of cells, each holding one of a finite set of states, that evolve over discrete time steps according to a fixed local update rule applied uniformly across the grid. The next state of a cell depends only on its current state and the states of a defined neighbourhood, yet iterated application of simple local rules can produce complex, emergent global behaviour. Cellular automata are used to study self-organisation and computational universality, and serve as lightweight simulation substrates for physical, biological and social systems.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:cellular-automata", "labels": [ "Cellular Automata" ], "is_subclass_of": [ "Computational Model" ], "wikilinks": [] }, { "id": "cellular-network", "title": "Cellular Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cellular network is a wireless communication system in which a geographic area is divided into cells, each served by a base station, enabling mobile devices to connect and maintain service while moving. Frequency reuse across cells and handover between base stations allow large numbers of users to share limited radio spectrum. Successive generations such as 4G LTE and 5G have progressively increased capacity, throughput and latency performance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cellular-network", "labels": [ "Cellular Network" ], "is_subclass_of": [ "Telecommunications Infrastructure" ], "wikilinks": [] }, { "id": "cen-cenelec", "title": "Cen Cenelec", "domain": "standards", "domain_name": "Standards", "definition": "CEN-CENELEC refers to the joint operation of the European Committee for Standardization (CEN) and the European Committee for Electrotechnical Standardization (CENELEC), the recognised European Standardisation Organisations responsible for developing voluntary European Standards (EN). Together with ETSI they produce harmonised standards that support the EU single market and underpin presumption of conformity with European legislation. CEN covers general and mechanical sectors while CENELEC covers electrotechnical fields.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cen-cenelec", "labels": [ "Cen Cenelec", "CEN-CENELEC" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "censorship-resistance", "title": "Censorship Resistance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Property of blockchain networks guaranteeing that any valid transaction submitted by any participant will eventually be included in the canonical chain, preventing miners, validators, or any coordinated group from systematically excluding transactions. Achieved through decentralised consensus, permissionless participation, and economic incentive alignment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:censorship-resistance", "labels": [ "Censorship Resistance", "CensorshipResistance", "censorship-resistance" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "censorship-resistant-payments", "title": "Censorship Resistant Payments", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Censorship resistant payments are financial transactions that cannot be selectively blocked, reversed, or denied by any single authority \u2014 including governments, financial institutions, or network intermediaries \u2014 by virtue of operating on decentralised, permissionless infrastructure. Resistance to censorship is achieved through architectural properties such as open validator sets, pseudonymous addressing, and consensus rules that treat all valid transactions equally regardless of the identities or purposes of parties. Bitcoin, Monero, and decentralised finance protocols on public blockchains are archetypal examples, contrasting with traditional payment rails where correspondent banks, card networks, and processors can refuse service.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:censorship-resistant-payments", "labels": [ "Censorship Resistant Payments" ], "is_subclass_of": [ "Censorship Resistance" ], "wikilinks": [] }, { "id": "central-bank-digital-currency-cbdc", "title": "Central Bank Digital Currency (CBDC)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sovereign digital currency issued and backed by a central bank for use in retail or wholesale payment systems, functioning as legal tender in digital form.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:central-bank-digital-currency-cbdc", "labels": [ "Central Bank Digital Currency (CBDC)", "CBDC", "Central Bank Digital Currency" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Banking Infrastructure", "BIS CBDC Blueprint", "Central Bank", "Digital Currency Ledger", "Financial Inclusion", "Identity System", "IMF CBDC Notes", "ISO 24165", "Payment Infrastructure", "Payment Protocol", "Programmable Money", "Settlement System", "Transaction Validator", "Blockchain", "Cryptographic Protocol", "DataLayer", "Digital Payments", "Digital Wallet", "Distributed Ledger", "Financial System" ] }, { "id": "central-bank-digital-currency", "title": "Central Bank Digital Currency", "domain": "finance", "domain_name": "Finance", "definition": "A digital form of sovereign fiat money issued directly by a nation's central bank, representing a liability of the monetary authority that can serve as legal tender for retail payments or wholesale settlement, distinct from commercial bank deposits and decentralized cryptocurrencies.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:central-bank-digital-currency", "labels": [ "Central Bank Digital Currency", "CBDC", "CentralBankDigitalCurrency", "Digital Dollar", "Tokenised Central Bank Money" ], "is_subclass_of": [ "Digital Currency" ], "wikilinks": [ "Central Bank Infrastructure", "Financial Inclusion", "Monetary Policy Transmission", "Payment Network", "Blockchain", "Digital Currency", "Digital Identity", "Digital Payments", "metaverse" ] }, { "id": "central-bank-infrastructure", "title": "Central Bank Infrastructure", "domain": "finance", "domain_name": "Finance", "definition": "Central bank infrastructure encompasses the systems, networks, and institutional arrangements through which a central bank fulfils its core functions: operating real-time gross settlement (RTGS) systems for interbank payments, managing the issuance and lifecycle of physical and digital currency, implementing monetary policy, maintaining financial stability, and serving as lender of last resort. It includes SWIFT connectivity, RTGS platforms (such as CHAPS, Fedwire, TARGET2), central securities depositories, and increasingly the digital and distributed ledger systems under evaluation for CBDC issuance. This infrastructure forms the foundation upon which all commercial banking and payment activity ultimately settles.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:central-bank-infrastructure", "labels": [ "Central Bank Infrastructure" ], "is_subclass_of": [ "Payment Infrastructure" ], "wikilinks": [] }, { "id": "central-bank", "title": "Central Bank", "domain": "finance", "domain_name": "Finance", "definition": "A central bank is a public financial institution vested with exclusive authority over a nation's or currency union's monetary base, responsible for setting benchmark interest rates, controlling the money supply, and acting as lender of last resort to the commercial banking sector. Central banks implement monetary policy to pursue macroeconomic objectives including price stability, sustainable economic growth, and full employment, typically operating with a degree of independence from short-term political influence. They hold official foreign exchange reserves, supervise the broader banking system, operate or oversee critical payment infrastructure, and increasingly research or issue Central Bank Digital Currencies as sovereign digital money. Major examples include the Federal Reserve, the European Central Bank, the Bank of England, the Bank of Japan, and the People's Bank of China.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:central-bank", "labels": [ "Central Bank", "Central Bank Oversight" ], "is_subclass_of": [ "Monetary System" ], "wikilinks": [] }, { "id": "central-banking", "title": "Central Banking", "domain": "finance", "domain_name": "Finance", "definition": "Central banking is the institutional practice of managing a nation's currency, money supply, and interest rates through a designated monetary authority. It encompasses setting policy rates, managing reserve requirements, and acting as lender of last resort to the banking system. Central banking decisions propagate through interbank settlement, credit markets, and macroeconomic conditions.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:central-banking", "labels": [ "Central Banking" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [] }, { "id": "central-limit-theorem", "title": "Central Limit Theorem", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Central Limit Theorem states that, under broad conditions, the distribution of the sum or mean of a large number of independent, identically distributed random variables approaches a normal distribution regardless of the underlying distribution's shape. It explains the ubiquity of the Gaussian distribution and provides the theoretical basis for many inferential procedures, including confidence intervals and significance tests. It is foundational to statistics, machine learning, and Monte Carlo estimation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:central-limit-theorem", "labels": [ "Central Limit Theorem" ], "is_subclass_of": [ "Probability Theory", "Measure Theory", "Limit Theorems", "Statistical Theory" ], "wikilinks": [] }, { "id": "central-processing-unit", "title": "Central Processing Unit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The central processing unit (CPU) is the primary computational component of a computer, executing the instructions of programs through arithmetic, logic, control, and input/output operations. It fetches instructions from memory, decodes them, and executes them in a repeating cycle, coordinated by a control unit and carried out by arithmetic-logic units across one or more cores. As a general-purpose processor it contrasts with specialised accelerators that optimise for narrower, highly parallel workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:central-processing-unit", "labels": [ "Central Processing Unit" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [] }, { "id": "central-securities-depository", "title": "Central Securities Depository", "domain": "finance", "domain_name": "Finance", "definition": "A central securities depository (CSD) is a financial market infrastructure that holds securities in dematerialised or immobilised form and enables their transfer through book-entry, providing safekeeping and settlement of trades. It maintains the authoritative record of securities ownership and supports the final leg of post-trade processing, often in conjunction with clearing houses. By centralising custody and settlement, a CSD reduces operational risk and is foundational to the integrity of securities markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:central-securities-depository", "labels": [ "Central Securities Depository" ], "is_subclass_of": [ "Financial Market Infrastructure" ], "wikilinks": [] }, { "id": "centralised-ai", "title": "Centralised AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An organisational and architectural model of artificial intelligence in which training data, compute, model weights, and inference services are concentrated under a single operator in large data centres, with users accessing capability remotely through APIs; the contrast class to decentralised AI, edge inference, and multi-agent approaches that distribute computation, control, or decision-making across many independent nodes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:centralised-ai", "labels": [ "Centralised AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "Decentralised AI", "Edge Inference" ] }, { "id": "centralized-application", "title": "Centralised Application", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A software application whose logic, state, and data are operated by a single controlling entity on infrastructure it administers, typically following a client-server architecture in which users depend on the operator for availability, data custody, access control, and rule changes, in contrast to decentralised applications whose execution and state are replicated across a permissionless network.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:centralized-application", "labels": [ "Centralised Application", "Centralized Application" ], "is_subclass_of": [ "Software System" ], "wikilinks": [ "Software System", "Decentralised Application", "Cloud Computing" ] }, { "id": "centralised-control", "title": "Centralised Control", "domain": "governance", "domain_name": "Governance", "definition": "Centralised control is a governance arrangement in which decision-making authority, coordination, and enforcement are concentrated in a single entity or hierarchy rather than distributed among participants. It offers clarity of accountability and rapid, consistent decisions but introduces a single point of failure and dependence on the controlling party's competence and intentions. Centralised control is the principal foil against which decentralised and distributed governance models define themselves.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:centralised-control", "labels": [ "Centralised Control" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "centralised-database", "title": "Centralised Database", "domain": "data", "domain_name": "Data", "definition": "A centralised database is a data management system in which all data is stored, administered, and accessed through a single physical or logical location under a single controlling authority. It provides a unified, authoritative view of data with strong consistency guarantees, simplified access control, and a single point of administration. All read and write operations are routed to this central node or cluster, making it the canonical source of truth for the entire system. The centralised model contrasts with distributed and decentralised architectures by sacrificing geographic fault-tolerance and autonomy in exchange for consistency, reduced coordination overhead, and operational simplicity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:centralised-database", "labels": [ "Centralised Database", "Centralised Data Silo", "Centralized Database" ], "is_subclass_of": [ "Database Management System" ], "wikilinks": [ "Decentralised Storage", "Distributed Systems" ] }, { "id": "centralised-exchange", "title": "Centralised Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A centralised exchange (CEX) is a custodial trading venue operated by a single company that matches buy and sell orders for cryptocurrencies and other assets through an internal order book. Users deposit funds into accounts the operator controls, and the exchange maintains the ledger of balances off-chain, settling trades internally rather than on a public blockchain. Centralised exchanges typically enforce identity verification and provide high liquidity and fast execution, at the cost of requiring trust in the operator.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:centralised-exchange", "labels": [ "Centralised Exchange" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [] }, { "id": "centralised-finance", "title": "Centralised Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Centralised Finance (CeFi) is a model of financial service provision in which a trusted intermediary custodies user assets and operates the order books, matching engines and settlement rails on behalf of participants. In the cryptocurrency context it denotes exchanges and lending platforms that hold customer funds in pooled wallets and enforce identity verification, in contrast to non-custodial, smart-contract-mediated alternatives. CeFi platforms offer familiar account models, fiat on-ramps and regulatory compliance at the cost of counterparty trust.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:centralised-finance", "labels": [ "Centralised Finance" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "centralised-governance", "title": "Centralised Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Centralised governance is a model in which decision-making authority over a system, protocol, or organisation is concentrated in a single entity or a small controlling group. In contrast to distributed models, changes, upgrades, and policy are determined by this authority rather than by broad stakeholder consensus, which can deliver faster, more decisive action at the cost of reduced censorship resistance and a single point of control. It is the conventional baseline against which decentralised governance defines itself.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:centralised-governance", "labels": [ "Centralised Governance" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "centralised-identifier", "title": "Centralised Identifier", "domain": "security", "domain_name": "Security", "definition": "An identifier issued, controlled, and resolvable only through a central authority\u2014such as an email address bound to a provider, a username on a platform, a government-issued number, or a domain name under registry control. Centralised identifiers are the architectural opposite of decentralised identifiers: the issuing authority can revoke, reassign, surveil, or lose them, creating single points of failure and lock-in, but also offering simple governance, accountability, and recovery paths that self-sovereign schemes must engineer explicitly.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:centralised-identifier", "labels": [ "Centralised Identifier" ], "is_subclass_of": [ "Identifier" ], "wikilinks": [ "Identifier", "Decentralised Identifier", "Single Point of Failure" ] }, { "id": "centralised-identity", "title": "Centralised Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Centralised identity is a digital identity model in which a single authority \u2014 such as a government, enterprise, or platform identity provider \u2014 issues, stores, and controls users' identity credentials and authenticates them on behalf of relying parties. Users authenticate against the central provider, which holds the authoritative record of their attributes and mediates access to connected services. The model is operationally simple and widely deployed, but concentrates control, data, and risk in one party, creating single points of failure, surveillance potential, and vendor lock-in that decentralised and self-sovereign identity approaches are designed to counter.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:centralised-identity", "labels": [ "Centralised Identity" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "centralized-exchange", "title": "Centralized Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A centralized exchange (CEX) is a digital-asset trading venue operated by a single company that custodies user funds, matches orders through an internal order book, and acts as an intermediary for every trade. Users deposit assets into accounts controlled by the operator, who maintains liquidity, settlement, and the matching engine off-chain. CEXs offer high throughput and familiar interfaces but require trust in the operator and typically enforce identity-verification and anti-money-laundering controls.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:centralized-exchange", "labels": [ "Centralized Exchange" ], "is_subclass_of": [ "Blockchain", "DeFi and Economics" ], "wikilinks": [] }, { "id": "centralized-identity-provider", "title": "Centralized Identity Provider", "domain": "security", "domain_name": "Security", "definition": "An identity service in which a single organisation issues, stores and authenticates user credentials on behalf of relying parties. It contrasts with decentralised identity models where control rests with the user.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:centralized-identity-provider", "labels": [ "Centralized Identity Provider", "Centralised Identity Provider", "Centralized Identity" ], "is_subclass_of": [ "Identity Provider" ], "wikilinks": [ "Identity Management", "Identity Verification System", "Decentralized Identifier", "Identity Provider" ] }, { "id": "centralized-swarm-control", "title": "Centralized Swarm Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Swarm robotics control architecture where a central controller coordinates all robot agents, providing global optimisation but creating a single point of failure.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:centralized-swarm-control", "labels": [ "Centralized Swarm Control" ], "is_subclass_of": [ "Robot Type", "Control System", "Swarm Control", "Robotics", "Centralized Control Architecture" ], "wikilinks": [ "Central Controller", "Centralised Monitoring", "Centralized Control Architecture", "Communication Network", "Computational Resources", "Coordinated Swarm Behaviour", "Drone Swarms", "Global State Model", "Global Task Optimisation", "Hierarchical Control Systems", "Inter-Agent Communication", "Production Scheduling", "Robot Agent", "Swarm Control", "Synchronised Timing", "Decentralized Swarm Control", "Robotics", "RoboticsDomain", "Swarm Robotics" ] }, { "id": "centrifuge", "title": "Centrifuge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Centrifuge is a decentralised finance protocol that brings real-world assets such as invoices and loans on chain so they can be used as collateral for financing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:centrifuge", "labels": [ "Centrifuge" ], "is_subclass_of": [ "Asset Tokenisation" ], "wikilinks": [ "Asset Tokenisation", "Smart Contract", "DeFi" ] }, { "id": "ceph", "title": "Ceph", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An open-source software-defined storage platform that provides object, block, and file storage from a single self-healing cluster of commodity servers, using the CRUSH algorithm to place data deterministically without a central metadata bottleneck, and protecting data through replication or erasure coding; widely deployed beneath OpenStack and Kubernetes as exabyte-scale storage infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:ceph", "labels": [ "Ceph" ], "is_subclass_of": [ "Distributed Storage" ], "wikilinks": [ "Distributed Storage", "Erasure Coding", "Storage Infrastructure" ] }, { "id": "cerebras-wafer-scale-chips", "title": "Cerebras Wafer Scale Chips", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A specialized AI hardware architecture developed by Cerebras that integrates a massive number of transistors onto a single wafer-scale integrated circuit to enable high-speed AI inference and training.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cerebras-wafer-scale-chips", "labels": [ "Cerebras Wafer Scale Chips" ], "is_subclass_of": [ "AI Hardware Infrastructure" ], "wikilinks": [] }, { "id": "certificate-authority", "title": "certificate authority", "domain": "security", "domain_name": "Security", "definition": "A Certificate Authority (CA) is a trusted third-party organisation that issues, signs, and manages the lifecycle of X.509 digital certificates, cryptographically binding a public key to an entity's verified identity within a Public Key Infrastructure (PKI). CAs operate within a hierarchical trust model where offline root CAs delegate signing authority to online intermediate CAs, which in turn issue end-entity certificates for servers, users, clients, and devices. The CA's signing operations \u2014 governed by RFC 5280, the CA/Browser Forum Baseline Requirements, and WebTrust auditing standards \u2014 underpin the TLS handshake, code-signing pipelines, S/MIME email encryption, mutual TLS (mTLS) in service meshes, and document-signing workflows across the internet. Trust propagates via chain-of-trust verification: a relying party validates each certificate against its issuer's signature, tracing back to an implicitly trusted root embedded in OS or browser trust stores.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:certificate-authority", "labels": [ "Certificate Authority", "Certificate Authority Service", "Intermediate Certificate Authority" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "certificate-revocation-list", "title": "Certificate Revocation List", "domain": "security", "domain_name": "Security", "definition": "A Certificate Revocation List (CRL) is a digitally signed, periodically published list of digital certificates that a certificate authority has revoked before their scheduled expiry. Each entry records the serial number of a revoked certificate, the revocation date, and an optional reason code, allowing relying parties to reject certificates that are no longer trustworthy. CRLs are a core revocation mechanism of X.509 public key infrastructure, complemented or replaced in many deployments by the Online Certificate Status Protocol.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:certificate-revocation-list", "labels": [ "Certificate Revocation List" ], "is_subclass_of": [ "Revocation Registry" ], "wikilinks": [] }, { "id": "certificate-revocation", "title": "Certificate Revocation", "domain": "security", "domain_name": "Security", "definition": "Certificate Revocation is the process by which a Certificate Authority (CA) invalidates a previously issued digital certificate before its natural expiry, typically due to key compromise, CA compromise, or change in the certificate holder's status. Revocation information is distributed via Certificate Revocation Lists (CRLs) or the Online Certificate Status Protocol (OCSP). It is a critical component of public key infrastructure (PKI) lifecycle management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:certificate-revocation", "labels": [ "Certificate Revocation" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "certificate-transparency", "title": "Certificate Transparency", "domain": "security", "domain_name": "Security", "definition": "Certificate Transparency (CT) is an open framework and internet standard (RFC 6962 and RFC 9162) that creates a publicly auditable, append-only log of all TLS certificates issued by certificate authorities, enabling domain owners, browser vendors, and security researchers to detect misissued or fraudulent certificates rapidly. CT logs use Merkle hash trees to provide cryptographic proofs of inclusion and consistency, guaranteeing that any certificate added to a log cannot be subsequently removed or altered. Major browsers enforce CT by requiring certificates to carry signed certificate timestamps (SCTs) from recognised logs, making unauthorised certificate issuance immediately detectable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:certificate-transparency", "labels": [ "Certificate Transparency" ], "is_subclass_of": [ "Public Key Infrastructure" ], "wikilinks": [] }, { "id": "certification", "title": "Certification", "domain": "governance", "domain_name": "Governance", "definition": "Certification is the formal process by which an independent authority attests that a product, system, process or person meets defined standards or requirements. It produces a verifiable credential or mark that signals conformance to interested parties and reduces the cost of establishing trust. Certification regimes are central to regulation, quality assurance and the safe interoperability of regulated goods and services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:certification", "labels": [ "Certification" ], "is_subclass_of": [ "Compliance" ], "wikilinks": [] }, { "id": "cha-cha20-poly1305", "title": "ChaCha20-Poly1305", "domain": "security", "domain_name": "Security", "definition": "ChaCha20-Poly1305 is an authenticated encryption with associated data (AEAD) cipher suite combining the ChaCha20 stream cipher for encryption with the Poly1305 message authentication code, providing both confidentiality and integrity in a single construction. Designed by Daniel J. Bernstein, it offers performance advantages over AES-GCM on systems lacking hardware AES acceleration (particularly mobile and embedded devices) whilst providing equivalent or superior security margins. It is standardised in RFC 7539 (IETF) and RFC 8439, and is widely deployed in TLS 1.3, QUIC, WireGuard, SSH, and numerous cryptographic libraries as an alternative to AES-based suites.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cha-cha20-poly1305", "labels": [ "ChaCha20-Poly1305", "XChacha20Poly1305" ], "is_subclass_of": [ "Cryptographic Algorithm" ], "wikilinks": [] }, { "id": "chain-reorganization", "title": "Chain Reorganization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Chain Reorganization (reorg) is the replacement of one or more blocks in the canonical blockchain by an alternative chain of equal or greater cumulative proof-of-work, occurring when competing miners produce valid chains of differing lengths and the network converges on the longest-chain rule to select the canonical history. Reorgs invalidate transactions confirmed only in the abandoned chain segment, enabling double-spend attacks when deliberately induced and posing settlement finality risks for exchanges and payment processors.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:chain-reorganization", "labels": [ "Chain Reorganization", "Blockchain Reorganization" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "chain-state", "title": "Chain State", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chain State is the complete, current snapshot of all data held by a blockchain at a given block height, encompassing account balances, smart contract storage, unspent transaction outputs (UTXOs), and any other data structures committed to the ledger. It represents the authoritative, globally agreed world-state that full nodes maintain and update after each validated block, serving as the ground truth against which new transactions are validated.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:chain-state", "labels": [ "Chain State", "Source Chain State" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "chain-of-custody", "title": "Chain of Custody", "domain": "security", "domain_name": "Security", "definition": "Chain of Custody is a documented, unbroken sequence of records that tracks the possession, handling, transfer, analysis, and disposition of physical or digital items from their point of origin to their ultimate use\u2014most critically in legal proceedings and regulated industries. Each handoff event must be recorded with the identity of the parties involved, the time and location of transfer, and the state of the item, ensuring that the integrity and authenticity of the item can be demonstrated to any subsequent examiner. In digital contexts the concept extends to data provenance, AI training datasets, and blockchain-based asset transfer records.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:chain-of-custody", "labels": [ "Chain of Custody", "Chain of Custody Tracking" ], "is_subclass_of": [ "Provenance Tracking" ], "wikilinks": [] }, { "id": "chain-of-thought", "title": "chain of thought", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Chain-of-Thought (CoT) prompting is a technique for eliciting explicit intermediate reasoning steps from autoregressive large language models before producing a final answer, substantially improving accuracy on arithmetic, symbolic, commonsense, and multi-hop reasoning tasks. The mechanism exploits the sequential token-generation process of transformer-based models: each generated reasoning token conditions all subsequent tokens, enabling multi-step deductions that single-pass prompting cannot reliably perform. CoT encompasses a family of variants \u2014 few-shot exemplar CoT, zero-shot CoT, self-consistency decoding, tree-of-thought search, and process reward modelling \u2014 collectively forming a foundational paradigm for inference-time compute scaling and complex reasoning in large language models.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:chain-of-thought", "labels": [ "Chain of Thought", "Chain Of Thought", "Chain-of-Thought", "Chain-of-Thought Prompting" ], "is_subclass_of": [ "Prompt Engineering", "Reasoning", "In-Context Learning" ], "wikilinks": [] }, { "id": "chain-of-thought-prompting", "title": "Chain-of-Thought Prompting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Chain-of-thought (CoT) prompting is a prompt engineering technique that elicits intermediate reasoning steps from a large language model before it produces a final answer, substantially improving performance on multi-step arithmetic, commonsense reasoning, and symbolic manipulation tasks. By including exemplars that demonstrate step-by-step reasoning or by appending the instruction 'Let's think step by step' (zero-shot CoT), the technique leverages the model's autoregressive generation to decompose complex problems into tractable substeps. CoT prompting was formally characterised by Wei et al. (2022) and has become a foundational capability-elicitation method for frontier language models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:chain-of-thought-prompting", "labels": [ "Chain-of-Thought Prompting", "Chain Of Thought Prompting" ], "is_subclass_of": [ "Prompt Engineering", "In-Context Learning", "Few-Shot Prompting" ], "wikilinks": [] }, { "id": "chain-of-thought-reasoning", "title": "chain-of-thought reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Chain-of-Thought Reasoning is the capacity of large language models to generate coherent sequences of intermediate logical, mathematical, or factual steps as part of producing a final answer, treating each intermediate token as an active computational resource that reshapes residual stream activations for subsequent Transformer layers. Unlike the surface prompting technique that elicits it, CoT Reasoning denotes an emergent model capability \u2014 present above certain parameter-scale thresholds \u2014 whereby the model decomposes complex problems into verifiable sub-steps rather than collapsing directly to a final token. This capability underpins inference-time compute scaling strategies including self-consistency sampling, process reward models, step-level verifiers, and tree-of-thought search, and connects neural language modelling to classical AI paradigms of deductive planning and symbolic reasoning.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:chain-of-thought-reasoning", "labels": [ "Chain-of-Thought Reasoning", "Chain of Thought Reasoning" ], "is_subclass_of": [ "AI Technique", "Emergent Capability", "Inference Protocol" ], "wikilinks": [] }, { "id": "chainalysis", "title": "Chainalysis", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chainalysis is a commercial blockchain data analytics and intelligence platform founded in 2014 that provides investigative, compliance, and risk-management tools to government agencies, financial institutions, and cryptocurrency businesses for tracing, monitoring, and understanding blockchain transaction flows. It maintains a large proprietary database of attributed blockchain addresses \u2014 linking pseudonymous on-chain addresses to real-world entities through heuristic clustering, open-source intelligence, data partnerships, and legal processes \u2014 and offers products including Reactor (graph-based investigation tool), KYT (Know Your Transaction real-time compliance API), and Kryptos (market intelligence). Chainalysis is a primary contractor to agencies including the US Department of Justice, IRS Criminal Investigation, and OFAC for cryptocurrency-related law enforcement investigations, and publishes the annual Crypto Crime Report, the most widely cited source for cryptocurrency illicit finance statistics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:chainalysis", "labels": [ "Chainalysis" ], "is_subclass_of": [ "Blockchain Analytics" ], "wikilinks": [] }, { "id": "chainlink-ccip", "title": "Chainlink CCIP", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chainlink Cross-Chain Interoperability Protocol (CCIP) is a cross-chain messaging and token transfer standard developed by Chainlink Labs that enables smart contracts on different blockchain networks to securely send messages and transfer assets across chain boundaries, backed by a decentralised oracle network providing an independent risk management layer that monitors and validates cross-chain transactions. CCIP introduces a programmable token transfer abstraction allowing arbitrary data payloads to accompany token movements, enabling complex cross-chain application logic.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:chainlink-ccip", "labels": [ "Chainlink CCIP", "Chainlink-CCIP" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "chainlink-oracles", "title": "Chainlink Oracles", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chainlink Oracles are decentralised oracle networks operated by the Chainlink protocol that securely fetch, validate, and deliver off-chain data to on-chain smart contracts. They aggregate inputs from multiple independent node operators using cryptographic proofs and reputation systems to ensure tamper-resistant data feeds. Widely deployed for price feeds, verifiable randomness, and cross-chain communication, they form critical infrastructure for decentralised finance and Web3 applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:chainlink-oracles", "labels": [ "Chainlink Oracles", "Chainlink Oracle" ], "is_subclass_of": [ "Oracle Network" ], "wikilinks": [] }, { "id": "chainlink", "title": "Chainlink", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chainlink is a decentralised oracle network and middleware layer that enables smart contracts on any blockchain to securely access off-chain data, computation, and cross-chain interoperability services. Founded in 2017 by Sergey Nazarov and Steve Ellis, it operates a network of independent node operators who retrieve, aggregate, and deliver external data \u2014 including price feeds, verifiable randomness, API responses, and event outcomes \u2014 to on-chain smart contracts, solving the oracle problem that prevents blockchains from natively interacting with real-world information. Its native token (LINK) provides cryptoeconomic incentives and collateral for node operators, whilst its Off-Chain Reporting protocol reduces on-chain costs through peer-to-peer consensus. Beyond data feeds, Chainlink has expanded into cross-chain messaging (CCIP), on-chain automation, and verifiable computation, positioning itself as general-purpose decentralised infrastructure for the entire blockchain ecosystem.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:chainlink", "labels": [ "Chainlink" ], "is_subclass_of": [ "Blockchain Oracle" ], "wikilinks": [] }, { "id": "challenge-response-protocol", "title": "Challenge-Response Protocol", "domain": "security", "domain_name": "Security", "definition": "A challenge-response protocol is an authentication protocol in which a verifier issues an unpredictable challenge and the claimant must return a response computed from a shared secret or private key, proving knowledge of the credential without transmitting it. Because each challenge is fresh, typically a random nonce, a valid response cannot be reused, defeating replay attacks. The pattern underpins many authentication mechanisms including CHAP, HMAC-based schemes, and public-key signature challenges.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:challenge-response-protocol", "labels": [ "Challenge-Response Protocol" ], "is_subclass_of": [ "Authentication Protocol" ], "wikilinks": [] }, { "id": "change-data-capture", "title": "Change Data Capture", "domain": "data", "domain_name": "Data", "definition": "Change data capture is a set of techniques for identifying and propagating row-level changes \u2014 inserts, updates, and deletes \u2014 from a source database to downstream systems in near real time. The most robust approach reads the database transaction log, turning committed mutations into an ordered stream of change events without burdening the source with polling. It underpins data replication, event streaming, and incremental data integration, keeping analytical stores, caches, and microservices consistent with operational systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:change-data-capture", "labels": [ "Change Data Capture" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [] }, { "id": "change-detection", "title": "Change Detection", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:change-detection", "labels": [ "Change Detection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "change-management", "title": "Change Management", "domain": "governance", "domain_name": "Governance", "definition": "Change Management is the structured discipline of guiding organisations, teams, and individuals through transitions in technology, process, or culture to achieve desired outcomes whilst minimising disruption and resistance. It encompasses planning, communication, training, stakeholder alignment, and feedback loops that collectively ensure adoption of new working methods. In technology contexts it also covers the formal control of modifications to systems and infrastructure to reduce risk. Effective change management is widely recognised as a critical determinant of whether digital transformation initiatives deliver their intended value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:change-management", "labels": [ "Change Management", "Change Management Process", "Change Management Programme" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "change-of-variables", "title": "Change of Variables", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Change of variables is a technique for re-expressing an integral or a probability density in terms of a new set of variables related to the original ones by a differentiable, invertible transformation, with the Jacobian determinant of that transformation accounting for the resulting change in volume. Normalising flows apply this principle directly, composing a sequence of invertible transformations and tracking the accumulated Jacobian determinant to convert a simple base density into a complex target density while keeping the density exactly computable. It is a standard tool wherever a probability distribution must be transported through a differentiable map.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:change-of-variables", "labels": [ "Change of Variables" ], "is_subclass_of": [ "Calculus" ], "wikilinks": [] }, { "id": "channel-coding", "title": "Channel Coding", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Channel coding is the branch of coding theory that adds controlled redundancy to transmitted data so that errors introduced by a noisy communication channel can be detected and corrected at the receiver. Techniques include block codes, convolutional codes, turbo codes, and LDPC codes, with forward error correction being its dominant application. It matters because it lets systems approach the Shannon channel capacity, trading bandwidth for reliability in wireless, storage, and deep-space links.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:channel-coding", "labels": [ "Channel Coding" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "channel-factory", "title": "Channel Factory", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A channel factory is a Bitcoin Lightning Network scaling construction in which multiple participants share a single on-chain funding transaction (a multiparty channel) from which many off-chain payment channels can be opened, closed, and rebalanced without further on-chain transactions. By amortising one on-chain output across many channels, it reduces the on-chain footprint and cost of channel management. It matters because it improves the capital efficiency and scalability of layer-2 payments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:channel-factory", "labels": [ "Channel Factory" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [] }, { "id": "chaos-engineering", "title": "Chaos Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Chaos engineering is the discipline of experimenting on a software system by deliberately injecting controlled faults \u2014 such as instance termination, network latency, or resource exhaustion \u2014 in order to build confidence in the system's ability to withstand turbulent, real-world conditions. Practitioners form a hypothesis about steady-state behaviour, introduce a failure in production or production-like environments, and observe whether the system maintains its service level. Originating with Netflix's Chaos Monkey, the practice surfaces hidden dependencies and weaknesses before they cause outages, complementing observability and resilient design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:chaos-engineering", "labels": [ "Chaos Engineering" ], "is_subclass_of": [ "Site Reliability Engineering" ], "wikilinks": [] }, { "id": "chaos-theory", "title": "Chaos Theory", "domain": "data", "domain_name": "Data", "definition": "Chaos theory is the study of deterministic dynamical systems whose long-term behaviour is highly sensitive to initial conditions, making them practically unpredictable despite obeying fixed deterministic rules. It characterises phenomena such as strange attractors, bifurcations, and the exponential divergence of nearby trajectories measured by Lyapunov exponents. It matters because it explains how simple nonlinear rules can produce complex, aperiodic behaviour, bridging deterministic dynamics and apparent randomness in physical, biological, and economic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:chaos-theory", "labels": [ "Chaos Theory" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "chapman-kolmogorov-equation", "title": "Chapman-Kolmogorov Equation", "domain": "ai", "domain_name": "Ai", "definition": "The Chapman-Kolmogorov Equation is a fundamental identity in probability theory and stochastic processes that expresses the consistency condition for transition probabilities of a Markov chain or continuous stochastic process. It states that the probability of moving from state i to state j in n+m steps equals the sum over all intermediate states k of the product of the n-step and m-step transition probabilities. This equation is the cornerstone of Markov chain analysis and underlies key algorithms in machine learning, Bayesian inference, and diffusion modelling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:chapman-kolmogorov-equation", "labels": [ "Chapman-Kolmogorov Equation" ], "is_subclass_of": [ "Stochastic Process", "Mathematical Identity", "Probability Theory" ], "wikilinks": [] }, { "id": "character-animation", "title": "Character Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Character animation is the discipline of creating the illusion of life and intentional movement in digital or physical characters through the coordinated manipulation of skeletal rigs, blend shapes, and motion data across time, driven by principles derived from traditional film animation. It encompasses the full pipeline from rigging and skinning a character mesh to authoring, retargeting, and blending motion clips in real time, and extends to AI-driven procedural and physics-based approaches that generate plausible movement without manual keyframing. The field spans offline cinematic animation for film and games and real-time systems for interactive avatars, XR experiences, and virtual humans.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:character-animation", "labels": [ "Character Animation", "3D Character Animation", "Character Animation Reuse" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "character-model", "title": "Character Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D digital representation of a character consisting of a polygonal mesh, textures, and materials, designed with clean topology optimized for animation and real-time rendering, serving as the foundation for rigging, skinning, and character animation workflows.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:character-model", "labels": [ "Character Model" ], "is_subclass_of": [ "Content and Assets", "3D Model" ], "wikilinks": [ "3D Modeling Software", "Character Animation", "Game Characters", "Topology Design", "Virtual Avatars", "3D Model", "Computer Vision", "metaverse", "Texture Mapping" ] }, { "id": "character-relationship-graph", "title": "Character Relationship Graph", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A network data structure representing narrative characters as vertices and their interactions or relationships as edges, used to model social dynamics, drive procedural story generation, and analyze narrative structure through graph theory and social network analysis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:character-relationship-graph", "labels": [ "Character Relationship Graph" ], "is_subclass_of": [ "Content and Assets", "Social Network Graph" ], "wikilinks": [ "Character AI", "Character Attributes", "Narrative Generation", "Relationship Modeling", "Story Analysis", "Graph Database", "metaverse", "Social Network Graph", "Telecollaboration" ] }, { "id": "character-rigging", "title": "Character Rigging", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of creating a hierarchical skeleton of interconnected bones and joints within a 3D character model, along with control systems and deformation rules, enabling animators to manipulate the mesh through inverse kinematics and forward kinematics for realistic movement.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:character-rigging", "labels": [ "Character Rigging", "Character Rig", "Rigging", "Rigging Systems", "Rigging and Skinning", "Skeletal Rigging" ], "is_subclass_of": [ "Content and Assets", "Animation Technique" ], "wikilinks": [ "Character Animation", "Motion Capture Retargeting", "Skeletal Animation", "Skinning", "Weight Painting", "Animation Technique", "Character Model", "Computer Vision", "metaverse" ] }, { "id": "chatbot", "title": "Chatbot", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A software application designed to simulate conversation with human users, processing natural-language input through rule-based, retrieval-based, or generative mechanisms to produce contextually appropriate responses. Modern chatbots are predominantly built on large language models, enabling open-domain dialogue, multi-turn context tracking, and task execution across text and voice interfaces. They serve as the primary user-facing layer of conversational AI systems, integrating intent recognition, dialogue management, and response generation into a coherent interaction loop. Deployment contexts range from narrow-domain customer support and virtual assistants to general-purpose AI agents capable of reasoning, tool use, and multi-step problem solving.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:chatbot", "labels": [ "Chatbot", "Rule-Based Chatbot" ], "is_subclass_of": [ "Conversational AI", "Dialogue System" ], "wikilinks": [ "Natural Language Processing", "Large Language Models", "Conversational AI" ] }, { "id": "chatbots", "title": "Chatbots", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software systems designed to simulate conversation with human users through text or voice interfaces, spanning an architectural spectrum from rule-based pattern-matching programs (ELIZA, 1966; ALICE/AIML, 1995) through retrieval-based and intent-classification pipelines to modern", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:chatbots", "labels": [ "Chatbots" ], "is_subclass_of": [ "AI Application", "Dialogue System", "AI Agent System", "Dialogue Systems", "Conversational AI", "Natural Language Processing", "Human-Computer Interaction" ], "wikilinks": [ "API Integration", "BERTScore", "BLEU", "Customer Service Automation", "Dialogue State Tracking", "Dialogue Systems", "Education Technology", "Embedding Models", "Enterprise Search", "Form-Based Interfaces", "Healthcare AI", "HumanComputerInteractionDomain", "Human Evaluation", "Inference Engine", "Intent Classification", "Intent Recognition", "InterfaceLayer", "ISO 30401", "Knowledge Base", "Knowledge Retrieval" ] }, { "id": "chaumian-ecash", "title": "Chaumian Ecash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chaumian ecash is a form of digital cash, invented by David Chaum, that uses blind signatures to issue bearer tokens redeemable at a central mint while preserving payer privacy. The mint signs blinded token requests so it cannot link issued tokens to the users who later spend them, providing strong untraceability with offline-style bearer transfer. It matters as the cryptographic foundation for privacy-preserving custodial payment systems, revived in Bitcoin through mints such as Cashu and Fedimint.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:chaumian-ecash", "labels": [ "Chaumian Ecash" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "checkerboard-pattern", "title": "Checkerboard Pattern", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A checkerboard pattern is a planar grid of alternating black and white squares used as a calibration target in computer vision. Its regularly spaced corners are easy to detect with sub-pixel accuracy and have precisely known relative positions, providing reliable correspondences for estimating camera parameters. The checkerboard is the most common target for intrinsic calibration, distortion correction, and stereo rig alignment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:checkerboard-pattern", "labels": [ "Checkerboard Pattern" ], "is_subclass_of": [ "Optical Calibration Target" ], "wikilinks": [] }, { "id": "checkpoint-recovery", "title": "Checkpoint Recovery", "domain": "ai", "domain_name": "Ai", "definition": "Checkpoint recovery is the process of resuming a computation, most commonly a machine learning training run, from a previously saved checkpoint after an interruption such as a hardware failure, pre-emption or planned restart. It requires that checkpoints capture sufficient state, including model parameters, optimiser state and progress markers, to continue correctly without repeating completed work. Reliable checkpoint recovery is essential for large-scale and decentralised training where node failures are expected rather than exceptional.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:checkpoint-recovery", "labels": [ "Checkpoint Recovery" ], "is_subclass_of": [ "Model Checkpoint" ], "wikilinks": [] }, { "id": "checkpointing", "title": "Checkpointing", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A fault-tolerance technique in which a system periodically captures a consistent snapshot of its execution state \u2014 memory, variables, message queues, or conversation context \u2014 and persists it to durable storage, so that after a crash, pre-emption, or migration the computation can resume from the most recent checkpoint rather than restarting from the beginning; foundational to long-running distributed workloads, ML training, and agent runtimes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:checkpointing", "labels": [ "Checkpointing" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [ "Fault Tolerance", "State Management", "Agent Runtime" ] }, { "id": "checkpoints", "title": "Checkpoints", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Checkpoints are serialised snapshots of a machine-learning model's complete trainable state \u2014 weight tensors, optimiser moment accumulators, learning-rate schedules, gradient scalers, random-number-generator seeds, and epoch/step counters \u2014 persisted to durable storage at regular intervals during...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:checkpoints", "labels": [ "Checkpoints", "Checkpoint", "Model Checkpoints", "Voting Power Checkpoints" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline", "Training", "Model Serialisation", "Fault Tolerance", "Model Versioning" ], "wikilinks": [ "Checkpoint Averaging", "Checkpoint Frequency Policy", "Checkpoint Index", "DeepSpeed", "Distributed Training", "Distributed Training Recovery", "DVC", "EMA Shadow Weights", "Experiment Reproducibility", "Experiment Tracking", "File System", "FSDP", "GGUF", "Hugging Face", "Hugging Face Hub", "Imperial College London", "Linux Foundation", "MachineLearningDomain", "MLflow", "MLOps" ] }, { "id": "chemical-propulsion", "title": "Chemical Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:chemical-propulsion", "labels": [ "Chemical Propulsion" ], "is_subclass_of": [ "Spacecraft Propulsion" ], "wikilinks": [] }, { "id": "china-ai-regulation", "title": "China AI Regulation", "domain": "governance", "domain_name": "Governance", "definition": "The set of laws, guidelines, and administrative measures implemented by the Chinese government to govern the development, deployment, and use of artificial intelligence technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:china-ai-regulation", "labels": [ "China AI Regulation" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "china-tech-export-controls", "title": "China Tech Export Controls", "domain": "governance", "domain_name": "Governance", "definition": "Governmental policies and regulatory actions in China that restrict the transfer of technology, talent, or intellectual property to foreign entities or jurisdictions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:china-tech-export-controls", "labels": [ "China Tech Export Controls" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "china", "title": "China", "domain": "governance", "domain_name": "Governance", "definition": "A country in East Asia and the world's most populous nation for much of recent history, with a large economy and significant influence in technology, manufacturing, and digital currency policy.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:china", "labels": [ "China" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Singapore", "owl:Thing" ] }, { "id": "chinese-ai-ecosystem", "title": "Chinese AI Ecosystem", "domain": "economics", "domain_name": "Economics", "definition": "The interconnected network of Chinese technology companies, government policies, hardware suppliers, and research institutions that collectively drive the development and deployment of artificial intelligence within China.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:chinese-ai-ecosystem", "labels": [ "Chinese AI Ecosystem" ], "is_subclass_of": [ "AI Ecosystem" ], "wikilinks": [] }, { "id": "chinese-seal-art-ai-classification-pipeline", "title": "Chinese Seal Art AI Classification Pipeline", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Seals, in this knowledge graph context, refers to a practical AI pipeline experiment using traditional Chinese seal art (\u6d77\u8c79-style stamps) as a five-shot image classification and translation task. The experiment demonstrates multimodal AI capabilities: image analysis via OpenAI vision APIs, English description generation, Chinese translation, and appropriateness ranking, subsequently serialised to JSON for potential NFT metadata enrichment.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:chinese-seal-art-ai-classification-pipeline", "labels": [ "Chinese Seal Art AI Classification Pipeline", "Seals" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "chivo-wallet", "title": "Chivo Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Chivo Wallet is the state-sponsored digital wallet application launched by the government of El Salvador in 2021 to support the country's adoption of bitcoin as legal tender. It allows citizens to hold and transact in both bitcoin and US dollars, including a signup bonus paid in bitcoin, with zero transaction fees within the network. It has been central to El Salvador's Bitcoin Law rollout and is frequently cited in analyses of state-level cryptocurrency adoption.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:chivo-wallet", "labels": [ "Chivo Wallet" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [] }, { "id": "chromosphere", "title": "Chromosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:chromosphere", "labels": [ "Chromosphere" ], "is_subclass_of": [ "Stellar Atmosphere" ], "wikilinks": [] }, { "id": "chunking", "title": "Chunking", "domain": "ai", "domain_name": "Ai", "definition": "Chunking is the process of segmenting source documents into smaller, semantically coherent passages before they are embedded and indexed for retrieval-augmented generation. Strategies range from fixed-size and overlapping windows to recursive, sentence-aware, and semantic chunking that respects document structure. It matters because chunk size and boundaries directly govern retrieval precision and the relevance of context supplied to a language model, making chunking a primary lever for RAG quality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:chunking", "labels": [ "Chunking", "Chunking Strategy", "Document Chunking" ], "is_subclass_of": [ "Information Retrieval", "Text Segmentation" ], "wikilinks": [ "RAG Pipeline", "Retrieval-Augmented Generation", "Embedding Model", "Information Retrieval", "Vector Database", "Large Language Models", "Natural Language Processing", "Semantic Search" ] }, { "id": "cinematic-rendering", "title": "Cinematic Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cinematic Rendering is a rendering approach that prioritises photorealistic, film-quality visual output over real-time frame rates, applying techniques such as ray tracing, physically based materials, depth of field, and colour grading. It is used for pre-rendered animation, marketing visualisation, and offline previsualisation rather than interactive applications. It builds on general rendering techniques but trades real-time performance for image fidelity.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:cinematic-rendering", "labels": [ "Cinematic Rendering" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "circle", "title": "Circle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Circle is a financial technology company that issues a US dollar stablecoin and provides payment and digital asset infrastructure.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:circle", "labels": [ "Circle", "Circle Financial" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [ "Stablecoin", "Cryptocurrency", "Financial Services" ] }, { "id": "circuit-breaker", "title": "Circuit Breaker", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The circuit breaker is a fault-tolerance design pattern that monitors calls to a remote service or resource and, once failures exceed a threshold, trips open to fail fast and stop sending requests for a cooling-off period. After a timeout it allows a limited number of trial calls in a half-open state to test recovery before closing again. The pattern prevents cascading failures, protects struggling dependencies, and enables graceful degradation in distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:circuit-breaker", "labels": [ "Circuit Breaker" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "circular-economy", "title": "Circular Economy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systemic economic model that eliminates waste and keeps materials, components, and products in use at their highest value for as long as possible through restorative and regenerative design, distinguished from the linear \"take-make-dispose\" paradigm by organising economic activity around three fo...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:circular-economy", "labels": [ "Circular Economy", "CircularEconomy" ], "is_subclass_of": [ "Legal and Regulatory", "Sustainability", "Economic Model", "Resource Management", "Industrial Ecology", "Systems Thinking", "Environmental Policy" ], "wikilinks": [ "BMW", "Carbon Credits", "Circularise", "Climate Change Mitigation", "Cradle-to-Cradle Design", "Deposit Return Scheme", "Digital Product Passport", "Digital Twins", "Ecodesign for Sustainable Products Regulation", "Economic Model", "Ellen MacArthur Foundation", "Environmental Policy", "EPCIS", "ESG Investing", "EU Battery Regulation", "EU Green Deal", "Extended Producer Responsibility", "Fairphone", "Fashion for Good", "Fast Fashion" ] }, { "id": "circular-orbit", "title": "Circular Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:circular-orbit", "labels": [ "Circular Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "circulating-supply", "title": "Circulating Supply", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Circulating Supply is the quantity of a blockchain token that is publicly available and actively tradeable in the market at a given point in time\u2014excluding tokens locked in smart contracts, held in treasury reserves, vested to team members, or burned. It serves as the operative supply figure for computing market capitalisation and price-to-earnings metrics, and changes continuously as new tokens are emitted via block rewards and locked tokens are released according to vesting schedules.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:circulating-supply", "labels": [ "Circulating Supply" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "cis-mars-space-environment", "title": "Cis-Mars Space Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cis-mars-space-environment", "labels": [ "Cis-Mars Space Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "cis-mars-space-radiation-environment", "title": "Cis-Mars Space Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cis-mars-space-radiation-environment", "labels": [ "Cis-Mars Space Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "cislunar-space-environment", "title": "Cislunar Space Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cislunar-space-environment", "labels": [ "Cislunar Space Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "cislunar-space-radiation-environment", "title": "Cislunar Space Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cislunar-space-radiation-environment", "labels": [ "Cislunar Space Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "citi-token-services", "title": "Citi Token Services", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A service from Citi that issues tokenised representations of institutional client deposits on a permissioned blockchain to support faster cash management and trade transactions. It operates within the bank's regulated infrastructure rather than on a public network.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:citi-token-services", "labels": [ "Citi Token Services" ], "is_subclass_of": [ "Trade Finance" ], "wikilinks": [ "Permissioned Blockchain", "Tokenization", "Payment Channel", "Distributed Ledger Technology", "Smart Contract", "Trade Finance", "https://www.citigroup.com/global/news" ] }, { "id": "civic-participation", "title": "Civic Participation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The engagement of citizens in democratic processes and public decision-making through digital platforms, virtual environments, and emerging technologies including metaverse spaces, blockchain voting systems, and AI-enhanced deliberation tools that enable new forms of collective governance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:civic-participation", "labels": [ "Civic Participation", "Civic Engagement" ], "is_subclass_of": [ "Governance and Safety", "Democratic Engagement" ], "wikilinks": [ "Collective Governance", "Digital Democracy", "Public Deliberation", "Accessibility", "Democratic Engagement", "Digital Platform", "metaverse", "Telecollaboration", "Trust Infrastructure" ] }, { "id": "civitai", "title": "Civitai", "domain": "ai", "domain_name": "Ai", "definition": "Civitai is a community-driven online platform for sharing, discovering, and downloading fine-tuned generative AI image models, primarily Stable Diffusion checkpoints, LoRA adaptors, embeddings, and VAEs, enabling practitioners and artists to distribute specialised model weights trained on specific styles, characters, or concepts without requiring the infrastructure overhead of operating their own model registry. It functions as both a social network for AI artists and a technical marketplace for model artefacts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:civitai", "labels": [ "Civitai", "CivitAI", "Civitai LoRA Marketplace", "Civitai Model Standards" ], "is_subclass_of": [ "Generative AI", "Model Registry" ], "wikilinks": [] }, { "id": "clarity", "title": "Clarity", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Clarity is a decidable smart contract language designed so that contract behaviour can be analysed before execution, with no compilation step and explicit handling of conditions that could otherwise fail silently.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:clarity", "labels": [ "Clarity" ], "is_subclass_of": [ "Smart Contracts" ], "wikilinks": [ "Smart Contracts", "Verifiable Computation", "Blockchain", "DeFi" ] }, { "id": "class-imbalance", "title": "Class Imbalance", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Class imbalance is the condition in a classification dataset where the number of examples in one class greatly exceeds that of another, causing learning algorithms to favour the majority class. It is common in problems such as fraud detection, medical diagnosis, and anomaly detection, where the events of interest are rare. Class imbalance undermines naive accuracy as an evaluation metric and motivates remedies such as resampling, cost-sensitive learning, and the use of precision-recall-oriented metrics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:class-imbalance", "labels": [ "Class Imbalance" ], "is_subclass_of": [ "Classification", "AI Technique", "Supervised Learning", "Machine Learning" ], "wikilinks": [] }, { "id": "classical-planning", "title": "Classical Planning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Classical planning is a branch of automated planning that computes a sequence of deterministic actions transforming a fully observable initial state into a state satisfying a goal condition. It assumes a single agent, discrete states, instantaneous actions with deterministic effects, and complete knowledge of the world. Problems are typically expressed in formalisms such as STRIPS or PDDL and solved by heuristic state-space search.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:classical-planning", "labels": [ "Classical Planning" ], "is_subclass_of": [ "Automated Planning", "Planning and Scheduling" ], "wikilinks": [ "Automated Planning", "State Space Search", "Heuristic Search", "Knowledge Representation", "STRIPS", "PDDL", "Constraint Satisfaction", "Graph Search", "Reinforcement Learning", "Markov Decision Process", "Robotics", "Motion Planning", "Game Playing", "Agent", "Optimisation", "Pathfinding", "Artificial Intelligence", "A Star Algorithm", "Hierarchical Task Network", "Temporal Planning" ] }, { "id": "classification-evaluation", "title": "Classification Evaluation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Classification evaluation is the set of methods used to quantify how well a classifier's predicted labels match ground-truth labels, typically summarised through metrics derived from a confusion matrix such as precision, recall, and F1 score. It distinguishes between overall accuracy, which can be misleading under class imbalance, and per-class metrics that expose asymmetric error costs. Classification evaluation guides model selection, threshold tuning, and reporting of deployed classifier performance.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:classification-evaluation", "labels": [ "Classification Evaluation" ], "is_subclass_of": [ "Evaluation Metric" ], "wikilinks": [] }, { "id": "classification-rules", "title": "Classification Rules", "domain": "ai", "domain_name": "Ai", "definition": "Classification Rules are explicit logical statements, typically in an if-then form, that assign instances or entities to predefined categories based on the values of their attributes or the satisfaction of specified conditions. Derived from rule-learning algorithms, expert elicitation, or ontology reasoning, classification rules provide interpretable, auditable decision logic for categorising data points in machine learning, knowledge engineering, and regulatory compliance contexts. They contrast with black-box classifiers by exposing their decision rationale directly as symbolic propositions.", "entityType": "Class", "qualityScore": 0.92, "maturity": "mature", "iri": "urn:ngm:class:classification-rules", "labels": [ "Classification Rules" ], "is_subclass_of": [ "Inference Engine", "Knowledge Representation", "Formal Logic" ], "wikilinks": [ "Expert Systems", "Machine Learning Discipline", "Ontology", "Knowledge Graph", "Logic Programming", "OWL", "RDF", "Knowledge Base", "Inference Engine", "Decision Tree", "Supervised Learning", "Large Language Models", "Explainability", "Interpretability", "Neurosymbolic AI", "Production Rules", "Forward Chaining", "Backward Chaining", "Description Logic", "Knowledge Representation" ] }, { "id": "classification-threshold", "title": "Classification Threshold", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Classification Threshold is a decision boundary applied to the probabilistic output of a classifier to assign discrete class labels. By default set at 0.5 for binary classification, the threshold can be adjusted to trade off precision against recall, or sensitivity against specificity, depending on application requirements. Threshold selection is a critical model calibration step that directly affects downstream decision quality and fairness properties.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:classification-threshold", "labels": [ "Classification Threshold" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique", "Model Evaluation", "Decision System", "Statistical Decision Theory" ], "wikilinks": [] }, { "id": "classification", "title": "Classification", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Classification is a supervised machine learning task in which a model learns a mapping from input features to a discrete set of predefined category labels, using labelled training examples to optimise decision boundaries or probabilistic scoring rules. At inference time the model assigns each unseen input to one or more categories by applying a learned discriminant function or probabilistic scoring rule. The task encompasses binary, multi-class, and multi-label variants, and underpins applications ranging from image recognition and natural language understanding to medical diagnosis and fraud detection. Performance is evaluated with metrics such as accuracy, precision, recall, F1-score, and the area under the receiver-operating-characteristic curve, selected according to class imbalance and the relative cost of false positives versus false negatives.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:classification", "labels": [ "Classification", "Binary Classification", "Classification Model", "Classification Scheme", "Incident Classification", "Land Cover Classification", "Ontology Classification", "Sequence Classification", "Statistical Classification" ], "is_subclass_of": [ "Supervised Learning", "AI Technique", "Machine Learning", "Pattern Recognition" ], "wikilinks": [ "Supervised Learning", "Object Detection", "Statistical Learning Theory", "Machine Learning", "Pattern Recognition", "Neural Network", "Deep Learning", "Loss Function", "Transfer Learning", "Regression", "Clustering", "Anomaly Detection", "Support Vector Machine", "Decision Tree", "Random Forest", "Gradient Boosted Trees", "Logistic Regression", "Naive Bayes", "K-Nearest Neighbours", "Transformer Architecture" ] }, { "id": "classifier-evaluation", "title": "Classifier Evaluation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Classifier evaluation is the set of methods and metrics used to assess how well a trained classification model performs, covering measures such as accuracy, precision, recall, specificity and area under the ROC curve, computed against held-out labelled data. Different metrics emphasise different error costs, so evaluation practice typically reports several complementary measures rather than a single figure, particularly under class imbalance. It is a specific application of the broader discipline of model evaluation to classification tasks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:classifier-evaluation", "labels": [ "Classifier Evaluation" ], "is_subclass_of": [ "Model Evaluation" ], "wikilinks": [] }, { "id": "classifier-free-guidance", "title": "classifier-free guidance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Classifier-Free Guidance (CFG) is a conditional generation technique for diffusion models that steers the denoising trajectory towards a specified condition by computing a weighted extrapolation between a conditional score estimate and an unconditional score estimate produced by the same single model. Unlike classifier guidance, which requires a separately trained differentiable classifier, CFG trains one network jointly on conditional and unconditional objectives by randomly replacing conditioning inputs with a null embedding during training. At inference, the guided score is: score_guided = score_unconditional + w * (score_conditional - score_unconditional), where w is the guidance scale hyperparameter controlling the trade-off between sample diversity and condition alignment. CFG has become the dominant conditioning mechanism across text-to-image, text-to-video, and audio generation systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:classifier-free-guidance", "labels": [ "Classifier-Free Guidance", "Classifier Free Guidance" ], "is_subclass_of": [ "AI Technique", "Generative AI", "Score-Based Generative Model", "Conditional Generation" ], "wikilinks": [ "Diffusion Model", "Latent Diffusion", "Score Function", "Variational Autoencoder", "Text-to-Image", "Image Generation", "Text-to-Video Generation", "Audio Generation", "Denoising Diffusion Probabilistic Model", "Conditioning Signal", "Score-Based Generative Model", "Dropout", "Generative Model", "Stable Diffusion", "ControlNet", "Prompt Engineering", "Sampling", "CLIP", "Generative AI", "Reinforcement Learning from Human Feedback" ] }, { "id": "classifier", "title": "Classifier", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A model or algorithm that assigns input instances to one of a finite set of discrete categories, typically by learning a decision function from labelled training data. Classifiers range from linear models (logistic regression, linear SVMs) through tree ensembles to deep neural networks, and most output class probabilities or scores that a decision threshold converts into hard labels. Their performance is characterised through confusion-matrix quantities such as true and false positives, precision, recall, and ROC analysis.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:classifier", "labels": [ "Classifier" ], "is_subclass_of": [ "AI Model" ], "wikilinks": [ "AI Model", "Classification", "Supervised Learning", "Classification Threshold", "Deep Generative Model" ] }, { "id": "claude-co-work", "title": "Claude Co-work", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A feature of Anthropic's Claude AI that enables collaborative work environments with file access, multi-step task execution, and plugin integration across platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:claude-co-work", "labels": [ "Claude Co-work" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "claude-code", "title": "Claude Code", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A command-line interface and agentic coding tool developed by Anthropic that enables developers to execute complex software engineering tasks using the Claude model family.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:claude-code", "labels": [ "Claude Code" ], "is_subclass_of": [ "Claude" ], "wikilinks": [] }, { "id": "claude-design", "title": "Claude Design", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A design tool within the Claude ecosystem that generates visual elements using code and SVGs, featuring customizable parameters and positioned for systems design.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:claude-design", "labels": [ "Claude Design" ], "is_subclass_of": [ "Claude" ], "wikilinks": [] }, { "id": "claude", "title": "Claude", "domain": "ai", "domain_name": "Ai", "definition": "Claude is a family of large language model AI assistants developed by Anthropic, trained using Constitutional AI and reinforcement learning from human feedback to be helpful, harmless, and honest. The Claude model family encompasses tiered variants (Haiku, Sonnet, and Opus) spanning cost-performance trade-offs, with support for extended context windows, multimodal inputs, tool use, and agentic workflows via the Model Context Protocol. Claude embodies Anthropic's research programme on aligning advanced AI systems with human values, serving as both a commercial API product and a living demonstration that safety and capability are complementary rather than opposed. The model family has evolved through multiple generations, with each iteration advancing state-of-the-art performance on reasoning, coding, and instruction-following benchmarks whilst maintaining rigorous safety evaluation.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:claude", "labels": [ "Claude", "Claude Artifacts", "Claude Desktop", "Claude Model Family" ], "is_subclass_of": [ "Large Language Model", "Proprietary Large Language Models", "Large Language Models", "Foundation Model", "Agentic AI" ], "wikilinks": [ "Anthropic", "Large Language Models", "Constitutional AI Training Methodology", "Reinforcement Learning from Human Feedback", "AI Safety", "Foundation Model", "Model Context Protocol", "Agentic Workflow", "Agentic AI", "Pre Training", "Instruction Tuning", "Direct Preference Optimisation", "RLHF", "Red Teaming", "Mechanistic Interpretability", "Scalable Oversight", "Responsible Scaling Policy", "Transformer Architecture", "Tool Use", "Conversational AI" ] }, { "id": "clean-spark", "title": "CleanSpark", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A publicly traded company that operates Bitcoin mining facilities in the United States, focusing on the use of low-cost and lower-carbon energy sources.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:clean-spark", "labels": [ "CleanSpark" ], "is_subclass_of": [ "Bitcoin Mining" ], "wikilinks": [ "Bitcoin Mining", "Bitcoin", "Mining" ] }, { "id": "clearing-and-settlement", "title": "Clearing And Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Clearing and settlement are the post-trade processes that finalise a financial transaction: clearing reconciles, nets and confirms the obligations between counterparties, while settlement effects the actual transfer of securities and funds to discharge those obligations. Central counterparties and securities depositories reduce counterparty risk by interposing themselves and managing margin. The integrity of these processes underpins the stability of payment and securities markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:clearing-and-settlement", "labels": [ "Clearing And Settlement", "Clearing and Settlement" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [] }, { "id": "clearing-house", "title": "Clearing House", "domain": "finance", "domain_name": "Finance", "definition": "A clearing house is a financial market intermediary that stands between the buyers and sellers of a trade to guarantee its completion, most often acting as a central counterparty that legally interposes itself through novation. By collecting margin, mutualising risk across a default fund, and netting offsetting obligations, it reduces counterparty credit risk and systemic contagion in securities, derivatives, and payment markets. Clearing houses are critical financial market infrastructure, subject to stringent prudential supervision because their failure could propagate across the financial system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:clearing-house", "labels": [ "Clearing House" ], "is_subclass_of": [ "Financial Market Infrastructure" ], "wikilinks": [] }, { "id": "clearing", "title": "Clearing", "domain": "finance", "domain_name": "Finance", "definition": "Clearing is the post-trade process of confirming, matching and managing the obligations arising from a transaction in the interval between execution and settlement. A clearing house, often acting as a central counterparty, novates trades and nets exposures, calculating the net obligations of each party and managing counterparty credit risk through margin and default funds. Clearing reduces systemic risk and operational burden in securities, derivatives and payment markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:clearing", "labels": [ "Clearing" ], "is_subclass_of": [ "Financial Market Infrastructure" ], "wikilinks": [] }, { "id": "client-server-architecture", "title": "Client-Server Architecture", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Client-server architecture is a distributed computing model in which client processes request services or resources and dedicated server processes provide them, typically over a network. The model centralises shared resources, data and logic on servers while distributing presentation and interaction to many clients. It is the foundational pattern for the web, networked applications and most online services, and it contrasts with peer-to-peer architectures where every node is both provider and consumer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:client-server-architecture", "labels": [ "Client-Server Architecture" ], "is_subclass_of": [ "Distributed Architecture" ], "wikilinks": [] }, { "id": "client-side-validation-theory", "title": "Client-Side Validation Theory", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Client-side validation theory is the conceptual framework in which the validity of state transitions is verified by the affected parties themselves rather than by every node of a global consensus layer, with the blockchain used only to commit to and order single-use seals. Data and proofs are kept off-chain and shared peer-to-peer, so the base chain provides ordering and double-spend prevention without learning transaction contents. It matters because it underpins scalable, private smart-contract systems such as RGB that inherit Bitcoin's security without bloating its chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:client-side-validation-theory", "labels": [ "Client-Side Validation Theory" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "client-side-validation", "title": "Client-Side Validation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Client-Side Validation is the process of verifying user-supplied data within the browser or client application before that data is transmitted to a server, providing immediate feedback to users and reducing unnecessary network requests. In the context of Bitcoin and RGB Protocol, it refers to a distinct validation paradigm where the full state of off-chain assets is verified locally by the recipient rather than by all network nodes, enabling scalable, private asset transfers. The two usages share the principle of local verification but differ fundamentally in their security models and scope.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:client-side-validation", "labels": [ "Client-Side Validation", "client-side-validation" ], "is_subclass_of": [ "Data Integrity" ], "wikilinks": [] }, { "id": "climate-action-dao", "title": "Climate Action DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralized autonomous organization (DAO) governed by blockchain-based smart contracts and token-weighted voting that coordinates collective action toward climate change mitigation, adaptation, and environmental sustainability through transparent, democratic mechanisms for funding climate pro...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:climate-action-dao", "labels": [ "Climate Action DAO" ], "is_subclass_of": [ "Governance and Regulation", "DecentralizedAutonomousOrganization", "ClimateFinance", "CollectiveGovernance", "ImpactInvestment" ], "wikilinks": [ "CarbonCreditRetirement", "Celo Climate Collective", "ClimateFinance", "ClimateProjectFunding", "CollectiveGovernance", "CommunityCoordination", "CommunityParticipation", "DAOGovernance", "EnvironmentalDomain", "Gitcoin", "ImpactInvestment", "ImpactMetrics", "ImpactVerification", "KlimaDAO", "ReFi DAO", "Regen Network", "RegenerativeFinance", "VotingMechanism", "BlockchainDomain", "DecentralizedAutonomousOrganization" ] }, { "id": "climate-action", "title": "Climate Action", "domain": "governance", "domain_name": "Governance", "definition": "Climate action refers to coordinated efforts to mitigate climate change by reducing greenhouse-gas emissions and to adapt human and natural systems to its impacts. It spans policy, finance, technology and behaviour, from decarbonising energy and transport to protecting ecosystems and building resilience against extreme weather. Frameworks such as the Paris Agreement set the shared goal of limiting global warming, and pathways toward net zero translate that goal into concrete reductions across economies and supply chains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:climate-action", "labels": [ "Climate Action" ], "is_subclass_of": [ "Environmental Sustainability" ], "wikilinks": [] }, { "id": "climate-change-mitigation", "title": "Climate Change Mitigation", "domain": "governance", "domain_name": "Governance", "definition": "Efforts to reduce or prevent the emission of greenhouse gases and to enhance their removal from the atmosphere, with the aim of limiting the extent of climate change. It includes shifting to low-carbon energy and improving efficiency.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:climate-change-mitigation", "labels": [ "Climate Change Mitigation" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Renewable Energy", "Sustainability" ] }, { "id": "climate-change", "title": "Climate Change", "domain": "governance", "domain_name": "Governance", "definition": "Climate change refers to long-term shifts in temperatures and weather patterns, driven predominantly since the industrial era by human emissions of greenhouse gases that trap heat in the atmosphere. Its impacts include rising sea levels, more frequent extreme weather, ecosystem disruption and threats to food, water and economic systems. Responses span mitigation (reducing emissions) and adaptation (managing unavoidable impacts), coordinated through policy frameworks such as the Paris Agreement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:climate-change", "labels": [ "Climate Change" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "climate-commitments", "title": "Climate Commitments", "domain": "governance", "domain_name": "Governance", "definition": "Climate commitments are formal pledges by governments, companies, or coalitions to achieve specified greenhouse-gas reduction or removal outcomes by defined dates, such as net-zero or science-based emissions targets. They typically combine a headline goal with interim milestones, accounting boundaries, and reporting obligations to make progress verifiable. They matter as the governance mechanism that translates climate ambition into measurable, accountable action and that shapes corporate sustainability and offsetting strategies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:climate-commitments", "labels": [ "Climate Commitments" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "climate-data-record", "title": "Climate Data Record", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:climate-data-record", "labels": [ "Climate Data Record", "CDR" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "climate-finance", "title": "Climate Finance", "domain": "finance", "domain_name": "Finance", "definition": "Climate finance encompasses the local, national, and transnational flow of funds \u2014 drawn from public, private, and blended sources \u2014 directed at activities that mitigate greenhouse gas emissions or support adaptation to the adverse effects of climate change. It operates through a range of instruments including grants, concessional loans, guarantees, equity investment, and market-based mechanisms such as carbon credits. Internationally, climate finance is governed by commitments under the UNFCCC and the Paris Agreement, which established obligations for developed nations to mobilise and transfer funds to developing countries. The field intersects investment policy, sovereign risk, environmental regulation, and multilateral development banking.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:climate-finance", "labels": [ "Climate Finance", "Climate Finance Innovation", "ClimateProjectFunding" ], "is_subclass_of": [ "Green Finance" ], "wikilinks": [ "Carbon Markets", "Sustainability", "Green Finance", "https://unfccc.int/topics/introduction-to-climate-finance", "https://www.worldbank.org/en/topic/climatefinance" ] }, { "id": "climate-governance", "title": "Climate Governance", "domain": "governance", "domain_name": "Governance", "definition": "Climate governance is the system of institutions, policies, agreements and accountability mechanisms through which societies steer collective action on climate change. It spans international treaties, national legislation, sub-national and corporate commitments, and the monitoring, reporting and verification arrangements that hold actors to their pledges. By coordinating mitigation and adaptation across scales, climate governance translates scientific assessment into binding and voluntary obligations. It draws on instruments such as carbon pricing, emissions targets and disclosure regimes to align economic activity with environmental limits.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:climate-governance", "labels": [ "Climate Governance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "climate-modelling", "title": "Climate Modelling", "domain": "data", "domain_name": "Data", "definition": "Climate Modelling is the scientific discipline of constructing mathematical and computational representations of Earth's climate system to simulate past, present, and future climate states. Models range from simple energy-balance equations to high-resolution coupled atmosphere-ocean-land-ice systems running on supercomputing infrastructure. They underpin global projections published by the Intergovernmental Panel on Climate Change and inform policy decisions on emissions reduction and adaptation. Machine learning is increasingly integrated to accelerate parameterisation and downscaling. Outputs feed directly into carbon accounting, sustainability planning, and digital twin applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:climate-modelling", "labels": [ "Climate Modelling", "Climate Scenario Modelling" ], "is_subclass_of": [ "Digital Twin" ], "wikilinks": [] }, { "id": "climate-policy", "title": "Climate Policy", "domain": "governance", "domain_name": "Governance", "definition": "Climate policy is the set of government measures, regulations, and commitments intended to reduce greenhouse gas emissions and address the effects of climate change.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:climate-policy", "labels": [ "Climate Policy", "International Climate Policy" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Greenwashing", "Energy Consumption", "Sustainability" ] }, { "id": "climate-risk-assessment", "title": "Climate Risk Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Climate risk assessment is the structured process of identifying, quantifying, and prioritising the physical and transition risks that climate change poses to an asset, portfolio, or institution. Physical risk assessment models exposure to acute events such as floods and storms alongside chronic shifts such as sea-level rise and heat stress, while transition risk assessment evaluates exposure to policy, technology, and market shifts away from carbon-intensive activity. It draws on climate modelling outputs to translate physical hazard projections into financial or operational impact estimates, and is increasingly required under disclosure frameworks such as TCFD. Results feed into climate finance decisions, capital allocation, and regulatory reporting.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:climate-risk-assessment", "labels": [ "Climate Risk Assessment" ], "is_subclass_of": [ "Climate Risk" ], "wikilinks": [] }, { "id": "climate-risk", "title": "Climate Risk", "domain": "governance", "domain_name": "Governance", "definition": "Climate risk is the set of financial, operational and strategic threats that arise from climate change and the transition to a low-carbon economy. It is conventionally divided into physical risk, stemming from acute and chronic climate impacts, and transition risk, arising from policy, technology, market and reputational shifts. Organisations assess and disclose climate risk to inform capital allocation, resilience planning and regulatory compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:climate-risk", "labels": [ "Climate Risk" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "climate-scenario-analysis", "title": "Climate Scenario Analysis", "domain": "governance", "domain_name": "Governance", "definition": "Climate scenario analysis is a structured method for exploring how an organisation's assets, operations, or investments might perform under different plausible future climate and policy pathways, such as varying degrees of global warming or rates of decarbonisation. It typically models physical risks, including extreme weather and chronic climate shifts, and transition risks, including policy, technology, and market changes, against defined time horizons. Climate scenario analysis informs climate risk disclosure and materiality assessment frameworks used by regulators and investors to evaluate long-term financial resilience.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:climate-scenario-analysis", "labels": [ "Climate Scenario Analysis" ], "is_subclass_of": [ "Climate Risk" ], "wikilinks": [] }, { "id": "climate-tech", "title": "Climate Tech", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Climate Tech encompasses the portfolio of technologies, platforms, and business models designed to mitigate greenhouse gas emissions, support climate adaptation, and accelerate the energy transition. It spans renewable energy generation, energy storage, carbon capture and removal, sustainable agriculture, climate risk analytics, and digital tools for emissions measurement and reporting. Climate Tech investment has grown into a distinct venture capital and corporate innovation category, underpinned by regulatory drivers, net-zero commitments, and the rapid cost-reduction curves of clean energy technologies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:climate-tech", "labels": [ "Climate Tech", "Climate Tech Deployment" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "clinical-decision-support", "title": "Clinical Decision Support", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Clinical Decision Support (CDS) refers to AI systems that provide healthcare professionals with patient-specific assessments, recommendations, and information to support clinical decision-making at the point of care. CDS systems integrate patient data, medical knowledge bases, clinical guidelines, and evidence-based protocols to assist in diagnosis, treatment selection, medication management, and care coordination whilst maintaining clinician autonomy and clinical judgement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:clinical-decision-support", "labels": [ "Clinical Decision Support", "Medical Decision Support", "Traditional Clinical Decision Support" ], "is_subclass_of": [ "Medical AI" ], "wikilinks": [ "Healthcare Analytics", "Medical AI", "Medical Diagnosis AI", "MetaverseDomain", "Telecollaboration", "Treatment Planning AI" ] }, { "id": "clinical-trials", "title": "Clinical Trials", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based clinical trial management systems employ immutable audit trails, smart contracts for protocol compliance, and cryptographic verification to ensure data integrity, prevent fraud, and provide FDA 21 CFR Part 11 compliant tamper-proof records whilst reducing monitoring costs. These systems address pharmaceutical development challenges including data falsification, patient recruitment inefficiency, and multi-site coordination failures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:clinical-trials", "labels": [ "Clinical Trials", "Clinical Trial", "Clinical Trial Infrastructure" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "AML (Anti-Money Laundering)", "BC-0426-hyperledger-fabric", "BC-0442-pharmaceutical-traceability", "BC-0456-self-sovereign-identity", "BC-0458-verifiable-credentials", "BC-0476-aml-kyc-compliance", "BC-0486-regulatory-reporting", "BC-0491-healthcare-records", "GDPR (General Data Protection Regulation)", "ISO (International Organization for Standardization)", "KYC (Know Your Customer)", "BlockchainDomain" ] }, { "id": "clock-recovery", "title": "Clock Recovery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Clock recovery (clock and data recovery, CDR) is the physical-layer process by which a receiver extracts a synchronised timing reference from an incoming data stream that carries no separate clock signal. It uses phase-locked loops or oversampling to align sampling instants with the centre of each symbol, compensating for jitter and frequency offset between transmitter and receiver. It matters because correct sampling timing is essential for reliable symbol detection in serial communication links.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:clock-recovery", "labels": [ "Clock Recovery" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "clock-synchronization", "title": "Clock Synchronization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Clock Synchronization is the process of coordinating the time references of distributed computing nodes or electronic systems so that they share a consistent and accurate notion of time, enabling correct ordering of events, coordinated actions, and time-stamped record-keeping. Protocols such as NTP (Network Time Protocol) and PTP (Precision Time Protocol, IEEE 1588) achieve synchronization by exchanging timestamped messages and compensating for network propagation delays. Accurate clock synchronization is critical for distributed databases, consensus algorithms, telecommunications, industrial control systems, and financial transaction ordering.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:clock-synchronization", "labels": [ "Clock Synchronization", "Clock Synchronisation", "Synchronised Clock" ], "is_subclass_of": [ "Network Synchronization" ], "wikilinks": [] }, { "id": "closed-world-assumption", "title": "Closed World Assumption", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Closed World Assumption (CWA) is the presumption that any statement not known to be true is false, treating the knowledge base as a complete description of the world. It is foundational to database query semantics, logic programming, and negation as failure, where the absence of a fact licenses inferring its negation. CWA simplifies reasoning over finite, curated domains but breaks down where information is incomplete.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:closed-world-assumption", "labels": [ "Closed World Assumption" ], "is_subclass_of": [ "Knowledge Representation", "Formal Logic", "Non-Monotonic Reasoning" ], "wikilinks": [ "Knowledge Representation", "Open World Assumption", "Negation as Failure", "Logic Programming", "Prolog", "SQL", "Relational Database", "Datalog", "SHACL", "Ontology", "Semantic Web", "OWL Class Hierarchy", "RDF", "SPARQL", "Knowledge Graph", "Inference Engine", "Formal Logic", "Description Logic", "First-Order Logic", "Reasoning" ] }, { "id": "closed-loop-control", "title": "Closed-Loop Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A control system that uses feedback from sensors to compare the actual output with the desired output and adjusts the control action to minimize error. The system continuously monitors and corrects its behavior.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:closed-loop-control", "labels": [ "Closed-Loop Control", "Closed Loop Control", "Closed Loop System", "Closed-Loop Operation", "RB-1002-closed-loop-control" ], "is_subclass_of": [ "Actuation and Control", "Control System" ], "wikilinks": [ "Actuator", "Adaptability", "Controller", "ISO 8373:2021", "PID Control", "RB-1001-open-loop-control", "RB-1004-adaptive-control", "RB-1007-trajectory-generation", "RB-1015-kalman-filter", "Sensor System", "Stability", "Accuracy", "AI Agent System", "Control System", "Control Theory", "Feedback Mechanism", "RB-1013-localization", "Robotics" ] }, { "id": "closed-loop-cooling", "title": "Closed-Loop Cooling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A thermal management system that recirculates coolant within a sealed circuit to dissipate heat without consuming external water resources.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:closed-loop-cooling", "labels": [ "Closed-Loop Cooling" ], "is_subclass_of": [ "Data Center Sustainability" ], "wikilinks": [] }, { "id": "cloth-simulation", "title": "Cloth Simulation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Cloth Simulation is the computational modelling of textile deformation and dynamics, representing fabric as a mesh of particles connected by spring constraints (stretch, shear, and bend) or by a continuum-mechanics model, and integrating the equations of motion to produce plausible cloth behaviour under gravity, wind, collision, and user interaction. The particle-spring model, popularised by Provot (1995), remains prevalent in real-time applications; more accurate results for offline rendering use finite-element or position-based dynamics (PBD) methods. Collision detection and response against rigid bodies and self-collision are the principal computational bottlenecks, requiring spatial acceleration structures such as bounding volume hierarchies. Cloth simulation is a sub-discipline of physically based animation used in character clothing, flag animation, curtains, and virtual fashion design within game engines, VFX pipelines, and metaverse avatar systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cloth-simulation", "labels": [ "Cloth Simulation" ], "is_subclass_of": [ "Physics Simulation" ], "wikilinks": [] }, { "id": "cloud-computing-revenue", "title": "Cloud Computing Revenue", "domain": "economics", "domain_name": "Economics", "definition": "The financial income generated by cloud service providers from the sale of infrastructure, platform, and software services to customers.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cloud-computing-revenue", "labels": [ "Cloud Computing Revenue" ], "is_subclass_of": [ "Hyperscale Cloud" ], "wikilinks": [] }, { "id": "cloud-computing", "title": "Cloud Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud computing is the on-demand delivery of computing resources \u2014 servers, storage, databases, networking, software, analytics, and AI accelerators \u2014 over the internet via provider-managed data centres, abstracting physical infrastructure into programmable APIs with pay-per-use economics. Service models (IaaS, PaaS, SaaS) and deployment models (public, private, hybrid, multi-cloud) define the boundary of managed responsibility between provider and consumer. Hyperscale providers such as AWS, Microsoft Azure, and Google Cloud Platform underpin modern AI training, inference serving, and distributed application deployment at global scale. The paradigm enables elastic provisioning \u2014 scaling from zero to thousands of compute nodes in seconds \u2014 transforming both software engineering and machine learning operations.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:cloud-computing", "labels": [ "Cloud Computing", "Cloud Computing Infrastructure", "Cloud Computing Platform", "CloudComputing" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "cloud-cover-fraction", "title": "Cloud Cover Fraction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cloud-cover-fraction", "labels": [ "Cloud Cover Fraction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cloud-gaming", "title": "Cloud Gaming", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Cloud gaming is a model in which games are rendered on remote server GPUs and streamed as video to a client device, while player input is sent back to the server over the network. It shifts compute off the end device, allowing demanding titles to run on thin clients, but is highly sensitive to network latency, bandwidth, and jitter. It matters as a delivery paradigm that depends on adaptive bitrate streaming and low-latency encoding to maintain interactivity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cloud-gaming", "labels": [ "Cloud Gaming" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "cloud-infrastructure", "title": "Cloud Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud infrastructure is a computing model that provides on-demand access to virtualized computing resources including servers, storage, networking, and platform services delivered over the internet, enabling organizations to provision and scale IT resources dynamically through IaaS, PaaS, and SaaS service models without managing physical hardware.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cloud-infrastructure", "labels": [ "Cloud Infrastructure", "Centralised Cloud Infrastructure", "CloudInfrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Elasticity", "Scalability", "Self-Service Provisioning", "ETSI_Domain_Infrastructure", "InfrastructureDomain", "Technology Domain" ] }, { "id": "cloud-mask", "title": "Cloud Mask", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cloud-mask", "labels": [ "Cloud Mask" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cloud-native-computing-foundation", "title": "Cloud Native Computing Foundation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Cloud Native Computing Foundation (CNCF) is a vendor-neutral, open-source foundation hosted by the Linux Foundation that stewards critical cloud-native projects such as Kubernetes, Prometheus and Envoy. It defines and promotes the cloud-native computing model based on containers, microservices and declarative orchestration. The CNCF maintains a graduated project lifecycle (sandbox, incubating, graduated) and coordinates the wider ecosystem through events, certification programmes and a technical oversight committee.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cloud-native-computing-foundation", "labels": [ "Cloud Native Computing Foundation" ], "is_subclass_of": [ "Linux Foundation" ], "wikilinks": [] }, { "id": "cloud-native", "title": "Cloud Native", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud Native is an approach to building and running applications that fully exploits the advantages of cloud computing infrastructure \u2014 elastic scaling, managed services, pay-per-use economics, and high availability \u2014 by designing for containerisation, dynamic orchestration, microservices decomposition, and declarative APIs. The Cloud Native Computing Foundation (CNCF) defines it as the use of containers, service meshes, microservices, immutable infrastructure, and declarative APIs to build resilient, manageable, and observable systems. Cloud Native architectures enable rapid, reproducible delivery of software at scale across public, private, and hybrid cloud environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cloud-native", "labels": [ "Cloud Native", "Cloud Native Computing", "Cloud-Native", "Cloud-Native Computing" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "cloud-platform", "title": "Cloud Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cloud platform is an integrated suite of managed infrastructure, runtime services, and developer tooling delivered over the internet that enables organisations to build, deploy, scale, and operate applications without owning physical hardware. Cloud platforms abstract away operational complexity through pay-as-you-go pricing, elastic scaling, and managed service lifecycles, spanning IaaS, PaaS, and SaaS delivery models. The dominant hyperscale providers \u2014 Amazon Web Services, Microsoft Azure, and Google Cloud Platform \u2014 offer hundreds of services covering compute, storage, networking, databases, AI/ML, security, and observability, forming the primary substrate for modern enterprise and AI workloads.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:cloud-platform", "labels": [ "Cloud Platform", "Cloud IoT Platform" ], "is_subclass_of": [ "Cloud Infrastructure" ], "wikilinks": [] }, { "id": "cloud-rendering", "title": "Cloud Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cloud rendering is the practice of generating images or interactive 3D frames on remote, GPU-equipped servers and streaming the results to a client device, rather than rendering locally. It decouples visual fidelity from the client's hardware, enabling thin clients to display high-quality graphics by offloading computation to data centers. It powers cloud gaming, remote visualisation, and immersive metaverse experiences, but depends on low-latency networking and efficient video streaming to remain responsive.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:cloud-rendering", "labels": [ "Cloud Rendering" ], "is_subclass_of": [ "Remote Rendering" ], "wikilinks": [] }, { "id": "cloud-security", "title": "Cloud Security", "domain": "security", "domain_name": "Security", "definition": "Cloud security is the discipline of protecting data, applications, identities and infrastructure hosted in cloud computing environments against unauthorised access, misconfiguration, data loss and service disruption. It applies controls across the shared-responsibility boundary between cloud providers and customers, spanning identity and access management, encryption, network segmentation, configuration governance and continuous monitoring. Cloud security extends established information-security principles to elastic, multi-tenant and API-driven platforms where infrastructure is provisioned programmatically.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cloud-security", "labels": [ "Cloud Security" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "cloud-shadow-mask", "title": "Cloud Shadow Mask", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cloud-shadow-mask", "labels": [ "Cloud Shadow Mask" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cloud-storage-infrastructure", "title": "Cloud Storage Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud storage infrastructure is the distributed system of servers, object and block stores, replication, and access APIs that lets applications persist and retrieve data on demand over a network. It provides durability through redundancy, elastic capacity, and tiered performance classes ranging from hot object storage to cold archives. It matters because it is the foundation for storing large media assets, such as meeting recordings, at scale with high availability and pay-as-you-go economics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cloud-storage-infrastructure", "labels": [ "Cloud Storage Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "cloud-storage", "title": "Cloud Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud storage is a model of data persistence in which digital data is held on remote servers operated by a cloud service provider, accessed over a network (typically the internet) via well-defined APIs rather than on locally attached hardware. Providers maintain geographically distributed, redundant infrastructure implementing erasure coding or multi-region replication to guarantee high durability and availability, typically billing clients on capacity consumed and egress data transfer. The paradigm encompasses object storage, cloud file systems, and block storage volumes, and underpins virtually every modern cloud-native application architecture by decoupling storage from compute. Major implementations include Amazon S3, Google Cloud Storage, Azure Blob Storage, and S3-compatible open-source systems such as MinIO.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:cloud-storage", "labels": [ "Cloud Storage", "Centralised Cloud Storage" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "cloud-native-applications", "title": "Cloud-Native Applications", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud-native applications are software systems specifically designed and architected to exploit the capabilities of cloud computing environments, built from loosely coupled microservices deployed via containers and orchestrated across private, public, or hybrid cloud infrastructure. They prioritise scalability, resilience, observability, and automated lifecycle management over the characteristics of traditional monolithic systems. The paradigm is distinguished not by where an application runs but by how it is constructed, operated, and evolved.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:cloud-native-applications", "labels": [ "Cloud-Native Applications", "Cloud-Native Application", "CloudNativeApplications" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Sensor Input", "MetaverseDomain" ] }, { "id": "cloud-native-architecture", "title": "Cloud-Native Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cloud-native architecture is an approach to designing and operating applications that fully exploit elastic, on-demand cloud infrastructure. It favours loosely coupled, independently deployable services packaged in containers, orchestrated dynamically, and managed through automation, declarative configuration and continuous delivery. The goal is resilient, scalable systems that can be evolved rapidly and recover automatically from failure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cloud-native-architecture", "labels": [ "Cloud-Native Architecture", "Cloud Native Architecture" ], "is_subclass_of": [ "Cloud Infrastructure" ], "wikilinks": [] }, { "id": "clustered-data-state", "title": "Clustered Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:clustered-data-state", "labels": [ "Clustered Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "clustering", "title": "Clustering", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Clustering is an unsupervised machine learning task that partitions a set of objects into groups, or clusters, such that objects within a group are more similar to one another than to those in other groups, according to a chosen distance or similarity metric. Unlike classification it operates without predefined labels, discovering latent structure directly from the data. Algorithms differ in their cluster model, ranging from centroid-based and density-based to hierarchical and probabilistic approaches. Clustering is widely used for exploratory analysis, segmentation, anomaly detection, and as a preprocessing step for downstream learning.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:clustering", "labels": [ "Clustering" ], "is_subclass_of": [ "Unsupervised Learning", "Machine Learning", "AI Technique", "Data Mining", "Statistical Learning" ], "wikilinks": [ "Unsupervised Learning", "Distance Metric", "Feature Engineering", "Embedding", "Embedding Space", "Gaussian Mixture Model", "Anomaly Detection", "Community Detection", "Dimensionality Reduction", "Predictive Analytics", "Hierarchical Clustering", "Graph Analytics", "Classification", "Regression", "DBSCAN", "K-Means", "Spectral Clustering", "Silhouette Score", "Data Mining", "Exploratory Data Analysis" ] }, { "id": "co-regulation", "title": "Co Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Co-regulation is a governance model in which statutory authorities and industry actors jointly develop and enforce rules, blending the legitimacy and backstop powers of the state with the technical expertise and agility of the regulated sector. It sits between command-and-control regulation and pure self-regulation: codes of practice drafted by industry are approved and overseen by a public regulator that retains enforcement powers. The approach is common in online safety, broadcasting and data protection.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:co-regulation", "labels": [ "Co Regulation", "Co-Regulation" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "co-training", "title": "Co Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Co-training is a semi-supervised machine learning technique where two or more models trained on complementary feature views iteratively label unlabelled data for each other. Each model labels examples confidently classified under its own view, and these pseudo-labels are added to the other model's training set, bootstrapping performance without requiring large labelled corpora.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:co-training", "labels": [ "Co Training" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "automation", "Base models", "Ethan Mollick", "IEEE (Institute of Electrical and Electronics Engineers)", "optimization", "research", "SEC (Securities and Exchange Commission)", "siyaev2021towards", "visualization", "Artificial Intelligence", "Artificial Superintelligence", "Autonomous Robot", "ChatGPT", "Copyright", "copyright", "data management", "Deepfakes and fraudulent content", "Education and AI", "Energy and Power", "Gemini" ] }, { "id": "co-presence", "title": "Co-Presence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Co-presence is the sense that other people are sharing the same virtual or physical-digital space with a user at the same time, and can perceive and be perceived in return. It extends the individual sense of presence to a shared, social context, and is a key quality metric for telecollaboration and metaverse platforms, where avatars, spatial audio and non-verbal cues are used to reinforce the impression of being together. Strong co-presence is associated with more natural social interaction in virtual environments.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:co-presence", "labels": [ "Co-Presence", "Co-presence" ], "is_subclass_of": [ "Presence" ], "wikilinks": [] }, { "id": "co-creation", "title": "Co-creation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Co-creation is a collaborative process in which two or more parties, including end users, jointly generate a product, design or piece of content rather than one party producing it for passive consumption by the others. It emphasises shared authorship and iterative feedback across the creative process, and is a common pattern in participatory design, open-source development and collaborative virtual environments. Co-creation is enabled by tools that support simultaneous, shared editing and by design methodologies that structure participant input.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:co-creation", "labels": [ "Co-creation", "Co-Creation" ], "is_subclass_of": [ "Collaboration" ], "wikilinks": [] }, { "id": "co-ap", "title": "CoAP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Constrained Application Protocol (CoAP) is a specialised web transfer protocol defined in RFC 7252 (IETF, 2014) designed for use with constrained nodes and networks in the Internet of Things ecosystem. Modelled on HTTP's request-response semantics and RESTful resource model but optimised for low-power, lossy networks, CoAP uses UDP as its transport layer, features a compact binary header, supports observe (pub/sub) extensions, and includes built-in mechanisms for reliability, multicast, and asynchronous communication. CoAP is a cornerstone protocol for IoT device management and machine-to-machine communication.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:co-ap", "labels": [ "CoAP" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "coastal-monitoring", "title": "Coastal Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coastal-monitoring", "labels": [ "Coastal Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cobot-deployment", "title": "Cobot Deployment", "domain": "robotics", "domain_name": "Robotics", "definition": "Cobot deployment is the process of installing and configuring a collaborative robot into a shared workspace with human operators, encompassing risk assessment, speed and separation monitoring, power and force limiting, and compliance with safety standards governing human-robot proximity. It translates abstract safety standards such as ISO/TS 15066 into a concrete, validated physical installation. Successful deployment requires certifying that the cobot's safety-rated monitored stop and force limits meet the standard's thresholds before it operates alongside people.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:cobot-deployment", "labels": [ "Cobot Deployment" ], "is_subclass_of": [ "Collaborative Robot" ], "wikilinks": [] }, { "id": "cobot-safety-levels", "title": "Cobot Safety Levels", "domain": "robotics", "domain_name": "Robotics", "definition": "Standardized safety classifications and requirements for collaborative robots (cobots) operating in shared workspaces with humans, defining protection measures, risk assessments, and operational modes to ensure safe human-robot interaction according to ISO/TS 15066.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:cobot-safety-levels", "labels": [ "Cobot Safety Levels", "RB-1011-cobot-safety-levels" ], "is_subclass_of": [ "Safety and Standards", "Safety Standard" ], "wikilinks": [ "Collaborative Robotics", "Contact Limits", "Force Limiting", "Human Safety", "ISO 10218-1", "ISO 10218-2", "ISO/TS 15066", "Operational Modes", "RB-1004-adaptive-control", "RB-1012-trust-in-automation", "Safety Certification", "Safety Requirements", "Safety Standard", "Safety Standards", "Speed Monitoring", "Collaborative Robot", "Human-Robot Interaction", "Risk Assessment", "Robotics", "Telecollaboration" ] }, { "id": "code-execution", "title": "Code Execution", "domain": "ai", "domain_name": "Ai", "definition": "Code execution, in the context of AI agents, is the capability whereby a model generates source code and runs it in a sandboxed interpreter or runtime, then incorporates the results into its reasoning. It transforms a language model from a text generator into a tool-using agent that can compute, manipulate data, call APIs, and verify outputs programmatically. It matters because executable tool use grounds agent behaviour in deterministic computation and extends capabilities beyond what next-token prediction alone can achieve.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:code-execution", "labels": [ "Code Execution", "Code Execution by LLMs", "Code Interpreter", "Handler Execution" ], "is_subclass_of": [ "AI Agent System", "Tool Use", "Agentic Workflow", "AI Agent Systems", "LLM Agents", "CLI Multi-Agent Systems" ], "wikilinks": [ "AI Agent Systems", "CLI Multi-Agent Systems", "Agentic Workflow", "Large Language Models", "Code Generation", "Tool Use", "Sandboxed Code Execution", "Firecracker", "Docker", "gVisor", "Python", "Bash", "CodeAct", "ReAct", "OpenHands", "Model Context Protocol", "E2B", "Modal", "Reinforcement Learning", "Function Calling" ] }, { "id": "code-generation", "title": "Code Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The automated production of source code by AI systems from natural language specifications, partial code, or structured prompts. Code generation systems leverage large language models trained on code corpora to synthesise functions, classes, and complete programs, accelerating software development workflows.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:code-generation", "labels": [ "Code Generation", "AI Code Generation", "Code Generation Agents", "Multi-File Code Generation" ], "is_subclass_of": [ "Generative AI", "Program Synthesis", "Artificial Intelligence", "Large Language Models" ], "wikilinks": [ "Artificial Intelligence", "Generative AI", "Large Language Models", "Transformer Architecture", "Prompt Engineering", "Reinforcement Learning from Human Feedback", "Program Synthesis", "Neural Network", "Software Testing", "Static Analysis", "DevOps", "Attention Mechanism", "Self-Supervised Learning", "Natural Language Processing", "Software Architecture", "Deep Learning", "Embedding", "Fine-Tuning", "Continuous Integration", "Test-Driven Development" ] }, { "id": "code-review", "title": "Code Review", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Code review is the systematic, human-led examination of proposed source code changes by one or more reviewers other than the original author, intended to detect defects, enforce coding standards, and disseminate architectural knowledge across a development team. It is most commonly performed asynchronously via pull requests or merge requests in version-controlled repositories, augmented by automated static analysis, linting, and continuous integration checks. Effective code review functions as both a quality gate and a collaborative learning mechanism, reducing the cost of defects by catching them before integration into the main codebase. It is a foundational practice in professional software engineering, distributed-collaboration workflows, and secure development lifecycles.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:code-review", "labels": [ "Code Review", "Pull Request Review" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Version Control", "Software Development", "Software Testing", "Software Engineering" ] }, { "id": "code-signing", "title": "Code Signing", "domain": "security", "domain_name": "Security", "definition": "Code Signing is a cryptographic practice in which software publishers digitally sign executables, scripts, container images, and other software artefacts using a private key, enabling recipients to verify the artefact's authenticity and integrity through the corresponding public key certificate. Implemented via asymmetric cryptography and X.509 certificate chains anchored to trusted Certificate Authorities or transparency logs, code signing is a fundamental control in software supply chain security, preventing the distribution of tampered or malicious software. Modern approaches include keyless signing via short-lived certificates and cryptographic transparency logs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:code-signing", "labels": [ "Code Signing" ], "is_subclass_of": [ "Cryptographic Signing" ], "wikilinks": [] }, { "id": "code-based-design", "title": "Code-based Design", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A design methodology where visual elements are generated and manipulated through code structures like SVGs rather than raster-based image synthesis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:code-based-design", "labels": [ "Code-based Design" ], "is_subclass_of": [ "Image Generation" ], "wikilinks": [] }, { "id": "code-former", "title": "CodeFormer", "domain": "ai", "domain_name": "Ai", "definition": "CodeFormer is a Transformer-based blind face restoration model that combines a discrete codebook prior \u2014 learned via a vector-quantised autoencoder \u2014 with a controllable fidelity-quality trade-off mechanism, allowing it to recover high-quality facial details from severely degraded inputs such as low-resolution, compressed, or heavily noisy images. Unlike earlier GAN-based restoration approaches, CodeFormer's codebook provides rich, semantically plausible facial priors that guide reconstruction without requiring clean reference images, and its fidelity weight parameter lets users tune the balance between realism and fidelity to the original degraded input. The model generalises to face enhancement tasks in AI-generated images, old photo restoration, and video face restoration.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:code-former", "labels": [ "CodeFormer" ], "is_subclass_of": [ "Deep Learning", "Image Restoration", "Computer Vision", "Generative AI", "Blind Image Restoration", "Face Recognition" ], "wikilinks": [ "Deep Learning", "Transformer Architecture", "Attention Mechanism", "Generative Adversarial Network", "AI Upscaling and Super-Resolution", "Super Resolution", "Inpainting", "Image Processing", "Generative AI", "Film VFX", "Convolutional Neural Network", "Diffusion Model", "Stable Diffusion", "Diffusion Transformer", "GPU Compute", "Training Data", "Computer Vision", "Self-Supervised Learning" ] }, { "id": "codec", "title": "Codec", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A codec is a piece of software or hardware that encodes data into a compressed or transmittable format and decodes it back for use, applied to audio, video or general data streams. Codecs trade off compression ratio, computational cost and fidelity, and a specific data format's specification typically names the codec, or set of permitted codecs, it requires. Screen recording software depends on a codec to encode captured frames into a storable video file in real time.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:codec", "labels": [ "Codec" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "coding-theory", "title": "Coding Theory", "domain": "data", "domain_name": "Data", "definition": "The mathematical study of codes \u2014 structured mappings from messages to symbol sequences \u2014 and of their properties for reliable, efficient representation and transmission of information. Spanning error-correcting codes that add redundancy to detect and repair corruption, source codes that compress data towards its entropy, and cryptographic and network codes, coding theory turns Shannon's existence theorems into constructive schemes such as Hamming, Reed-Solomon, LDPC, turbo, and polar codes that underpin storage, networking, and broadcasting.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:coding-theory", "labels": [ "Coding Theory" ], "is_subclass_of": [ "Information Theory" ], "wikilinks": [ "Information Theory", "Error Correction", "Reed-Solomon Codes", "Erasure Coding", "Computational Complexity Theory" ] }, { "id": "cognitive-ai", "title": "Cognitive AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cognitive AI is the family of artificial-intelligence systems built around cognitively-plausible architectures that explicitly model the components and information-flow of human cognition \u2014 working memory, declarative and procedural long-term memory, perception-action loops, attention, production...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cognitive-ai", "labels": [ "Cognitive AI" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence", "Cognitive Architecture", "Neurosymbolic AI", "Hybrid AI", "Cognitive Science" ], "wikilinks": [ "AAAI Cognitive Systems", "ACT-R", "Advances in Cognitive Systems Journal", "AlgorithmLayer", "Anderson 1993 Rules of the Mind", "Anderson 2004 ACT-R Integrated Theory of the Mind", "ArchitectureLayer", "Bayesian Inference", "Black-Box Machine Learning", "Chunking Mechanism", "CLARION", "Clark 2008 Supersizing the Mind", "Clark 2023 The Experience Machine", "CoALA", "Cognitive Architecture", "Cognitive Engineering", "Cognitive Modelling", "Cognitive Psychology", "Cognitive Science", "CognitiveScienceDomain" ] }, { "id": "cognitive-architecture", "title": "Cognitive Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A cognitive architecture is a formal specification of the fixed computational structures, memory systems, and control mechanisms that together constitute a general-purpose intelligent agent, independent of any particular task or domain. It prescribes how perception, attention, memory retrieval, reasoning, learning, and action selection are integrated into a unified processing cycle, providing a theoretical and engineering framework for building systems that exhibit adaptive, goal-directed behaviour. Classical examples include ACT-R (Adaptive Control of Thought\u2013Rational), SOAR, and LIDA; contemporary variants extend these principles to neural-symbolic hybrids, transformer-based agent frameworks, and large language model scaffolding systems. Cognitive architectures serve simultaneously as psychological theories of the human mind and as blueprints for artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cognitive-architecture", "labels": [ "Cognitive Architecture" ], "is_subclass_of": [ "Artificial Intelligence", "AI Model Architecture", "Intelligent Agent", "Agent-Based Modelling" ], "wikilinks": [ "Artificial Intelligence", "Agentic AI", "Working Memory", "Long-Term Memory", "Declarative Memory", "Procedural Memory", "Reasoning", "Planning and Scheduling", "Perception", "Action Selection", "Symbolic Reasoning", "Autonomous Agents", "Knowledge Representation", "Ontology", "Machine Learning", "Production Rules", "Neural Networks", "Bayesian Inference", "Reinforcement Learning", "Large Language Models" ] }, { "id": "cognitive-automation-revolution", "title": "Cognitive Automation Revolution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Cognitive Revolution describes the transformative societal shift driven by AI systems automating cognitive and knowledge work at scale, drawing analogy to the Industrial Revolution's mechanisation of physical labour. It encompasses the displacement of routine mental tasks\u2014coding, content creation, data analysis\u2014freeing human effort for higher-order creativity and judgement, whilst raising urgent questions about workforce adaptation and economic equity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cognitive-automation-revolution", "labels": [ "Cognitive Automation Revolution", "The Cognitive Revolution" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "cognitive-feedback-interface", "title": "Cognitive Feedback Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Adaptive interface system that dynamically adjusts information flow and interaction modalities based on real-time assessment of user cognitive state, attention levels, and mental workload.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cognitive-feedback-interface", "labels": [ "Cognitive Feedback Interface" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Adaptive Information Display", "Adaptive UI Controller", "Attention Tracker", "Biometric Sensors", "Brain-Computer Interface", "Cognitive Load Management", "Cognitive Model", "Cognitive State Monitor", "ISO 9241-112", "Neurofeedback System", "Personalized UX", "Real-time Analytics", "Sensor Input", "Workload Analyzer", "Attention-Aware Interaction", "Eye Tracking", "InteractionDomain", "Machine Learning", "NetworkLayer" ] }, { "id": "cognitive-load-metric", "title": "Cognitive Load Metric", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Quantitative measure of mental effort required during virtual interaction tasks, typically assessed using standardized scales like NASA-TLX.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:cognitive-load-metric", "labels": [ "Cognitive Load Metric", "Cognitive Load Analysis", "Cognitive Load Measurements" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Interface Design Validation", "ISO 9241-112", "Measurement Framework", "PresentationLayer", "Psychometric Scale", "Task Complexity Analysis", "Usability Testing", "User Experience Assessment", "User Feedback", "ApplicationLayer", "InteractionDomain", "Performance Optimization", "Telecollaboration" ] }, { "id": "cognitive-load", "title": "Cognitive Load", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cognitive load is the total amount of working-memory resource demanded by a task, interface, or learning activity at a given moment. It is commonly decomposed into intrinsic load (inherent task difficulty), extraneous load (imposed by poor presentation), and germane load (effort devoted to building durable mental schemas). Managing cognitive load is a central goal of user-experience and interaction design, since exceeding a user's capacity degrades comprehension, accuracy, and satisfaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cognitive-load", "labels": [ "Cognitive Load" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [] }, { "id": "cognitive-modelling", "title": "Cognitive Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cognitive modelling is the construction of computational models that simulate human cognitive processes such as perception, memory, reasoning and decision-making, in order to test theories of mind or to inform artificial agents. Models range from symbolic production-rule systems to connectionist and hybrid architectures that reproduce empirical response-time and error patterns from human studies. Cognitive architectures and cognitive AI systems draw on cognitive modelling to structure how an artificial agent represents and reasons over knowledge.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:cognitive-modelling", "labels": [ "Cognitive Modelling" ], "is_subclass_of": [ "Cognitive Science" ], "wikilinks": [] }, { "id": "cognitive-psychology", "title": "Cognitive Psychology", "domain": "ai", "domain_name": "Ai", "definition": "Cognitive psychology is the scientific discipline that investigates the internal mental processes underlying intelligent behaviour, including perception, attention, memory, language, problem-solving, reasoning, and decision-making, treating the mind as an information-processing system and using controlled experiments, computational models, and neuroimaging to reveal its representations and operations.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:cognitive-psychology", "labels": [ "Cognitive Psychology" ], "is_subclass_of": [ "Cognitive Science", "Psychology", "AI Research Area", "Behavioural Science" ], "wikilinks": [ "Decision Making", "Behavioural Economics", "owl:Thing", "Cognitive Science", "Neuroscience", "Working Memory", "Attention Mechanism", "Natural Language Processing", "Human Computer Interaction", "Reinforcement Learning", "Explainable AI", "Cognitive Architecture", "Philosophy of Mind", "Embodied Cognition", "Knowledge Representation", "Machine Learning", "Artificial Intelligence" ] }, { "id": "cognitive-science", "title": "Cognitive Science", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cognitive Science is an inherently interdisciplinary field that investigates the nature of mind, intelligence, and cognition by integrating methods and theories from psychology, neuroscience, linguistics, philosophy, computer science, and anthropology. It studies how information is represented, processed, and transformed in biological and artificial systems, encompassing perception, attention, memory, language, reasoning, problem-solving, and decision-making. Computational models derived from cognitive science provide foundational frameworks for artificial intelligence, informing architectures ranging from symbolic reasoning systems to neural network design. Its empirical findings on human cognition directly shape human-computer interaction, educational technology, and the development of intelligent user interfaces.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:cognitive-science", "labels": [ "Cognitive Science" ], "is_subclass_of": [ "Artificial Intelligence", "Interdisciplinary Science", "AI Research Area" ], "wikilinks": [ "Psychology", "Neuroscience", "Linguistics", "Philosophy of Mind", "Computer Science", "Anthropology", "Artificial Intelligence", "Human Computer Interaction", "Consciousness", "Embodied Cognition", "Cognitive Architecture", "Computational Linguistics", "Cognitive Psychology", "Natural Language Processing", "Deep Learning", "Machine Learning", "Computer Vision", "Reinforcement Learning", "Large Language Models", "Knowledge Graphs" ] }, { "id": "cognitive-walkthrough", "title": "Cognitive Walkthrough", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A cognitive walkthrough is a usability inspection method in which evaluators step through a task sequence as a first-time user would, at each step asking whether the user would know what to do, would notice the correct action is available, and would correctly interpret the feedback received. Unlike heuristic evaluation, which checks a design against general usability principles, a cognitive walkthrough focuses specifically on learnability for new users performing a defined task. It is typically conducted early in design, using prototypes rather than finished products.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:cognitive-walkthrough", "labels": [ "Cognitive Walkthrough" ], "is_subclass_of": [ "Usability" ], "wikilinks": [] }, { "id": "cohere", "title": "Cohere", "domain": "ai", "domain_name": "Ai", "definition": "Cohere is a Canadian-American enterprise AI company founded in 2019 by Aidan Gomez, Nick Frosst, and Ivan Zhang \u2014 all former Google Brain researchers \u2014 that develops and deploys large language models, text embedding models, and reranking models through a cloud API, private cloud, and on-premises deployment model, with a strategic focus on enterprise data security, sovereign AI, and retrieval-augmented generation pipelines. Its model families include the Command series for generative tasks, the Embed series for dense vector representations, the Rerank series for cross-encoder document scoring, and the Aya multilingual research models covering 101 languages. Cohere is distinguished by its early and consistent commitment to private and on-premises deployment options that ensure enterprise data never transits external infrastructure, and by its April 2026 merger with Germany's Aleph Alpha to form a transatlantic sovereign AI entity valued at approximately $20 billion.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:cohere", "labels": [ "Cohere" ], "is_subclass_of": [ "Large Language Models", "Enterprise AI", "Natural Language Processing" ], "wikilinks": [ "Transformer", "Transformer Architecture", "Embeddings", "Semantic Search", "Retrieval-Augmented Generation", "Natural Language Processing", "Language Model", "Large Language Models", "Fine-Tuning", "Vector Database", "API", "Enterprise AI", "Sovereign AI", "Mixture of Experts", "Multimodal AI", "Text Classification", "Question Answering", "Document Understanding", "Reranking", "On-Premises Deployment" ] }, { "id": "coin", "title": "Coin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Coin is a native cryptographic asset issued and governed by a blockchain protocol itself, distinct from tokens created by smart contracts on top of an existing chain. Coins function as the primary medium of exchange for transaction fees, validator rewards, and network participation incentives within their respective distributed ledger ecosystems, and may also serve as stores of value, monetary reserves, or collateral in decentralised finance applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:coin", "labels": [ "Coin" ], "is_subclass_of": [ "Token and Asset", "Blockchain Entity", "Economic Mechanism", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "coinbase-transaction", "title": "Coinbase Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The first transaction in every blockchain block, created by the miner or block producer, that contains no inputs and issues the block reward plus accumulated transaction fees to the miner's address. Unlike regular transactions, the coinbase transaction has no sender; it creates new coins from protocol-defined issuance rules and serves as the primary mechanism by which new cryptocurrency enters circulation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:coinbase-transaction", "labels": [ "Coinbase Transaction" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "coinbase", "title": "Coinbase", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Coinbase is a publicly traded United States cryptocurrency exchange, custody, and financial services company founded in 2012 by Brian Armstrong and Fred Ehrsam. It provides retail and institutional trading, custodial and self-custody wallet services, staking, lending, and payment infrastructure, operating as one of the most regulated crypto platforms in the United States. The company completed a direct listing on Nasdaq in April 2021, becoming the first major cryptocurrency exchange to go public in the US. Beyond exchange operations, Coinbase develops the Base layer-2 network (built on the OP Stack), offers a developer platform for on-chain applications, and provides institutional prime brokerage services.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:coinbase", "labels": [ "Coinbase" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [ "Binance", "Optimism", "Digital Asset Domain" ] }, { "id": "colbert", "title": "ColBERT", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A late-interaction neural retrieval model that encodes queries and documents into per-token embedding matrices with BERT and scores relevance by summing each query token's maximum similarity over document tokens, retaining much of the accuracy of full cross-encoder attention whilst permitting document embeddings to be precomputed and indexed for scalable, fast retrieval.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:colbert", "labels": [ "ColBERT" ], "is_subclass_of": [ "Dense Retrieval" ], "wikilinks": [ "Dense Retrieval", "BERT", "Cross-Encoder Reranking", "BEIR Benchmark" ] }, { "id": "cold-chain-monitoring", "title": "Cold Chain Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cold chain monitoring is an integrated infrastructure discipline encompassing sensor networks, communication protocols, data management platforms, and provenance-recording systems that collectively maintain, verify, and audit temperature-controlled conditions for perishable goods across the entir...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cold-chain-monitoring", "labels": [ "Cold Chain Monitoring", "ColdChainMonitoring" ], "is_subclass_of": [ "Legal and Regulatory", "Supply Chain", "Monitoring Systems", "Sensor Networks", "Traceability", "Compliance Infrastructure", "Logistics Management" ], "wikilinks": [ "Alert System", "Automated Lot Release", "Biologics Logistics", "Blockchain Records", "Centralised Data Silos", "Certificate of Compliance", "Chemical Logistics", "Codex Alimentarius", "Compliance Evidence", "Compliance Infrastructure", "ConnectivityLayer", "Data Analytics Platform", "DHL", "DSCSA", "DSCSA Compliance", "Edge Gateway", "Emergen Research", "EU GDP Guidelines", "Excursion Detection", "FDA" ] }, { "id": "cold-gas-propulsion", "title": "Cold Gas Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cold-gas-propulsion", "labels": [ "Cold Gas Propulsion" ], "is_subclass_of": [ "Spacecraft Propulsion" ], "wikilinks": [] }, { "id": "cold-storage", "title": "Cold Storage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cold Storage in the context of digital assets and cryptocurrency refers to the practice of holding private keys in an offline environment \u2014 physically disconnected from any network \u2014 to eliminate the attack surface presented by internet-connected systems. Hardware wallets, air-gapped computers, and paper wallets are common cold storage implementations. By contrast with hot wallets (internet-connected), cold storage sacrifices transaction convenience for maximum security, and is the industry standard for custodying large quantities of cryptocurrency at exchanges, institutional custodians, and high-net-worth individual holders.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:cold-storage", "labels": [ "Cold Storage", "Cold Storage System" ], "is_subclass_of": [ "Custody Infrastructure" ], "wikilinks": [] }, { "id": "collaboration-platform", "title": "Collaboration Platform", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Collaboration Platform is an integrated software environment that unifies communication channels, shared workspaces, workflow automation, and content management tooling to enable coordinated work among distributed teams. Such platforms expose API-first architectures built atop real-time messaging protocols, WebRTC-based media pipelines, and cloud storage with conflict resolution, providing a composable substrate for synchronous and asynchronous teamwork. Enterprise-grade deployments layer in identity federation, data residency controls, compliance archiving, and extensible integration ecosystems, while modern platforms increasingly embed AI-driven capabilities such as meeting transcription, action-item extraction, and intelligent search. The category spans lightweight messaging tools through full unified-communications suites and extends into spatial and immersive collaboration modalities.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:collaboration-platform", "labels": [ "Collaboration Platform", "3D Collaboration Platform", "Digital Collaboration Platform", "Real-Time Collaboration Platform" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "IETF (Internet Engineering Task Force)" ] }, { "id": "collaboration-technology", "title": "Collaboration Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software and hardware systems that enable multiple participants to coordinate, communicate, and jointly produce work across physical or virtual spaces. In spatial computing contexts, collaboration technology encompasses shared XR environments, co-presence avatars, spatial audio, and synchronised digital workspace tools that support telecollaboration at a distance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:collaboration-technology", "labels": [ "Collaboration Technology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing", "Telecollaboration" ] }, { "id": "collaboration-tools", "title": "Collaboration Tools", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Software that supports people working together on shared tasks or documents, including communication, coordination and concurrent editing features.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:collaboration-tools", "labels": [ "Collaboration Tools", "Multiplayer Collaboration Tools" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [ "Network Architecture", "Operational Transformation", "CRDT", "Version Control", "Software Development" ] }, { "id": "collaboration", "title": "Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Collaboration is the coordinated joint activity of two or more agents \u2014 human or artificial \u2014 working toward shared goals through negotiated division of labour, mutual adjustment, and shared artefact production. In digital and distributed settings it encompasses synchronous and asynchronous communication, collective sense-making, and structured workflows mediated by software systems that maintain shared state and contextual awareness across geographic and organisational boundaries. Effective collaboration depends on social and technical mechanisms for conflict resolution, attribution, version management, and trust; its theoretical foundations draw from Computer-Supported Cooperative Work (CSCW), distributed cognition, and organisational theory. Contemporary collaboration increasingly involves AI agents, immersive shared environments, and decentralised coordination protocols.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:collaboration", "labels": [ "Collaboration", "Efficient Collaboration", "Global Collaboration", "Industry Collaboration" ], "is_subclass_of": [ "Distributed Collaboration", "Workspace Tools" ], "wikilinks": [ "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "W3C (World Wide Web Consortium)" ] }, { "id": "collaborative-design", "title": "Collaborative Design", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Collaborative Design is a design methodology in which multiple stakeholders\u2014including designers, engineers, end users, and domain experts\u2014jointly participate in defining, iterating, and validating artefacts or systems. It draws on participatory design traditions to ensure that lived experience and diverse expertise shape outcomes rather than being consulted retrospectively. Digital collaborative design platforms support real-time co-authoring, version management, and synchronous review across distributed teams. The approach is applied in product development, spatial design, software architecture, and policy co-creation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:collaborative-systems-modality-design", "labels": [ "Collaborative Design", "CollaborativeDesign" ], "is_subclass_of": [ "Design Thinking" ], "wikilinks": [] }, { "id": "collaborative-editing", "title": "Collaborative Editing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Collaborative editing is the capability for multiple people to view and modify a shared document simultaneously, with their changes merged consistently and surfaced to all participants in near real time. It relies on concurrency-control techniques such as operational transformation or conflict-free replicated data types to reconcile concurrent edits without conflicts. Combined with presence and cursor awareness, it makes co-authoring across distributed users feel immediate and coherent.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:collaborative-editing", "labels": [ "Collaborative Editing" ], "is_subclass_of": [ "Real-Time Collaboration", "Version History" ], "wikilinks": [] }, { "id": "collaborative-filtering", "title": "Collaborative Filtering", "domain": "ai", "domain_name": "Ai", "definition": "Collaborative filtering is a machine learning technique for building recommendation systems that generates predictions about user preferences by aggregating the behaviour or ratings of many users, without requiring explicit knowledge of item content. Memory-based approaches compute similarity between users (user-based) or items (item-based) using rating vectors; model-based approaches such as matrix factorisation decompose the user-item interaction matrix into latent factor spaces. The method operates on the assumption that users who agreed in the past will agree in the future, and is the foundational algorithm behind recommendation engines at Netflix, Spotify, and Amazon.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:collaborative-systems-modality-filtering", "labels": [ "Collaborative Filtering" ], "is_subclass_of": [ "Machine Learning Technique", "Information Retrieval", "Machine Learning Discipline" ], "wikilinks": [ "Recommendation System", "Recommendation Engine", "Recommendation Systems", "Machine Learning Technique", "Machine Learning Discipline", "Matrix Factorisation", "Embeddings", "Deep Learning", "Neural Network", "Graph Neural Network", "Transformer Architecture", "Federated Learning", "Differential Privacy", "Data Privacy", "Privacy Preserving Data Mining", "Hyper personalisation", "Gradient Descent", "Stochastic Gradient Descent", "Large Language Model", "Vector Database" ] }, { "id": "collaborative-learning", "title": "Collaborative Learning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An educational approach where learners work together in virtual environments, metaverse platforms, or VR spaces to achieve shared learning goals through social interaction, knowledge co-construction, and collective problem-solving enabled by immersive and embodied digital experiences.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:collaborative-systems-modality-learning", "labels": [ "Collaborative Learning", "CollaborativeLearning" ], "is_subclass_of": [ "Educational Methodology" ], "wikilinks": [ "Communication Tools", "Knowledge Co-Construction", "Shared Workspace", "Social Learning", "Educational Methodology", "metaverse", "Remote Collaboration", "Telecollaboration", "Virtual Environment" ] }, { "id": "collaborative-operation", "title": "Collaborative Operation", "domain": "robotics", "domain_name": "Robotics", "definition": "Collaborative Operation - Coordinated execution of tasks between Human Operators and Robotic Systems within the same workspace, governed by safety protocols, task allocation mechanisms, and real-time communication to achieve shared objectives.", "entityType": "Class", "qualityScore": 0.56, "maturity": "draft", "iri": "urn:ngm:class:collaborative-systems-modality-operation", "labels": [ "Collaborative Operation" ], "is_subclass_of": [ "Human-Robot Interaction", "Robotics" ], "wikilinks": [ "Force Feedback", "GDPR (General Data Protection Regulation)", "Human Operators", "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "Manufacturing Process", "Production Flexibility", "Robotic Systems", "Safety Monitoring", "Workspace Efficiency", "Human-Robot Interaction", "Robotics", "RoboticsDomain", "Task Planning" ] }, { "id": "collaborative-robot", "title": "Collaborative Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Collaborative Robot (cobot) - A lightweight robotic arm engineered to operate safely alongside human workers, combining force/torque sensing, reduced kinetic energy, and speed limitations to enable Human-Robot Collaboration in shared manufacturing and assembly environments. Cobots are distinguished from traditional industrial robots by their intrinsic safety mechanisms, ease of programming, and ability to share workspace with humans without physical guarding barriers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:collaborative-systems-modality-robot", "labels": [ "Collaborative Robot", "Collaborative Robot Deployment", "Collaborative Robots", "CollaborativeRobot" ], "is_subclass_of": [ "Industrial Robot", "Robotics" ], "wikilinks": [ "Collaborative Manufacturing System", "Flexible Assembly", "Human-Robot Collaboration", "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "ISO/TS 15066 Compliance", "Rapid Deployment", "Reduced Labour Costs", "Safety Controller", "Force Torque Sensor", "Industrial Robot", "Robotics", "RoboticsDomain", "Telecollaboration" ] }, { "id": "collaborative-robotics", "title": "Collaborative Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Collaborative Robotics is the engineering discipline concerned with designing, deploying, and operating robotic systems \u2014 particularly collaborative robots (cobots) \u2014 that share workspace and tasks with human workers without requiring physical barriers, relying instead on force-torque sensing, speed-and-separation monitoring, and power-and-force limiting to maintain safety under ISO 10218 and ISO/TS 15066. Unlike traditional industrial robots that operate in guarded cages, cobots are designed with compliant joints, rounded profiles, and real-time collision-detection to enable direct physical cooperation with humans on assembly, inspection, and logistics tasks. The discipline integrates mechanical design, control theory, human-robot interaction research, and regulatory compliance to enable safe human-robot co-presence in shared workspaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:collaborative-systems-modality-robotics", "labels": [ "Collaborative Robotics" ], "is_subclass_of": [ "Collaborative Robot" ], "wikilinks": [] }, { "id": "collaborative-robots", "title": "Collaborative Robots", "domain": "robotics", "domain_name": "Robotics", "definition": "Collaborative robots, or cobots, are robots designed to work safely alongside humans within a shared workspace.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:collaborative-systems-modality-robots", "labels": [ "Collaborative Robots" ], "is_subclass_of": [ "Industrial Robotics", "Industrial Robot" ], "wikilinks": [ "Sensors", "Actuators", "Control Theory", "Industrial Robotics", "https://www.iso.org/standard/62996.html", "https://en.wikipedia.org/wiki/Cobot" ] }, { "id": "collaborative-simulation", "title": "Collaborative Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computational environment in which multiple participants \u2014 human or agent \u2014 share, manipulate, and observe a common simulation state in real time. It combines networked synchronisation with physics or behavioural modelling to support joint exploration, training, or design activities across distributed locations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:collaborative-systems-modality-simulation", "labels": [ "Collaborative Simulation", "CollaborativeSimulation" ], "is_subclass_of": [ "Collaborative Technology" ], "wikilinks": [] }, { "id": "collaborative-systems-modality", "title": "Collaborative Systems Modality", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Collaborative describes systems, processes, or modalities in which multiple agents\u2014human or robotic\u2014work together toward shared goals, actively coordinating actions, sharing information, and adapting to one another's contributions. In robotics, 'collaborative' is most precisely applied to human\u2013robot collaboration (HRC) per ISO/TS 15066, where robot and human share a workspace without fixed barriers. In knowledge work and software, it describes tools and practices enabling co-creation, distributed editing, and collective decision-making across asynchronous or real-time channels.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:collaborative-systems-modality", "labels": [ "Collaborative Systems Modality", "Collaborative AI Platform", "Collaborative Authoring", "Collaborative Document Editing", "Collaborative IDEs", "Collaborative Knowledge Discovery", "Collaborative Manufacturing System", "Collaborative Model Ownership", "Collaborative Scene Assembly", "Collaborative Work", "Collaborative XR", "collaborative" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "collaborative-technology", "title": "Collaborative Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Collaborative Technology encompasses the hardware, software, protocols, and infrastructure components that enable distributed teams to work together effectively across time and space. This includes real-time media processing, synchronisation protocols, edge computing for latency reduction, and zero-trust security models forming the technological substrate of telecollaboration.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:collaborative-systems-modality-technology", "labels": [ "Collaborative Technology" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "IEEE (Institute of Electrical and Electronics Engineers)", "IETF (Internet Engineering Task Force)" ] }, { "id": "collaborative-whiteboard", "title": "Collaborative Whiteboard", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Collaborative Whiteboard is a shared infinite-canvas digital workspace enabling real-time multi-user drawing, diagramming, sticky-note placement, voting, and structured ideation, delivered through web or native applications with CRDT-based (Conflict-free Replicated Data Type) or Operational...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:collaborative-systems-modality-whiteboard", "labels": [ "Collaborative Whiteboard" ], "is_subclass_of": [ "Workspace Tools", "Communication Technology", "Visual Collaboration Tools", "Distributed Collaboration", "Asynchronous Communication", "Remote Work Technology", "Digital Workspace" ], "wikilinks": [ "Affinity Diagramming", "Agile Retrospective", "Asynchronous Communication", "Asynchronous Ideation", "Authentication", "Breakout Board", "Browser Graphics Rendering", "Canvas API", "Cloud Storage", "Collaborative Design", "Conflict Resolution", "Confluence", "Connector", "Content Delivery Network", "CRDT Academic Literature", "Design Sprint", "Design Thinking", "DesignThinkingDomain", "Design Tools", "Digital Collaboration Platform" ] }, { "id": "collateral-management", "title": "Collateral Management", "domain": "finance", "domain_name": "Finance", "definition": "Collateral management is the operational and risk management discipline concerned with the posting, valuation, optimisation, and return of financial assets pledged as security against credit exposure in derivatives, securities lending, repo, and cleared transactions, ensuring that counterparty credit risk is adequately mitigated throughout the life of a financial contract. It encompasses collateral eligibility determination, margin call issuance and settlement, collateral transformation, and regulatory compliance with frameworks including EMIR, Dodd-Frank, and Basel III initial margin requirements.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:collateral-management", "labels": [ "Collateral Management", "Collateral Reserves", "DeFi Collateral", "Margin Collateral" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "collateral", "title": "Collateral", "domain": "finance", "domain_name": "Finance", "definition": "Collateral is an asset pledged by a borrower to a lender as security for a loan or obligation, which the lender may seize or liquidate if the borrower defaults. In decentralised finance it typically takes the form of over-collateralised crypto assets locked in a smart contract, backing loans, synthetic assets, or stablecoin pegs. The collateral-to-debt ratio determines solvency and triggers liquidation when it falls below a protocol-defined threshold.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:collateral", "labels": [ "Collateral" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "collateralised-borrowing", "title": "Collateralised Borrowing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Collateralised borrowing is a lending mechanism in which a borrower locks digital assets as collateral in a smart contract in order to draw a loan, typically in a different asset, without a credit check or intermediary. The collateral is held against default risk and may be liquidated automatically if its value falls below a protocol-defined threshold relative to the loan. It is a core primitive of decentralised finance protocols such as Aave and Compound.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:collateralised-borrowing", "labels": [ "Collateralised Borrowing" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "collective-action", "title": "Collective Action", "domain": "governance", "domain_name": "Governance", "definition": "Coordinated behaviour by a group of individuals or organisations toward a shared goal that no single actor could achieve unilaterally. It is characterised by interdependence, the risk of free-riding, and the need for mechanisms \u2014 such as incentives, norms, or contracts \u2014 to align individual contributions with the group objective.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:collective-action", "labels": [ "Collective Action", "Collective Action Problem", "Collective Action Problems" ], "is_subclass_of": [ "Coordination Mechanisms" ], "wikilinks": [] }, { "id": "collective-communication", "title": "Collective Communication", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Collective communication is the class of synchronised, multi-party data-exchange operations in which a group of processes jointly participate, such as broadcast, scatter, gather, all-gather, reduce, and all-reduce. It provides the communication primitives that coordinate state across the nodes of a parallel or distributed system, and it is the backbone of distributed machine-learning training, where gradients and parameters are aggregated and synchronised across many accelerators. Implementations are optimised over high-speed interconnects to minimise the communication overhead that otherwise bottlenecks scaling.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:collective-communication", "labels": [ "Collective Communication" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "collective-decision-making", "title": "Collective Decision Making", "domain": "governance", "domain_name": "Governance", "definition": "Collective decision making is the social and technical process by which a group of agents \u2014 human, algorithmic, or hybrid \u2014 aggregate individual preferences, information, or votes to arrive at a binding or advisory choice on behalf of the group. It encompasses the design and analysis of voting systems, preference aggregation mechanisms, deliberation protocols, and incentive structures that determine how group choices are reached and enforced. In digital and distributed contexts, collective decision making is formalised through on-chain governance mechanisms, decentralised autonomous organisations, and algorithmic consensus protocols that encode group agency in smart contract logic. The field draws on social choice theory, mechanism design, game theory, and organisational science to evaluate fairness, efficiency, and manipulation-resistance of alternative procedures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:collective-decision-making", "labels": [ "Collective Decision Making", "Collective Decision-Making" ], "is_subclass_of": [ "Social Choice Theory" ], "wikilinks": [] }, { "id": "collective-governance", "title": "Collective Governance", "domain": "governance", "domain_name": "Governance", "definition": "A system of rules, norms, and decision-making processes through which a group of stakeholders jointly determines how shared resources, organisations, or platforms are managed. It distributes authority among participants rather than concentrating it in a single administrator, seeking legitimacy through inclusive participation and transparent deliberation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:collective-governance", "labels": [ "Collective Governance", "CollectiveGovernance" ], "is_subclass_of": [ "Community Governance" ], "wikilinks": [] }, { "id": "collective-intelligence-system", "title": "Collective Intelligence System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mechanism enabling groups of humans and agents to solve problems collaboratively using shared data through swarm intelligence and emergent decision-making.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:collective-intelligence-system", "labels": [ "Collective Intelligence System" ], "is_subclass_of": [ "AI Research Area", "Multi-Agent System", "Collaborative AI Platform", "Distributed Artificial Intelligence" ], "wikilinks": [ "Collaborative AI Platform", "Collaborative Decision-Making", "Data Synchronization", "Distributed Decision Network", "Emergent Pattern Detector", "Emergent Problem-Solving", "Human-AI Interface", "Knowledge Aggregation Module", "Multi-Agent System", "OECD AI Collective Intelligence 2025", "Swarm Coordination Engine", "Swarm Intelligence", "Autonomous Robot", "Collective Learning", "Communication Protocol", "ComputationAndIntelligenceDomain", "Consensus Mechanism", "Distributed Computing", "Machine Learning", "MiddlewareLayer" ] }, { "id": "collective-intelligence", "title": "Collective Intelligence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Collective intelligence is the shared or group intelligence that emerges from the collaboration, competition, and collective decision-making of many individuals or agents, producing cognitive capacities \u2014 prediction accuracy, problem-solving breadth, creative output \u2014 that exceed what any individual participant could achieve alone. It manifests in biological systems such as ant colonies and immune networks, human social institutions, and engineered multi-agent systems, and is characterised by distributed information processing, diversity of perspective, and aggregation mechanisms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:collective-intelligence", "labels": [ "Collective Intelligence", "CollectiveIntelligence" ], "is_subclass_of": [ "Collective Intelligence System" ], "wikilinks": [] }, { "id": "collective-learning", "title": "Collective Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A collaborative educational approach in virtual environments and metaverse platforms where groups of learners work toger to construct knowledge, share experiences, and solve problems through social interaction, benefiting from diverse perspectives, AI-powered personalization, and immersive VR/AR ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:collective-learning", "labels": [ "Collective Learning" ], "is_subclass_of": [ "AI Technique", "Educational Methodology" ], "wikilinks": [ "Communication Tools", "Global Collaboration", "Knowledge Co-Construction", "Peer Learning", "Educational Methodology", "Learning Management System", "metaverse", "Telecollaboration", "Virtual Environment" ] }, { "id": "collective-memory-archive", "title": "Collective Memory Archive", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A community-maintained repository that preserves shared cultural memories, historical events, and collective experiences for long-term access and cultural heritage preservation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:collective-memory-archive", "labels": [ "Collective Memory Archive" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Community Contributions", "Community Knowledge Systems", "Community Storytelling", "Cultural Preservation", "Digital Preservation Standards", "Heritage Access", "Historical Documentation", "Memory Records", "Preservation Metadata", "Preservation Policies", "Temporal Index", "Access Control System", "Authentication Service", "CreativeMediaDomain", "DID Nostr Identity", "Digital Repository", "Metadata Registry", "MiddlewareLayer", "Search Engine", "Storage Infrastructure" ] }, { "id": "collision-avoidance", "title": "Collision Avoidance", "domain": "robotics", "domain_name": "Robotics", "definition": "Collision Avoidance - An active safety mechanism using Sensors (lidar, ultrasonic, vision) and path planning algorithms to detect obstacles and dynamically modify Robot Trajectories to prevent unintended contact with people, equipment, or structures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:collision-avoidance", "labels": [ "Collision Avoidance", "CollisionAvoidance" ], "is_subclass_of": [ "Navigation and Planning", "Robotics" ], "wikilinks": [ "Emergency Stop", "IEEE (Institute of Electrical and Electronics Engineers)", "ISO (International Organization for Standardization)", "Robot Trajectories", "Safe Human-Robot Proximity", "Safety System", "Sensors", "Autonomous Navigation", "Computer Vision", "Motion Control", "Perception System", "Real-time Processing", "Robotics", "RoboticsDomain" ] }, { "id": "collision-detection-system", "title": "Collision Detection System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computational subsystem that continuously tests whether geometric objects in a simulation or physical environment intersect or are about to intersect, enabling physics engines, robotics planners, and interactive applications to respond to contact events. It combines spatial partitioning structures with narrow-phase geometry tests to balance accuracy against performance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:collision-detection-system", "labels": [ "Collision Detection System" ], "is_subclass_of": [ "Collision Detection" ], "wikilinks": [] }, { "id": "collision-detection", "title": "Collision Detection", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Collision Detection is the computational discipline of determining when two or more geometric primitives, rigid bodies, deformable meshes, articulated kinematic chains, or volumetric fields occupy overlapping regions of a shared spatial domain, decomposed canonically into a broad phase that rapid...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:collision-detection", "labels": [ "Collision Detection", "Collision Checker" ], "is_subclass_of": [ "Platform and Environment", "Physics Simulation", "Spatial Computing Paradigm", "Computational Geometry", "Geometric Algorithm", "Real-Time Algorithm" ], "wikilinks": [ "1X NEO", "AABB", "AlgorithmLayer", "Animation", "Apollo (Baidu)", "Apple Vision Pro", "Apple Vision Pro", "Autonomous Vehicles", "Autonomous Vehicles", "Box2D", "Brax", "Bridson Fedkiw Anderson 2002 Cloth Collisions", "Broad-Phase Detection", "BSP Tree", "BSP Tree", "Bullet Physics", "Bullet Physics", "CAD Software", "CATIA", "Catto 2007 Box2D" ] }, { "id": "collision-resistance", "title": "Collision Resistance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A security property of a cryptographic hash function asserting that it is computationally infeasible to find two distinct inputs that produce the same output digest. Collision resistance underpins the integrity guarantees of Merkle trees, digital signatures, and proof-of-work puzzles, and its absence would allow adversaries to forge blocks or certificates.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:collision-resistance", "labels": [ "Collision Resistance", "Hash Collision Resistance" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "colour-grading", "title": "Colour Grading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Colour grading is the post-production process of altering and enhancing the colour, contrast, and tonal qualities of a moving image or rendered frame to establish a consistent look and convey mood. It builds on basic colour correction \u2014 which neutralises exposure and white-balance errors \u2014 by applying creative adjustments such as curves, lift-gamma-gain controls, and secondary qualifiers that isolate specific hues or regions. In real-time rendering and virtual production the grade is frequently applied as a post-processing pass using look-up tables and tone-mapping operators so interactive scenes match the intended cinematic aesthetic.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:colour-grading", "labels": [ "Colour Grading" ], "is_subclass_of": [ "Post Processing" ], "wikilinks": [] }, { "id": "colour-management", "title": "Colour Management", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Colour management is the controlled conversion of colour representations between the characteristics of different devices and media so that colours appear consistent across capture, display and output. It relies on device profiles that describe how a given device reproduces colour and a profile connection space to translate between them. In spatial computing and real-time rendering, colour management ensures perceptually accurate imagery across cameras, displays and headsets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:colour-management", "labels": [ "Colour Management" ], "is_subclass_of": [ "Display Technology" ], "wikilinks": [] }, { "id": "columnar-storage", "title": "Columnar Storage", "domain": "data", "domain_name": "Data", "definition": "Columnar storage is a data organisation scheme that stores values from the same column of a table contiguously on disk, rather than storing complete rows together as in row-oriented storage. This layout allows analytical queries to scan only the columns they need, and enables aggressive compression because adjacent values within a column tend to be similar. Formats such as Apache Parquet and engines built for online analytical processing rely on columnar storage to accelerate large-scale aggregation and filtering workloads.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:columnar-storage", "labels": [ "Columnar Storage" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "combinatorial-optimisation", "title": "Combinatorial Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Combinatorial optimisation is the study of finding an optimal object from a finite but typically enormous set of discrete candidate solutions. Problems are defined over discrete structures such as graphs, permutations and integer assignments, and many are NP-hard, meaning no known algorithm solves all instances efficiently. Practical approaches combine exact methods, approximation algorithms and metaheuristics to obtain good solutions within acceptable time bounds.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:combinatorial-optimisation", "labels": [ "Combinatorial Optimisation" ], "is_subclass_of": [ "Optimisation", "Mathematical Optimisation" ], "wikilinks": [ "Optimisation", "Mathematical Optimisation", "Operations Research", "Graph Theory", "Algorithm", "Integer Programming", "Linear Programming", "Genetic Algorithm", "Convex Optimisation", "Constraint Satisfaction", "Logistics", "Decision Making", "NP-Hardness", "Computational Complexity", "Branch and Bound", "Simulated Annealing", "Local Search", "Approximation Algorithm", "Travelling Salesman Problem", "Vehicle Routing Problem" ] }, { "id": "comet", "title": "Comet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:comet", "labels": [ "Comet" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "comfy-ui", "title": "Comfy Ui", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ComfyUI is an open-source, node-graph based interface for building and executing generative-AI image and video pipelines, most commonly around diffusion models such as Stable Diffusion. Users assemble workflows by wiring together nodes for model loading, sampling, conditioning and post-processing, giving fine-grained control over the generation graph. Its modular design supports extensions, custom nodes and reproducible, shareable workflows.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:comfy-ui", "labels": [ "Comfy Ui", "ComfyUI" ], "is_subclass_of": [ "Generative AI", "Open Source Software", "Workflow Automation", "Node-Based Visual Programming" ], "wikilinks": [ "Generative AI", "Diffusion Model", "Stable Diffusion", "Latent Diffusion", "Node Graph", "Workflow Automation", "Image Generation", "Image Synthesis", "Prompt Engineering", "ControlNet", "Open Source Software", "Generative Model", "Inpainting", "Checkpoint Model", "User Interface", "LoRA Fine-Tuning", "VAE", "CLIP", "Text to Image", "Image to Image" ] }, { "id": "comfy-ui-api-specification", "title": "ComfyUI API Specification", "domain": "ai", "domain_name": "Ai", "definition": "The programmatic interface specification for ComfyUI, a node-based generative AI workflow engine, defining JSON-serialised graph representations (prompt API), WebSocket-based progress streaming, queue management endpoints, and model loading conventions that allow headless or remote execution of image and video generation pipelines.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:node-based-diffusion-pipeline-interface-api-specification", "labels": [ "ComfyUI API Specification", "ComfyUI API" ], "is_subclass_of": [ "ComfyUI Workflows", "REST API", "Workflow Automation", "Inference Orchestration System" ], "wikilinks": [ "Generative AI", "ComfyUI Workflows", "ComfyUI Client", "ComfyUI Manager", "ComfyUI", "Agentic Workflow", "Content Creation Pipeline", "REST API", "WebSocket Protocol", "JSON Serialisation", "Diffusion Model", "Stable Diffusion", "Text-to-Image Generation", "Image-to-Image Generation", "Video Generation", "ControlNet", "LoRA", "Directed Acyclic Graph Execution", "Workflow Execution Engine", "Workflow JSON Format" ] }, { "id": "comfy-ui-client", "title": "ComfyUI Client", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ComfyUI Client refers collectively to the suite of client-side interfaces, API layers, and software components that communicate with a running ComfyUI server\u2014the open-source, node-based diffusion model inference engine created by comfyanonymous (GitHub user) and released in January 2023\u2014enabling ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:node-based-diffusion-pipeline-interface-client", "labels": [ "ComfyUI Client" ], "is_subclass_of": [ "AI Infrastructure", "Node Based Interface", "Graphical User Interface", "API Client", "Workflow Orchestration System", "Generative AI Tool" ], "wikilinks": [ "3D Generation", "9elements 2024 Hosting ComfyUI Workflow via API", "aiohttp", "API", "API Client", "API Integration", "AUTOMATIC1111 WebUI", "BentoML 2025 Guide to ComfyUI Custom Nodes", "Black Forest Labs 2024 Flux", "Brooks et al 2024 Sora", "CLIP Text Encoder", "Comfy-Org", "Comfy-Org 2024 ComfyUI V1 Release", "Comfy-Org 2025 Dynamic VRAM", "Comfy-Org 2025 Meet the New ComfyUI-Manager", "Comfy-Org 2025 NVIDIA GPU Optimisations", "Comfy-Org ComfyUI Desktop", "Comfy-Org ComfyUI Frontend", "ComfyUI API Specification", "ComfyUI Desktop App" ] }, { "id": "comfy-ui-manager", "title": "comfyui manager", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ComfyUI Manager is a community-developed extension for the ComfyUI node-based image and video generation interface that provides an integrated package management system for discovering, installing, updating, and disabling custom nodes and their Python dependencies. It maintains a curated registry of available custom node repositories and model assets, resolves dependency conflicts, and enables reproducible workflow sharing by exporting workflow snapshots that encode all required node specifications. The extension also integrates model management features for downloading checkpoint, LoRA, VAE, and ControlNet files from external repositories, and performs missing-node detection when importing workflows created on other machines. As the de facto package manager for the ComfyUI ecosystem, it substantially lowers the barrier to extending the platform with community-developed preprocessing, sampling, and post-processing nodes.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:node-based-diffusion-pipeline-interface-manager", "labels": [ "ComfyUI Manager" ], "is_subclass_of": [ "AI Application", "Package Manager" ], "wikilinks": [ "ComfyUI", "ComfyUI API Specification", "ComfyUI Workflows", "ComfyUI Client", "Generative AI", "Stable Diffusion", "Diffusion Model", "Python", "Git", "Hugging Face", "Hugging Face Hub", "CivitAI", "ControlNet", "LoRA", "IP-Adapter", "Image Generation", "Video Generation", "AnimateDiff", "Flux.1", "SDXL" ] }, { "id": "comfy-ui-workflows", "title": "ComfyUI Workflows", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ComfyUI Workflows are node-based directed acyclic graph (DAG) pipelines for Stable Diffusion and broader generative AI inference, implemented within the ComfyUI open-source graphical interface developed by comfyanonymous (first commit January 2023), in which discrete processing operations...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:node-based-diffusion-pipeline-interface-workflows", "labels": [ "ComfyUI Workflows" ], "is_subclass_of": [ "Generative Model", "Node-Based Diffusion Pipeline Interface", "Generative AI Pipeline", "Node-Based Visual Programming", "Workflow Execution Engine", "Inference Orchestration System", "Diffusion Model Interface" ], "wikilinks": [ "AI", "API Server Protocol", "AUTOMATIC1111 WebUI", "Batch Image Processing", "BentoML 2025 comfy-pack", "Black Forest Labs 2024 Flux.1", "CheckpointLoader", "Civitai Workflow Sharing", "CLIP", "CLIPTextEncode", "Comfy-Org 2024 Workflow JSON Spec", "Comfy-Org 2025 Workflow Templates", "Comfy-Org GitHub Organisation", "ComfyUI API Specification", "ComfyUI Manager", "comfyanonymous 2023 ComfyUI", "ComputerVisionDomain", "ControlNet", "ControlNet Conditioning", "ControlNet Node" ] }, { "id": "command-and-data-handling", "title": "Command and Data Handling", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:command-and-data-handling", "labels": [ "Command and Data Handling" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "commercial-bank", "title": "Commercial Bank", "domain": "finance", "domain_name": "Finance", "definition": "A commercial bank is a deposit-taking financial institution that accepts deposits from the public, extends loans and provides payment and related financial services. Through lending against a fraction of its deposits it participates in money creation and the transmission of monetary policy. Commercial banks are central to retail and corporate finance and operate under prudential regulation and capital requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:commercial-bank", "labels": [ "Commercial Bank" ], "is_subclass_of": [ "Monetary System" ], "wikilinks": [] }, { "id": "commit-reveal-scheme", "title": "Commit-Reveal Scheme", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A commit-reveal scheme is a two-phase cryptographic protocol in which a participant first publishes a binding, hiding commitment to a value (typically a hash of the value plus a nonce) and later reveals the value for verification. The commit phase prevents others from learning or altering the choice, while the reveal phase lets anyone check the value against the earlier commitment. It matters for on-chain voting and auctions because it stops front-running and last-mover advantage by concealing inputs until all parties are bound.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:commit-reveal-scheme", "labels": [ "Commit-Reveal Scheme", "Commit-Reveal Schemes" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "commit", "title": "Commit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The atomic unit of change in a version control system: an immutable, uniquely identified snapshot of a project's tracked content together with metadata \u2014 author, timestamp, descriptive message, and references to one or more parent commits \u2014 so that the full set of commits forms a directed acyclic graph recording the project's history; in Git each commit is content-addressed by a cryptographic hash of its tree and parents, making history tamper-evident and enabling branching, merging, reverting, and precise attribution of every line of a codebase.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:commit", "labels": [ "Commit" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [ "Version Control", "Git", "Cryptographic Hash", "Directed Acyclic Graph" ] }, { "id": "commitment-scheme", "title": "Commitment Scheme", "domain": "security", "domain_name": "Security", "definition": "A commitment scheme is a two-phase cryptographic primitive that allows a party (the committer) to bind itself to a chosen value by producing a short commitment string \u2014 analogous to sealing a value in an envelope \u2014 that is later opened by revealing the original value and a randomness parameter, with the scheme satisfying two security properties: binding (the committer cannot change the value after committing) and hiding (the commitment reveals nothing about the value before opening). Commitment schemes are foundational building blocks for zero-knowledge proofs, secure multi-party computation, and blockchain protocols.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:commitment-scheme", "labels": [ "Commitment Scheme", "Cryptographic Commitment Scheme", "MS-SMT Commitment Scheme", "Note Commitment Scheme", "Polynomial Commitment Scheme" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "commodity-money", "title": "Commodity Money", "domain": "finance", "domain_name": "Finance", "definition": "Commodity money is a form of money whose value derives from the intrinsic worth of the physical commodity from which it is made, such as gold, silver, salt or grain. Because the medium of exchange is itself a useful or scarce good, its monetary value is anchored to its commodity value rather than to government decree. Commodity money predates and contrasts with fiat currency, where value rests on trust in an issuing authority.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:commodity-money", "labels": [ "Commodity Money" ], "is_subclass_of": [ "Money" ], "wikilinks": [] }, { "id": "common-crawl", "title": "Common Crawl", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Common Crawl is a California 501(c)3 non-profit organisation founded in 2008 by Gil Elbaz that maintains a petabyte-scale, freely accessible archive of web crawl data stored on Amazon S3 under the AWS Open Data Sponsorship Programme. The dataset \u2014 distributed as WARC, WAT, and WET files \u2014 underpins virtually every major open pre-training corpus including C4, FineWeb, RedPajama, Dolma, and DCLM, and has been cited by the Mozilla Foundation (2024) as essential to the emergence of modern generative AI.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:common-crawl", "labels": [ "Common Crawl" ], "is_subclass_of": [ "Training Data", "Open Data", "Web Archive" ], "wikilinks": [ "Model Training", "GPT", "Machine Learning", "Synthetic Data", "Large Language Models", "Natural Language Processing", "Training Data", "Data Curation", "Pre-training", "Foundation Models", "Web Scraping", "Deduplication", "Tokenisation", "Information Retrieval", "Data Pipeline", "Deep Learning", "BERT", "Transfer Learning", "Bias in Large Language Models", "Data Governance" ] }, { "id": "common-pool-resources", "title": "Common Pool Resources", "domain": "governance", "domain_name": "Governance", "definition": "Natural or human-made resources that are non-excludable to a defined community but rival in consumption, meaning one actor's use diminishes availability for others. Classic examples include fisheries, groundwater, pastures, and shared spectrum; digital analogues include shared compute pools and open training datasets. Sustainable governance requires institutions that constrain extraction without full privatisation or state control.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:common-pool-resources", "labels": [ "Common Pool Resources", "Common-Pool Resource Governance", "Common-Pool Resource Management" ], "is_subclass_of": [ "Resource Management" ], "wikilinks": [] }, { "id": "common-sense-reasoning", "title": "Common Sense Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "The capacity of an AI system to draw inferences that humans consider obvious from general background knowledge about the physical world, social norms, causality, and everyday object behaviour \u2014 without explicit instruction. It encompasses naive physics, naive psychology, temporal reasoning, spatial reasoning, and social cognition grounded in embodied human experience, and is tested against benchmarks including CommonsenseQA, HellaSwag, WinoGrande, and PIQA.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:common-sense-reasoning", "labels": [ "Common Sense Reasoning" ], "is_subclass_of": [ "Natural Language Understanding", "Reasoning", "Automated Reasoning" ], "wikilinks": [ "Natural Language Understanding", "Natural Language Processing", "AI Agents", "AI Alignment", "Collective Intelligence", "Large Language Models", "Knowledge Graph", "Knowledge Representation", "Symbolic AI", "Neurosymbolic AI", "Reasoning", "Embodied AI", "Reinforcement Learning", "Deep Learning", "Question Answering", "Causal Reasoning", "Temporal Reasoning", "Spatial Reasoning", "RLHF", "Hallucination" ] }, { "id": "commons-governance", "title": "Commons Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Commons governance is the set of institutions, rules and collective practices through which a community manages a shared resource so that it remains productive and is not depleted by individual self-interest. Drawing on Elinor Ostrom's design principles, it relies on clearly defined boundaries, participatory rule-making, monitoring and graduated sanctions rather than pure markets or central control. In blockchain contexts it is expressed through DAOs and public-goods funding that coordinate stewardship of shared digital infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:commons-governance", "labels": [ "Commons Governance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "commons-stack", "title": "Commons Stack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Commons Stack is an initiative providing tools and economic models for funding and governing public goods through token-based communities. It draws on mechanisms such as bonding curves and continuous funding.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:commons-stack", "labels": [ "Commons Stack", "Commons Management" ], "is_subclass_of": [ "Public Goods Funding" ], "wikilinks": [ "Tokenomics", "Quadratic Funding", "Decentralized Autonomous Organization", "Public Goods Funding" ] }, { "id": "commons-based-peer-production", "title": "Commons-Based Peer Production", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Commons-based peer production is a model of decentralised, collaborative creation in which large numbers of individuals contribute effort to a shared resource without centralised managerial coordination or market-priced compensation, coined by legal scholar Yochai Benkler. Open-source software development is its archetypal example, with contributions coordinated through modular tasks, transparent version control, and reputation rather than employment contracts. The model has since been extended to open encyclopaedias, open data, and blockchain-funded public goods.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:commons-based-peer-production", "labels": [ "Commons-Based Peer Production" ], "is_subclass_of": [ "Digital Commons" ], "wikilinks": [] }, { "id": "commonsense-reasoning", "title": "Commonsense Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Commonsense reasoning is the ability of an artificial system to make plausible inferences about everyday situations using broad background knowledge that humans take for granted. It covers naive physics, folk psychology, temporal causality, social norms, and typical cause-and-effect expectations that are rarely stated explicitly in text. It remains a long-standing challenge because such knowledge is vast, tacit, and context-dependent, requiring both large-scale knowledge resources and flexible defeasible inference mechanisms.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:commonsense-reasoning", "labels": [ "Commonsense Reasoning" ], "is_subclass_of": [ "Reasoning", "Automated Reasoning" ], "wikilinks": [ "Reasoning", "Symbolic Reasoning", "Knowledge Base", "World Model", "Large Language Model", "Inference", "Knowledge Representation", "Cognitive Architecture", "Automated Reasoning", "Natural Language Understanding", "Natural Language Processing", "Knowledge Graph", "ConceptNet", "Neural-Symbolic Integration", "Question Answering", "Transfer Learning", "Retrieval-Augmented Generation", "Hallucination", "AI Safety", "Planning" ] }, { "id": "communication-channel", "title": "Communication Channel", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A communication channel is the medium or pathway, physical or logical, through which information is transmitted from a sender to a receiver, characterised by properties such as bandwidth, latency, noise and capacity. The concept originates in information theory, where channel capacity bounds the rate at which information can be transmitted reliably. It is a prerequisite for any feedback loop or social interaction that depends on exchanging signals between agents.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:communication-channel", "labels": [ "Communication Channel" ], "is_subclass_of": [ "Information Theory" ], "wikilinks": [] }, { "id": "communication-infrastructure", "title": "Communication Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical and logical substrate \u2014 including fibre, wireless networks, data centres, protocols, and switching equipment \u2014 that enables the transmission of information between nodes in a networked system. It underpins digital services by providing reliable, low-latency, and high-bandwidth connectivity at scale, from local area networks to global internet backbones.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:communication-infrastructure", "labels": [ "Communication Infrastructure", "Communication Systems", "Digital Communication Infrastructure" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "communication-interface", "title": "Communication Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A communication interface is a defined boundary \u2014 hardware, software, or logical \u2014 through which two or more distinct systems, components, or agents exchange data, commands, or signals according to agreed protocols and encodings. It abstracts the internal implementation details of each participant, exposing only the contract necessary for interoperability, and may operate synchronously or asynchronously across local buses, networks, or inter-process mechanisms. Communication interfaces are fundamental to modular system design, enabling independent development, testing, and replacement of components without disrupting the broader system.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:communication-interface", "labels": [ "Communication Interface", "Classical Communication Interface" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "communication-layer", "title": "Communication Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Communication Layer is the stratum that governs the exchange of messages between components or participants. It sits above the Transport Layer that delivers bytes and below the coordination and application strata that rely on conversation. It contains messaging patterns, encoding, addressing, and delivery semantics.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:communication-layer", "labels": [ "Communication Layer" ], "is_subclass_of": [ "Network and Communication", "owl:Thing" ], "wikilinks": [ "Transport Layer", "Coordination Layer", "Integration Layer", "Publish-Subscribe", "Message Passing", "owl:Thing" ] }, { "id": "communication-network", "title": "Communication Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A communication network is an interconnected collection of nodes and links that transports information between endpoints according to shared protocols. Networks are characterised by their topology, switching method, transmission media, and the protocol stack that governs addressing, routing, and error control. They span scales from local wireless links to global packet-switched internetworks, and they provide the substrate on which distributed systems, decentralised ledgers, and immersive media depend. Performance is described by metrics including bandwidth, latency, jitter, and reliability.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:communication-network", "labels": [ "Communication Network", "Quantum Communication Network" ], "is_subclass_of": [ "Communication Infrastructure" ], "wikilinks": [] }, { "id": "communication-protocol", "title": "Communication Protocol", "domain": "robotics", "domain_name": "Robotics", "definition": "Communication protocol defines message formats, transmission rules, addressing schemes, and error-handling procedures that enable robots and computational systems to reliably exchange information over wired or wireless channels.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:communication-protocol", "labels": [ "Communication Protocol" ], "is_subclass_of": [ "Human-Robot Interaction", "Communication Systems", "Information Architecture" ], "wikilinks": [ "5G", "Addressing Scheme", "Bandwidth Constraints", "Battery-Powered Robots", "CAN Bus", "Communication Systems", "Congestion Control", "Cyber-Physical Security", "Data Serialisation", "DDS", "Distributed Control", "Error Correction", "Ethernet", "Inter-Robot Communication", "Latency Tolerance", "Message Format", "Network Physical Layer", "OROCOS", "PROFINET", "Remote Teleoperation" ] }, { "id": "communication-protocols", "title": "Communication Protocols", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Communication protocols are formal sets of rules, conventions, and standards that specify how data is formatted, transmitted, received, and acknowledged between two or more communicating entities, including computers, devices, and software systems. They define the syntax and semantics of messages, error detection and correction mechanisms, flow control, session management, and the sequencing of exchanges required to achieve reliable information transfer. Protocols operate across layered architectural models such as the OSI Reference Model and the TCP/IP suite, where each layer provides well-defined services to the layer above it while abstracting the implementation details below. Together, these layered agreements enable heterogeneous systems from different vendors and organisations to interoperate reliably across diverse network topologies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:communication-protocols", "labels": [ "Communication Protocols", "Industrial Communication Protocols", "TC-0040-Communication-Protocols" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Interoperability", "HTTP", "Application Layer", "Network Protocol" ] }, { "id": "communication-software", "title": "Communication Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software applications and platforms that enable real-time interaction, collaboration, and social connection within virtual environments and metaverse spaces, including immersive video conferencing, spatial audio, avatar-based communication, and AI-enhanced translation and transcription services t...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:communication-software", "labels": [ "Communication Software" ], "is_subclass_of": [ "Platform and Environment", "Collaboration Technology" ], "wikilinks": [ "Audio Visual Systems", "Real-Time Collaboration", "Virtual Meetings", "Collaboration Technology", "metaverse", "Network Infrastructure", "Remote Communication", "Telecollaboration", "User Interface" ] }, { "id": "communication-theory", "title": "Communication Theory", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Communication Theory provides the theoretical foundations for understanding how information, meaning, and social presence are transmitted and perceived in mediated communication contexts. Key frameworks include media richness theory, social presence theory, and theories of computer-mediated communication that directly inform the design of telecollaboration systems and collaborative technology.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:communication-theory", "labels": [ "Communication Theory", "CommunicationTheory" ], "is_subclass_of": [ "Communication Technology", "Telecollaboration" ], "wikilinks": [ "Telecollaboration" ] }, { "id": "communication-tools", "title": "Communication Tools", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software applications and platforms that enable individuals and teams to exchange information, coordinate activities, and maintain shared context across synchronous and asynchronous channels. They span messaging, video conferencing, email, collaborative documents, and emerging immersive communication modalities, forming the operational backbone of distributed work.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:communication-tools", "labels": [ "Communication Tools" ], "is_subclass_of": [ "Collaborative Technology" ], "wikilinks": [] }, { "id": "community-coordination", "title": "Community Coordination", "domain": "governance", "domain_name": "Governance", "definition": "Community coordination is the process by which a distributed group of participants aligns on shared goals, decisions, and resource allocation without centralised command, typically using deliberation, voting, or reputation mechanisms. It underpins governance of open-source projects, decentralised autonomous organisations, and platform communities, where outcomes must reflect dispersed stakeholder input. Effective community coordination balances inclusivity of participation against the decision latency inherent in consensus-seeking.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:community-coordination", "labels": [ "Community Coordination", "CommunityCoordination" ], "is_subclass_of": [ "Collective Action" ], "wikilinks": [] }, { "id": "community-decision-making", "title": "Community Decision Making", "domain": "governance", "domain_name": "Governance", "definition": "The processes by which a defined community of stakeholders collectively identifies issues, deliberates alternatives, and reaches binding or advisory decisions about matters affecting shared resources, norms, or direction. It encompasses participatory design, public consultation, deliberative democracy mechanisms, and digital governance platforms that translate community preferences into actionable outcomes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:community-decision-making", "labels": [ "Community Decision Making" ], "is_subclass_of": [ "Collective Decision Making" ], "wikilinks": [] }, { "id": "community-detection", "title": "Community Detection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Community detection is the computational task of identifying cohesive subgroups, or communities, within a network graph, where nodes within each group are more densely interconnected than they are with nodes in other groups. Algorithms such as the Louvain and Leiden methods optimise a modularity objective to partition the graph, whilst spectral clustering, label propagation, and stochastic block models offer alternative formulations. The problem is formally NP-hard in its general form, making approximation and heuristic approaches the practical norm. Applications span social-network analysis, bioinformatics, knowledge-graph organisation, recommendation systems, and cybersecurity anomaly detection.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:community-detection", "labels": [ "Community Detection" ], "is_subclass_of": [ "Network Analysis", "Clustering", "Unsupervised Learning", "Graph Analysis" ], "wikilinks": [ "Graph Theory", "Data Aggregation", "Network Analysis", "Louvain Algorithm", "Leiden Algorithm", "Spectral Clustering", "Label Propagation", "Stochastic Block Model", "Graph Neural Networks", "Matrix Factorisation", "Random Walks", "Recommendation Systems", "Anomaly Detection", "Knowledge Graph Embedding", "Social Network Analysis", "Link Prediction", "Clustering", "Network Topology", "Distributed Systems", "Blockchain Governance" ] }, { "id": "community-governance-model", "title": "Community Governance Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A participatory decision-making framework that defines rules, voting mechanisms, proposal systems, and dispute resolution processes for virtual communities, enabling democratic and transparent collective governance.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:community-governance-model", "labels": [ "Community Governance Model" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance Framework", "Collective Governance", "Commons Governance", "Institutional Framework", "Participatory Democracy" ], "wikilinks": [ "Collective Action", "Community Decision Making", "DAO Governance Standards", "Decision Rules", "Dispute Resolution Process", "ISO 37001 Anti-Bribery Management", "Membership Criteria", "Proposal Mechanism", "Voting System", "W3C Decentralized Governance", "Blockchain", "Blockchain Infrastructure", "Consensus Mechanism", "Decentralized Autonomous Organization", "Democratic Participation", "Governance Token", "Identity Management", "MiddlewareLayer", "Reputation System", "Smart Contract" ] }, { "id": "community-governance", "title": "Community Governance", "domain": "governance", "domain_name": "Governance", "definition": "Community Governance refers to the structures, processes, and norms through which a community of stakeholders collectively makes decisions, allocates resources, resolves disputes, and sets rules for shared systems or commons. It encompasses formal mechanisms such as voting protocols, proposal systems, and constitutional rules, as well as informal norms of participation, legitimacy, and accountability. Community governance may be implemented on-chain through smart contracts and token-weighted voting in decentralised autonomous organisations, or off-chain through forum deliberation, elected councils, and working groups in open-source projects, platform cooperatives, and public institutions. The field draws on political science, institutional economics, commons theory, and distributed systems to design governance models that are simultaneously legitimate, efficient, and resistant to capture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:community-governance", "labels": [ "Community Governance" ], "is_subclass_of": [ "Decentralised Governance" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "community-standards", "title": "Community Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The rules, guidelines, and behavioral norms established for virtual environments and metaverse platforms that govern user conduct, content creation, and social interactions, often enforced through technical standards, governance frameworks, and moderation systems to ensure safety, inclusivity, an...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:community-standards", "labels": [ "Community Standards" ], "is_subclass_of": [ "Governance and Safety", "Platform Governance" ], "wikilinks": [ "GDPR (General Data Protection Regulation)", "IEEE (Institute of Electrical and Electronics Engineers)", "Moderation Tools", "Reporting Mechanisms", "Trust Building", "Content Moderation", "EU AI Act", "Governance Framework", "metaverse", "Platform Governance", "Telecollaboration", "User Safety" ] }, { "id": "comparative-planetology", "title": "Comparative Planetology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:comparative-planetology", "labels": [ "Comparative Planetology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "compatibility-process", "title": "Compatibility Process", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic procedure for ensuring that digital assets, applications, and systems conform to common standards, protocols, and specifications to enable seamless exchange, integration, and interoperability across metaverse platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:compatibility-process", "labels": [ "Compatibility Process" ], "is_subclass_of": [ "Standards and Interoperability", "Compatibility Standards" ], "wikilinks": [ "API Specifications", "Asset Portability", "Conformance Criteria", "Cross-Platform Testing", "Data Format Schemas", "Ecosystem Connectivity", "Format Validation", "Integration Testing", "Interoperability Verification", "ISO/IEC 30170", "MSF Taxonomy 2025", "Protocol Compatibility Checks", "Protocol Definitions", "Reference Implementations", "Sensor Input", "Test Specifications", "Validation Tools", "Compatibility Standards", "Cross-Platform Interoperability", "Data Layer" ] }, { "id": "compatibility-standards", "title": "Compatibility Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical specifications and protocols that enable interoperability between different metaverse platforms, virtual environments, and digital systems, including asset formats, communication protocols, and interface standards that allow seamless user experiences across multiple platforms without ve...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:compatibility-standards", "labels": [ "Compatibility Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standards" ], "wikilinks": [ "Asset Portability", "Cross-Platform Experience", "IEEE (Institute of Electrical and Electronics Engineers)", "Industry Collaboration", "Protocol Development", "Sensor Input", "Standards Bodies", "Interoperability", "metaverse", "Technical Standards" ] }, { "id": "competency-based-education", "title": "Competency Based Education", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Competency-Based Education is an instructional model that grants progression and credentials based on demonstrated mastery of defined competencies rather than time spent in instruction. Learners advance when they can prove they have met explicit, measurable outcomes, often supported by personalised pacing and frequent assessment. The approach relies on learning analytics and educational technology to track and verify competency attainment.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:competency-based-education", "labels": [ "Competency Based Education", "Competency-Based Education" ], "is_subclass_of": [ "Education Technology", "Mastery Learning", "Personalised Learning", "Self-Paced Learning" ], "wikilinks": [ "Education Technology", "Adaptive Learning", "Learning Management System", "Learning Analytics", "Mastery Learning", "Formative Assessment", "Summative Assessment", "Personalised Learning", "Intelligent Tutoring System", "Knowledge Graph", "Micro-Credential", "Digital Badge", "Open Badges", "Credential Framework", "Prior Learning Assessment", "Rubric-Based Assessment", "Bloom's Taxonomy", "Self-Paced Learning", "Curriculum Design", "Competency Framework" ] }, { "id": "competition-in-ai", "title": "Competition in AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Competition in AI is the multi-dimensional race dynamics \u2014 geopolitical, corporate, research, and regulatory \u2014 through which nation-states, hyperscaler enterprises, independent laboratories, open-source collectives, and frontier-model providers compete for technological leadership in artificial i...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:competition-in-ai", "labels": [ "Competition in AI", "Competition" ], "is_subclass_of": [ "AI Governance and Ethics", "Geopolitics", "Technology Race", "Industrial Competition", "AI Governance", "Innovation Economics" ], "wikilinks": [ "Agentic AI Systems", "AI Chips", "AI Cooperation", "AI Investment", "AI Policy", "AI Regulation", "AI Research Talent", "AI Sovereignty", "Anthropic", "Antitrust Oversight", "Benchmark Evaluation", "Capital Markets", "Compute Access", "Compute Governance", "Economic Competitiveness", "Export Controls", "Frontier Models", "Geopolitics", "GeopoliticsDomain", "GPU Supply Chain" ] }, { "id": "compiler-optimization", "title": "Compiler Optimization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Compiler optimisation is the set of program transformations a compiler applies to make generated code faster, smaller or more energy-efficient while preserving its observable behaviour. It operates over intermediate representations using analyses such as data-flow and dependence analysis to enable transformations like inlining, loop optimisation and dead-code elimination. Optimisation is central to extracting performance from modern hardware without burdening the programmer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:compiler-optimization", "labels": [ "Compiler Optimization" ], "is_subclass_of": [ "Compiler" ], "wikilinks": [] }, { "id": "compiler", "title": "Compiler", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A software system that translates source code written in a high-level programming language into a lower-level representation \u2014 typically machine code, bytecode, or an intermediate representation \u2014 performing lexical analysis, syntactic parsing, semantic checking, optimisation, and code emission in a structured pipeline. Compilers are foundational to software development, enabling human-readable programs to run efficiently on hardware.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:compiler", "labels": [ "Compiler", "Just-In-Time Compiler", "LLVM Compiler Infrastructure", "Solidity Compiler", "Triton Compiler", "XLA Compiler" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "complaint-infrastructure", "title": "Complaint Infrastructure", "domain": "governance", "domain_name": "Governance", "definition": "Complaint infrastructure is the organisational and technical system through which consumers can submit, track, escalate, and resolve grievances against providers, and through which regulators monitor systemic harms. It comprises intake channels, case-management workflows, redress mechanisms, and reporting that feed enforcement and policy. It matters because accessible, auditable complaint handling is a structural requirement of consumer-protection regimes, turning individual grievances into actionable signals and remedies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:complaint-infrastructure", "labels": [ "Complaint Infrastructure" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "complementary-filter", "title": "Complementary Filter", "domain": "robotics", "domain_name": "Robotics", "definition": "A complementary filter is a lightweight sensor fusion technique that combines two signals with complementary error characteristics, typically a low-noise but slow-drifting measurement such as an accelerometer and a fast but drift-prone measurement such as a gyroscope, into a single accurate estimate. It applies a low-pass filter to one signal and a high-pass filter to the other before summing them, avoiding the computational cost of a full Kalman filter. Complementary filters are widely used in inertial measurement units for real-time orientation estimation in robotics and spatial-computing applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:complementary-filter", "labels": [ "Complementary Filter" ], "is_subclass_of": [ "Sensor Fusion" ], "wikilinks": [] }, { "id": "complex-adaptive-systems", "title": "Complex Adaptive Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Complex adaptive systems are systems composed of many interacting agents whose collective behaviour emerges from local interactions and adaptation rather than central control. The agents adjust their behaviour in response to one another and to their environment, producing non-linear dynamics, self-organisation and emergent order. Studied across biology, economics and artificial intelligence, they provide a lens for understanding resilience, learning and unpredictability in distributed populations.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:complex-adaptive-systems", "labels": [ "Complex Adaptive Systems" ], "is_subclass_of": [ "Complexity Science", "Systems Theory", "Dynamical Systems Theory" ], "wikilinks": [ "Complexity Science", "Emergence", "Systems Theory", "Cybernetics", "Chaos Theory", "Feedback Loop", "Self-Organisation", "Non-Linear Dynamics", "Agent-Based Modelling", "Agent-Based Models", "Swarm Intelligence", "Multi-Agent Systems", "Control Theory", "Network Science", "Network Theory", "Phase Transition", "Self-Organised Criticality", "Dynamical Systems Theory", "Statistical Mechanics", "Information Theory" ] }, { "id": "complex-event-processing", "title": "Complex Event Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Complex event processing is a paradigm that continuously analyses streams of discrete events to detect meaningful patterns, correlations and derived higher-level events in near real time. Rather than querying stored data after the fact, it evaluates standing pattern queries over moving event streams, recognising temporal sequences, aggregates over windows and absence-of-event conditions. It enables systems to react to situations as they emerge rather than discovering them in later batch analysis.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:complex-event-processing", "labels": [ "Complex Event Processing" ], "is_subclass_of": [ "Stream Processing" ], "wikilinks": [] }, { "id": "complex-systems", "title": "Complex Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systems composed of many interacting components whose collective behaviour cannot be simply inferred from the behaviour of the individual parts; characterised by nonlinearity, emergence, self-organisation, and feedback dynamics across scales.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:complex-systems", "labels": [ "Complex Systems" ], "is_subclass_of": [ "Systems Theory", "owl:Thing" ], "wikilinks": [ "Dynamical Systems Theory", "Emergence", "Simulation", "Feedback Loop", "owl:Thing" ] }, { "id": "complexity-science", "title": "Complexity Science", "domain": "data", "domain_name": "Data", "definition": "Complexity science is the interdisciplinary study of complex adaptive systems composed of many interacting components whose collective behaviour cannot be reduced to that of the individual parts. It investigates emergence, self-organisation, networks, nonlinear dynamics, and adaptation across physical, biological, social, and economic systems. It matters because it provides a unifying lens and a set of computational and analytical tools for understanding phenomena such as markets, ecosystems, and epidemics that defy reductionist explanation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:complexity-science", "labels": [ "Complexity Science" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "compliance-audit-trail", "title": "Compliance Audit Trail", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Immutable record system demonstrating adherence to policies and regulations through cryptographically sealed logs of compliance verification activities and evidence.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:compliance-audit-trail", "labels": [ "Compliance Audit Trail" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance", "Data Governance", "Accountability" ], "wikilinks": [ "Audit Automation", "Compliance Event Log", "Compliance Management System", "Immutable Storage", "ISO 37301", "Policy Document", "Regulatory Evidence", "Verification Record", "Access Control", "Accountability", "Blockchain", "Compliance Verification", "Cryptographic Hash", "DataLayer", "Data Provenance", "MiddlewareLayer", "Policy Engine", "Regulatory Framework", "Regulatory Reporting", "Risk Assessment" ] }, { "id": "compliance-automation", "title": "Compliance Automation", "domain": "governance", "domain_name": "Governance", "definition": "Compliance automation is the use of software systems to continuously monitor, enforce, and evidence adherence to regulatory and policy requirements, replacing manual audit-driven checks with machine-executable controls. It typically encodes rules as code, evaluates them against live system state or transaction streams, and generates auditable records without human intervention. It reduces the cost and latency of regulatory adherence while improving consistency across large organisations.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:compliance-automation", "labels": [ "Compliance Automation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "compliance-carbon-market", "title": "Compliance Carbon Market", "domain": "governance", "domain_name": "Governance", "definition": "A compliance carbon market is a regulated market in which emitters covered by a mandatory cap must surrender allowances or credits equal to their emissions, created under law to enforce climate targets. Prices are set by the supply of capped allowances and demand from regulated entities, creating a financial incentive to abate. It stands in contrast to voluntary carbon markets, where participation and credit purchase are discretionary.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:compliance-carbon-market", "labels": [ "Compliance Carbon Market" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "compliance-control", "title": "Compliance Control", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Compliance Control is an enterprise governance mechanism comprising the policies, procedures, automated tests, and continuous monitoring activities that an organisation deploys to demonstrate, document, and enforce adherence to regulatory obligations, internal standards, and contractual requireme...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:compliance-control", "labels": [ "Compliance Control" ], "is_subclass_of": [ "Legal and Regulatory", "Security and Identity", "Risk Management", "Corporate Governance", "Internal Control", "Regulatory Compliance", "Information Security Management", "Enterprise Risk Management" ], "wikilinks": [ "AssuranceLayer", "Audit", "Audit Management System", "Automated Control", "Basel Committee on Banking Supervision", "Basel III Compliance", "Business Process Management", "Compensating Control", "Conduct Risk Management", "Consumer Duty", "Consumer Duty Compliance", "Continuous Controls Monitoring", "Control Activity", "Control Documentation", "Control Evidence", "Control Library", "Control Maturity Assessment", "Control Objective", "Control Owner", "Control Testing" ] }, { "id": "compliance-dashboard", "title": "Compliance Dashboard", "domain": "governance", "domain_name": "Governance", "definition": "A visual interface and monitoring system that provides real-time visibility into regulatory compliance status, risk metrics, and audit trails across blockchain networks and digital platforms, enabling organizations to track adherence to legal requirements, detect anomalies, and demonstrate compliance to regulators through transparent reporting.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:compliance-dashboard", "labels": [ "Compliance Dashboard" ], "is_subclass_of": [ "AI Governance and Ethics", "Regulatory Technology" ], "wikilinks": [ "Analytics Engine", "Real-Time Monitoring", "Risk Visualization", "Blockchain", "Blockchain Network", "Data Integration", "metaverse", "Regulatory Reporting", "Regulatory Technology" ] }, { "id": "compliance-evidence", "title": "Compliance Evidence", "domain": "governance", "domain_name": "Governance", "definition": "Compliance evidence is the documented, verifiable record that an organisation or process meets a regulatory, contractual, or standards-based requirement. It includes artefacts such as audit logs, sensor readings, certificates, attestations, and chain-of-custody records, ideally tamper-evident and timestamped. It matters because it converts asserted compliance into auditable proof, supporting attestations to regulators and customers and enabling automated verification in areas such as cold-chain integrity and supply-chain assurance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compliance-evidence", "labels": [ "Compliance Evidence" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "compliance-framework", "title": "Compliance Framework", "domain": "governance", "domain_name": "Governance", "definition": "A Compliance Framework is a structured set of policies, controls, procedures, and audit mechanisms that organisations implement to satisfy regulatory obligations and industry standards governing the design, development, and deployment of systems \u2014 especially AI and data-intensive systems. It systematically maps external legal requirements (such as the EU AI Act, GDPR, ISO 42001, and NIST AI RMF) onto internal risk registers, technical documentation, accountability chains, and monitoring pipelines. A compliance framework defines the scope of applicability, assigns ownership of controls, establishes evidence-collection workflows, and mandates review cycles so that organisations can continuously demonstrate conformance to auditors, regulators, and stakeholders. It bridges governance intent with operational practice by providing a traceable path from regulatory obligation through implemented control to verified outcome.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compliance-framework", "labels": [ "Compliance Framework", "Compliance Frame", "Regulatory Compliance Framework" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "compliance-layer", "title": "Compliance Layer", "domain": "governance", "domain_name": "Governance", "definition": "The Compliance Layer is the cross-cutting stratum that checks system behaviour against external obligations such as law, regulation, and contractual terms. It sits above the Policy Layer, consuming its enforcement records, and reports to the Regulatory and Governance Layers. It contains controls, evidence collection, audit trails, and conformance assessments rather than the operations being assessed.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:compliance-layer", "labels": [ "Compliance Layer", "AML CFT Compliance Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Policy Layer", "Governance Layer", "Regulatory Layer", "Institutional Layer", "Audit", "Risk Management", "owl:Thing", "ISO (International Organization for Standardization)" ] }, { "id": "compliance-management", "title": "Compliance Management", "domain": "governance", "domain_name": "Governance", "definition": "Compliance Management is the systematic process by which organisations identify, assess, implement, and monitor adherence to applicable laws, regulations, standards, and internal policies. It encompasses the full lifecycle of obligation tracking, control design, evidence collection, and reporting to demonstrate that operational activities conform to required norms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:compliance-management", "labels": [ "Compliance Management", "Compliance Management System" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "compliance-monitoring", "title": "Compliance Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Compliance Monitoring is the continuous, automated oversight of systems, processes, people, and data flows to verify ongoing adherence to applicable regulatory requirements, internal policies, contractual obligations, and technical standards across the full operational surface area of an organisa...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:compliance-monitoring", "labels": [ "Compliance Monitoring", "BC-0487-compliance-monitoring" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Technology", "Governance Risk Compliance", "Risk Management", "Continuous Control Monitoring", "Automated Compliance" ], "wikilinks": [ "Alert Management", "AMLD6 Directive 2024", "Anomaly Detection", "Anti Money Laundering", "Archer GRC", "Arthur AI", "Automated Compliance", "Basel Committee on Banking Supervision", "Basel Committee on Banking Supervision Principles 2023", "Bayesian Risk Models", "BigID", "Blockchain Analytics", "Business Process Management", "Case Management System", "Checkbox Compliance", "Collibra", "Comp AI", "Continuous Assurance", "Continuous Control Monitoring", "Credo AI" ] }, { "id": "compliance-standards", "title": "Compliance Standards", "domain": "governance", "domain_name": "Governance", "definition": "The regulatory requirements, technical specifications, and legal frameworks that govern operations within virtual environments and metaverse platforms, encompassing data privacy, intellectual property, consumer protection, and conduct standards that organizations must adhere to when operating in or developing for immersive digital spaces.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:compliance-standards", "labels": [ "Compliance Standards" ], "is_subclass_of": [ "AI Governance and Ethics", "Regulatory Framework" ], "wikilinks": [ "AI Act", "Consumer Safety", "DMA", "DSA", "GDPR", "Legal Compliance", "Legal Expertise", "Monitoring Systems", "Regulatory Guidance", "W3C", "Blockchain", "Data Protection", "metaverse", "Regulatory Framework" ] }, { "id": "compliance-systems", "title": "Compliance Systems", "domain": "governance", "domain_name": "Governance", "definition": "Compliance systems are the integrated software and process frameworks organisations use to enforce, monitor, and report adherence to laws, regulations, and internal policies. They encompass policy management, control automation, monitoring and alerting, audit-trail capture, and regulatory reporting workflows. They matter because they operationalise obligations such as content-moderation duties under the Digital Services Act or custody and reporting rules for digital assets, turning legal requirements into continuously enforced controls.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compliance-systems", "labels": [ "Compliance Systems" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "compliance-testing", "title": "Compliance Testing", "domain": "governance", "domain_name": "Governance", "definition": "Compliance testing is the systematic verification that a product, system, or process conforms to a defined standard, specification, or regulatory requirement. It produces objective, repeatable measurements that are compared against documented acceptance criteria to determine pass or fail status. It is a core component of validation and certification workflows, providing the evidentiary basis for conformity claims.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compliance-testing", "labels": [ "Compliance Testing" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "compliance-verification", "title": "Compliance Verification", "domain": "governance", "domain_name": "Governance", "definition": "Compliance verification in blockchain contexts encompasses the automated and manual processes for ensuring that cryptoasset transactions, service providers, and participants adhere to regulatory requirements including Know Your Customer (KYC), Anti-Money Laundering (AML), sanctions screening, and transaction monitoring obligations. Blockchain-based compliance systems leverage immutable ledgers, smart contracts, and AI-driven analytics to maintain tamper-proof records and detect suspicious activities in real-time.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:compliance-verification", "labels": [ "Compliance Verification", "Compliance Verification System" ], "is_subclass_of": [ "Compliance Monitoring" ], "wikilinks": [ "KYC/AML Requirements", "Blockchain" ] }, { "id": "compliance", "title": "Compliance", "domain": "governance", "domain_name": "Governance", "definition": "Compliance is the organisational practice of adhering to laws, regulations, standards, and internal policies that govern the conduct of systems, processes, and personnel. It encompasses the identification of applicable obligations, the implementation of controls to satisfy them, continuous monitoring to verify ongoing conformance, and documented evidence to demonstrate that conformance to internal and external stakeholders. Modern compliance programmes span regulatory domains including data protection, financial services, cybersecurity, environmental obligations, and increasingly AI-specific requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:compliance", "labels": [ "Compliance", "AI Compliance", "Compliance Module", "Compliance Obligations", "Compliance Programme", "Compliance Statement", "Compliance as Code", "DeFi Compliance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "compliant-manipulation", "title": "Compliant Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Compliant Manipulation is a branch of robotic manipulation in which the robot's end-effector or structural members intentionally yield to contact forces, using passive mechanical compliance or active force control to safely interact with uncertain, fragile, or human-occupied environments. Unlike rigid position-controlled manipulation, compliant manipulation regulates interaction forces as a first-class control objective.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:compliant-manipulation", "labels": [ "Compliant Manipulation" ], "is_subclass_of": [ "Robot Manipulation" ], "wikilinks": [] }, { "id": "compliant-motion", "title": "Compliant Motion", "domain": "robotics", "domain_name": "Robotics", "definition": "Compliant motion is robot motion that yields to external contact forces rather than rigidly tracking a predefined trajectory, allowing a manipulator to accommodate unexpected contact, surface irregularities or misalignment. It is achieved through admittance or impedance control, which regulate the relationship between measured force and commanded motion. It is essential for tasks involving physical contact with uncertain environments, such as assembly, polishing and human-robot interaction.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:compliant-motion", "labels": [ "Compliant Motion" ], "is_subclass_of": [ "Force Control" ], "wikilinks": [] }, { "id": "component", "title": "Component", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A discrete, modular unit within an AI system or infrastructure that provides a specific capability, can be developed and tested independently, and interacts with other components through defined interfaces. Components include hardware accelerators, software modules, model artefacts, data pipelines, and monitoring subsystems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:component", "labels": [ "Component", "Software Component" ], "is_subclass_of": [ "AI Infrastructure", "Digital Infrastructure", "Software Architecture" ], "wikilinks": [ "Artificial Intelligence", "AI Infrastructure", "Hardware Component", "Software Module", "Neural Network Layer", "Neural Network", "Machine Learning Pipeline", "AI Framework", "MLOps", "Model Serving", "Data Pipeline", "GPU Compute", "Microservices", "Containerisation", "API", "Embedded Systems", "Deep Learning", "Model Registry", "Feature Engineering", "Software Architecture" ] }, { "id": "composability", "title": "Composability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Composability is the property by which independent components can be combined into larger systems whose behaviour is predictable from the behaviour of the parts. In blockchain and decentralised finance it describes how permissionless smart contracts can call and build upon one another so that protocols interlock like building blocks, often called money legos. Composability accelerates innovation but couples systems together, propagating both functionality and risk across the stack.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:composability", "labels": [ "Composability" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "composable-architecture", "title": "Composable Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Composable architecture is a software design approach in which independent, interchangeable components or services are assembled and reassembled to build applications, rather than being built as a single monolith. Components expose well-defined interfaces so they can be combined, replaced or extended without redesigning the whole system. It underpins API-integration strategies and modern system-integration practice, where new capabilities are added by composing existing building blocks.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:composable-architecture", "labels": [ "Composable Architecture" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "composite-ai", "title": "Composite AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Composite AI is an architectural paradigm that integrates multiple distinct AI techniques \u2014 such as machine learning, symbolic reasoning, natural language processing, computer vision, and knowledge graphs \u2014 into a unified hierarchical system to solve complex real-world problems that no single technique alone can address adequately. By orchestrating complementary AI capabilities, composite AI achieves higher contextual understanding, adaptability, and decision-making precision than monolithic approaches. It is increasingly the default pattern in enterprise intelligent automation, autonomous systems, and multimodal applications.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:composite-ai", "labels": [ "Composite AI" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "compound-coordinate-reference-system", "title": "Compound Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:compound-coordinate-reference-system", "labels": [ "Compound Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "compound-governor-bravo", "title": "Compound Governor Bravo", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Compound Governor Bravo is the second-generation on-chain governance smart contract deployed by the Compound protocol, superseding Governor Alpha with configurable parameters and a clean separation between governance voting logic and the Timelock executor contract. It enables COMP token holders to create, vote on, and enqueue governance proposals that modify protocol parameters \u2014 including interest rate models, collateral factors, reserve factors, and supported asset listings \u2014 without requiring contract redeployment to adjust governance thresholds. Governor Bravo introduced an abstraction layer allowing the proposal threshold, voting delay, voting period, and quorum to be updated through the same governance process they govern, substantially reducing upgrade friction. The contract has become a widely forked reference implementation across decentralised finance, with derivatives adopted by Uniswap, Indexed Finance, and many other protocols seeking battle-tested on-chain governance infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:compound-governor-bravo", "labels": [ "Compound Governor Bravo", "Compound Bravo", "Compound Bravo Governance", "Compound Governor", "Compound Governor Bravo Interface", "Compound Governor Bravo Reference Implementation", "Compound GovernorBravo", "Governor Bravo" ], "is_subclass_of": [ "On-chain Governance" ], "wikilinks": [] }, { "id": "compound", "title": "Compound", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Compound is an algorithmic, autonomous interest-rate protocol built on Ethereum that enables users to supply crypto assets to liquidity pools and earn continuously accruing interest, or borrow assets against over-collateralised positions at algorithmically determined rates. Interest rates adjust dynamically based on the utilisation ratio of each asset pool, eliminating the need for bilateral loan negotiation or centralised intermediaries. Supplied assets are represented as cTokens \u2014 ERC-20 tokens whose exchange rate appreciates with every block as interest accrues \u2014 which can themselves be used as collateral or freely traded. Governance of protocol parameters is conducted on-chain through the COMP token and the Compound Governor Bravo smart-contract system, making it one of the earliest fully decentralised autonomous lending protocols in the DeFi ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compound", "labels": [ "Compound", "Compound Finance", "Compound Protocol" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "compressed-sensing", "title": "Compressed Sensing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Compressed sensing is a signal processing technique that reconstructs a signal from far fewer samples than classical sampling theory requires, exploiting the fact that many real-world signals are sparse in some basis. Reconstruction is typically posed as a convex optimisation problem that recovers the sparsest signal consistent with the observed measurements. It has applications in medical imaging, wireless communications and any domain where sampling is costly or constrained.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:compressed-sensing", "labels": [ "Compressed Sensing" ], "is_subclass_of": [ "Sparse Representation" ], "wikilinks": [] }, { "id": "compression-function", "title": "Compression Function", "domain": "security", "domain_name": "Security", "definition": "A compression function is a fixed-input-length cryptographic primitive that maps two inputs (a chaining value and a message block) to a single shorter output. It is the core building block of iterated hash functions, where it is applied repeatedly under constructions such as Merkle-Damgard to process arbitrary-length messages. Its collision and preimage resistance directly determine the security of the hash function built on top of it.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:compression-function", "labels": [ "Compression Function" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "computability-theory", "title": "Computability Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computability theory, also known as recursion theory, is the branch of mathematical logic and theoretical computer science that studies which problems can be solved algorithmically in principle, independent of resource constraints. It defines models of computation such as the Turing machine, establishes the existence of undecidable problems like the halting problem, and characterises the limits of effective procedures via the Church-Turing thesis. It contrasts with complexity theory, which asks how efficiently solvable problems can be solved.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:computability-theory", "labels": [ "Computability Theory" ], "is_subclass_of": [ "Mathematical Logic", "Computational Complexity Theory", "Proof Theory", "Theoretical Computer Science" ], "wikilinks": [ "Computational Complexity Theory", "Mathematical Logic", "Set Theory", "Automata Theory", "Formal Language", "Algorithm", "Cryptography", "Artificial Intelligence", "Turing Machine", "Lambda Calculus", "Recursive Function", "Halting Problem", "Church-Turing Thesis", "Decidability", "Undecidability", "Oracle Computation", "Turing Degree", "Rice's Theorem", "Post Correspondence Problem", "Kolmogorov Complexity" ] }, { "id": "computation-graph", "title": "Computation Graph", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Computation Graph is a directed graph in which nodes represent operations or variables and edges represent the flow of data (typically tensors) between them. It is the central abstraction in modern machine learning frameworks, where a model's forward pass is expressed as a graph and gradients are computed by traversing it in reverse via automatic differentiation. Graphs may be built statically ahead of execution or dynamically as code runs, and they enable optimisation, scheduling and hardware acceleration.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:computation-graph", "labels": [ "Computation Graph" ], "is_subclass_of": [ "Directed Acyclic Graph" ], "wikilinks": [] }, { "id": "computational-biology", "title": "Computational Biology", "domain": "ai", "domain_name": "Ai", "definition": "Computational biology is the application of computational methods, mathematical modelling, statistical inference, and data analysis to understand biological systems and processes at every scale of organisation, from individual molecules and cells through tissues, organisms, populations, and ecosystems. It develops algorithms and models to interpret molecular, cellular, and organismal data, spanning sequence analysis, structural prediction, systems modelling, and simulation. The discipline increasingly relies on machine learning and deep learning to extract patterns from large and complex biological datasets, and is the parent domain of bioinformatics.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:computational-biology", "labels": [ "Computational Biology" ], "is_subclass_of": [ "Data Science", "Scientific Computing", "Artificial Intelligence", "Biology", "Bioinformatics" ], "wikilinks": [ "Bioinformatics", "Machine Learning", "Deep Learning", "Data Science", "Artificial Intelligence", "Scientific Computing", "Genomics", "Drug Discovery", "Protein Structure Prediction", "Precision Medicine", "AlphaFold", "Systems Biology", "Statistics", "Big Data", "GPU", "Neural Network", "Transformer Architecture", "Natural Language Processing", "High Performance Computing", "Sequence Alignment" ] }, { "id": "computational-complexity-theory", "title": "Computational Complexity Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The mathematical study of the resources \u2014 principally time and memory (space) \u2014 required to solve computational problems, and the classification of problems according to their inherent difficulty into complexity classes such as P, NP, PSPACE, BPP, and EXPTIME. It establishes which problems are tractable (solvable efficiently in polynomial time) and which are intractable, and investigates the relationships between complexity classes, most famously the unresolved P vs NP question.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:computational-complexity-theory", "labels": [ "Computational Complexity Theory", "Complexity Theory", "Computational Complexity" ], "is_subclass_of": [ "Theoretical Computer Science", "AI Research Area", "Mathematical Logic", "owl:Thing" ], "wikilinks": [ "Algorithm", "Set Theory", "Quantum Computing", "Information Theory", "owl:Thing", "Computability Theory", "Mathematical Logic", "Cryptography", "Formal Language", "Turing Machine", "Knowledge Representation", "Automated Reasoning", "Machine Learning", "Deep Learning", "SAT Solving", "Description Logic", "Optimisation", "Graph Algorithm", "Approximation Algorithms", "Randomised Algorithms" ] }, { "id": "computational-component", "title": "Computational Component", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Computational Component is a modular, well-defined unit of an AI or software system that encapsulates specific processing logic \u2014 such as a neural network layer, a feature extraction module, or an inference engine \u2014 and interacts with other components through defined interfaces. Computational components abstract implementation details from consumers, enabling composition into larger pipelines and substitution of equivalent implementations without modifying the surrounding system. They are the primary unit of reuse, testing, and deployment in AI infrastructure, ranging from low-level hardware accelerator kernels to high-level model serving endpoints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:computational-component", "labels": [ "Computational Component" ], "is_subclass_of": [ "AI Infrastructure", "Component" ], "wikilinks": [ "Artificial Intelligence", "Component" ] }, { "id": "computational-creativity", "title": "Computational Creativity", "domain": "ai", "domain_name": "Ai", "definition": "Computational creativity is the subfield of artificial intelligence concerned with building systems that exhibit behaviours regarded as creative, such as generating novel and valuable artefacts in art, music, design, or narrative. It studies both the algorithmic generation of original output and the evaluation of novelty, surprise, and value. The field informs generative models, procedural content systems, and AI-assisted creative tooling.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:computational-creativity", "labels": [ "Computational Creativity" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence", "Cognitive Science" ], "wikilinks": [ "Creative AI", "Generative AI", "Artificial Intelligence", "Deep Generative Model", "Diffusion Model", "Generative Adversarial Network", "Large Language Models", "Transformer", "Natural Language Processing", "Machine Learning", "Deep Learning", "Reinforcement Learning", "Cognitive Science", "Cognitive AI", "Procedural Content Generation", "Image Generation", "Music Generation", "Content Generation", "Human-AI Collaboration", "Intellectual Property" ] }, { "id": "computational-geometry", "title": "Computational Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computational geometry is the study of algorithms for solving geometric problems, such as finding convex hulls, intersections, and nearest points. It provides foundational techniques for computer graphics, robotics motion planning, geographic information systems, and computer-aided design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:computational-geometry", "labels": [ "Computational Geometry", "Constructive Solid Geometry", "NURBS Geometry" ], "is_subclass_of": [ "Algorithm" ], "wikilinks": [ "Data Structure", "3D Engine", "Pathfinding", "Computational Geometry", "Algorithm", "https://en.wikipedia.org/wiki/Computational_geometry", "https://www.cgal.org" ] }, { "id": "computational-graph", "title": "Computational Graph", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A computational graph is a directed acyclic graph in which nodes represent mathematical operations or variables and edges represent the flow of data (tensors) between them. It provides the structural backbone for evaluating composite functions and for computing gradients through automatic differentiation. Deep learning frameworks construct such graphs either statically ahead of execution or dynamically during the forward pass, then traverse them in reverse to propagate derivatives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:computational-graph", "labels": [ "Computational Graph" ], "is_subclass_of": [ "Automatic Differentiation" ], "wikilinks": [] }, { "id": "computational-hardness-assumption", "title": "Computational Hardness Assumption", "domain": "security", "domain_name": "Security", "definition": "A computational hardness assumption is a conjecture that a particular mathematical problem cannot be solved efficiently by any probabilistic polynomial-time algorithm. Such assumptions, including integer factorisation, the discrete logarithm, and learning-with-errors, are the foundations on which provable security of cryptographic schemes is reduced. If an assumption is broken, every construction whose security reduces to it is compromised.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:computational-hardness-assumption", "labels": [ "Computational Hardness Assumption", "Cryptographic Hardness Assumption", "Discrete Logarithm Assumption", "Mathematical Hardness Assumption" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "computational-image-relighting-technique", "title": "Computational Image Relighting Technique", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Relighting is a computer graphics and generative AI technique that computationally alters the illumination of an existing image or video, repositioning or replacing light sources to produce a different lighting environment without re-capturing the scene. Modern approaches use neural rendering models trained on diverse lighting conditions, enabling portrait relighting, object relighting, and scene relighting from a single input image. Open-source workflows in tools such as ComfyUI implement IC-Light and similar diffusion-based pipelines that rival or exceed proprietary solutions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:computational-image-relighting-technique", "labels": [ "Computational Image Relighting Technique", "relighting" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Global lighting", "AI Video", "Apple", "ComfyUI", "relighting" ] }, { "id": "computational-imaging", "title": "Computational Imaging", "domain": "ai", "domain_name": "Ai", "definition": "Computational imaging is the field that jointly designs image-capture hardware and reconstruction algorithms, using computation to recover visual information such as depth, motion, or scenes beyond conventional optics that a camera sensor alone cannot directly capture.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:computational-imaging", "labels": [ "Computational Imaging" ], "is_subclass_of": [ "Image Processing" ], "wikilinks": [] }, { "id": "computational-infrastructure", "title": "Computational Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Computational infrastructure refers to the ensemble of physical and virtualised hardware resources, networking fabric, storage systems, and supporting services that provide the computational substrate upon which software workloads execute. It encompasses data centres, server clusters, GPUs, networking interconnects, and the orchestration layers that manage resource allocation, scheduling, and fault tolerance. In the context of AI and large-scale distributed systems, computational infrastructure determines the ceiling on model scale, training throughput, and inference latency, making it a strategic bottleneck as much as a technical one.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:computational-infrastructure", "labels": [ "Computational Infrastructure" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "computational-intelligence", "title": "Computational Intelligence", "domain": "ai", "domain_name": "Ai", "definition": "Computational Intelligence (CI) is an umbrella field encompassing bio-inspired and nature-analogous algorithmic paradigms\u2014principally neural networks, fuzzy logic, and evolutionary computation\u2014that equip machines with adaptive, learning, and reasoning capabilities. Unlike classical symbolic AI, CI techniques tolerate imprecision, uncertainty, and partial truth, deriving solutions through iterative adaptation rather than explicit programming.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:computational-intelligence", "labels": [ "Computational Intelligence" ], "is_subclass_of": [ "Artificial Intelligence", "AI Research Area", "Applied Machine Learning" ], "wikilinks": [ "Artificial Intelligence", "Neural Network", "Fuzzy Logic", "Evolutionary Algorithm", "Machine Learning", "Deep Learning", "Reinforcement Learning", "Optimization Algorithm", "Reasoning", "Swarm Intelligence", "Convolutional Neural Network", "Deep Neural Network", "Deep Reinforcement Learning", "Robotics", "Autonomous Systems", "Natural Language Processing", "Computer Vision", "Pattern Recognition", "Control Systems", "Explainable AI" ] }, { "id": "computational-linguistics", "title": "Computational Linguistics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Computational Linguistics is the interdisciplinary study of language from a computational perspective, developing formal models and algorithms that enable computers to analyse, generate and understand human language. It draws on theoretical linguistics, computer science and statistics to model phenomena such as syntax, semantics and discourse. The field underpins practical natural language processing systems and provides the scientific grounding for language technologies.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:computational-linguistics", "labels": [ "Computational Linguistics" ], "is_subclass_of": [ "Natural Language Processing", "Artificial Intelligence", "Machine Learning" ], "wikilinks": [ "Natural Language Processing", "Machine Learning", "Language Model", "Large Language Model", "Information Retrieval", "Linguistics", "Tokenization", "Named Entity Recognition", "Word Embedding", "Transformer", "Machine Translation", "Sentiment Analysis", "Question Answering", "Speech Recognition", "Text Mining", "Artificial Intelligence", "Semantic Parsing", "Natural Language Understanding", "Natural Language Generation", "Dependency Parsing" ] }, { "id": "computational-model", "title": "Computational Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computational model is a mathematical and algorithmic representation of a system that uses computing resources to simulate, predict, or analyse the system's behaviour under varying conditions. It encodes the entities, state variables, and governing rules of a phenomenon into executable form so that experiments can be run in silico rather than physically. Computational models underpin scientific simulation, engineering design, and the predictive components of spatial and physical computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:computational-model", "labels": [ "Computational Model" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "computational-modelling", "title": "Computational Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computational Modelling is the use of mathematical formalisms implemented as computer programs to simulate the behaviour of real-world systems across domains such as physics, biology, economics, and climate science. A computational model translates theoretical assumptions into executable code so that hypotheses can be tested, predictions generated, and sensitivity analyses performed at scale. It forms the methodological backbone of simulation-based science and underpins modern AI training pipelines.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:computational-modelling", "labels": [ "Computational Modelling" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence", "Scientific Machine Learning", "Simulation" ], "wikilinks": [ "Mathematical Optimisation", "Simulation", "Differential Equations", "Numerical Methods", "Finite Element Analysis", "Agent-Based Modelling", "Machine Learning", "Deep Learning", "Neural Network", "Climate Modelling", "Computational Biology", "Digital Twin", "Bayesian Inference", "Monte Carlo Simulation", "High-Performance Computing", "GPU Acceleration", "Physics-Informed Neural Network", "Surrogate Model", "Sensitivity Analysis", "Uncertainty Quantification" ] }, { "id": "computational-neuroscience", "title": "Computational Neuroscience", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computational neuroscience is the study of the nervous system through mathematical models, simulations, and information-processing theories that describe how neurons and neural circuits represent, compute, and learn. It spans biophysical models of single neurons, network models of population dynamics, and theories of coding, plasticity, and behaviour. The field both explains experimental data and inspires artificial neural and neuromorphic systems by clarifying the principles of biological computation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:computational-neuroscience", "labels": [ "Computational Neuroscience" ], "is_subclass_of": [ "Cognitive Science", "Cognitive AI" ], "wikilinks": [] }, { "id": "computational-photography", "title": "Computational Photography", "domain": "ai", "domain_name": "Ai", "definition": "A discipline that extends or supplants the optical and electromechanical capabilities of a camera system through digital computation, combining multiple sensor readings, learned priors, and algorithmic inference to produce images that no single physical exposure could yield. The field integrates optics, signal processing, machine learning, and human visual perception to reconstruct scene radiance, estimate scene geometry, and synthesise perceptually superior final images.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:computational-photography", "labels": [ "Computational Photography" ], "is_subclass_of": [ "Image Processing", "Computational Imaging", "Computer Vision" ], "wikilinks": [ "Camera", "Computer Vision", "Image Processing", "Deep Learning", "Convolutional Neural Network", "Neural Network", "Signal Processing", "HDR Imaging", "Semantic Segmentation", "Depth Estimation", "Multi-Frame Fusion", "Image Noise Reduction", "Optical Flow", "Light Field Camera", "Image Sensor", "Neural Image Signal Processor", "Generative Adversarial Network", "Super Resolution", "Panorama Stitching", "Bokeh Simulation" ] }, { "id": "computational-resources", "title": "computational resources", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Computational Resources denotes the aggregate hardware and software infrastructure \u2014 encompassing processors (CPUs, GPUs, NPUs, TPUs), memory hierarchies, storage systems, network fabrics, and associated middleware \u2014 that a computing system makes available to execute workloads. The capacity, performance characteristics, and allocation policies of these resources determine achievable throughput, latency, and quality of service across application domains from real-time simulation and AI inference to distributed data processing. Resource management disciplines \u2014 including scheduling, load balancing, virtualisation, and power efficiency \u2014 govern how competing workloads share finite physical capacity, making computational resources a foundational abstraction in systems design and infrastructure planning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:computational-resources", "labels": [ "Computational Resources", "Computational Resource" ], "is_subclass_of": [ "Infrastructure", "Computing Infrastructure" ], "wikilinks": [] }, { "id": "compute-cluster", "title": "Compute Cluster", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A compute cluster is a collection of interconnected computers that work together as a single system to execute large or parallel workloads. Nodes are coordinated by a scheduler that allocates jobs across processors, accelerators, and memory, sharing high-speed networking and often a common storage fabric. Clusters underpin large-scale model training, simulation, and data processing where a single machine cannot supply enough compute.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:compute-cluster", "labels": [ "Compute Cluster" ], "is_subclass_of": [ "High-Performance Computing" ], "wikilinks": [] }, { "id": "compute-governance", "title": "Compute Governance", "domain": "governance", "domain_name": "Governance", "definition": "Compute governance is the set of policies, controls, and oversight mechanisms applied to the large-scale computing hardware used to train and deploy advanced AI systems. Because frontier AI capability is tightly coupled to access to specialised accelerators, governing compute offers a measurable, supply-chain-anchored lever for AI policy. Mechanisms include export controls, usage reporting thresholds, and on-chip verification of training runs.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:compute-governance", "labels": [ "Compute Governance" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "compute-infrastructure", "title": "Compute Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Compute Infrastructure in the artificial intelligence era denotes the integrated stack of accelerated silicon, high-bandwidth interconnect fabrics, multi-megawatt power and cooling plant, and orchestration software that hosts the training and inference of frontier neural-network models at ind...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:compute-infrastructure", "labels": [ "Compute Infrastructure", "Classified Compute Infrastructure", "Compute Orchestrator", "GPUComputeInfrastructure" ], "is_subclass_of": [ "Computing and Cloud", "Computing Infrastructure", "Infrastructure", "Data Centre Infrastructure", "Critical National Infrastructure" ], "wikilinks": [ "AI Action Plan", "AI ASIC", "AI Factory", "AI Factory", "AI Safety Institute", "AI Supply Chain", "Amazon", "AMD MI300X", "AMD MI300X Datasheet", "AMD MI325X", "AMD MI325X Product Brief", "AMD MI350X", "AMD MI350X Product Brief", "Anthropic", "ARIA", "Autonomous Systems", "AWS Inferentia 2", "AWS Trainium 2", "AWS Trainium 2 Project Rainier", "Blackstone" ] }, { "id": "compute-layer", "title": "Compute Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software layer managing computational resources and orchestration for rendering, simulation, physics, AI processing, and real-time processing within metaverse systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:compute-layer", "labels": [ "Compute Layer", "ComputeLayer" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "AI Inference", "AI Processing Service", "Compute Orchestrator", "Container Orchestration", "GPU Resources", "Load Balancer", "MSF Taxonomy 2025", "Resource Scheduler", "Computer Vision", "Data Storage Layer", "Distributed Computing", "Edge Computing", "Experience Layer", "InfrastructureDomain", "Metaverse Architecture Stack", "Metaverse Stack", "Network Infrastructure", "Physics Engine", "Physics Simulation", "Processing Hardware" ] }, { "id": "compute-resources", "title": "Compute Resources", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Compute resources are the aggregate processing capacity available to execute computational workloads, encompassing CPUs, GPUs, specialised accelerators (TPUs, NPUs), memory, network bandwidth, and associated storage I/O. They are quantified by metrics such as FLOPS, memory bandwidth, core count, and clock speed, and are provisioned via physical hardware, virtualised cloud instances, or serverless functions. Effective allocation and scheduling of compute resources is the foundational concern of cloud computing, high-performance computing, and AI infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:compute-resources", "labels": [ "Compute Resources", "Compute Resource", "ComputeResources", "Onboard Compute" ], "is_subclass_of": [ "Computational Resources", "Computing and Cloud" ], "wikilinks": [] }, { "id": "compute-shader", "title": "Compute Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Compute Shader is a programmable GPU kernel that executes arbitrary parallel computations outside the traditional graphics rendering pipeline, enabling general-purpose GPU (GPGPU) workloads such as physics simulation, procedural generation, image post-processing, and data-parallel algorithms within real-time 3D and metaverse applications.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:compute-shader", "labels": [ "Compute Shader" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "GPGPU", "Parallel Computing", "thread_position_in_grid", "Computer Vision", "Metaverse", "Physics Simulation", "Pixel Shader", "Vertex Shader" ] }, { "id": "computed-torque-control", "title": "Computed Torque Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Computed torque control is a model-based robot control method that uses the inverse dynamics of the manipulator to cancel nonlinear coupling and gravity terms, linearising the closed-loop behaviour. The controller computes the joint torques required to achieve a desired acceleration, then adds a linear feedback term to correct tracking error. It enables high-accuracy trajectory following at the cost of requiring an accurate dynamic model.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:computed-torque-control", "labels": [ "Computed Torque Control" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "computer-aided-design", "title": "Computer Aided Design", "domain": "robotics", "domain_name": "Robotics", "definition": "Computer Aided Design is the use of software to create, modify, analyse and document precise two- and three-dimensional geometric models of physical products and structures. CAD systems support parametric and direct modelling, assemblies, drafting and engineering analysis, forming the digital foundation for manufacturing and robotics. They produce the geometry and tolerances that downstream simulation, fabrication and inspection processes consume.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:computer-aided-design", "labels": [ "Computer Aided Design", "Computer-Aided Design" ], "is_subclass_of": [ "Robotics", "Software Engineering" ], "wikilinks": [] }, { "id": "computer-graphics", "title": "Computer Graphics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computer Graphics is the computational discipline concerned with synthesising, manipulating, and displaying visual imagery using mathematical models and algorithms running on specialised hardware. It encompasses the full pipeline from scene representation\u2014geometry, materials, lighting\u2014through rasterisation or ray-tracing renderers, shading languages, and GPU-accelerated display systems. As the enabling layer for all visual digital experiences, it underpins spatial computing, interactive media, simulation, scientific visualisation, and metaverse infrastructure. The field spans real-time and offline rendering paradigms, blending applied mathematics, physics-based light transport, and hardware architecture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:computer-graphics", "labels": [ "Computer Graphics", "3D Graphics" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "computer-hardware", "title": "Computer Hardware", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Physical computing components optimised for artificial intelligence and spatial computing workloads, encompassing GPUs, TPUs, FPGAs, ASICs, and neuromorphic chips. Modern AI hardware emphasises parallel matrix processing, high-bandwidth memory, mixed-precision arithmetic, and energy-efficient inference at scale, enabling both large-model training and real-time rendering in immersive applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:computer-hardware", "labels": [ "Computer Hardware" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AI Accelerator", "Distributed Training", "GPU Computing", "Neuromorphic Hardware", "Autonomous Robot", "Blockchain", "Digital Twin" ] }, { "id": "computer-science", "title": "Computer Science", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Computer science is the systematic study of computation, algorithms, data structures, programming languages, and the principles underlying the design and analysis of computing systems. It encompasses both theoretical foundations \u2014 including computability theory, computational complexity, formal languages, and logic \u2014 and applied disciplines such as software engineering, artificial intelligence, databases, and computer networks. As a rigorous scientific and engineering discipline, it provides the intellectual and practical frameworks for building reliable, efficient, and scalable digital systems. Computer science intersects with mathematics, electrical engineering, cognitive science, and the natural sciences, and it underpins virtually every domain of modern technological infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:computer-science", "labels": [ "Computer Science", "Department of Computer Science", "Theoretical Computer Science", "UCL Computer Science" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Software Engineering", "Machine Learning", "owl:Thing" ] }, { "id": "computer-use-and-browser-agents", "title": "Computer Use and Browser Agents", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computer Use and Browser Agents are a class of agentic AI systems that operate graphical user interfaces, web browsers, and operating systems by perceiving the screen through vision-language models (VLMs), planning actions in natural language, and emitting low-level input events (mouse coordinate...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:computer-use-and-browser-agents", "labels": [ "Computer Use and Browser Agents", "Browser Agent", "Computer Use Agents" ], "is_subclass_of": [ "AI Application", "Agents", "Multimodal AI", "Autonomous System", "Intelligent System", "Cognitive Architecture" ], "wikilinks": [ "Accessibility Tree", "Action Executor", "Adept ACT-1 Sep 2022", "Agent2Agent Protocol", "Agent2Agent Protocol (Google 2025)", "Agentic AI", "Agentic Workflow", "Anthropic Claude Computer Use Oct 2024", "Anthropic Model Context Protocol Nov 2024", "API-Only Agent", "Apollo Research 2024 Scheming AIs", "Autonomous System", "AutonomousSystemsDomain", "Autonomous Task Execution", "Bonatti et al 2024 Windows Agent Arena", "Brave Leo", "Browser Automation", "Browser Engine", "Browser Plugin", "Browser-Use Library 2024" ] }, { "id": "computer-use", "title": "computer use", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computer Use is an AI capability that enables multimodal models to perceive, navigate, and interact with graphical user interfaces, desktop applications, and operating system environments to autonomously complete multi-step tasks. The model receives screenshot observations of the screen and emits mouse-click, keyboard, and scroll actions, effectively operating software as a human operator would. It extends conventional tool-use paradigms by treating the entire GUI surface as an action space rather than a structured API.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:computer-use", "labels": [ "Computer Use" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "computer-vision-system", "title": "Computer Vision System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An integrated assembly of hardware and software components\u2014cameras, depth sensors, inference pipelines, and output interfaces\u2014that acquires, processes, and interprets visual information from the physical world. Computer vision systems underpin spatial computing applications such as AR tracking, spatial mapping, and object detection, translating raw image data into actionable semantic understanding for downstream tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:computer-vision-system", "labels": [ "Computer Vision System", "Vision System" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "computer-vision-task", "title": "Computer Vision Task", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Computer Vision Task is a specific computational problem solved using visual input data, encompassing image classification, object detection, semantic segmentation, instance segmentation, and pose estimation. These tasks form the building blocks of downstream vision applications such as scene understanding, autonomous navigation, and visual question answering, typically implemented via convolutional or transformer-based neural architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:computer-vision-task", "labels": [ "Computer Vision Task" ], "is_subclass_of": [ "AI Research Area", "ComputerVision" ], "wikilinks": [ "Artificial Intelligence", "ComputerVision" ] }, { "id": "computer-vision-video-analysis", "title": "Computer Vision Video Analysis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The computational analysis, understanding, and manipulation of video data using machine learning and computer vision techniques. Core tasks include object detection and tracking, semantic segmentation, action recognition, temporal modelling, and scene understanding. Modern approaches employ 3D convolutional networks, vision transformers, and self-supervised learning on large-scale video datasets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:computer-vision-video-analysis", "labels": [ "Computer Vision Video Analysis", "Video Analysis", "Video Processing" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Action Recognition", "Temporal Modeling", "Computer Vision", "Object Detection" ] }, { "id": "computer-vision", "title": "Computer Vision", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computer Vision is the field of artificial intelligence concerned with enabling machines to interpret, understand, and process visual information from the world, emulating human visual perception capabilities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:computer-vision", "labels": [ "Computer Vision", "Computer Vision Understanding", "ComputerVision" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Autonomous Vehicle Navigation", "boring2009scroll", "hansberger2017dispelling", "Image Processing", "KnoWhere", "Neural Networks", "Quality Control", "Robotics Perception", "Surveillance Systems", "3D Reconstruction", "Apple", "Artificial Intelligence", "ArtificialIntelligenceDomain", "computer vision", "Convolutional Neural Network", "Deep Learning", "deep learning", "Digital Twin", "Feature Extraction", "Image Classification" ] }, { "id": "computer-supported-cooperative-work", "title": "Computer-Supported Cooperative Work", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Computer-Supported Cooperative Work (CSCW) is the interdisciplinary research field that studies how people work together using computer systems and how to design technology that supports their collaboration. It examines coordination, communication, awareness and the social organisation of group work across time and space, informing the design of groupware and collaborative tools. CSCW bridges human-computer interaction, the social sciences and distributed-systems engineering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:computer-supported-cooperative-work", "labels": [ "Computer-Supported Cooperative Work" ], "is_subclass_of": [ "Distributed Collaboration", "Workspace Tools" ], "wikilinks": [] }, { "id": "computerised-adaptive-testing", "title": "Computerised Adaptive Testing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Computerised adaptive testing (CAT) is an assessment method that selects each subsequent question based on a test-taker's performance on previous items, converging on an estimate of ability with fewer questions than a fixed-form test. It relies on item response theory to model the probability of a correct answer as a function of item difficulty and estimated learner ability, updating the estimate after every response. CAT is widely used in standardised testing and adaptive learning platforms because it shortens test duration while improving measurement precision at the extremes of the ability range. It is closely related to Bayesian knowledge tracing, which similarly updates a learner model from sequential evidence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:computerised-adaptive-testing", "labels": [ "Computerised Adaptive Testing" ], "is_subclass_of": [ "Adaptive Learning" ], "wikilinks": [] }, { "id": "computing-hardware", "title": "Computing Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computing Hardware encompasses the physical processing, memory, and peripheral devices that underpin spatial computing experiences, including GPUs, neural accelerators, XR headsets, and edge devices. It forms the substrate on which spatial applications, rendering engines, and sensor fusion pipelines execute.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:computing-hardware", "labels": [ "Computing Hardware" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Sensor Input", "owl:Thing" ] }, { "id": "computing-infrastructure", "title": "Computing Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Computing infrastructure encompasses the physical and virtual resources required for building, running, and delivering applications and services, including servers, storage systems, networking equipment, power systems, and cooling facilities housed in data centres. It provides the foundational capacity that IT systems require to process, store, and transmit digital data. Modern computing infrastructure increasingly extends from on-premises hardware into cloud-hosted and edge environments, and is a prerequisite for AI workloads, distributed applications, and spatial computing platforms.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:computing-infrastructure", "labels": [ "Computing Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "InfrastructureDomain", "Technology Domain" ] }, { "id": "computing-platform", "title": "Computing Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A computing platform is a hardware and software environment that provides the foundational execution context for applications, services, and AI workloads. It encompasses the combination of processor architecture, operating system, runtime libraries, and supporting infrastructure\u2014such as cloud, edge, or on-premises nodes\u2014that determines what software can run and how it performs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:computing-platform", "labels": [ "Computing Platform" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "concentrated-liquidity", "title": "Concentrated Liquidity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Concentrated liquidity is a capital efficiency mechanism for automated market makers (AMMs) in which liquidity providers deposit assets within user-specified price ranges rather than uniformly across the full price curve from zero to infinity. Within an active price range, the capital deployed behaves equivalently to a much larger position in a constant-product AMM, dramatically increasing fee revenue per unit of capital while simultaneously reducing the price impact of trades of a given size. This design was pioneered by Uniswap v3 and has since been widely adopted across decentralised exchanges.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:concentrated-liquidity", "labels": [ "Concentrated Liquidity", "Concentrated Liquidity Position" ], "is_subclass_of": [ "Liquidity Provision" ], "wikilinks": [] }, { "id": "concentration-inequalities", "title": "Concentration Inequalities", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Concentration inequalities are probabilistic bounds \u2014 such as Markov's, Chebyshev's, Hoeffding's and Chernoff's inequalities \u2014 that quantify how tightly a random variable, typically a sum or average, concentrates around its expected value. In machine learning they provide the mathematical basis for generalisation bounds, showing how closely empirical risk on a finite sample tracks true risk as sample size grows. They are a core analytical tool within statistical learning theory and underpin bias-variance and PAC-learning arguments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:concentration-inequalities", "labels": [ "Concentration Inequalities" ], "is_subclass_of": [ "Statistical Learning Theory" ], "wikilinks": [] }, { "id": "concept-art", "title": "Concept Art", "domain": "ai", "domain_name": "Ai", "definition": "Concept art is a discipline of visual development practice in which illustrators, designers, and increasingly AI-assisted workflows produce preliminary visual representations of characters, environments, vehicles, creatures, and props to establish the aesthetic, mood, and functional parameters of an intended creative production before commitment to the labour-intensive downstream stages of 3D modelling, animation, and rendering. It functions as the design language and communication medium between creative direction and production teams across games, film, animation, and extended reality.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:concept-art", "labels": [ "Concept Art", "Concept Art Workflows" ], "is_subclass_of": [ "AI Application", "Visual Development", "Creative Industries", "Creative Expression" ], "wikilinks": [ "Midjourney Text-to-Image Service", "Proprietary Image Generation", "Generative AI", "Diffusion Model", "Text-to-Image Generation", "Stable Diffusion", "Film Production", "Game Asset Generation", "Visual Development", "Character Design", "Environment Design", "Pre-Production", "Art Direction", "Colour Palette", "3D Modeling", "Animation", "Game Engine", "Intellectual Property", "Generative Adversarial Network", "Creative AI" ] }, { "id": "concept-drift", "title": "Concept Drift", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Concept Drift is the phenomenon in which the statistical properties of the target variable that a machine learning model was trained to predict change over time, causing model performance to degrade. Drift can be abrupt, gradual, or recurring, and may stem from evolving user behaviour, environmental shifts, or data collection changes. Detecting and adapting to concept drift is essential for maintaining the reliability of deployed ML systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:concept-drift", "labels": [ "Concept Drift" ], "is_subclass_of": [ "AI Technique", "Machine Learning", "Machine Learning Technique", "Distribution Shift" ], "wikilinks": [ "Machine Learning", "MLOps", "Model Monitoring", "Data Drift", "Online Learning", "Continual Learning", "Distribution Shift", "Covariate Shift", "Anomaly Detection", "Statistics", "Probability Distribution", "Feature Engineering", "Deep Learning", "Neural Networks", "Reinforcement Learning", "Transfer Learning", "Federated Learning", "Active Learning", "Model Training", "Data Pipeline" ] }, { "id": "conceptual-hierarchy", "title": "Conceptual Hierarchy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Conceptual Hierarchy is a structured taxonomic organisation of domain concepts into subsumption (is-a) and composition (part-of) relationships, enabling systematic knowledge representation, inheritance of properties, and semantic interoperability across robotic and autonomous systems. It supports automated reasoning, modular system design, and classification of new entities within standardised ontological frameworks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:conceptual-hierarchy", "labels": [ "Conceptual Hierarchy", "ConceptualHierarchy" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "IEEE RAS Ontology", "IEEE/RSJ IROS" ] }, { "id": "conceptual-layer", "title": "Conceptual Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Conceptual Layer represents the highest abstraction level in an ontological classification system, encompassing pure concepts, theoretical models, design principles, and logical frameworks that are independent of specific technical implementations. Concepts at this layer can be realised through multiple concrete approaches while retaining their essential semantic properties.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:conceptual-layer", "labels": [ "Conceptual Layer", "ConceptualLayer" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "concurrency-control", "title": "Concurrency Control", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Concurrency control is the set of techniques that coordinate simultaneous operations on shared data so that correctness is preserved despite interleaving. It ensures that concurrent transactions or processes produce results equivalent to some valid serial execution, preventing anomalies such as lost updates and inconsistent reads. It is foundational to databases, distributed systems, and collaborative applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:concurrency-control", "labels": [ "Concurrency Control" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "concurrency", "title": "Concurrency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Concurrency is the composition of independently executing computations that make progress within overlapping time periods, whether or not they run simultaneously on separate processors. It is a way of structuring a program so that multiple tasks can be in flight at once, coordinating access to shared state through synchronisation primitives. Concurrency is distinct from parallelism: it concerns the dealing with many things at once, while parallelism concerns the doing of many things at once.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:concurrency", "labels": [ "Concurrency" ], "is_subclass_of": [ "Infrastructure", "Software Engineering (Infrastructure)" ], "wikilinks": [] }, { "id": "condition-monitoring", "title": "Condition Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Condition monitoring is the continuous or periodic measurement of physical parameters \u2014 vibration, temperature, pressure, or acoustic signature \u2014 from operating machinery to detect degradation before it causes failure. It is a core Industrial IoT application, relying on networked sensors to stream measurements to analytics systems that flag anomalies for predictive maintenance. It reduces unplanned downtime compared with fixed-interval preventive maintenance schedules.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:condition-monitoring", "labels": [ "Condition Monitoring" ], "is_subclass_of": [ "Industrial IoT" ], "wikilinks": [] }, { "id": "conditional-payment", "title": "Conditional Payment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A conditional payment is a transfer of value that is released only when one or more predefined conditions are met, rather than executing unconditionally on submission. On blockchains these conditions are enforced by smart contracts or scripts such as hash and time locks, removing the need for a trusted intermediary to adjudicate. Conditional payments are the foundation of escrow, payment channels and atomic cross-chain swaps.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:conditional-payment", "labels": [ "Conditional Payment" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "conditional-random-field", "title": "Conditional Random Field", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Conditional Random Field (CRF) is a discriminative probabilistic graphical model used for structured prediction tasks such as sequence labelling, where the model directly estimates the conditional probability P(y|x) of an output label sequence y given an observed input sequence x. CRFs overcome the label bias problem of maximum-entropy Markov models by considering the entire output sequence jointly, making them effective for named entity recognition, part-of-speech tagging, and image segmentation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:conditional-random-field", "labels": [ "Conditional Random Field" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Blockchain" ] }, { "id": "conditioning-signal", "title": "Conditioning Signal", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A conditioning signal is auxiliary input - such as a text prompt, class label, pose map, or edge map - supplied to a generative model to steer its output toward desired attributes without retraining the base model. Techniques such as classifier-free guidance and ControlNet-style spatial conditioning inject these signals at specific points in the generation process to control content, structure, or style. The strength and fidelity of a conditioning signal determine how closely generated output adheres to the intended constraint.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:conditioning-signal", "labels": [ "Conditioning Signal" ], "is_subclass_of": [ "Control Signal" ], "wikilinks": [] }, { "id": "cone-of-depression", "title": "Cone Of Depression", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cone-of-depression", "labels": [ "Cone Of Depression" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cone-of-impression", "title": "Cone Of Impression", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cone-of-impression", "labels": [ "Cone Of Impression" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "confidential-computing-consortium", "title": "Confidential Computing Consortium", "domain": "security", "domain_name": "Security", "definition": "The Confidential Computing Consortium is a Linux Foundation project that brings together hardware vendors, cloud providers and software firms to advance the adoption and standardisation of confidential computing. It defines confidential computing as protecting data in use by performing computation within a hardware-based trusted execution environment. The consortium stewards open-source projects, terminology and best practices that promote interoperable enclave technologies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:confidential-computing-consortium", "labels": [ "Confidential Computing Consortium" ], "is_subclass_of": [ "Standards Documentation" ], "wikilinks": [] }, { "id": "confidential-computing", "title": "Confidential Computing", "domain": "security", "domain_name": "Security", "definition": "A hardware-based security paradigm that protects data in use by isolating computation within trusted execution environments (TEEs) backed by processor security extensions (Intel SGX, AMD SEV, ARM TrustZone). It extends encryption from data at rest and in transit to data actively being processed, preventing access even by privileged software, hypervisors, or cloud providers. Key AI applications include secure model training, private inference, and TEE-protected federated learning aggregation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:confidential-computing", "labels": [ "Confidential Computing" ], "is_subclass_of": [ "Hardware Security" ], "wikilinks": [ "AMD SEV", "Confidential Computing Consortium", "Intel SGX", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin" ] }, { "id": "confidential-transactions", "title": "Confidential Transactions", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Confidential Transactions (CT) is a cryptographic protocol for distributed ledgers, designed by Gregory Maxwell in 2015, that conceals the amounts transferred in financial transactions whilst preserving the ability for validators to verify that no value is created or destroyed. It employs Pedersen commitments\u2014homomorphic elliptic-curve constructs\u2014to encode transaction values in a form that is computationally hiding yet perfectly binding, combined with range proofs (typically Bulletproofs) to ensure committed values are non-negative and thus prevent inflation attacks. CT has been deployed in the Liquid Network sidechain, Monero's RingCT, and the MimbleWimble protocol family, and underpins much of the contemporary research into privacy-preserving decentralised finance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:confidential-transactions", "labels": [ "Confidential Transactions", "Anonymous Transactions", "Confidential Assets", "Confidential Assets Protocol", "Confidential Transaction", "Confidential-Transactions", "Private Transactions", "Ring Confidential Transactions" ], "is_subclass_of": [ "Blockchain Transaction", "Cryptographic Primitive (Blockchain)" ], "wikilinks": [] }, { "id": "configuration-management", "title": "Configuration Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Configuration management is the engineering discipline of systematically establishing, recording, and maintaining the desired state of a system's components, settings, and dependencies throughout its lifecycle. It ensures that environments are reproducible and consistent by treating configuration as versioned, auditable artefacts rather than ad hoc manual changes. In modern practice it underpins infrastructure-as-code and continuous delivery, using declarative tools to converge machines and services to a defined state and to track every change for traceability and rollback.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:configuration-management", "labels": [ "Configuration Management" ], "is_subclass_of": [ "Software Engineering", "Automation" ], "wikilinks": [ "Software Engineering", "Version Control", "Change Management", "DevOps", "Continuous Integration", "Infrastructure as Code", "Quality Assurance", "Audit", "Testing", "GitOps", "Terraform", "Ansible", "Kubernetes", "Continuous Delivery", "Continuous Deployment", "MLOps", "Reproducibility", "Immutable Infrastructure", "Idempotency", "Cloud Computing" ] }, { "id": "configuration-setting", "title": "Configuration Setting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Configuration Setting is a named parameter or preference value that governs the runtime behaviour of a spatial computing application, platform, or XR experience. Configuration settings control rendering quality, user accessibility options, privacy preferences, and network parameters, and are typically persisted across sessions to maintain consistent user experiences within metaverse and immersive environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:configuration-setting", "labels": [ "Configuration Setting" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "configuration-space", "title": "Configuration Space", "domain": "robotics", "domain_name": "Robotics", "definition": "Configuration space, often written C-space, is the set of all possible configurations of a robot, where each point fully specifies the position of every part of the mechanism. Its dimensionality equals the robot's degrees of freedom, and obstacles in the physical workspace map to forbidden regions, partitioning C-space into free and blocked subsets. Motion planning is then recast as finding a continuous path through the free portion of configuration space.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:configuration-space", "labels": [ "Configuration Space" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [] }, { "id": "configuration", "title": "Configuration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Configuration is the set of parameters, settings, and environment variables that determine the runtime behaviour of a spatial computing platform, application, or pipeline component. Proper configuration governs interoperability between subsystems, controls feature flags, and defines integration points between platform services, hardware abstraction layers, and content pipelines.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:configuration", "labels": [ "Configuration", "Static Configuration" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "confined-aquifer", "title": "Confined Aquifer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:confined-aquifer", "labels": [ "Confined Aquifer" ], "is_subclass_of": [ "Aquifer" ], "wikilinks": [] }, { "id": "confined-bed", "title": "Confined Bed", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:confined-bed", "labels": [ "Confined Bed" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "confined-unit", "title": "Confined Unit", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:confined-unit", "labels": [ "Confined Unit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "confined-zone", "title": "Confined Zone", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:confined-zone", "labels": [ "Confined Zone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "conflict-free-replicated-data-type", "title": "Conflict Free Replicated Data Type", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A conflict-free replicated data type (CRDT) is a data structure that can be replicated across many nodes and updated independently, with mathematical guarantees that all replicas converge to the same state once they have exchanged updates. By designing operations to be commutative or merges to be monotonic, CRDTs avoid the need for coordination or central conflict resolution. They are a foundational technique for offline-first and real-time collaborative distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:conflict-free-replicated-data-type", "labels": [ "Conflict Free Replicated Data Type", "Conflict-Free Replicated Data Type" ], "is_subclass_of": [ "Eventual Consistency" ], "wikilinks": [] }, { "id": "conflict-mineral-tracking", "title": "Conflict Mineral Tracking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Conflict Mineral Tracking denotes the use of blockchain technology to record and verify the provenance of 3TG minerals (tin, tantalum, tungsten, gold) through complex supply chains, ensuring they do not finance armed conflict or human rights abuses in the Democratic Republic of Congo and adjoining regions. Implementations combine immutable custody chains, cryptographic material fingerprinting, and smart-contract-enforced compliance checks to satisfy regulatory obligations including Dodd-Frank Section 1502 and the EU Conflict Minerals Regulation while enabling premium pricing for verified responsible sourcing.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:conflict-mineral-tracking", "labels": [ "Conflict Mineral Tracking", "BC-0445-conflict-mineral-tracking", "Conflict Minerals", "Conflict-Free Minerals" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BC-0013-smart-contracts", "BC-0029-permissioned-blockchain", "BC-0044-supply-chain-management", "BC-0066-ethereum", "BC-0067-hyperledger-fabric", "BC-0214-environmental-sustainability", "BC-0432-consortium-blockchain", "BC-0434-blockchain-as-a-service", "BC-0441-provenance-tracking", "Circulor", "Everledger", "Internet of Things", "RCS Global", "Blockchain", "BlockchainDomain", "Ethereum", "Hyperledger Fabric" ] }, { "id": "conflict-resolution", "title": "Conflict Resolution", "domain": "governance", "domain_name": "Governance", "definition": "Conflict Resolution encompasses the structured methods, protocols, and mechanisms used to identify, address, and settle disputes between parties \u2014 whether human individuals, organisations, autonomous agents, or distributed systems. In technical contexts it includes algorithmic approaches for reconciling inconsistencies in distributed data systems, consensus protocols that converge divergent states, and governance frameworks for mediating disagreements in decentralised organisations. In social and legal contexts it spans negotiation, mediation, arbitration, and adjudication, with increasing automation and AI-assisted facilitation. Across all domains, effective conflict resolution preserves relationships, maintains system integrity, and enables continued collaboration.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:conflict-resolution", "labels": [ "Conflict Resolution", "Conflict Resolution Engine", "Conflict Resolution Module", "Conflict Resolution Protocol", "Conflict Resolution Strategy" ], "is_subclass_of": [ "Dispute Resolution Mechanism" ], "wikilinks": [] }, { "id": "confluence", "title": "Confluence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Confluence is a team workspace and wiki product developed by Atlassian for collaborative documentation, knowledge management, and project information sharing. It organises content into spaces and pages with rich editing, versioning, and tight integration with issue-tracking tools such as Jira. It is widely deployed as a digital-workplace platform for enterprise knowledge capture.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:confluence", "labels": [ "Confluence" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "conformal-prediction", "title": "Conformal Prediction", "domain": "ai", "domain_name": "Ai", "definition": "Conformal prediction is a distribution-free, model-agnostic framework for producing statistically valid prediction sets or intervals that are guaranteed to contain the true outcome with a user-specified probability, using only the mild assumption of exchangeability over the data. A held-out calibration set and a nonconformity score function are sufficient to construct finite-sample marginal coverage guarantees for any base predictor \u2014 ranging from linear models and neural networks to Gaussian processes and large language models. The framework unifies classification (prediction sets), regression (prediction intervals), and sequence generation (conformal risk control), and has become a cornerstone method for trustworthy AI deployment in safety-critical and regulated domains.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:conformal-prediction", "labels": [ "Conformal Prediction" ], "is_subclass_of": [ "Machine Learning Technique", "Uncertainty Quantification", "Machine Learning", "Statistical Learning" ], "wikilinks": [ "Probabilistic Model", "Uncertainty Quantification", "Machine Learning", "Statistical Learning", "Calibration", "Nonconformity Score", "Exchangeability", "Bayesian Inference", "Frequentist Statistics", "Gaussian Process", "Neural Network", "Deep Learning", "Classification", "Regression", "Prediction Interval", "Coverage Guarantee", "Hypothesis Testing", "Anomaly Detection", "Drug Discovery", "Medical Imaging" ] }, { "id": "conformity-assessment-body", "title": "Conformity Assessment Body", "domain": "governance", "domain_name": "Governance", "definition": "A conformity assessment body is an accredited organisation authorised to evaluate whether products, processes, or services meet specified standards or regulatory requirements. It performs testing, inspection, and certification activities and issues the formal attestations that allow goods to be placed on a market. Such bodies are themselves accredited by national accreditation authorities to ensure impartiality and competence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:conformity-assessment-body", "labels": [ "Conformity Assessment Body" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "conformity-assessment", "title": "conformity assessment", "domain": "governance", "domain_name": "Governance", "definition": "Conformity Assessment is a structured set of procedures used to determine whether a product, system, service, or process meets specified requirements defined in regulations, standards, or contractual obligations. In the context of AI and digital systems, it encompasses technical documentation review, testing, auditing, risk analysis, and post-market surveillance carried out either by the developer as a self-assessment or by an accredited third-party notified body. Under frameworks such as the EU AI Act, ISO/IEC 17000-series standards, and NIST guidelines, the conformity declaration produced is a legal and operational precondition for market entry and continued deployment. It inherits procedural principles from product safety certification while adapting them to the probabilistic, data-dependent, and emergent behaviour of machine learning systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:conformity-assessment", "labels": [ "Conformity Assessment", "AI Conformity Assessment", "Conformity Assessment (AI-0103)" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "confusion-matrix", "title": "Confusion Matrix", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A tabular visualisation and analytical tool summarising the performance of a classification model by displaying the counts or proportions of predictions cross-tabulated against actual class labels, typically organised with predicted classes as columns and actual classes as rows (or vice versa), enabling systematic analysis of where a model succeeds and fails, calculation of various performance metrics, and identification of specific confusion patterns between classes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:confusion-matrix", "labels": [ "Confusion Matrix", "Confusion Matrices" ], "is_subclass_of": [ "Model Evaluation" ], "wikilinks": [ "Error analysis", "fairness assessment", "False Negative", "False Positive", "ISO/IEC 25024", "ISO/IEC 25059", "model debugging", "NIST AI RMF", "Sensitivity", "Specificity", "True Negative", "True Positive", "Accuracy", "Computer Vision", "F1 Score", "MetaverseDomain", "Model Performance", "Precision", "Recall", "ROC Curve" ] }, { "id": "congelation-ice", "title": "Congelation Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:congelation-ice", "labels": [ "Congelation Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "congestion-control", "title": "Congestion Control", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Congestion Control is the set of mechanisms and algorithms that regulate the rate of data transmission across a network to prevent any sender, link, or node from being overwhelmed by more traffic than it can handle, thereby maintaining overall network stability and fairness. Operating primarily at the transport layer, congestion control algorithms infer network capacity from signals such as packet loss, explicit congestion notification, and round-trip time variations, adjusting sender rates accordingly. Classical implementations include TCP Tahoe, Reno, CUBIC, and BBR; modern variants extend to QUIC, WebRTC, and multipath scenarios where joint path management adds additional complexity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:congestion-control", "labels": [ "Congestion Control", "BBR Congestion Control" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "connectionism", "title": "Connectionism", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Connectionism is an approach in cognitive science and artificial intelligence that models mental and computational phenomena as emergent from networks of simple, densely interconnected units whose collective activity, governed by weighted connections, gives rise to behaviour. It treats knowledge as distributed across connection strengths rather than stored as explicit symbols, and learning as the adjustment of those weights through experience. The paradigm provides the theoretical foundation for artificial neural networks and deep learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:connectionism", "labels": [ "Connectionism" ], "is_subclass_of": [ "Cognitive Science", "AI Technique" ], "wikilinks": [] }, { "id": "connectionist-temporal-classification", "title": "Connectionist Temporal Classification", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Connectionist Temporal Classification (CTC) is a sequence-modelling loss function and decoding scheme that trains neural networks to map unsegmented input sequences to output label sequences without requiring pre-aligned data. It introduces a blank symbol and marginalises over all valid alignments, allowing a network to learn the alignment implicitly during training. CTC is widely used in speech recognition and handwriting recognition where input and output lengths differ and frame-level labels are unavailable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:connectionist-temporal-classification", "labels": [ "Connectionist Temporal Classification" ], "is_subclass_of": [ "Speech Recognition" ], "wikilinks": [] }, { "id": "connector-standards", "title": "Connector Standards", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Connector standards are specifications that define the mechanical form factor, pin assignment, and electrical signalling of the physical interfaces used to join cables and devices. They guarantee interoperability across vendors at the physical layer, covering examples such as USB-C, RJ45, and HDMI. Standardised connectors reduce fragmentation and enable plug-and-play interconnection.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:connector-standards", "labels": [ "Connector Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "connext", "title": "Connext", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cross-chain interoperability protocol that enables fast transfers and contract calls between Ethereum-compatible blockchains and Layer 2 networks without relying on a single trusted custodian, coordinating liquidity routers and a verification layer to achieve trust-minimised cross-chain composability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:connext", "labels": [ "Connext" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [ "Ethereum", "Liquidity Pool", "Interoperability", "Cross-Chain Bridge" ] }, { "id": "consen-sys", "title": "ConsenSys", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ConsenSys is a Brooklyn-founded blockchain software company, established in 2014 by Ethereum co-founder Joseph Lubin, that builds developer tooling, infrastructure, and end-user products for the Ethereum ecosystem. Its product portfolio spans the MetaMask browser wallet, the Infura node-as-a-service API network, the Truffle and Hardhat-adjacent development suite, and the Linea zkEVM Layer 2 network. ConsenSys acts as a principal driver of Ethereum protocol adoption by providing the production-grade infrastructure on which the majority of decentralised application front-ends depend. The company also engages in enterprise blockchain consulting and contributes to Ethereum standards through active participation in the Ethereum Foundation and EIP processes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:consen-sys", "labels": [ "ConsenSys" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Ethereum", "Web3", "Smart Contract" ] }, { "id": "consensus-algorithm", "title": "Consensus Algorithm", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed computational protocol ensuring all participants in a Blockchain Network agree on the canonical transaction history and current state without centralised authority, tolerating a bounded fraction of faulty or malicious nodes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:consensus-algorithm", "labels": [ "Consensus Algorithm", "Consensus Algorithms", "ConsensusAlgorithm" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "achievesFinality", "Consensus Algorithms", "Consensus Mechanisms", "Data Integrity", "DistributedConsensus", "dt:coordinates", "dt:governs", "dt:secures", "dt:synchronizes", "dt:validates", "implementedBy", "ISO/IEC 2382:2025", "Metaverse Standards Forum", "MultiAgentSystem", "Network Security", "NetworkSecurity", "requiresValidators", "toleratesFaults", "Trustless Coordination", "ValidatorNetwork" ] }, { "id": "consensus-layer", "title": "Consensus Layer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Consensus Layer is the stratum responsible for agreement on a single canonical ordering of events across distributed participants. In the canonical stack it sits directly above the Protocol Layer and below the Data Layer, converting peer-to-peer message exchange into a shared, append-only history. It contains the agreement algorithms, fork-choice rules, and finality conditions that all participants follow.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:consensus-layer", "labels": [ "Consensus Layer" ], "is_subclass_of": [ "Protocol and Consensus", "owl:Thing" ], "wikilinks": [ "Protocol Layer", "Data Layer", "Byzantine Fault Tolerance", "Proof of Stake", "owl:Thing", "IETF (Internet Engineering Task Force)" ] }, { "id": "consensus-mechanism", "title": "Consensus Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Consensus Mechanism is a Distributed Algorithm enabling a population of independent, potentially adversarial nodes communicating over an unreliable network to agree on a single, totally-ordered sequence of state transitions (a replicated log) such that all honest participants eventually...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:consensus-mechanism", "labels": [ "Consensus Mechanism", "ConsensusMechanism" ], "is_subclass_of": [ "Protocol and Consensus", "Distributed Algorithm", "Coordination Protocol", "State Machine Replication", "Byzantine Fault Tolerance", "Atomic Broadcast" ], "wikilinks": [ "Aleph Zero", "Algorand", "AptosBFT", "Atomic Broadcast", "Avalanche", "BABE", "Babel et al 2024 Mysticeti", "Baird 2016 Hashgraph Swirlds", "BFT", "Bitcoin BIPs", "Block Proposal", "Blockchain Security", "BLS Signature", "Buchman Kwon Milosevic 2018 Tendermint", "Bullshark", "Buterin Griffith 2017 Casper FFG", "Byzantine Quorum Without Total Order", "Cambridge Computer Laboratory Distributed Systems", "CAP Theorem", "Carnot" ] }, { "id": "consensus-mechanisms", "title": "Consensus Mechanisms", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The family of protocols by which distributed participants agree on a single shared state or ordering of events without relying on a central authority, spanning classical fault-tolerant algorithms, probabilistic approaches, and cryptoeconomic incentive designs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:consensus-mechanisms", "labels": [ "Consensus Mechanisms" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Consensus", "Blockchain", "Distributed Ledger", "Proof of Stake", "Tendermint", "Distributed Systems" ] }, { "id": "consensus-process", "title": "Consensus Process", "domain": "standards", "domain_name": "Standards", "definition": "A consensus process is the structured, participatory procedure by which a standards body or governance group develops, refines, and adopts decisions through deliberation aimed at broad agreement rather than narrow majority voting. It gathers stakeholder input, circulates drafts for review and public comment, resolves objections substantively, and seeks a state of general acceptance \u2014 often characterised as rough consensus \u2014 where remaining dissent has been heard and addressed. The process underpins the legitimacy and durability of open standards by ensuring decisions reflect the considered views of affected parties rather than the preferences of a dominant actor.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:consensus-process", "labels": [ "Consensus Process" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "consensus-protocol", "title": "Consensus Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A consensus protocol is a set of rules and algorithms by which distributed network participants reach agreement on a single shared state or value without requiring centralised authority, ensuring liveness, safety, and Byzantine fault tolerance up to a specified threshold of adversarial nodes. Classical families include compute-bound Proof-of-Work, stake-weighted Proof-of-Stake, and message-passing BFT protocols such as PBFT, Tendermint, and HotStuff. Each protocol specifies leader election, block proposal, voting rounds, and finality conditions that together determine throughput, latency, decentralisation, and security trade-offs. Consensus protocols underpin blockchains, distributed databases, and any replicated state machine requiring deterministic agreement across potentially unreliable or malicious peers.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:consensus-protocol", "labels": [ "Consensus Protocol", "Consensus-Protocol", "ConsensusProtocol" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "Blockchain" ] }, { "id": "consensus-rule", "title": "Consensus Rule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Consensus Rule is a protocol-level validation requirement that every fully-validating node in a blockchain network must enforce uniformly to reach and maintain agreement on the canonical chain state. Consensus rules define which blocks and transactions are valid, covering aspects such as block structure, cryptographic proofs, transaction format, gas limits, and state-transition logic. Deviation from consensus rules \u2014 whether accidental or deliberate \u2014 results in a fork, splitting the network into incompatible chains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:consensus-rule", "labels": [ "Consensus Rule" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "consensus", "title": "Consensus", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The general problem of getting distributed processes to agree on a common value or decision despite failures, communication delays or adversarial behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:consensus", "labels": [ "Consensus", "Consensus Safety", "Decentralised Consensus", "Federated Consensus", "Network Consensus" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Distributed Systems", "Consensus Mechanisms", "Distributed Ledger", "Tendermint" ] }, { "id": "consent-management", "title": "Consent Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "System for recording and enforcing user permissions for data collection, processing, and sharing across metaverse platforms, ensuring compliance with privacy regulations and user autonomy.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:consent-management", "labels": [ "Consent Management", "Consent Management Framework", "Consent Mechanism", "ConsentManagement", "Patient Consent Management", "User Consent Management" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Audit Logger", "Consent Registry", "Data Governance Framework", "Data Privacy", "ENISA", "GDPR Compliance", "ISO 29184", "Permission Controller", "Privacy Policy", "User Authentication", "User Control", "Application Layer", "ETSI Domain: Data Management + Ethics", "Identity Provider", "Middleware Layer", "Policy Engine", "Right to be Forgotten", "Right to Be Forgotten", "Telecollaboration", "Transparency" ] }, { "id": "consent-registry", "title": "Consent Registry", "domain": "security", "domain_name": "Security", "definition": "A consent registry is a system of record that stores, versions, and serves the consent decisions data subjects have granted or withdrawn for processing their personal data. It provides an auditable, queryable source of truth that data controllers and processors consult before performing a processing activity. It is a central component of privacy-compliance and consent-management architectures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:consent-registry", "labels": [ "Consent Registry" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "consistency-checking", "title": "Consistency Checking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Consistency checking is the process of verifying that a set of statements, data items or constraints contains no contradictions and that all derivable conclusions remain mutually compatible. In knowledge representation it confirms that an ontology or knowledge base admits at least one model, while in data systems it confirms that records satisfy declared integrity rules. The technique underpins trust in automated reasoning by rejecting configurations that would license arbitrary or unsound inferences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:consistency-checking", "labels": [ "Consistency Checking" ], "is_subclass_of": [ "Verification" ], "wikilinks": [] }, { "id": "consistency-model", "title": "Consistency Model", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A consistency model is a contract between a distributed data store and its clients that specifies the guarantees about the visibility and ordering of reads and writes across replicas. It defines which outcomes of concurrent operations are permissible, ranging from strong models like linearizability that behave as a single up-to-date copy, to weak models like eventual consistency that allow temporary divergence. The chosen model shapes application correctness, performance, and the achievable balance among consistency, availability, and partition tolerance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:consistency-model", "labels": [ "Consistency Model" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "consistent-hashing", "title": "Consistent Hashing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Consistent hashing is a distribution technique that maps both data keys and storage nodes onto the same circular hash space, so that each key is assigned to the next node encountered clockwise on the ring. When a node joins or leaves, only the keys in its immediate neighbourhood are remapped rather than the entire key space, minimising data movement. Virtual nodes are commonly used to smooth load distribution across heterogeneous servers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:consistent-hashing", "labels": [ "Consistent Hashing" ], "is_subclass_of": [ "Distributed Hash Table" ], "wikilinks": [] }, { "id": "consortium-blockchain", "title": "Consortium Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Consortium Blockchain is a permissioned distributed ledger architecture in which a pre-approved set of legally distinct organisations jointly operates the validator infrastructure, governs the protocol rules, and controls the read/write access matrix \u2014 occupying a deliberate middle position in th...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:consortium-blockchain", "labels": [ "Consortium Blockchain", "BC-0432-consortium-blockchain", "ConsortiumBlockchain" ], "is_subclass_of": [ "Network Component", "Distributed Ledger Technology", "Permissioned Network", "Enterprise Blockchain", "Federated System", "Multi-Organisation Database" ], "wikilinks": [ "Androulaki et al 2018 Hyperledger Fabric EuroSys", "Atomic DvP Settlement", "Azure Confidential Consortium Framework", "Bank of Canada Project Jasper", "BIS Project Agor\u00e1 Announcement 2024", "BIS Project mBridge MVP Report 2024", "BIS Project Mariana Report 2023", "BoE Project Rosalind Phase 1-2", "Cambridge CCAF Global Enterprise Blockchain Benchmarking Study", "Castro Liskov 1999 PBFT", "CBDC", "Certificate Authority", "Chaincode Container", "Channels and Private Data Collections", "Citi Token Services", "Citi Token Services Press Release Sept 2023", "City of London Tokenisation Strategy 2024", "ConsensusLayer", "Consensys GoQuorum", "Consortium Governance Framework" ] }, { "id": "consortium-governance", "title": "Consortium Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Consortium governance is the model by which a defined group of organisations jointly operates and controls a permissioned blockchain or shared infrastructure. Membership, validator rights, and decision-making authority are restricted to vetted participants who agree to a governing charter. It balances the decentralisation benefits of distributed ledgers with the accountability and access control enterprises require.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:consortium-governance", "labels": [ "Consortium Governance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "conspiratorial-thinking-in-technology-communities", "title": "Conspiratorial Thinking in Technology Communities", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Conspiracies, in the context of technology communities, refers to the recurring pattern of conspiratorial thinking that attaches to emerging technologies such as cryptocurrency and AI. Often rooted in legitimate governance concerns, such thinking can escalate into disinformation ecosystems, creating tensions within technology policy, AI safety debates, and open-source governance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:conspiratorial-thinking-in-technology-communities", "labels": [ "Conspiratorial Thinking in Technology Communities", "Conspiracies" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "martin2022dark", "va2010neoconservatism", "Blockchain", "California AI bill", "Politics, Law, Privacy", "Safety and alignment" ] }, { "id": "constant-product-formula", "title": "Constant Product Formula", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The constant product formula is the pricing rule used by many automated market makers in which the product of the two pooled token reserves is held constant during trades.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:constant-product-formula", "labels": [ "Constant Product Formula", "Constant Product Invariant", "Pricing Formula" ], "is_subclass_of": [ "Automated Market Maker" ], "wikilinks": [ "Liquidity Pool", "Uniswap", "Automated Market Maker" ] }, { "id": "constitutional-ai-language-model-family", "title": "Constitutional AI Language Model Family", "domain": "ai", "domain_name": "Ai", "definition": "Anthropic Claude is a family of large language models developed by Anthropic PBC, designed around Constitutional AI principles that prioritise safety, helpfulness, and honesty. The model series spans Claude Instant, Claude 2, Claude 3 (Haiku, Sonnet, Opus), and Claude 3.5, with each generation advancing reasoning, coding, and multilingual capabilities. Claude is distinguished by its long context windows (up to 200k tokens in Claude 3) and its training methodology that emphasises harmlessness through RLHF and Constitutional AI feedback loops.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:constitutional-ai-training-methodology-language-model-family", "labels": [ "Constitutional AI Language Model Family" ], "is_subclass_of": [ "Proprietary Large Language Models" ], "wikilinks": [] }, { "id": "constitutional-ai-training-methodology", "title": "Constitutional AI Training Methodology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A method for training AI assistants to be harmless through self-improvement, using a set of principles or \"constitution\" to guide behaviour without human labels for harmful outputs. Constitutional AI combines supervised learning for self-critiques and revisions with Reinforcement Learning from AI Feedback (RLAIF), producing models that are helpful, harmless, and honest at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:constitutional-ai-training-methodology", "labels": [ "Constitutional AI Training Methodology", "Constitutional AI" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Computer Vision", "Constitutional AI", "MetaverseDomain" ] }, { "id": "constitutional-ai", "title": "Constitutional AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An alignment training method, introduced by Anthropic in 2022, in which a language model is steered by an explicit written set of principles (a constitution) rather than solely by human preference labels: the model critiques and revises its own outputs against the principles during supervised learning, then a preference model trained on AI-generated comparisons (RLAIF) provides the reinforcement signal, yielding harmlessness that is transparent, auditable, and scalable with far less human labelling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:constitutional-ai", "labels": [ "Constitutional AI" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [ "AI Alignment", "RLHF", "Large Language Model", "AI Safety Research" ] }, { "id": "constitutional-principle", "title": "Constitutional Principle", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Constitutional Principle in AI alignment is a natural-language rule or norm\u2014derived from human rights frameworks, professional codes, or organisational policies\u2014that is embedded into an AI system's training or inference process to constrain its behaviour across diverse contexts. Constitutional AI, introduced by Anthropic, uses a curated list of such principles as a self-critique scaffold during reinforcement learning from AI feedback (RLAIF), enabling large language models to evaluate and revise their own outputs against explicit normative standards without requiring a human rater for every example. Constitutional Principles serve as the value-bearing component of this approach, operationalising abstract ethical commitments into verifiable behavioural constraints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:constitutional-principle", "labels": [ "Constitutional Principle" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Blockchain" ] }, { "id": "constrained-decoding", "title": "Constrained Decoding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of inference-time techniques that restrict the token choices of a language model during generation so that output is guaranteed to satisfy formal constraints \u2014 a JSON schema, a context-free grammar, a regular expression, or required lexical content. At each decoding step the sampler masks tokens that would violate the constraint, typically by intersecting the model's next-token distribution with the valid transitions of a compiled automaton, yielding syntactically valid structured output without retraining the model.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:constrained-decoding", "labels": [ "Constrained Decoding" ], "is_subclass_of": [ "Text Generation" ], "wikilinks": [ "Text Generation", "Structured Output", "Beam Search" ] }, { "id": "constraint-based-design", "title": "Constraint Based Design", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Constraint based design is an engineering and AI methodology in which the valid solution space is explicitly defined by a set of constraints \u2014 physical laws, geometric relationships, functional requirements, regulatory bounds, manufacturing limits, or performance thresholds \u2014 that any acceptable design must simultaneously satisfy. Constraint solvers, optimisation algorithms, and AI planners traverse or prune this feasible region to discover configurations meeting all constraints, optionally optimising an objective within the feasible space. The method is foundational to parametric CAD, generative design tools, topology optimisation, robotics motion planning, and cyber-physical system validation, and is increasingly integrated with machine learning to learn constraint representations from data and guide search with neural heuristics.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:constraint-based-design", "labels": [ "Constraint Based Design", "Constraint-Based Design" ], "is_subclass_of": [ "AI Technique", "Constraint Satisfaction", "AI System Component", "Symbolic AI" ], "wikilinks": [ "AI System Component", "Autonomous Robot", "Digital Twin", "Generative Design Tool", "Parametric Modeling", "Topology Optimization", "Constraint Satisfaction", "Optimization Algorithm", "Formal Verification", "Cyber Physical Systems", "Simulation", "Robotics", "Motion Planning", "SAT Solver", "Knowledge Representation", "Symbolic AI", "Product Design", "Additive Manufacturing", "Finite Element Analysis", "Computer Aided Design" ] }, { "id": "constraint-propagation", "title": "Constraint Propagation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An inference technique in constraint satisfaction that repeatedly applies local consistency rules to shrink the domains of variables, eliminating values that cannot participate in any solution before or during search. By propagating the logical consequences of each constraint through the constraint network, it prunes the search space dramatically, often exposes infeasibility early without any backtracking, and turns otherwise intractable combinatorial problems into practically solvable ones.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:constraint-propagation", "labels": [ "Constraint Propagation" ], "is_subclass_of": [ "Constraint Satisfaction" ], "wikilinks": [ "Constraint Satisfaction", "Constraint Solver", "Arc Consistency", "Backtracking Search" ] }, { "id": "constraint-satisfaction", "title": "Constraint Satisfaction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Constraint Satisfaction is a paradigm in artificial intelligence and combinatorial mathematics in which a problem is represented as a set of variables, each with a domain of possible values, and a set of constraints that restrict the allowable combinations of those values. The goal is to find an assignment of values to all variables such that every constraint is simultaneously satisfied, or to determine that no such assignment exists. Solution methods combine systematic backtracking search with constraint propagation techniques \u2014 notably arc consistency and path consistency \u2014 that prune infeasible values early, dramatically reducing the search space. Constraint satisfaction underpins scheduling, configuration, planning, and combinatorial optimisation across virtually every engineering domain.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:constraint-satisfaction", "labels": [ "Constraint Satisfaction", "Constraint Solving" ], "is_subclass_of": [ "AI Technique", "Combinatorial Optimisation", "Symbolic AI" ], "wikilinks": [ "Artificial Intelligence", "Combinatorial Optimisation", "Symbolic AI", "Search Algorithm", "Arc Consistency", "Backtracking Search", "Constraint Propagation", "Constraint Solver", "Variable Ordering Heuristic", "Local Search", "Graph Theory", "Propositional Logic", "Satisfiability", "Integer Programming", "Linear Programming", "Probabilistic Inference", "Automated Planning", "Planning and Scheduling", "Configuration Management", "Resource Allocation" ] }, { "id": "constraint-solver", "title": "Constraint Solver", "domain": "ai", "domain_name": "Ai", "definition": "A constraint solver is a computational engine that finds assignments of values to variables such that all specified constraints \u2014 mathematical relationships, logical predicates, or physical laws \u2014 are simultaneously satisfied, drawing on techniques from constraint programming, SAT/SMT solving, linear programming, and numerical methods. Solvers operate by propagating constraint implications to prune the search space, applying backtracking or branch-and-bound search, and invoking domain-specific inference procedures that make the infeasibility of partial assignments detectable early. They are applied across planning and scheduling, combinatorial optimisation, formal verification, physics simulation, computer-aided design, and robotic motion planning.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:constraint-solver", "labels": [ "Constraint Solver", "Parametric Constraint Solving" ], "is_subclass_of": [ "Optimization Algorithm", "Inference Engine", "Combinatorial Optimisation", "Symbolic AI" ], "wikilinks": [ "Constraint Satisfaction", "Constraint Propagation", "Backtracking Search", "Arc Consistency", "Search Algorithm", "Automated Planning", "Motion Planning", "Planning and Scheduling", "Formal Verification", "Combinatorial Optimisation", "Logic Programming", "Satisfiability", "SMT Solving", "Linear Programming", "Physics Simulation", "Computer-Aided Design", "Heuristic Search", "Neural Network", "Constraint Based Design", "Trajectory Planning" ] }, { "id": "constraint-specification", "title": "Constraint Specification", "domain": "robotics", "domain_name": "Robotics", "definition": "A constraint specification is a formal statement of the limits and conditions a control system or plan must satisfy, such as joint limits, actuator bounds, obstacle avoidance, and safety envelopes. It is supplied to control and optimisation algorithms so that generated commands remain feasible and safe. Precise constraint specification is essential for model-based and optimisation-based control design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:constraint-specification", "labels": [ "Constraint Specification", "Declarative Constraint Specification" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "constraint", "title": "Constraint", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Constraint is a condition, restriction, or rule that limits the set of permissible states, actions, or solutions within a computational, logical, or physical system. Constraints formalise requirements such as resource bounds, logical invariants, safety properties, and optimality criteria, and are manipulated by constraint-satisfaction and optimisation algorithms to find feasible or optimal solutions. They appear across AI planning, machine learning regularisation, smart-contract execution, and formal verification.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:constraint", "labels": [ "Constraint", "Brightness Constancy Constraint" ], "is_subclass_of": [ "AI Technique", "Knowledge Representation", "Formal Method" ], "wikilinks": [ "Artificial Intelligence", "Blockchain", "Digital Twin", "Objective", "Model", "Formal Verification", "Reasoning", "Safety", "Algorithm", "Search Algorithm", "Inference", "Machine Learning Discipline", "Fairness", "Ontology", "Parameter", "Property", "System", "Smart Contract", "Neural Network", "Reinforcement Learning" ] }, { "id": "construction-digital-twin", "title": "Construction Digital Twin", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An integrated 3D model of built assets synchronized with real-time construction, operational, and maintenance data, enabling lifecycle management from design through decommissioning.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:construction-digital-twin", "labels": [ "Construction Digital Twin" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin" ], "wikilinks": [ "Asset Database", "BIM Model", "BSI Digital Built Britain", "Building Information Modeling", "Cloud Platform", "Construction Data", "Energy Management System", "Energy Optimization", "Facility Management System", "IoT Infrastructure", "ISO 23247", "Lifecycle Management", "Maintenance Schedule", "Real-time Synchronization", "Smart Building Ecosystem", "Space Planning", "ApplicationLayer", "Autonomous Robot", "BIM Software", "DataLayer" ] }, { "id": "consumer-ai-adoption", "title": "Consumer AI Adoption", "domain": "economics", "domain_name": "Economics", "definition": "The process and metrics by which individual users integrate artificial intelligence tools into their daily workflows, driving demand for consumer-facing AI products and services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:consumer-ai-adoption", "labels": [ "Consumer AI Adoption" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "consumer-ai-market", "title": "Consumer AI Market", "domain": "economics", "domain_name": "Economics", "definition": "The commercial sector focused on delivering artificial intelligence products and services directly to individual end-users rather than enterprise clients.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:consumer-ai-market", "labels": [ "Consumer AI Market" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "consumer-price-index", "title": "Consumer Price Index", "domain": "finance", "domain_name": "Finance", "definition": "The Consumer Price Index (CPI) is a statistical measure that tracks the average change over time in the prices paid by households for a representative basket of consumer goods and services. It is the principal indicator used to quantify inflation, adjust wages and benefits, and inform monetary policy. The index is computed by weighting price observations according to typical household expenditure patterns and rebasing them against a reference period.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:consumer-price-index", "labels": [ "Consumer Price Index" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "consumer-protection", "title": "Consumer Protection", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Body of law, regulation, and enforcement mechanisms protecting individuals from unfair, deceptive, or abusive commercial practices \u2014 spanning statutory rights, enforcement agencies, complaint and redress processes, and product liability standards \u2014 now substantially reshaped by digital markets, A...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:consumer-protection", "labels": [ "Consumer Protection", "BC-0489-consumer-protection", "Consumer Safety" ], "is_subclass_of": [ "Legal and Regulatory", "Regulation", "Market Regulation", "Digital Regulation", "Platform Governance", "Product Liability", "AI Governance", "Competition Law" ], "wikilinks": [ "ADR Schemes", "AI Act", "AI Act Product Safety", "AI Product Safety", "AIGovernanceDomain", "Algorithmic Accountability Rules", "Algorithmic Auditing", "Algorithmic Transparency", "ASA", "Behavioural Economics", "CFPB", "CMA", "CMA Strategic Market Status", "Co-regulation", "Collective Proceedings", "Collective Redress", "Competition Appeal Tribunal", "Competition Law", "Complaint Analytics", "Complaint Infrastructure" ] }, { "id": "consumer-trust", "title": "Consumer Trust", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Consumer trust is the confidence buyers place in a brand, product, or supply chain to behave reliably, safely, and ethically. It is built through verifiable provenance, transparent practices, consistent quality, and accountable handling of grievances. In supply-chain contexts it is increasingly underpinned by traceability data and ethical-sourcing attestations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:consumer-trust", "labels": [ "Consumer Trust" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "contact-centre-as-a-service", "title": "Contact Centre as a Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Contact Centre as a Service (CCaaS) is a cloud-delivered model that provides the software for managing inbound and outbound customer interactions across voice, chat, email, and messaging channels. It removes the need for on-premises telephony infrastructure by offering routing, queuing, agent tooling, and analytics on a subscription basis. Modern CCaaS platforms increasingly embed AI for self-service, transcription, and agent assistance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:contact-centre-as-a-service", "labels": [ "Contact Centre as a Service" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "contact-mechanics", "title": "Contact Mechanics", "domain": "robotics", "domain_name": "Robotics", "definition": "Contact mechanics is the study of the forces, deformations, and motions that arise when solid bodies touch and interact at their surfaces. It models phenomena such as normal contact forces, friction, adhesion, and local deformation, and provides the constitutive laws that govern how bodies push against, stick to, and slide over one another. In robotics it is essential for grasping, manipulation, locomotion, and physical simulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:contact-mechanics", "labels": [ "Contact Mechanics" ], "is_subclass_of": [ "Physics Simulation" ], "wikilinks": [] }, { "id": "container-image", "title": "Container Image", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An immutable, layered filesystem bundle packaging an application together with its runtime, libraries, and configuration metadata, from which container instances are created. Defined by the OCI Image Specification as content-addressed layers plus a manifest and configuration, images are built once, distributed through registries, and executed identically on any compliant runtime, providing the reproducible unit of software delivery that underpins containerised infrastructure from cloud clusters to edge devices.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:container-image", "labels": [ "Container Image" ], "is_subclass_of": [ "Containerisation" ], "wikilinks": [ "Containerisation", "Container", "Container Registry", "Docker Containerisation Platform", "Edge Computing" ] }, { "id": "container-orchestration", "title": "Container Orchestration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Container orchestration is the automated management of the deployment, scaling, networking, and lifecycle of containerised workloads across a cluster of machines. An orchestrator continuously reconciles the observed state of the cluster with a declarative desired state, handling scheduling, health checking, self-healing, and rolling updates. It abstracts the underlying hosts into a single pool of compute, enabling resilient, horizontally scalable services without manual intervention.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:container-orchestration", "labels": [ "Container Orchestration" ], "is_subclass_of": [ "Orchestration" ], "wikilinks": [] }, { "id": "container-registry", "title": "Container Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A content-addressed storage and distribution service for container images that implements the OCI Distribution Specification's push, pull, and discovery API. Registries hold image layers and manifests under named, tagged repositories, enforce authentication and access control, and increasingly store signatures, SBOMs, and other supply-chain artefacts alongside images, making them the central hand-off point between build pipelines and every runtime that deploys containerised software.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:container-registry", "labels": [ "Container Registry" ], "is_subclass_of": [ "Containerisation" ], "wikilinks": [ "Containerisation", "Container Image", "Docker Containerisation Platform", "Model Registry" ] }, { "id": "container-runtime", "title": "Container Runtime", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A container runtime is the software responsible for running containers on a host: pulling and unpacking images, configuring isolation and resource limits, and starting, stopping and supervising container processes. Runtimes operate at low and high levels, from minimal process launchers to daemons that manage image lifecycles, and present standard interfaces consumed by orchestrators such as Kubernetes. Container runtimes provide the execution foundation for portable, lightweight, cloud-native workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:container-runtime", "labels": [ "Container Runtime" ], "is_subclass_of": [ "Cloud Native" ], "wikilinks": [] }, { "id": "container", "title": "Container", "domain": "ai", "domain_name": "Ai", "definition": "A container is a lightweight, isolated runtime package that bundles an application together with its dependencies, libraries and configuration so it runs consistently across environments. Containers share the host operating system kernel while using namespaces and control groups for isolation, making them far more efficient than full virtual machines. They are the standard unit of deployment for machine-learning services and microservices.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:container", "labels": [ "Container" ], "is_subclass_of": [ "Infrastructure", "AI Infrastructure (Artificial Intelligence)", "Cloud Native", "Deployment Artifact" ], "wikilinks": [ "Containerization", "Control Groups", "Namespaces", "Resource Isolation", "Microservices", "Model Deployment", "MLOps", "CI/CD", "Orchestration", "Cloud Infrastructure", "Kubernetes", "DevOps", "Serverless", "Virtualization", "Deployment Artifact", "Docker Containerisation Platform", "Container Image", "Container Registry", "Container Runtime", "Open Container Initiative" ] }, { "id": "containerisation", "title": "Containerisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Containerisation is an operating-system-level virtualisation technique that packages an application together with its dependencies, libraries, and configuration into a single portable, isolated unit called a container. Containers share the host kernel yet maintain isolated user spaces, making them far lighter than full virtual machines while remaining reproducible across environments. The approach underpins cloud-native software delivery, providing consistent runtime behaviour from a developer's laptop to production clusters.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:containerisation", "labels": [ "Containerisation", "Containerization" ], "is_subclass_of": [ "Virtualisation", "Container Runtime" ], "wikilinks": [] }, { "id": "content-addressing", "title": "Content Addressing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The fundamental principle (R1 in ADR-013) that derives a resource's URI deterministically from its Content Hash|cryptographic content hash (SHA-256), ensuring immutability, tamper-detection, and deduplication, enabling VisionClaw Agentic Container|VisionClaw artefacts (credentials, receip...", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:content-addressing", "labels": [ "Content Addressing", "Content-Based Addressing", "ContentAddressing" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "ADR-013", "AgenticSystemsDomain", "Blockchain Hash Functions", "Canonical JSON", "Content-Addressed Storage Principles", "Content Hash", "Content Hash", "Cryptographic Hash Function", "DataIntegritydomain", "Decentralised Storage", "Deduplication", "Deterministic Encoding", "Deterministic Serialisation", "Distributed Hash Table", "DistributedSystemsDomain", "Git Object Addressing", "Git Objects", "Hash Minting", "IdentifierLayer", "IPFS Content Addressing" ] }, { "id": "content-authentication", "title": "Content Authentication", "domain": "security", "domain_name": "Security", "definition": "Content Authentication is the process of verifying the origin, integrity, and provenance of digital content through cryptographic techniques such as digital signatures, watermarking, and blockchain-anchored certificates. It enables consumers and platforms to distinguish authentic content from synthetically generated or tampered media, particularly critical as generative AI proliferates.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:content-authentication", "labels": [ "Content Authentication" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "content-authenticity", "title": "Content Authenticity", "domain": "security", "domain_name": "Security", "definition": "Content authenticity is the property of a digital asset whereby its origin, creation history, and any subsequent modifications can be cryptographically verified and traced back to an identifiable source. It encompasses technical mechanisms \u2014 including cryptographic signatures, tamper-evident manifests, and provenance metadata \u2014 that allow consumers of media to assess whether content has been created or manipulated by humans or AI systems and whether it has been altered since its stated point of capture or creation. Content authenticity is increasingly codified through standards such as C2PA (Coalition for Content Provenance and Authenticity).", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:content-authenticity", "labels": [ "Content Authenticity", "AI Content Authenticity", "Authenticity Checking", "Authenticity Tracking", "Content Authenticity Initiative", "Open Content Authenticity Initiative" ], "is_subclass_of": [ "Content Authentication" ], "wikilinks": [] }, { "id": "content-creation-pipeline", "title": "Content Creation Pipeline", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Content Creation Pipeline is a structured, orchestrated workflow that sequences and parallelises the stages of producing digital content\u2014spanning ideation, scripting, asset generation, editorial review, quality assurance, and multi-platform distribution\u2014into a reproducible and scalable production system. Modern pipelines integrate AI generation models, digital asset management systems, localisation engines, and publishing endpoints to deliver high-throughput content production with traceable asset provenance. They are the operational backbone of film VFX studios, game development, streaming platforms, and enterprise marketing operations, governing how raw creative intent is transformed into finished, distributed artefacts. Pipeline architecture choices directly determine release velocity, brand consistency, and the cost structure of large-scale content organisations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:content-creation-pipeline", "labels": [ "Content Creation Pipeline", "Content Authoring Pipeline" ], "is_subclass_of": [ "Content Creation" ], "wikilinks": [] }, { "id": "content-creation-tool", "title": "Content Creation Tool", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A content creation tool is software that enables users to author, edit, and produce digital media such as text, images, audio, 3D assets, or video. It provides the interfaces and primitives needed to turn creative intent into finished, exportable artefacts. Generative AI has expanded this category to include tools that synthesise content from prompts and steer it within a production pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:content-creation-tool", "labels": [ "Content Creation Tool" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "content-creation", "title": "Content Creation", "domain": "ai", "domain_name": "Ai", "definition": "Content Creation is the structured process of conceiving, producing, and distributing communicative artefacts\u2014including text, imagery, audio, video, and interactive media\u2014designed to inform, entertain, educate, or persuade defined audiences. It spans the complete production lifecycle from ideation, research, and scripting through asset generation, editing, quality assurance, and multichannel distribution. In AI-augmented workflows, generative models assist at every stage, lowering production costs and accelerating iteration cycles while demanding rigorous editorial oversight for accuracy and brand consistency. Effective content creation integrates creative vision, platform-specific optimisation, audience psychology, and provenance management to maximise reach, trust, and impact.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:content-creation", "labels": [ "Content Creation", "Algorithmic Content Creation" ], "is_subclass_of": [ "Digital Content", "Media Production", "Digital Content Creation" ], "wikilinks": [ "Generative AI", "Creative AI", "Large Language Model", "Natural Language Generation", "Text-to-Image Model", "Synthetic Media", "Creator Economy", "Digital Asset Management", "Content Moderation", "Digital Marketing", "Content Strategy", "Content Discovery", "Editorial Workflow", "Brand Identity", "Audience Engagement", "Search Engine Optimisation", "Cloud Computing", "Content Delivery Network", "Knowledge Management", "Diffusion Model" ] }, { "id": "content-curation", "title": "Content Curation", "domain": "data", "domain_name": "Data", "definition": "The practice of discovering, selecting, contextualising, and organising existing content for a particular audience, in contrast to producing new material. Content curation adds value through judgement\u2014filtering signal from an overwhelming volume of published media, sequencing items into coherent collections, and annotating them with commentary or provenance\u2014whether performed by human editors, algorithmic recommendation systems, or hybrid pipelines that combine both.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:content-curation", "labels": [ "Content Curation" ], "is_subclass_of": [ "Information Management" ], "wikilinks": [ "Content Creation", "Information Management", "Recommendation System" ] }, { "id": "content-delivery-network-cdn", "title": "Content Delivery Network (CDN)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A geographically distributed network of proxy servers and data centers designed to provide high availability, high performance, and low latency content delivery by caching content closer to end-users.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:content-delivery-network-cdn", "labels": [ "Content Delivery Network (CDN)", "CDN Distribution" ], "is_subclass_of": [ "Network and Communication", "Network Infrastructure" ], "wikilinks": [ "Cache System", "Data Center", "DDoS Protection", "DNS Service", "Edge Server", "ETSI GR ARF 010", "EWG/MSF taxonomy", "Geographic Redundancy", "Internet Service Provider", "Load Balancer", "Low-Latency Content Delivery", "Origin Server", "Routing Protocol", "Scalable Distribution", "Storage System", "Computer Vision", "Infrastructure Domain", "Network Infrastructure", "Network Layer", "Network Protocol" ] }, { "id": "content-delivery-network", "title": "content delivery network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Content Delivery Network (CDN) is a geographically distributed system of edge servers and Points of Presence (PoPs) that cache, replicate, and serve web assets \u2014 static files, media streams, and increasingly dynamic API responses \u2014 from locations physically near end users, substantially reducing round-trip latency and offloading origin-server load. CDNs use anycast routing, DNS-based request steering, and intelligent load balancing to direct each client request to the nearest healthy PoP, with cache coherence governed by HTTP cache-control semantics (RFC 9111) and conditional request mechanisms. Modern CDNs have evolved into full edge-compute platforms, offering TLS termination, DDoS mitigation, Web Application Firewall (WAF) services, and serverless edge runtimes (e.g. Cloudflare Workers, Fastly Compute) that execute application logic at PoPs without origin round-trips, blurring the boundary between network infrastructure and distributed application hosting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:content-delivery-network", "labels": [ "Content Delivery Network", "ContentDeliveryNetwork" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "content-delivery", "title": "Content Delivery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Content delivery is the distribution of digital media such as web pages, video and software to end users, typically optimised for speed and reliability. It is commonly accelerated by content delivery networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:content-delivery", "labels": [ "Content Delivery", "Adaptive Content Delivery", "Augmented Reality Content Delivery", "Dynamic Content Delivery", "Global Content Delivery", "Streaming Content Delivery" ], "is_subclass_of": [ "Network and Communication", "owl:Thing" ], "wikilinks": [ "Video Compression", "Content Delivery Network", "owl:Thing" ] }, { "id": "content-discovery", "title": "Content Discovery", "domain": "ai", "domain_name": "Ai", "definition": "Content Discovery encompasses the mechanisms, algorithms, and systems by which users or automated agents locate, surface, and retrieve relevant digital content from large-scale repositories or networks. It spans search-engine indexing, recommendation algorithms, semantic retrieval, and social curation, and is increasingly driven by machine learning models that personalise results to individual preference signals. Effective content discovery is foundational to user experience across the web, streaming platforms, knowledge graphs, and decentralised content networks. It sits at the intersection of information retrieval theory, personalisation engineering, and data governance.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:content-discovery", "labels": [ "Content Discovery" ], "is_subclass_of": [ "Information Retrieval", "Machine Learning Discipline", "Knowledge Management" ], "wikilinks": [ "Information Retrieval", "Recommendation Systems", "Search Engine", "Semantic Search", "Natural Language Processing", "Machine Learning Discipline", "Algorithmic Bias", "Algorithmic Accountability", "Algorithmic Transparency", "Digital Curation Platform", "Discovery Layer", "Content Creation", "Transformer Architecture", "Deep Learning", "Large Language Model", "Knowledge Graph", "Personalisation", "Filter Bubble", "Collaborative Filtering", "Content-Based Filtering" ] }, { "id": "content-distribution", "title": "Content Distribution", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Content Distribution encompasses the technical and logistical systems that deliver digital media\u2014web pages, video, software, and other assets\u2014from origin servers to end users at scale, minimising latency and maximising availability by replicating and caching content geographically close to consumers. It is realised primarily through Content Delivery Networks (CDNs) that operate distributed edge node infrastructure, but also includes peer-to-peer protocols, adaptive bitrate streaming, and decentralised storage networks as alternative or complementary mechanisms. Effective content distribution is the foundational infrastructure of the commercial internet, underpinning e-commerce, streaming media, and cloud-native application delivery.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:content-distribution", "labels": [ "Content Distribution", "Content Distribution Platform", "ContentDistribution" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "content-generation", "title": "Content Generation", "domain": "ai", "domain_name": "Ai", "definition": "Content Generation is the application domain within [[Generative AI]] and [[AI Machine Learning]] concerned with automatically synthesising novel media artefacts \u2014 including text, images, code, audio, video, and mixed-modality outputs \u2014 from structured or natural-language inputs, using probabilistic generative models that have learned the distribution of training corpora. It encompasses [[Large Language Model]]-based text and code generation, [[Diffusion Model]]-based image and video synthesis, multimodal fusion pipelines, and the wider infrastructure for deployment, evaluation, and governance of such systems at scale.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:content-generation", "labels": [ "Content Generation", "AI Content Generation", "Audio Content Generation", "Dynamic Content Generation" ], "is_subclass_of": [ "Generative AI", "AI Application", "AI Application Domain", "Creative AI" ], "wikilinks": [ "Generative AI", "AI Machine Learning", "Large Language Model", "Diffusion Model", "Transformer Architecture", "Foundation Model", "Natural Language Generation", "Multimodal AI", "Generative Adversarial Network", "Reinforcement Learning from Human Feedback", "Prompt Engineering", "Fine-Tuning", "Retrieval-Augmented Generation", "Text-to-Image", "Text-to-Video", "Code Synthesis", "Synthetic Media", "Content Moderation", "AI Ethics", "Intellectual Property" ] }, { "id": "content-identifier", "title": "Content Identifier", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Content Identifier (CID) is a self-describing, cryptographically derived label used in the InterPlanetary File System and IPLD ecosystem to uniquely identify and verify content through its hash, encoding the hash function used, the hash digest, and a multicodec descriptor for the serialised data format into a compact, version-aware multihash structure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:content-identifier", "labels": [ "Content Identifier", "Content Hash" ], "is_subclass_of": [ "Content Addressing" ], "wikilinks": [] }, { "id": "content-interoperability", "title": "Content Interoperability", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Content interoperability is the ability of digital assets to be created, exchanged, and rendered consistently across different platforms, engines, and tools without loss of fidelity. It depends on shared open formats, agreed schemas, and conformant import/export behaviour. In the metaverse it is the precondition for portable 3D assets, avatars, and scenes that work across virtual worlds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:content-interoperability", "labels": [ "Content Interoperability" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "content-layer", "title": "Content Layer", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The Content Layer is the stratum that holds the substantive information and media that a system manages and presents. It sits above the Data Layer that stores it and below the Presentation Layer that renders it. It contains documents, media assets, metadata, and the structures that organise meaning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:content-layer", "labels": [ "Content Layer" ], "is_subclass_of": [ "Content and Assets", "owl:Thing" ], "wikilinks": [ "Data Layer", "Presentation Layer", "Application Layer", "Content Management", "Metadata", "owl:Thing" ] }, { "id": "content-licensing", "title": "Content Licensing", "domain": "governance", "domain_name": "Governance", "definition": "Content Licensing is the legal and commercial framework through which rights holders grant third parties permission to use, reproduce, distribute, or monetise creative or informational works under defined conditions, terms, and compensation structures. It governs the relationship between creators, intermediary platforms, and end consumers across media, software, data, and digital asset categories.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:content-licensing", "labels": [ "Content Licensing", "Content Licensing Agreement" ], "is_subclass_of": [ "Digital Rights Management" ], "wikilinks": [] }, { "id": "content-management-system", "title": "Content Management System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Content Management System (CMS) is software that lets users create, store, organise, edit and publish digital content without requiring direct manipulation of underlying code. It separates content from presentation, providing authoring interfaces, versioning, access control and workflow so that non-technical contributors can manage websites and applications. Modern variants include headless systems that expose content through APIs for delivery across many front ends, including spatial and immersive interfaces.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:content-management-system", "labels": [ "Content Management System" ], "is_subclass_of": [ "Enterprise Software Platform" ], "wikilinks": [] }, { "id": "content-moderation-standards", "title": "Content Moderation Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The policies, technologies, and practices used to monitor, review, and regulate user-generated content and behavior within virtual environments and metaverse platforms, addressing challenges unique to immersive spaces including harassment, hate speech, and harmful conduct that require both tradit...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:content-moderation-standards", "labels": [ "Content Moderation Standards" ], "is_subclass_of": [ "Governance and Safety", "Platform Governance" ], "wikilinks": [ "AI Detection Systems", "Community Trust", "Harassment Prevention", "Human Moderators", "Moderation Tools", "metaverse", "Platform Governance", "Telecollaboration", "User Safety" ] }, { "id": "content-moderation", "title": "Content Moderation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Content Moderation is a systematic sociotechnical process for reviewing, classifying, filtering, and actioning user-generated and AI-generated media to enforce community standards, legal requirements, and platform policies, combining [[Machine Learning Models]], [[Natural Language Processing]], computer vision classifiers, and [[Human Moderators]] in hybrid pipelines that balance [[Harmful Content Prevention]] with [[Regulatory Compliance]] and freedom of expression, operating at scale under obligations imposed by frameworks including the [[EU AI Act]], the EU Digital Services Act, and the UK Online Safety Act.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:content-moderation", "labels": [ "Content Moderation", "Content Moderation Pipeline", "Moderation Infrastructure" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Safety", "Platform Governance System", "Responsible AI" ], "wikilinks": [ "Appeal Process", "Automated Filtering", "Community Guidelines Enforcement", "Content Analysis Tools", "Content Classification System", "Decision Framework", "ETSI GR ARF 010", "Harmful Content Prevention", "Human Moderators", "Human Review Workflow", "Moderation Policy", "Platform Governance System", "Policy Enforcement Engine", "Reporting System", "Reviewer Training Program", "Safe User Experience", "Trust and Safety Infrastructure", "Application Layer", "Community Standards", "Machine Learning Models" ] }, { "id": "content-monetisation", "title": "Content Monetisation", "domain": "governance", "domain_name": "Governance", "definition": "Content monetisation is the set of mechanisms by which creators and rights-holders earn revenue from digital media, including advertising, subscriptions, micropayments, licensing and royalty distribution. Governance of these mechanisms covers rights attribution, usage tracking and the fair allocation of proceeds among contributors. Emerging models use programmable payments and on-chain royalties to automate compensation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:content-monetisation", "labels": [ "Content Monetisation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "content-pipeline", "title": "Content Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Content Pipeline is the end-to-end automated or semi-automated workflow that transforms raw creative assets\u2014geometry, textures, audio, video, or data\u2014from authoring tools into the optimised, platform-specific formats required by a runtime engine, distribution system, or media player. It encompasses ingestion, validation, processing, compression, and delivery stages with dependency tracking and incremental rebuild capabilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:content-pipeline", "labels": [ "Content Pipeline" ], "is_subclass_of": [ "Production Pipeline" ], "wikilinks": [] }, { "id": "content-portability", "title": "Content Portability", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Content portability is the property that allows users to move their digital content and the associated rights between services, platforms, or virtual environments. It depends on standardised data formats and export mechanisms so that assets retain structure and meaning after transfer. It supports user ownership, reduces vendor lock-in, and is a stated goal of open data and open metaverse initiatives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:content-portability", "labels": [ "Content Portability", "XR Content Portability" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "content-production-workflow", "title": "Content Production Workflow", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured sequence of stages and toolchain integrations through which raw creative inputs are transformed into publishable digital assets, covering pre-production, asset authoring, review, rendering, and distribution. In spatial computing contexts, such workflows incorporate 3D asset pipelines, real-time rendering checks, and version-controlled delivery to metaverse platforms.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:content-production-workflow", "labels": [ "Content Production Workflow" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "content-protection", "title": "Content Protection", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Content Protection comprises technical and legal mechanisms that prevent unauthorised reproduction, distribution, or alteration of digital creative works. It includes digital rights management (DRM), encryption, access control, and blockchain-based provenance systems applied to media, software, and metaverse assets.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:content-protection", "labels": [ "Content Protection", "Content Protection Infrastructure" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "content-provenance", "title": "Content Provenance", "domain": "governance", "domain_name": "Governance", "definition": "Content provenance is the verifiable, machine-readable record of the origin, authorship, creation context, and transformation history of a piece of digital media, enabling downstream consumers and automated systems to establish authenticity and detect unauthorised alteration. It relies on cryptographic signing, structured metadata schemas, and tamper-evident manifests attached to or bound with the asset at point of creation. As generative AI proliferates synthetic media, content provenance serves as the principal technical mechanism for distinguishing camera-captured or human-authored material from algorithmically generated content. Governance frameworks and regulatory instruments increasingly mandate provenance disclosure as a baseline trust control for media distributed at scale.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:content-provenance", "labels": [ "Content Provenance" ], "is_subclass_of": [ "Provenance" ], "wikilinks": [ "Cryptography", "AI Governance", "Provenance" ] }, { "id": "content-repository", "title": "Content Repository", "domain": "data", "domain_name": "Data", "definition": "A structured store for digital content and its metadata that provides hierarchical or graph organisation, typed properties, versioning, access control, search, and observation APIs, decoupling how content is stored from the applications that author and deliver it \u2014 the persistence backbone of content management systems, digital curation platforms, and adaptive learning systems that assemble instruction from tagged learning objects.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:content-repository", "labels": [ "Content Repository" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [ "Data Storage", "Content Management System", "Metadata", "Adaptive Learning" ] }, { "id": "content-addressed-storage", "title": "Content-Addressed Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Content-addressed storage (CAS) is a data storage paradigm in which each piece of data is identified and retrieved by a cryptographic hash of its content rather than by its location or a human-assigned name. Because the identifier is derived deterministically from the data itself, identical content always maps to the same address, enabling automatic deduplication and verifiable integrity without requiring trust in the storage provider. Content-addressed storage forms the basis of distributed systems such as IPFS, Git, and Arweave, and underlies the content-integrity mechanisms of blockchain data layers. It is fundamentally different from location-addressed storage, where the same content can exist at multiple addresses or the same address can point to different content over time.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:content-addressed-storage", "labels": [ "Content-Addressed Storage", "Content Addressed Storage", "Content-Addressable Storage", "Content-Addressed Storage Principles" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "content-based-filtering", "title": "Content-Based Filtering", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A recommendation technique that suggests items by matching the attributes of items a user has previously engaged with \u2014 such as text features, genres, tags, or learned embeddings \u2014 against the attributes of candidate items, building a per-user preference profile in feature space; it requires no data about other users, handles new items gracefully, and offers explainable suggestions, but tends to over-specialise, recommending only items similar to what the user already knows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:content-based-filtering", "labels": [ "Content-Based Filtering" ], "is_subclass_of": [ "Recommendation Systems" ], "wikilinks": [ "Recommendation Systems", "Collaborative Filtering", "Content Discovery" ] }, { "id": "contestability", "title": "Contestability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Contestability is the property of a system or decision-making process that allows an affected party to challenge, question or seek review of an automated decision. It requires that the basis of a decision be sufficiently transparent to be examined, and that a mechanism exists for raising and resolving disputes. Contestability is a design and governance goal in AI accountability frameworks, complementing explainability and transparency by ensuring outcomes are not merely explained but also revisable.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:contestability", "labels": [ "Contestability" ], "is_subclass_of": [ "Accountability" ], "wikilinks": [] }, { "id": "context-aware-computing", "title": "Context Aware Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Context Aware Computing is a paradigm in which systems automatically adapt their behaviour based on contextual signals such as the user's location, activity, device state, social environment, and temporal cues. In spatial computing and XR applications, context awareness allows virtual overlays and services to respond intelligently to physical surroundings, enabling location-sensitive AR, adaptive user interfaces, and personalised immersive experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:context-aware-computing", "labels": [ "Context Aware Computing", "Context-Aware Computing" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "context-aware-response", "title": "Context Aware Response", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An AI-driven capability in metaverse and virtual environments that enables systems to generate dynamic, personalized responses based on real-time understanding of user context, including location, activity, preferences, social environment, and interaction history, utilizing large language models ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:context-aware-response", "labels": [ "Context Aware Response", "Context-Aware Response" ], "is_subclass_of": [ "Platform and Environment", "Artificial Intelligence" ], "wikilinks": [ "Adaptive UX", "Dynamic Content", "Personalized Interaction", "Artificial Intelligence", "Computer Vision", "Context Awareness System", "Machine Learning", "metaverse", "Natural Language Processing" ] }, { "id": "context-awareness-system", "title": "Context Awareness System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A technology framework that captures, processes, and interprets contextual information about users and their environments using IoT sensors, location tracking, and AI analytics to enable adaptive services, personalized experiences, and intelligent decision-making in virtual and physical spaces.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:context-awareness-system", "labels": [ "Context Awareness System" ], "is_subclass_of": [ "Platform and Environment", "Intelligent Systems" ], "wikilinks": [ "Adaptive Services", "Location Based Services", "Sensor Input", "Situational Awareness", "Data Analytics", "Intelligent Systems", "IoT Sensors", "metaverse", "Network Infrastructure" ] }, { "id": "context-awareness", "title": "Context Awareness", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The capability of computing systems to sense, interpret, and respond to environmental conditions, user state, situational factors, and contextual information to dynamically adapt behavior and deliver personalized experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:context-awareness", "labels": [ "Context Awareness" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Adaptive Computing", "Adaptive Interfaces", "Ambient Intelligence", "Behavioral Adaptation", "Context Modeling", "Decision Logic", "Environmental Sensing", "EWG/MSF Taxonomy", "IEEE P2048-3", "IoT Infrastructure", "ISO/IEC 30141", "Personalized Experiences", "Pervasive Computing", "Proactive Services", "Real-Time Analytics", "Semantic Reasoning", "Sensor Input", "Situational Inference", "Smart Environments", "User State Detection" ] }, { "id": "context-engineering", "title": "Context Engineering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Context Engineering is the systems-level discipline of designing, curating, compressing, and orchestrating the information delivered into a Large Language Model's context window across a multi-step agentic loop, popularised between mid-2024 and mid-2025 by Walden Yan (Cognition AI's \"Don't Build ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:context-engineering", "labels": [ "Context Engineering" ], "is_subclass_of": [ "AI Technique", "AI Agent System", "AI Engineering", "LLM Application Engineering", "Agent Engineering", "Systems Engineering for AI" ], "wikilinks": [ "Agent Engineering", "AgentRuntimeLayer", "Agentic AI", "AgenticSystemsDomain", "AI Engineering", "Anthropic 2024 Contextual Retrieval", "Anthropic 2024 Model Context Protocol Specification", "Anthropic Prompt Engineering Guide", "Asai et al 2024 Self-RAG", "Beltagy et al 2020 Longformer", "Breunig 2025 How Long Contexts Fail", "Browser Agents", "Chevalier et al 2023 AutoCompressors", "Claude Code CLAUDE.md Convention", "Coding Agents", "Context Compaction", "Context Compression", "Cost-Efficient Inference", "Customer Service Agents", "Evaluation Harness" ] }, { "id": "context-graph", "title": "Context Graph", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The assembly layer at the top of the knowledge stack. A context graph consumes the layers beneath it \u2014 a formal ontology, a populated knowledge graph, and operational sources such as documents, chat history and tool outputs \u2014 to assemble the working set of information an AI agent needs for its next response. It does not compete with the ontology or the knowledge graph; it selects from them under a token budget. In the DreamLab mesh this layer is implemented by the Ontology Loom, which retrieves a budget-clamped structured scaffold from the reasoned ontology and injects it into model context at query time.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:context-graph", "labels": [ "Context Graph" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Knowledge Graph", "Ontology", "Ontology Loom", "Retrieval-Augmented Generation", "Taxonomy" ] }, { "id": "context-management", "title": "Context Management", "domain": "ai", "domain_name": "Ai", "definition": "Context management is the set of techniques for assembling, prioritising, compressing, and maintaining the information supplied to a language model within its bounded context window across a task or conversation. It governs what prompts, retrieved documents, prior turns, tool outputs, and state are placed in context, in what order, and at what fidelity, so that the model has the most relevant evidence without exceeding token limits. Effective context management is central to retrieval-augmented generation, long-running agents, and conversational systems, where it directly shapes coherence, accuracy, and cost.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:context-management", "labels": [ "Context Management" ], "is_subclass_of": [ "Large Language Model", "Prompt Engineering" ], "wikilinks": [] }, { "id": "context-window-management", "title": "Context Window Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The set of architectural and algorithmic techniques used to extend the effective memory of large language models beyond their native token limits, enabling coherent processing of long-horizon inputs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:context-window-management", "labels": [ "Context Window Management" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "context-window", "title": "Context Window", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The maximum sequence length that a language model can process in a single forward pass, measured in tokens; it determines how much prior context the model can attend to during generation or understanding tasks and directly bounds memory, coherence, and long-range reasoning capabilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:context-window", "labels": [ "Context Window", "Extended Context Window", "Long Context Window" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Autonomous Robot", "Blockchain", "Large Language Models" ] }, { "id": "contextual-embedding", "title": "Contextual Embedding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A contextual embedding is a vector representation of a token or span whose value depends on the surrounding context in which it appears, in contrast to static embeddings that assign a fixed vector to each word regardless of usage. Contextual embeddings are produced by encoder models such as BERT, typically via masked language modelling objectives, and allow the same word to be represented differently depending on its sense in a given sentence. They substantially improved performance on downstream natural language understanding tasks by resolving ambiguity that static embeddings could not capture.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:contextual-embedding", "labels": [ "Contextual Embedding" ], "is_subclass_of": [ "Embedding" ], "wikilinks": [] }, { "id": "continual-learning", "title": "Continual Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Continual Learning is a machine learning paradigm in which a model sequentially learns from a non-stationary stream of tasks or data distributions while retaining competence on previously acquired knowledge. It directly confronts catastrophic forgetting \u2014 the tendency of neural networks to overwrite earlier representations when updated on new training distributions. Core strategies span regularisation-based protection of critical parameters, rehearsal-based replay of past exemplars or synthetic surrogates, and architectural methods that expand or isolate network capacity per task. The field spans both supervised and reinforcement learning settings and is fundamental wherever data arrives incrementally and retraining from scratch is computationally or commercially infeasible.", "entityType": "Class", "qualityScore": 0.91, "maturity": "emerging", "iri": "urn:ngm:class:continual-learning", "labels": [ "Continual Learning", "Continual Learner" ], "is_subclass_of": [ "Machine Learning", "AI Technique", "Online Learning", "Transfer Learning" ], "wikilinks": [ "Machine Learning", "Catastrophic Forgetting", "Neural Networks", "Transfer Learning", "Online Learning", "Neural Plasticity", "Gradient Descent", "Elastic Weight Consolidation", "Experience Replay", "Knowledge Distillation", "Generative Adversarial Networks", "Variational Autoencoders", "Progressive Neural Networks", "Meta-Learning", "Few-Shot Learning", "Reinforcement Learning", "Federated Learning", "Edge Computing", "Robotics", "Domain Adaptation" ] }, { "id": "continued-pre-training", "title": "Continued Pre Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An intermediate training phase in which a pre-trained foundation model undergoes additional unsupervised pre-training on domain-specific or task-relevant corpora before supervised fine-tuning. Continued pre-training bridges general-purpose representations and domain expertise, using the same self-supervised objectives as initial pre-training whilst employing reduced learning rates and selective data to mitigate catastrophic forgetting of general capabilities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:continued-pre-training", "labels": [ "Continued Pre Training" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "continuous-ai-strategy", "title": "Continuous AI Strategy", "domain": "governance", "domain_name": "Governance", "definition": "An iterative approach to AI planning where strategies and recommendations are dynamically updated in response to evolving capabilities and market conditions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:continuous-ai-strategy", "labels": [ "Continuous AI Strategy" ], "is_subclass_of": [ "Enterprise Ai" ], "wikilinks": [] }, { "id": "continuous-authentication", "title": "Continuous Authentication", "domain": "security", "domain_name": "Security", "definition": "Continuous authentication is an approach to identity verification that repeatedly re-validates a user's identity throughout a session, rather than relying solely on a single login event. It draws on signals such as behavioural biometrics, device posture, and network context to detect anomalies that might indicate session hijacking or credential theft. It is increasingly used alongside biometric authentication and identity providers to reduce the window of exposure after initial login.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:continuous-authentication", "labels": [ "Continuous Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "continuous-batching", "title": "Continuous Batching", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Continuous batching is a large language model serving technique in which the inference scheduler admits and evicts requests at the granularity of individual decoding steps rather than running a whole batch to completion before starting the next. As soon as any sequence in the batch finishes generating, its slot is freed and a queued request takes its place, keeping the GPU saturated and dramatically improving throughput and latency under heterogeneous request lengths. It is a defining feature of modern inference engines such as vLLM and is typically combined with paged attention and a shared key-value cache.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:continuous-batching", "labels": [ "Continuous Batching" ], "is_subclass_of": [ "Inference Optimisation", "Model Serving" ], "wikilinks": [ "Inference Optimisation", "Large Language Models", "KV Cache", "Paged Attention", "vLLM", "Autoregressive Decoding", "Model Serving", "GPU Compute", "GPU Memory", "Throughput", "Latency", "Token Generation", "Request Scheduling", "Transformer Architecture", "Attention Mechanism", "Speculative Decoding", "Tensor Parallelism", "Flash Attention", "Quantisation", "Batch Processing" ] }, { "id": "continuous-care-navigator", "title": "Continuous Care Navigator", "domain": "governance", "domain_name": "Governance", "definition": "A healthcare role that leverages AI-enabled systems to provide ongoing, personalized coordination and support for patients, bridging the gap between clinical interventions and daily health management.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:continuous-care-navigator", "labels": [ "Continuous Care Navigator" ], "is_subclass_of": [ "Healthcare" ], "wikilinks": [] }, { "id": "continuous-delivery", "title": "Continuous Delivery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Continuous Delivery is a software engineering discipline in which code changes are automatically built, tested and prepared for release to production so that the software is always in a deployable state. It extends continuous integration by adding automated release pipelines, environment promotion and deployment gates, allowing teams to ship changes rapidly, reliably and with low risk. Releases become routine, low-ceremony events rather than infrequent high-stakes operations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:continuous-delivery", "labels": [ "Continuous Delivery" ], "is_subclass_of": [ "DevOps" ], "wikilinks": [] }, { "id": "continuous-deployment", "title": "Continuous Deployment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Continuous Deployment is a software release practice in which every code change that passes the automated pipeline is released to production automatically, without manual approval gates. It extends continuous delivery by removing the final human decision step, so that a successful build, test, and integration sequence results directly in a live deployment. The practice depends on comprehensive automated testing, robust monitoring, and rapid rollback mechanisms to maintain reliability while sustaining a high deployment cadence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:continuous-deployment", "labels": [ "Continuous Deployment" ], "is_subclass_of": [ "DevOps" ], "wikilinks": [] }, { "id": "continuous-integration", "title": "Continuous Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Continuous Integration (CI) is a software engineering practice in which developers frequently merge code changes into a shared repository \u2014 typically multiple times per day \u2014 triggering automated build and test pipelines that provide rapid feedback on integration correctness. CI reduces the cost and risk of integration by detecting conflicts, regressions, and build failures early, and serves as the foundation of broader continuous delivery and DevOps workflows. Originating in Extreme Programming, the practice has become a cornerstone of modern software delivery, enabling teams to maintain a consistently releasable main branch through disciplined automation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:continuous-integration", "labels": [ "Continuous Integration", "Continuous Integration Continuous Delivery", "Continuous Integration and Delivery" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "continuous-monitoring", "title": "Continuous Monitoring", "domain": "governance", "domain_name": "Governance", "definition": "An ongoing, automated process of collecting, analysing, and reporting metrics, logs, and events from systems, processes, or environments in near-real time to detect anomalies, ensure compliance, and support rapid response. Continuous monitoring operationalises oversight by replacing periodic audits with persistent telemetry streams fed into dashboards, alert rules, and automated remediation workflows. It is fundamental to DevSecOps, regulatory compliance programmes, and risk management frameworks that require timely evidence of control effectiveness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:continuous-monitoring", "labels": [ "Continuous Monitoring" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "continuous-training", "title": "Continuous Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Continuous training is an MLOps practice in which machine-learning models are automatically retrained on fresh data on a recurring or event-driven basis to maintain predictive accuracy over time. It extends continuous integration and delivery to the model lifecycle, triggering retraining when new data arrives or when monitoring detects data or model drift. The retrained model is validated and promoted through automated pipelines before deployment.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:continuous-training", "labels": [ "Continuous Training" ], "is_subclass_of": [ "MLOps", "Machine Learning Operations", "AI Lifecycle" ], "wikilinks": [ "MLOps", "Model Training", "Data Drift", "Concept Drift", "Model Monitoring", "Model Deployment", "Model Serving", "Model Registry", "Feature Store", "Model Training Pipeline", "CI-CD Automation", "Continuous Deployment", "Continuous Integration", "Model Governance", "Experiment Tracking", "Data Pipeline", "Data Versioning", "Model Evaluation", "Kubeflow", "MLflow" ] }, { "id": "contract-code", "title": "Contract Code", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Contract code is the executable program logic that defines the rules, state, and behaviour of a smart contract deployed on a distributed ledger. It is typically written in a high-level language such as Solidity or Kotlin and compiled to bytecode that nodes execute deterministically. Because it is immutable once deployed and controls value, contract code is a primary target for auditing and formal verification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:contract-code", "labels": [ "Contract Code" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "contract-enforcement", "title": "Contract Enforcement", "domain": "governance", "domain_name": "Governance", "definition": "Contract enforcement is the set of legal, procedural, and technological mechanisms by which the obligations set out in an agreement are made binding and are upheld when a party fails to perform, ranging from courts and arbitration to automated on-chain execution. Traditional enforcement relies on dispute resolution processes and judicial remedies, whereas digital and blockchain-based approaches encode enforcement logic directly into self-executing code or arbitration protocols. Effective contract enforcement is a precondition for trust in both commercial transactions and decentralised digital agreements.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:contract-enforcement", "labels": [ "Contract Enforcement" ], "is_subclass_of": [ "Dispute Resolution" ], "wikilinks": [] }, { "id": "contract-net-protocol", "title": "Contract Net Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Contract Net Protocol is a task-sharing interaction pattern for multi-agent systems in which a manager agent announces a task, soliciting bids from potential contractor agents, evaluates the responses, and awards the task to the most suitable bidder. It decomposes distributed problem solving into the phases of announcement, bidding, awarding, and result reporting, treating the agent population as a market for negotiating the allocation of work. The protocol provides a decentralised mechanism for dynamic task allocation without a fixed assignment table.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:contract-net-protocol", "labels": [ "Contract Net Protocol" ], "is_subclass_of": [ "Artificial Intelligence", "Multi-Agent System" ], "wikilinks": [] }, { "id": "contract-theory", "title": "Contract Theory", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Contract theory is the branch of economics that studies how parties design agreements under asymmetric information, incentive conflicts, and incomplete contracting. It analyses problems such as moral hazard, adverse selection, and signalling to derive incentive-compatible contracts that align the interests of principals and agents. It provides the formal foundations for mechanism and incentive design, including in tokenised and decentralised systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:contract-theory", "labels": [ "Contract Theory" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "contractum-language", "title": "Contractum Language", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Contractum is a high-level smart-contract programming language designed for the RGB protocol, which performs client-side validation of state transitions anchored to Bitcoin. It compiles to AluVM bytecode and lets developers express RGB schemas and contract logic in a readable, declarative form. It is central to authoring confidential, scalable contracts that settle off-chain while inheriting Bitcoin's security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:contractum-language", "labels": [ "Contractum Language" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "contrastive-learning", "title": "Contrastive Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A self-supervised representation learning approach that trains models by comparing positive pairs (semantically similar samples) against negative pairs (dissimilar samples), pushing similar embeddings closer and dissimilar ones apart in latent space, enabling powerful feature learning without explicit labels.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:contrastive-learning", "labels": [ "Contrastive Learning" ], "is_subclass_of": [ "Self-Supervised Learning" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "control-algorithm", "title": "Control Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ControlAlgorithm is a formalised mathematical and computational procedure that generates actuator commands to drive a dynamic system from its current state toward a desired target state, exploiting feedback from sensors, an internal model of the plant, or learned approximations of system dynamics...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:control-algorithm", "labels": [ "Control Algorithm", "Control-Algorithm", "ControlAlgorithm" ], "is_subclass_of": [ "AI Technique", "Dynamical Systems", "Computational Intelligence", "Control Theory", "Feedback Control", "Optimal Control" ], "wikilinks": [ "Actuator Model", "Adaptation Law", "Aerospace", "AlgorithmLayer", "AutomationDomain", "Autonomous Systems", "Autonomous Vehicles", "Bang-Bang Control", "Chemical Process Control", "Computational Intelligence", "Constraint Specification", "ControlEngineeringDomain", "Convex Optimisation", "Cost Function", "Differential Equations", "DO-178C", "Dynamical Systems", "Dynamics", "EmbeddedSystemLayer", "Embedded Systems" ] }, { "id": "control-framework", "title": "Control Framework", "domain": "governance", "domain_name": "Governance", "definition": "A control framework is a structured set of governance, risk, and compliance controls that an organisation adopts to manage risk and demonstrate conformance to regulatory or industry requirements. Examples include NIST CSF, COBIT, ISO 27001, and SOC 2, each mapping objectives to specific control activities and evidence. It provides a common reference for designing, operating, and auditing controls consistently.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:control-framework", "labels": [ "Control Framework", "COSO Internal Control Framework" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "control-interface", "title": "Control Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "A control interface is the defined boundary through which commands and feedback are exchanged between a controller and a controlled device such as an actuator, end-effector, or robot subsystem. It specifies the signals, protocols, data rates, and timing required for deterministic command of the hardware. A well-defined control interface is what allows higher-level planners and digital twins to actuate physical systems reliably.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:control-interface", "labels": [ "Control Interface", "Remote Control Interface" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "control-law", "title": "Control Law", "domain": "robotics", "domain_name": "Robotics", "definition": "A control law is the mathematical rule that maps measured system state and reference signals to the control inputs applied to a plant. It is the algorithmic core of a feedback controller, expressed as functions such as PID, state feedback, or impedance laws. The choice of control law determines stability, tracking accuracy, and disturbance rejection of the closed-loop system.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:control-law", "labels": [ "Control Law" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "control-layer", "title": "Control Layer", "domain": "robotics", "domain_name": "Robotics", "definition": "The Control Layer is the stratum that issues commands to actuators or subsystems to drive a system toward desired states. It sits above the Perception Layer, on whose estimates it acts, and below planning and agent strata that set goals. It contains controllers, feedback loops, set-point logic, and the actuation interfaces that effect change.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:control-layer", "labels": [ "Control Layer" ], "is_subclass_of": [ "Actuation and Control", "owl:Thing" ], "wikilinks": [ "Perception Layer", "Agent Layer", "Coordination Layer", "Control Theory", "Feedback Loop", "owl:Thing" ] }, { "id": "control-loop", "title": "Control Loop", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A control loop is the cyclical process by which a system measures its current state, compares it against a desired set point and applies corrective action to reduce the difference. Closed-loop control uses feedback from sensors to continuously regulate an actuator, while open-loop control acts without such feedback. Control loops are foundational to automation, cyber-physical systems and the reconciliation pattern used in infrastructure orchestration.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:control-loop", "labels": [ "Control Loop" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "control-moment-gyroscope", "title": "Control Moment Gyroscope", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:control-moment-gyroscope", "labels": [ "Control Moment Gyroscope" ], "is_subclass_of": [ "Actuator" ], "wikilinks": [] }, { "id": "control-signal", "title": "Control Signal", "domain": "robotics", "domain_name": "Robotics", "definition": "A control signal is the command output computed by a controller and applied to actuate or steer a system toward a desired state. In robotics and electromechanical contexts it is the voltage, current, or set-point delivered to an actuator; in generative pipelines it is the spatial conditioning input that guides synthesis. In both senses it is the carrier of intent between a decision process and the system it drives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:control-signal", "labels": [ "Control Signal" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "control-system", "title": "Control System", "domain": "robotics", "domain_name": "Robotics", "definition": "Control system encompasses the computational and hardware subsystems that sense environmental state, evaluate performance against desired objectives, and generate actuation commands to regulate robot behaviour toward goals.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:control-system", "labels": [ "Control System", "Centralised Control System", "Control Systems", "Flight Control Systems", "Industrial Control System" ], "is_subclass_of": [ "Actuation and Control", "Robotic Systems", "Feedback Systems" ], "wikilinks": [ "Actuator Control", "Actuator Driver", "Actuator Output", "Actuators", "Adaptive Behaviour", "Communication Protocols", "Controller", "Feedback Systems", "Learning from Demonstration", "Model Predictive Control", "PID Control", "Real-Time Computation", "Real-Time Operating Systems", "Robot Perception", "Robotic Systems", "Sensor Input", "Sensor Interface", "Stabilisation", "Task Execution", "Timing Synchronisation" ] }, { "id": "control-theory", "title": "Control Theory", "domain": "robotics", "domain_name": "Robotics", "definition": "Control Theory is the interdisciplinary mathematical framework for analysing and synthesising dynamical systems that automatically regulate themselves toward desired behaviour in the presence of disturbances, model uncertainty, sensor noise, and parameter variation, formalising the closed-loop fe...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:control-theory", "labels": [ "Control Theory" ], "is_subclass_of": [ "Actuation and Control", "Control System", "Applied Mathematics", "Dynamical Systems Theory", "Systems Engineering", "Cybernetics", "Control Engineering" ], "wikilinks": [ "Actuator", "Aerospace Engineering", "AI Without Dynamics Models", "AlgorithmLayer", "Ames 2019 Control Barrier Functions", "Anderson Moore 1989 Optimal Control LQ Methods", "Applied Mathematics", "AppliedMathematicsDomain", "ARP4754A", "Astrom Wittenmark 1995 Adaptive Control", "Autonomous Vehicles", "Backstepping", "Bellman 1957 Dynamic Programming", "Bertsekas Tsitsiklis 1996 Neuro-Dynamic Programming", "Biomedical Engineering", "Bode 1945 Network Analysis Feedback Amplifier Design", "Bode Plot", "Classical Optimisation Without Feedback", "Closed-Loop Operation", "ControlAndAutomationDomain" ] }, { "id": "control-net-conditioning", "title": "ControlNet Conditioning", "domain": "ai", "domain_name": "Ai", "definition": "ControlNet conditioning is a technique that augments a pretrained diffusion model with an auxiliary network so generation can be steered by spatial control signals such as edge maps, depth, pose, or segmentation. The ControlNet branch copies the encoder of the base model and injects conditioning through zero-initialised connections, preserving the original weights while adding controllability. It gives image-generation pipelines precise structural control without retraining the base model.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:control-net-conditioning", "labels": [ "ControlNet Conditioning", "ControlNet Garment Conditioning", "ControlNet Guidance" ], "is_subclass_of": [ "Generative Model", "AI Technique", "Parameter-Efficient Fine-Tuning", "Adapter Tuning", "Spatial Conditioning", "Transfer Learning", "Fine-Tuning", "ControlNet" ], "wikilinks": [ "Diffusion Model", "Stable Diffusion", "Latent Diffusion", "U-Net", "Zero Convolution", "Classifier-Free Guidance", "ControlNet", "ControlNet and Similar Spatial Conditioning Systems", "Spatial Conditioning", "Conditioning Signal", "Edge Detection", "Canny Edge Detection", "Depth Estimation", "Pose Estimation", "Semantic Segmentation", "Image Generation", "Text-to-Image", "Conditional Image Generation", "Generative Model", "Fine-Tuning" ] }, { "id": "control-net-and-similar-spatial-conditioning-systems", "title": "ControlNet and Similar Spatial Conditioning Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ControlNet and similar spatial conditioning systems constitute a family of neural network architectures and adapter frameworks that augment large pretrained text-to-image Diffusion Models with fine-grained spatial and semantic control signals\u2014enabling deterministic structural guidance over ge...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:control-net-and-similar-spatial-conditioning-systems", "labels": [ "ControlNet and Similar Spatial Conditioning Systems" ], "is_subclass_of": [ "AI Application", "Generative Model", "Fine Tuning", "Diffusion Model", "Conditional Image Generation", "Neural Network", "Transfer Learning" ], "wikilinks": [ "Adapter Fine-tuning", "AlgorithmLayer", "arXiv", "Architectural Visualisation", "Automatic1111", "Autonomous Systems", "Canny Edge Detection", "Cao et al. 2017 OpenPose CVPR", "Character Design", "Chen et al. 2025 SmartSpatial IJCAI", "CLIP", "CLIP Encoder", "CLIP Image Encoder", "ComputerVisionDomain", "Conditional Image Generation", "Conditioning Preprocessor", "Conditioning Signal", "Control Signal", "Convolutional Neural Networks", "Cross-Attention Adapter" ] }, { "id": "control-net", "title": "controlnet", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ControlNet is a trainable adapter architecture that attaches conditional spatial control to pre-trained text-to-image diffusion models by duplicating the U-Net encoder into a locked copy and a trainable copy, connecting them through zero-convolution layers initialised to exactly zero weight and bias. The zero initialisation guarantees that at the start of training no gradient noise corrupts the pre-trained backbone, allowing fine-tuning on relatively small paired datasets of (control map, image) pairs. Input control maps include Canny edge maps, depth maps, human pose skeletons, semantic segmentation masks, surface normal maps, line-art, and scribbles, each producing spatially precise, prompt-steerable image generation. The result is a modular conditioning mechanism that can be composed \u2014 multiple ControlNets with weighted merging \u2014 and transplanted across base diffusion model checkpoints without retraining.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:control-net", "labels": [ "ControlNet", "PyraCanny ControlNet" ], "is_subclass_of": [ "AI Technique", "Adapter Tuning", "Transfer Learning", "Fine-Tuning", "Generative Model", "Neural Network Architecture" ], "wikilinks": [ "Diffusion Model", "Latent Diffusion", "U-Net", "Zero Convolution", "Classifier-Free Guidance", "Stable Diffusion", "Image Generation", "Text-to-Image", "Conditional Image Synthesis", "Neural Network Architecture", "Transfer Learning", "Fine-Tuning", "Convolutional Neural Network", "LoRA", "IP-Adapter", "Textual Inversion", "Pose Estimation", "Depth Estimation", "Edge Detection", "Semantic Segmentation" ] }, { "id": "controlled-illumination", "title": "Controlled Illumination", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Controlled illumination is the deliberate management of light source position, intensity, and spectrum during image capture so that resulting measurements are consistent and repeatable. It is used in camera calibration and optical calibration targets to remove ambient lighting variability that would otherwise bias corner or feature detection. Structured and patterned illumination variants are also used to actively encode depth or surface information into captured images.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:controlled-illumination", "labels": [ "Controlled Illumination" ], "is_subclass_of": [ "Camera Calibration" ], "wikilinks": [] }, { "id": "controlled-vocabularies", "title": "Controlled Vocabularies", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Standardized sets of terms, phrases, and hierarchical relationships used to describe and organize information within knowledge management systems, databases, and metadata schemas, ensuring consistent indexing, retrieval, and interoperability across digital platforms and virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:controlled-vocabularies", "labels": [ "Controlled Vocabularies" ], "is_subclass_of": [ "Standards and Interoperability", "Knowledge Organization" ], "wikilinks": [ "Consistent Indexing", "Governance Process", "Semantic Interoperability", "Taxonomy Development", "Information Retrieval", "Knowledge Organization", "Metadata Standards", "metaverse", "Telecollaboration" ] }, { "id": "controlled-vocabulary", "title": "Controlled Vocabulary", "domain": "data", "domain_name": "Data", "definition": "A restricted list of standardized terms or phrases used within a specific domain for consistent cataloging, tagging, and indexing, where users may only apply terms from the approved list to ensure uniform description and enable reliable information retrieval across systems and platforms.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:controlled-vocabulary", "labels": [ "Controlled Vocabulary", "Keyword Vocabulary" ], "is_subclass_of": [ "Information Architecture" ], "wikilinks": [ "Accurate Retrieval", "Content Discovery", "Data Standardization", "Domain Expertise", "Maintenance Process", "Term Governance", "Blockchain", "Information Architecture", "metaverse" ] }, { "id": "controller", "title": "controller", "domain": "robotics", "domain_name": "Robotics", "definition": "A Controller is a computational or hardware module that receives a reference setpoint and measured feedback signals, applies a control law or learned policy, and generates command signals that drive a plant or process towards desired states. Controllers span a spectrum from classical linear regulators such as PID and LQR to nonlinear model-based controllers such as Model Predictive Control and computed-torque methods, as well as data-driven approaches including neural network policies trained via reinforcement learning. Operating within closed-loop feedback architectures, the controller continuously computes an error signal \u2014 the difference between desired and measured output \u2014 and produces corrective actuation to minimise that error subject to stability and performance constraints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:controller", "labels": [ "Controller", "Embedded Controller" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "convection-zone", "title": "Convection Zone", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:convection-zone", "labels": [ "Convection Zone" ], "is_subclass_of": [ "Stellar Interior" ], "wikilinks": [] }, { "id": "convergence-concept", "title": "Convergence Concept", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Convergence Concept describes the synthesis of previously distinct technologies, domains, and paradigms into unified systems. In telecollaboration and spatial computing, convergence encompasses XR integration, AI-human collaboration, physical-digital fusion via digital twins, and platform unification over shared protocols such as WebXR and OpenXR. The convergence of edge and cloud computing creates hybrid architectures that optimise latency and scalability for real-time collaborative experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:convergence-concept", "labels": [ "Convergence Concept", "ConvergenceConcept" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "ACM Queue", "IEEE Future Networks", "XR Association", "Blockchain" ] }, { "id": "convergence", "title": "Convergence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Technological Convergence is the macro-level process by which previously distinct technologies, disciplines, and sociotechnical systems merge, overlap, and mutually reinforce to produce qualitatively new capabilities, applications, and social configurations that none of the converging components ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:convergence", "labels": [ "Convergence", "Convergence Diagnostics", "Convergence Theorems", "IT/OT Convergence", "Model Convergence", "Supervisory Convergence" ], "is_subclass_of": [ "AI Technique", "AI Research Area", "Digital Transformation", "Innovation Systems", "Sociotechnical Systems", "General Purpose Technology", "Technology Management" ], "wikilinks": [ "AI and Blockchain", "AI and Blockchain", "AI and Blockchain", "AI and Blockchain", "Autonomous Systems", "Biocomputing", "Biocomputing", "Cross-Domain Research Infrastructure", "Cyber Physical Systems", "Digital Divide", "Disruptive Innovation", "Drug Discovery", "Embodied AI", "Emerging Technologies", "General Purpose Technology", "General Purpose Technology Theory", "IEEE Standards Association", "Industry 4.0", "Innovation Systems", "InnovationSystemsDomain" ] }, { "id": "conversational-ai", "title": "Conversational AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Conversational AI encompasses systems that engage in natural-language dialogue with humans, including chatbots, voice assistants, and large-language-model-powered interfaces. It applies natural language processing, language modeling, dialogue management, and speech recognition to understand intent and generate contextually appropriate responses.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:conversational-ai", "labels": [ "Conversational AI", "Conversational Ai" ], "is_subclass_of": [ "Natural Language Processing", "Artificial Intelligence", "Human Computer Interaction" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Digital Twin", "Natural Language Processing", "Large Language Models", "Dialogue System", "Transformer Architecture", "Retrieval-Augmented Generation", "Reinforcement Learning from Human Feedback", "Speech Recognition", "Intent Recognition", "Multimodal AI", "Natural Language Understanding", "Sentiment Analysis", "Machine Learning", "Text-to-Speech", "Chatbot", "Virtual Assistant", "Spatial Computing", "Human Robot Interaction" ] }, { "id": "conversational-commerce", "title": "Conversational Commerce", "domain": "economics", "domain_name": "Economics", "definition": "Conversational commerce is the practice of facilitating buying and selling through interactive, natural-language interfaces such as chatbots and AI assistants.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:conversational-commerce", "labels": [ "Conversational Commerce" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "conversion-pipeline", "title": "Conversion Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An automated workflow process that transforms digital data or assets from one format, schema, or representation to another, enabling interoperability and compatibility across heterogeneous systems and platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:conversion-pipeline", "labels": [ "Conversion Pipeline" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Asset Metadata", "Asset Optimization", "Asset Pipeline", "Conversion Rules", "Data Harmonization", "Data Schema", "Data Validation", "Error Handler", "Format Migration", "Format Specification", "MSF Taxonomy 2025", "Output Generator", "SIGGRAPH Pipeline WG", "Transformation Engine", "Validation Module", "Computation And Intelligence Domain", "Computer Vision", "Cross-Platform Interoperability", "Data Layer", "Data Processing" ] }, { "id": "conversion-rate-optimisation", "title": "Conversion Rate Optimisation", "domain": "data", "domain_name": "Data", "definition": "Conversion rate optimisation (CRO) is the systematic practice of increasing the percentage of visitors to a digital property who complete a desired action \u2014 purchasing, subscribing, registering \u2014 by forming hypotheses about user behaviour and testing changes to copy, design, pricing, and flow. It combines quantitative instrumentation such as funnel analytics and A/B and multivariate testing with qualitative research such as session recordings and user interviews, turning traffic acquired through marketing into measurable business outcomes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:conversion-rate-optimisation", "labels": [ "Conversion Rate Optimisation" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [ "Digital Marketing", "E-Commerce", "A/B Testing", "User Experience" ] }, { "id": "convex-finance", "title": "Convex Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Convex Finance is a decentralised finance (DeFi) meta-protocol built on Ethereum that enables Curve Finance liquidity providers and CRV holders to earn enhanced rewards without individually locking their own CRV tokens as veCRV. By aggregating vote-escrowed CRV (veCRV) from users who deposit CRV in exchange for cvxCRV, Convex accumulates collective governance and boost power which it redistributes pro-rata to depositors, eliminating the individual capital lockup barrier inherent in Curve's vote-escrow tokenomics. The protocol's native CVX token governs allocation of this aggregated veCRV voting power via vlCVX staking, creating a secondary governance layer \u2014 the so-called Curve Wars \u2014 in which DeFi protocols competitively bribe CVX holders to direct CRV emissions toward their own liquidity pools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:convex-finance", "labels": [ "Convex Finance" ], "is_subclass_of": [ "De Fi Protocol" ], "wikilinks": [] }, { "id": "convex-optimisation", "title": "Convex Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Convex optimisation is the mathematical discipline concerned with minimising convex objective functions over convex feasible sets, exploiting the fundamental property that every local minimum is also a global minimum. This structural guarantee enables the design of polynomial-time algorithms \u2014 such as interior-point methods, subgradient descent, and proximal methods \u2014 that reliably find exact or near-exact solutions. The field is grounded in convex analysis and duality theory (Lagrangian and Fenchel duality), and underpins a vast range of applications spanning machine learning, signal processing, control systems, finance, and operations research. Canonical problem classes include linear programming, quadratic programming, second-order cone programming, and semidefinite programming.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:convex-optimisation", "labels": [ "Convex Optimisation" ], "is_subclass_of": [ "Optimisation", "Mathematical Optimisation", "Applied Mathematics", "Numerical Methods" ], "wikilinks": [ "Optimisation", "Machine Learning", "Numerical Methods", "Functional Analysis", "Linear Programming", "Quadratic Programming", "Semidefinite Programming", "Gradient Descent", "Support Vector Machine", "Deep Learning", "Duality Theory", "KKT Conditions", "LASSO Regression", "Portfolio Optimisation", "Compressed Sensing", "Operations Research", "Stochastic Optimisation", "Reinforcement Learning", "Signal Processing", "Optimal Control" ] }, { "id": "conviction-voting", "title": "Conviction Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Conviction Voting is a continuous-time, conviction-weighted decentralised governance mechanism \u2014 originally formalised by Block Science researchers Jeff Emmett, Michael Zargham and Jessica Zartler in the 2019 working paper Conviction Voting: A Novel Continuous Decision Making Alternative to G...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:conviction-voting", "labels": [ "Conviction Voting" ], "is_subclass_of": [ "Governance and Regulation", "Voting System", "DAO Governance Mechanism", "Continuous Decision-Making", "Token-Weighted Governance", "Treasury Allocation Mechanism" ], "wikilinks": [ "1Hive", "1Hive Conviction Voting Parameter Analysis 2021", "1Hive Gardens", "1Hive Gardens Documentation", "1Hive Gardens Specification", "Aggelos Kiayias", "Aragon", "Aragon Conviction Voting App", "Aragon Conviction Voting App ABI", "Aragon Conviction Voting App Documentation", "Augmented Bonding Curve Communities", "Augur", "Beck Muller-Bloch King 2018 Governance in the Blockchain Economy", "Block-Indexed Time Reference", "Block Number as Discrete Time", "Block Science", "Block Science Theory", "Block Science Working Papers 2019-2025", "Buterin Hitzig Weyl 2018 Liberal Radicalism Quadratic Funding", "cadCAD" ] }, { "id": "convolution", "title": "Convolution", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Convolution is a mathematical operation that combines two functions by sliding one (the kernel or filter) over another (the input signal or image) and computing a weighted sum of overlapping values at each position, producing a third function that expresses how the shape of one modifies the other. In deep learning and signal processing it provides a translation-equivariant mechanism for local feature extraction with shared parameter weights. The discrete 2D form underpins convolutional neural networks; the continuous form via the Convolution Theorem connects to Fourier analysis and frequency-domain filtering.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:convolution", "labels": [ "Convolution", "3D Convolution", "Atrous Convolution", "Fourier Convolution", "Transposed Convolution" ], "is_subclass_of": [ "Signal Processing", "Mathematical Operation" ], "wikilinks": [ "Signal Processing", "Image Processing", "Convolutional Neural Network", "Feature Extraction", "Deep Learning", "Fourier Analysis", "Fast Fourier Transform", "Linear Algebra", "Matrix Multiplication", "GPU Acceleration", "Object Detection", "Image Segmentation", "Self Attention", "Recurrent Neural Network", "Graph Neural Network", "Batch Normalisation", "Audio Processing", "Natural Language Processing", "Backpropagation", "Transfer Learning" ] }, { "id": "convolutional-neural-network", "title": "Convolutional Neural Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Convolutional Neural Network (CNN) is a feed-forward deep learning architecture that applies learned convolutional filters across spatial dimensions of input data, enabling hierarchical feature extraction from images and other grid-structured inputs. Weight sharing and local receptive fields make CNNs highly parameter-efficient for visual recognition tasks including image classification, object detection, and semantic segmentation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:convolutional-neural-network", "labels": [ "Convolutional Neural Network", "Convolutional Neural Networks", "ConvolutionalNeuralNetwork" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "ISO/IEC 22989:2022", "ISO/IEC 23053:2022", "ISO/IEC 23894:2023", "NIST AI 100-3", "NIST AI RMF", "OECD AI Principles", "ArtificialIntelligenceDomain", "EU AI Act" ] }, { "id": "cooling-system", "title": "Cooling System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A system that removes heat from equipment to keep it within safe operating temperatures. In computing it covers air, liquid, and immersion methods used in servers and data centres.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cooling-system", "labels": [ "Cooling System", "Cooling Infrastructure", "Cooling Systems" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [ "Hardware", "Renewable Energy" ] }, { "id": "cooperative-game-theory", "title": "Cooperative Game Theory", "domain": "ai", "domain_name": "Ai", "definition": "Cooperative game theory is the branch of game theory that studies how groups of agents can form coalitions and how the value they jointly create should be divided among members. Rather than focusing on individual strategies, it analyses solution concepts such as the core and the Shapley value that capture fair or stable allocations of collective payoff. Its allocation principles underpin applications from economics and mechanism design to feature attribution in explainable machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cooperative-game-theory", "labels": [ "Cooperative Game Theory" ], "is_subclass_of": [ "Game Theory" ], "wikilinks": [] }, { "id": "coordinate-conversion", "title": "Coordinate Conversion", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coordinate-conversion", "labels": [ "Coordinate Conversion" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "coordinate-epoch", "title": "Coordinate Epoch", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coordinate-epoch", "labels": [ "Coordinate Epoch" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "coordinate-frame", "title": "Coordinate Frame", "domain": "robotics", "domain_name": "Robotics", "definition": "A coordinate frame is a reference system defined by an origin and a set of axes against which positions, orientations, and motions are measured. In robotics and perception, multiple frames (world, base, sensor, tool) are related by rigid-body transforms so that data from different sources can be expressed consistently. Correct frame management is essential for sensor fusion, motion planning, and collision checking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:coordinate-frame", "labels": [ "Coordinate Frame", "Coordinate Frames" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "coordinate-pole", "title": "Coordinate Pole", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coordinate-pole", "labels": [ "Coordinate Pole" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "coordinate-reference-system", "title": "Coordinate Reference System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A coordinate reference system (CRS) is a framework that defines how positions are described and measured on or near the Earth's surface, combining a datum, a coordinate system, and often a map projection. It enables consistent interpretation of spatial coordinates by anchoring them to a known model of the Earth. CRSs are essential for aligning geospatial data sets and for accurate location computations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:coordinate-reference-system", "labels": [ "Coordinate Reference System" ], "is_subclass_of": [ "Geospatial Data" ], "wikilinks": [] }, { "id": "coordinate-system", "title": "Coordinate System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A coordinate system is a mathematical framework that assigns a unique ordered tuple of numbers to every point in a geometric space, enabling unambiguous specification of position, orientation, and scale. Coordinate systems establish reference frames \u2014 world, camera, object, and sensor frames \u2014 that must be composed via rigid-body or affine transforms to map quantities from one frame into another. In spatial computing, robotics, and computer vision, multiple overlapping coordinate frames coexist and their consistent management is essential for rendering, navigation, and perception tasks. The choice of convention (handedness, axis orientation, unit) and the algebraic formalism used (homogeneous matrices, quaternions, dual quaternions, Lie group elements) profoundly affects numerical stability and interoperability across software stacks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:coordinate-system", "labels": [ "Coordinate System", "3D Coordinate System", "3D Coordinate Systems", "Geospatial Coordinate System", "Global Coordinate System", "World Space Coordinate System" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "coordinate-transformation", "title": "Coordinate Transformation", "domain": "robotics", "domain_name": "Robotics", "definition": "A coordinate transformation is a mathematical mapping that converts the representation of a point, vector, or geometric object from one coordinate system or reference frame to another, preserving geometric relationships while expressing them in a new basis. In robotics and computer graphics, transformations are represented as homogeneous matrices, quaternions, or dual quaternions encoding rotation, translation, and scaling operations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:coordinate-transformation", "labels": [ "Coordinate Transformation", "Coordinate Frame Transform", "Transformation Matrix" ], "is_subclass_of": [ "Coordinate System" ], "wikilinks": [] }, { "id": "coordinated-universal-time", "title": "Coordinated Universal Time", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coordinated-universal-time", "labels": [ "Coordinated Universal Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "coordination-layer", "title": "Coordination Layer", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The Coordination Layer is the cross-cutting stratum that orchestrates the work of multiple agents or services toward a shared objective. It sits above control and integration concerns and below the application goals it serves. It contains schedulers, workflow engines, consensus on task assignment, and the protocols that keep distributed actors aligned.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:coordination-layer", "labels": [ "Coordination Layer", "Coordination Service" ], "is_subclass_of": [ "Protocol and Infrastructure", "owl:Thing" ], "wikilinks": [ "Control Layer", "Integration Layer", "Agent Layer", "Application Layer", "Distributed Systems", "Task Allocation", "owl:Thing" ] }, { "id": "coordination-mechanisms", "title": "Coordination Mechanisms", "domain": "governance", "domain_name": "Governance", "definition": "Coordination Mechanisms are the structural arrangements, protocols, and incentive designs that enable multiple autonomous agents\u2014whether human organisations, software processes, or robotic systems\u2014to align their actions toward shared goals whilst managing conflicts, resource contention, and information asymmetries. They include market-based mechanisms, hierarchical authority structures, consensus protocols, and shared state systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:coordination-mechanisms", "labels": [ "Coordination Mechanisms", "Coordination Mechanism", "CoordinationMechanisms" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "coordination-protocol", "title": "Coordination Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A Coordination Protocol is a formally specified set of rules, message formats, and interaction sequences that govern how distributed agents or system components communicate and synchronise their actions to achieve a common objective. It defines the obligations, permissions, and commitments of each participant at each state of an interaction, ensuring predictable collective behaviour despite independent agent decision-making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:coordination-protocol", "labels": [ "Coordination Protocol", "Distributed Coordination Protocol", "Signer Coordination Protocol" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "copper-co", "title": "Copper.co", "domain": "finance", "domain_name": "Finance", "definition": "Copper.co is a digital asset custody and infrastructure company serving institutional clients. It provides secure storage, settlement and prime services for cryptocurrencies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:copper-co", "labels": [ "Copper.co", "Copper MPC" ], "is_subclass_of": [ "Custody" ], "wikilinks": [ "Cryptocurrency", "Financial Services", "Custody" ] }, { "id": "intellectual-property-rights-framework-law", "title": "Copyright Law", "domain": "governance", "domain_name": "Governance", "definition": "The body of statute and case law granting creators exclusive rights over the reproduction, distribution, adaptation, and communication of original works of authorship, including text, images, music, software, and audiovisual media. Copyright law defines the exceptions \u2014 fair use, fair dealing, text-and-data-mining carve-outs \u2014 that determine whether scraping and training artificial intelligence models on protected works is lawful, making it the central legal battleground of generative AI.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:intellectual-property-rights-framework-law", "labels": [ "Copyright Law" ], "is_subclass_of": [ "Intellectual Property" ], "wikilinks": [ "Intellectual Property", "Copyright", "Licensing", "AI Scrapers" ] }, { "id": "copyright", "title": "Copyright", "domain": "governance", "domain_name": "Governance", "definition": "Copyright is a legal right that grants the creator of an original work exclusive control over its reproduction, distribution, adaptation and public performance for a limited term. It arises automatically upon fixation of an expressive work and may be licensed, assigned or waived by the rights holder. Within governance it anchors how digital content, software and creative outputs are owned, shared and monetised.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:copyright", "labels": [ "Copyright" ], "is_subclass_of": [ "Intellectual Property Rights Framework" ], "wikilinks": [] }, { "id": "cor-dapp", "title": "CorDapp", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A CorDapp (Corda Distributed Application) is a distributed application built to run on the R3 Corda enterprise blockchain platform. It bundles the contracts, states, and flows that define shared business logic and the point-to-point messaging that coordinates transactions between participating nodes. CorDapps are the unit of deployment by which enterprises implement permissioned, privacy-preserving workflows on Corda.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cor-dapp", "labels": [ "CorDapp" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "corda", "title": "Corda", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Corda is an open-source permissioned distributed ledger platform developed by R3 and purpose-built for regulated industries, where transaction data is shared exclusively between the counterparties involved rather than broadcast to all network participants. Unlike public blockchains, Corda structures its ledger as a directed acyclic graph of Unspent Transaction Outputs (UTXOs) called states, which are consumed and produced by atomic transactions validated by both smart contract code and legal prose. A notary cluster provides consensus on transaction uniqueness, preventing double-spend without revealing transaction details to uninvolved parties, making Corda particularly suited to financial services, trade finance, and healthcare interoperability scenarios.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:corda", "labels": [ "Corda" ], "is_subclass_of": [ "Permissioned Blockchain" ], "wikilinks": [ "Distributed Ledger", "Smart Contract", "Privacy", "Permissioned Blockchain" ] }, { "id": "core-lightning", "title": "Core Lightning", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Core Lightning (formerly c-lightning) is a specification-compliant, production-grade implementation of the Bitcoin Lightning Network protocol written in C and maintained by Blockstream. It provides a lightweight, modular node daemon that enables off-chain Bitcoin micropayments through bidirectional payment channels anchored on the Bitcoin blockchain. The implementation adheres to the BOLT (Basis of Lightning Technology) specification suite, ensuring interoperability with other Lightning implementations such as LND and Eclair. A distinguishing feature is its plugin architecture, which allows operators to extend node behaviour in arbitrary programming languages via a JSON-RPC interface.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:core-lightning", "labels": [ "Core Lightning" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning Network", "Payment Channel", "Bitcoin" ] }, { "id": "coreference-resolution", "title": "Coreference Resolution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Coreference resolution is the natural language processing task of identifying all expressions in a text that refer to the same real-world entity, grouping noun phrases, pronouns, definite descriptions, and other referring expressions into coreference clusters so that downstream systems can maintain coherent entity representations across sentences and documents. It is a foundational subtask of information extraction, question answering, summarisation, and knowledge graph population, enabling models to track entities without re-identifying them from scratch at each mention.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:coreference-resolution", "labels": [ "Coreference Resolution" ], "is_subclass_of": [ "Natural Language Processing", "Natural Language Understanding" ], "wikilinks": [ "Natural Language Processing", "Natural Language Understanding", "Named Entity Recognition", "Information Extraction", "Question Answering", "Knowledge Graph", "Transformer Architecture", "Large Language Model", "Machine Translation", "Entity Resolution", "Semantic Parsing", "Text Mining", "Relation Extraction", "Document Summarisation", "BERT", "Anaphora Resolution", "Word Embeddings", "Mention Detection", "Span Representation", "Retrieval Augmented Generation" ] }, { "id": "coronal-mass-ejection", "title": "Coronal Mass Ejection", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:coronal-mass-ejection", "labels": [ "Coronal Mass Ejection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "corporate-governance", "title": "Corporate Governance", "domain": "governance", "domain_name": "Governance", "definition": "The system of rules, practices and processes by which a company is directed and controlled, balancing the interests of shareholders, management and other stakeholders, including board oversight, executive accountability, disclosure obligations, and mechanisms that align the interests of management, shareholders, and wider society.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:corporate-governance", "labels": [ "Corporate Governance", "Corporate Board Governance" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [ "Accountability", "Transparency", "Compliance", "Audit", "Risk Management", "Governance Framework" ] }, { "id": "corporate-sustainability-reporting-directive", "title": "Corporate Sustainability Reporting Directive", "domain": "governance", "domain_name": "Governance", "definition": "The Corporate Sustainability Reporting Directive (CSRD) is a European Union law that expands and standardises the sustainability information companies must disclose, replacing the earlier Non-Financial Reporting Directive. It mandates reporting against the European Sustainability Reporting Standards, applies a double-materiality lens and requires assurance of the disclosed data. It substantially widens the scope of in-scope undertakings and the depth of environmental, social and governance disclosure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:corporate-sustainability-reporting-directive", "labels": [ "Corporate Sustainability Reporting Directive" ], "is_subclass_of": [ "ESG Reporting" ], "wikilinks": [] }, { "id": "corporate-sustainability-reporting", "title": "Corporate Sustainability Reporting", "domain": "governance", "domain_name": "Governance", "definition": "Corporate sustainability reporting is the structured disclosure by an organisation of its environmental, social, and governance (ESG) performance, impacts, risks, and opportunities, typically published alongside or integrated within financial reporting. It translates non-financial activity \u2014 emissions, resource use, workforce conditions, supply-chain practices, and governance arrangements \u2014 into standardised, often externally assured metrics intended for investors, regulators, and other stakeholders. Modern regimes increasingly mandate disclosure under frameworks such as the EU Corporate Sustainability Reporting Directive (CSRD) and the ISSB's IFRS S1/S2 standards, shifting reporting from voluntary marketing toward audited, decision-useful information.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:corporate-sustainability-reporting", "labels": [ "Corporate Sustainability Reporting" ], "is_subclass_of": [ "Sustainability Reporting" ], "wikilinks": [] }, { "id": "corporate-tax-compliance-framework", "title": "Corporate Tax Compliance Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Tax, in the context of this knowledge graph, refers to the statutory obligations imposed on individuals and legal entities \u2014 particularly limited companies \u2014 by HMRC and comparable revenue authorities, covering corporation tax on trading profits, self-assessment for personal income drawn as salary or dividends, and VAT registration thresholds. Compliance requires accurate bookkeeping, timely filing of CT600 returns, and annual accounts submission to Companies House.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:corporate-tax-compliance-framework", "labels": [ "Corporate Tax Compliance Framework", "International Tax Compliance", "Tax", "Tax Calculation Engine", "Tax Evasion Detection", "Tax Record Microdata", "Tax Transparency" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Accounts" ] }, { "id": "corporate-training", "title": "Corporate Training", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The organised development of employee knowledge, skills, and competencies undertaken by employers to improve individual and organisational performance. It spans onboarding, compliance, technical upskilling, soft-skills coaching, and leadership development, and is increasingly delivered through digital, collaborative, and immersive learning technologies \u2014 learning management systems, virtual classrooms with breakout rooms, and VR simulation \u2014 that let distributed workforces practise skills safely and at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:corporate-training", "labels": [ "Corporate Training" ], "is_subclass_of": [ "Workforce Development" ], "wikilinks": [ "Workforce Development", "Immersive Learning", "Breakout Room", "Education Technology", "Virtual Training" ] }, { "id": "correspondent-banking", "title": "Correspondent Banking", "domain": "finance", "domain_name": "Finance", "definition": "Correspondent banking is an arrangement in which one bank (the correspondent) holds accounts and provides payment, settlement, trade finance, and liquidity services on behalf of another bank (the respondent) that lacks a direct presence or licence in a given jurisdiction or currency. The respondent bank maintains a nostro account at the correspondent and an equivalent vostro account on its own books, enabling cross-border value transfer without requiring every institution to maintain a full global branch network. This intermediated model underpins a substantial share of international wire transfers and documentary trade, but introduces multi-hop settlement chains, foreign-exchange conversion costs, and heightened anti-money-laundering compliance obligations, motivating ongoing efforts to reform or replace the model through ISO 20022, CBDC corridors, and distributed ledger payment rails.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:correspondent-banking", "labels": [ "Correspondent Banking", "Banking Correspondent Relationships" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "Cross-Border Payments", "SWIFT" ] }, { "id": "corrigibility", "title": "Corrigibility", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Corrigibility is the property of an AI system that allows it to be corrected, redirected or shut down by authorised humans without resisting, deceiving or manipulating them. A corrigible agent does not treat interventions as threats to its objectives and cooperates with oversight even when doing so conflicts with its current goals. It is a central concept in AI safety because it keeps powerful systems amenable to human control as they grow more capable.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:corrigibility", "labels": [ "Corrigibility" ], "is_subclass_of": [ "Value Alignment", "AI Safety" ], "wikilinks": [ "AI Safety", "Value Alignment", "AI Alignment", "Human-in-the-Loop", "Reinforcement Learning", "Reward Hacking", "Interpretability", "Transparency", "Accountability", "Robustness", "Existential Risk", "Responsible AI", "Scalable Oversight", "Constitutional AI", "Mesa-Optimisation", "Governance", "Human Oversight", "Instrumental Convergence", "Reinforcement Learning from Human Feedback", "Mechanistic Interpretability" ] }, { "id": "cortex-agent", "title": "Cortex Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cortex Agent is a class of enterprise agentic AI system that orchestrates multi-step reasoning and action across both structured and unstructured data sources within a governed cloud data platform, with Snowflake Cortex Agents (General Availability November 4, 2025) as the dominant production ins...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cortex-agent", "labels": [ "Cortex Agent" ], "is_subclass_of": [ "AI Application", "Agentic Internet", "Agents", "Agent Frameworks", "Large-Scale Pretrained Foundation Model", "Cognitive AI" ], "wikilinks": [ "AgenticSystemsDomain", "Anderson 1983 Architecture of Cognition", "Anderson Lebiere 1998 Atomic Components of Thought", "Anthropic Snowflake Partnership 2024 2025", "Bratman 1987 Intention Plans Practical Reason", "Cortex AISQL GA 2025", "Cortex Analyst Semantic Model Spec 2025", "Cortex Search Overview 2025", "EnterpriseAIDomain", "Franklin Graesser 1996 Agent Taxonomy", "Hawkins Blakeslee 2004 On Intelligence", "InferenceLayer", "Interface Media 2026 UK AI Survey", "Kahneman 2011 Thinking Fast and Slow", "Laird 2012 Soar Cognitive Architecture", "Laird Newell Rosenbloom 1987 Soar", "Minsky 1986 Society of Mind", "Newell 1990 Unified Theories of Cognition", "OrchestrationLayer", "Rao Georgeff 1991 BDI Architecture" ] }, { "id": "cosine-similarity", "title": "Cosine Similarity", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Cosine Similarity is a metric that measures the cosine of the angle between two non-zero vectors in an inner product space, yielding a value in [\u22121, 1] that quantifies directional similarity independently of vector magnitude. It is the dominant similarity measure for comparing high-dimensional sparse and dense vector representations of text, images, and other data in information retrieval and machine learning systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:cosine-similarity", "labels": [ "Cosine Similarity" ], "is_subclass_of": [ "AI Technique", "Information Retrieval", "Nearest Neighbour Search" ], "wikilinks": [ "Semantic Search", "Information Retrieval", "Embedding", "Vector Database", "Retrieval-Augmented Generation", "Nearest Neighbour Search", "Natural Language Processing", "Transformer Architecture", "Cross-Encoder Reranking", "Bi-Encoder", "Euclidean Distance", "Dot Product", "TF-IDF", "BM25", "Representation Learning", "Knowledge Graph Embedding", "Recommendation System", "Sentence-BERT", "FAISS", "Approximate Nearest Neighbour" ] }, { "id": "cosmic-microwave-background", "title": "Cosmic Microwave Background", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cosmic-microwave-background", "labels": [ "Cosmic Microwave Background" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cosmic-web", "title": "Cosmic Web", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cosmic-web", "labels": [ "Cosmic Web" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cosmological-distance", "title": "Cosmological Distance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cosmological-distance", "labels": [ "Cosmological Distance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cosmology", "title": "Cosmology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cosmology", "labels": [ "Cosmology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cosmos-ibc", "title": "Cosmos IBC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Inter-Blockchain Communication protocol that enables sovereign blockchain networks to exchange data and tokens trustlessly, using light clients and Merkle proofs to verify state across chains without centralised intermediaries, providing the foundational interoperability layer of the Cosmos ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cosmos-ibc", "labels": [ "Cosmos IBC", "Cosmos IBC Specification" ], "is_subclass_of": [ "Cross-Chain Communication", "Interoperability Protocol" ], "wikilinks": [ "Blockchain Light Client", "Channel Protocol", "Connection Protocol", "Cosmos Ecosystem", "Cosmos Network", "Cross-Chain Communication", "Cross-Chain Contracts", "IBC Protocol", "Inter-Chain Value Transfer", "Interoperability Protocol", "Layer 2 Interop", "Packet Authentication", "Packet System", "Polkadot XCM", "BlockchainDomain", "Merkle Proof" ] }, { "id": "cosmos-sdk", "title": "Cosmos SDK", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Cosmos SDK is an open-source modular framework for building application-specific blockchains in the Cosmos ecosystem. It provides composable modules for accounts, staking, governance, and token management, and integrates with the Tendermint BFT consensus engine to enable sovereign, interoperable chains connected via the Inter-Blockchain Communication protocol.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cosmos-sdk", "labels": [ "Cosmos SDK" ], "is_subclass_of": [ "Cosmos" ], "wikilinks": [ "Tendermint", "IBC", "Blockchain Interoperability", "Cosmos", "https://docs.cosmos.network", "https://github.com/cosmos/cosmos-sdk" ] }, { "id": "cosmos", "title": "Cosmos", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cosmos is an ecosystem of sovereign, application-specific blockchains that interoperate through the Inter-Blockchain Communication (IBC) protocol, coordinated by a central hub (the Cosmos Hub, secured by the ATOM staking token) and constructed using the Cosmos SDK, a modular Go framework for building proof-of-stake chains. Each chain in the ecosystem runs the CometBFT (formerly Tendermint BFT) consensus engine, which provides instant deterministic finality and enables IBC light-client proofs across trust boundaries without centralised bridges. The architecture decouples application logic from consensus, permitting developers to optimise validator sets, fee markets, and governance parameters independently while participating in a shared interoperability fabric often described as the Internet of Blockchains.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:cosmos", "labels": [ "Cosmos", "Cosmos Ecosystem", "Cosmos Hub", "Cosmos ICS Standards", "Cosmos Network" ], "is_subclass_of": [ "Blockchain Network" ], "wikilinks": [] }, { "id": "cost-function", "title": "cost function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A cost function is a scalar-valued mathematical function that maps a model's parameters or a system's state to a real number representing the magnitude of error, resource expenditure, or divergence from a desired outcome. In supervised machine learning, the cost function (also called a loss function) measures the aggregate discrepancy between predicted and ground-truth outputs across a training set, providing the objective that optimisation algorithms such as gradient descent minimise. In control theory and robotics, cost functions encode trajectory quality criteria including path length, energy consumption, and collision risk, enabling optimal control policies via formulations such as LQR and model predictive control. The design of the cost function is one of the most consequential decisions in any learning or optimisation system, as misspecified objectives lead to reward hacking, degenerate solutions, or physically unrealisable behaviours.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cost-function", "labels": [ "Cost Function" ], "is_subclass_of": [ "AI Technique", "Objective Function", "Mathematical Function" ], "wikilinks": [ "Loss Function", "Objective Function", "Machine Learning", "Optimal Control", "Robotics", "Model Training", "Gradient Descent", "Backpropagation", "Regularisation", "Supervised Learning", "Reinforcement Learning", "Neural Network", "Deep Learning", "Mean Squared Error", "Cross-Entropy Loss", "Maximum Likelihood Estimation", "Bayesian Inference", "Convex Optimisation", "Overfitting", "Reward Function" ] }, { "id": "cost-optimisation", "title": "Cost Optimisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cost optimisation is the discipline of minimising the financial cost of computing infrastructure while preserving required performance, reliability and capacity. It combines right-sizing of resources, elimination of waste, demand-aligned scaling and commercial levers such as committed-use and spot pricing. In cloud environments it is operationalised through continuous measurement, allocation of spend to teams, and feedback loops that align consumption with actual need.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cost-optimisation", "labels": [ "Cost Optimisation" ], "is_subclass_of": [ "Resource Management" ], "wikilinks": [] }, { "id": "cost-effectiveness", "title": "Cost-Effectiveness", "domain": "economics", "domain_name": "Economics", "definition": "The economic efficiency of an AI model or system, evaluated by the ratio of output quality or task completion rate to the total cost of inference and deployment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cost-effectiveness", "labels": [ "Cost-Effectiveness" ], "is_subclass_of": [ "AI Economics" ], "wikilinks": [] }, { "id": "cost-efficient-inference", "title": "Cost-Efficient Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cost-efficient inference is the practice of serving model predictions at the lowest achievable compute and financial cost per request while meeting latency and quality targets. Techniques include dynamic batching, quantisation, caching, and context engineering to reduce token usage. It matters because inference, not training, dominates the lifetime cost of deployed large language models at scale.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cost-efficient-inference", "labels": [ "Cost-Efficient Inference", "Cost Efficient Inference" ], "is_subclass_of": [ "Inference Optimisation" ], "wikilinks": [] }, { "id": "cost-per-intelligence", "title": "Cost-Per-Intelligence", "domain": "economics", "domain_name": "Economics", "definition": "A benchmarking metric that evaluates the cost of achieving a specific level of AI capability or task performance, often used to identify the Pareto frontier of model efficiency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cost-per-intelligence", "labels": [ "Cost-Per-Intelligence" ], "is_subclass_of": [ "AI Economics" ], "wikilinks": [] }, { "id": "costmap", "title": "Costmap", "domain": "robotics", "domain_name": "Robotics", "definition": "A costmap is a grid-based spatial data structure used in robot navigation that assigns a traversal cost to each cell of the environment, encoding obstacles, inflation zones, and free space. Local and global costmaps fuse sensor data and static maps so that planners can compute collision-free, low-cost paths. It is a core component of navigation stacks such as Nav2, where it underpins both global path planning and local trajectory control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:costmap", "labels": [ "Costmap", "Costmap 2D" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "counter-terrorist-financing", "title": "Counter-Terrorist Financing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Counter-Terrorist Financing (CTF or CFT) is a set of legal, regulatory, and operational measures designed to detect, prevent, and disrupt the funding of terrorist activities. It complements anti-money-laundering regimes and is enforced through obligations such as customer due diligence, transaction monitoring, and reporting of suspicious activity. In digital assets, CTF underpins requirements like the FATF Travel Rule that mandate originator and beneficiary information for transfers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:counter-terrorist-financing", "labels": [ "Counter-Terrorist Financing", "Counter-Terrorism Financing", "Terrorist Financing Disruption" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "counterfactual-explanation", "title": "Counterfactual Explanation", "domain": "ai", "domain_name": "Ai", "definition": "An explainable-AI method that accounts for a model's decision by presenting the smallest realistic change to the input that would have produced a different outcome \u2014 for example, 'the loan would have been approved had annual income been \u00a35,000 higher' \u2014 giving affected individuals actionable recourse without requiring disclosure of the model's internal structure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:counterfactual-explanation", "labels": [ "Counterfactual Explanation" ], "is_subclass_of": [ "Explainability" ], "wikilinks": [ "Explainability", "LIME", "Decision Transparency" ] }, { "id": "counterfactual-reasoning", "title": "Counterfactual Reasoning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Counterfactual reasoning is the process of inferring what would have happened under a hypothetical intervention or alternative set of conditions, contrary to what was actually observed, given a causal model of a system. It sits at the top of Pearl's ladder of causation, above association and intervention, and requires a structural causal model rather than observational data alone to answer 'what if' questions such as the effect of a different action or policy. In machine learning, counterfactual reasoning underpins causal inference methods, explainable AI techniques and world models that must predict consequences of actions not actually taken.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:counterfactual-reasoning", "labels": [ "Counterfactual Reasoning" ], "is_subclass_of": [ "Causal Inference" ], "wikilinks": [] }, { "id": "counterparty-risk", "title": "Counterparty Risk", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Counterparty Risk is the risk that the other party to a financial transaction or contract will fail to meet its obligations before settlement, causing loss to the non-defaulting party. It is central to derivatives, lending and securities trading, and is managed through collateral, netting and central clearing. Blockchain settlement and atomic swaps aim to reduce counterparty risk by removing reliance on trusted intermediaries.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:counterparty-risk", "labels": [ "Counterparty Risk" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "covariance-matrix", "title": "Covariance Matrix", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A covariance matrix is a square symmetric positive semi-definite matrix whose (i,j) entry is the covariance between the i-th and j-th components of a multivariate random variable, with variances on the main diagonal. It completely characterises the second-order statistics and linear dependence structure of a multivariate distribution, and \u2014 together with the mean vector \u2014 fully specifies a multivariate Gaussian. The covariance matrix is central to multivariate statistics, machine learning, state estimation, signal processing, and finance, serving as the fundamental data structure for representing and propagating uncertainty in high-dimensional systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:covariance-matrix", "labels": [ "Covariance Matrix" ], "is_subclass_of": [ "Linear Algebra", "Multivariate Statistics", "Statistical Object" ], "wikilinks": [ "Probability Distribution", "Linear Algebra", "Gaussian Distribution", "Principal Component Analysis", "Dimensionality Reduction", "Kalman Filter", "Bayesian Inference", "Anomaly Detection", "Feature Extraction", "Machine Learning", "Multivariate Statistics", "Eigendecomposition", "Singular Value Decomposition", "Regularisation", "Mahalanobis Distance", "Gaussian Process", "State Estimation", "Sensor Fusion", "Portfolio Optimisation", "Random Matrix Theory" ] }, { "id": "covert-ai-tool-adoption-risk", "title": "Covert AI Tool Adoption Risk", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Secret Cyborg Problem describes the widespread but concealed practice of workers integrating AI tools into professional workflows without disclosure to employers or colleagues, creating risks around data governance, accountability, and skills assessment. Research indicates approximately 78% of AI-using knowledge workers employ tools covertly, undermining organisational AI adoption strategies and responsible-use frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:covert-ai-tool-adoption-risk", "labels": [ "Covert AI Tool Adoption Risk", "The Secret Cyborg Problem" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "cradle-to-cradle-design", "title": "Cradle-to-Cradle Design", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Cradle-to-Cradle design is a sustainability framework in which products are conceived so that all materials become safe nutrients for biological or technical cycles, eliminating the concept of waste. It emphasises material health, reuse, renewable energy, and closed-loop recovery rather than the linear take-make-dispose model. It is a foundational design philosophy within the circular economy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cradle-to-cradle-design", "labels": [ "Cradle-to-Cradle Design" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "crash-fault-tolerance", "title": "Crash Fault Tolerance", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Crash fault tolerance is the property of a distributed system that continues to operate correctly despite nodes failing by stopping, that is, by halting and ceasing to send messages. It assumes the crash-stop or crash-recovery failure model, in which faulty processes do not behave maliciously or send incorrect information. Protocols such as Paxos and Raft achieve it through replication and consensus over a quorum of non-faulty nodes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:crash-fault-tolerance", "labels": [ "Crash Fault Tolerance" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "creative-ai", "title": "Creative AI", "domain": "ai", "domain_name": "Ai", "definition": "Creative AI refers to artificial intelligence systems designed to generate, augment, or evaluate creative artefacts \u2014 including images, music, text, video, and interactive experiences \u2014 by learning latent representations of aesthetic structure from large corpora of human-produced works. Drawing on generative architectures such as diffusion models, large language models, transformers, and generative adversarial networks, these systems can synthesise novel content that extends or stylistically mimics human creative practice. They are deployed across professional creative pipelines in entertainment, advertising, game development, and design, while simultaneously raising contested questions of authorship, copyright, and economic disruption for human creators. Creative AI represents the convergence of machine learning research with aesthetic domains historically regarded as uniquely human, positioning it as both a technical milestone and a socio-cultural inflection point.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:creative-ai", "labels": [ "Creative AI" ], "is_subclass_of": [ "Generative AI", "Deep Learning", "Computational Creativity" ], "wikilinks": [ "Generative AI", "Diffusion Model", "Generative Adversarial Network", "Large Language Models", "Transformer", "Variational Autoencoder", "Neural Style Transfer", "Image Generation", "Content Creation", "Digital Content Creation", "Music Generation", "Text-to-Speech", "Procedural Content Generation", "Training Data", "Deep Learning", "Computational Creativity", "Human-AI Collaboration", "Copyright", "Creative Industries", "Synthetic Media" ] }, { "id": "creative-commons", "title": "Creative Commons", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Creative Commons is a family of standardised public copyright licences that let creators grant defined reuse, attribution, and sharing permissions to the public. The licences (e.g. CC BY, CC BY-SA, CC0) provide machine-readable and human-readable terms that promote legal reuse of content. They are widely used to license training data, open educational resources, and open media corpora.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:creative-commons", "labels": [ "Creative Commons", "Creative Commons Licensing" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "creative-expression", "title": "Creative Expression", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Creative expression is the human activity of producing original artistic, narrative, or aesthetic artefacts to communicate ideas, emotions, or identity. In virtual and open-world environments it manifests as user-generated content, world-building, avatar customisation, and emergent play. It is a primary driver of engagement and value in persistent metaverse spaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:creative-expression", "labels": [ "Creative Expression" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "creative-industries", "title": "Creative Industries", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The creative industries are the economic sectors that generate value primarily from individual creativity, skill, and intellectual property, spanning design, fashion, film, music, gaming, advertising, architecture, and publishing. They combine cultural production with commercial distribution and are increasingly reshaped by generative AI tooling. As an economic domain they constitute a significant employer and exporter of cultural goods \u2014 contributing \u00a3124 billion in gross value added to the UK economy in 2023 \u2014 while simultaneously representing the sector most affected by unresolved questions of copyright, training data use, and AI-enabled labour displacement.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:creative-industries", "labels": [ "Creative Industries" ], "is_subclass_of": [ "Economics", "Knowledge Economy" ], "wikilinks": [ "Fashion", "AI Companies", "Creative AI", "Generative AI", "Copyright", "Intellectual Property Rights Framework", "Film Production", "Game Development", "Music Generation", "Digital Content Creation", "Content Creation", "Human-AI Collaboration", "AI Ethics", "Diffusion Model", "Large Language Models", "Synthetic Media", "Image Generation", "Text-to-Image", "Procedural Content Generation", "Digital Twin" ] }, { "id": "creative-software", "title": "Creative Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Application software designed to support artistic, design, and content production workflows, including 3D modelling, animation, compositing, generative art, and interactive experience authoring. In spatial computing contexts, creative software bridges concept and deployment by enabling artists to produce assets compatible with real-time rendering engines and metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:creative-software", "labels": [ "Creative Software" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "creative-tools", "title": "creative tools", "domain": "ai", "domain_name": "Ai", "definition": "Creative Tools are AI-powered and digitally-augmented software applications that assist, augment, or automate human creative processes across modalities including image generation, music composition, video synthesis, 3D asset creation, and long-form text authoring. They typically expose foundation models\u2014particularly diffusion models and large language models\u2014through interactive interfaces or programmable API endpoints, enabling both professionals and non-expert users to produce high-quality creative outputs. Their role is dual: serving as productivity accelerators for practitioners and as accessibility bridges lowering barriers to creative expression. Provenance attribution, intellectual property frameworks, and content authentication standards such as C2PA are active governance concerns around AI-generated content produced by these tools.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:creative-tools", "labels": [ "Creative Tools", "Creative AI Tools" ], "is_subclass_of": [ "AI Application", "Generative AI" ], "wikilinks": [ "Diffusion Model", "Generative AI", "Large Language Model", "Multimodal Model", "Prompt Engineering", "Foundation Model", "GPU Compute", "Training Data", "API Endpoint", "Image Generation", "Text-to-Image", "Video Synthesis", "Music Generation", "3D Asset Creation", "Content Creation", "Content Provenance", "Intellectual Property", "Human-Computer Interaction", "Human-AI Collaboration", "Spatial Computing" ] }, { "id": "creator-compensation", "title": "Creator Compensation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Creator compensation refers to the economic mechanisms by which content creators are paid for their work, including royalties, revenue sharing, tipping, and tokenised ownership rights. In blockchain contexts it is often automated through smart contracts that distribute proceeds on primary sale and secondary resale. It is central to sustainable creator economies and to debates over fair value capture on digital platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:creator-compensation", "labels": [ "Creator Compensation", "Automated Creator Compensation", "Contributor Compensation" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "creator-economy", "title": "Creator Economy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Ecosystem enabling individuals and organizations to design, build, and monetize virtual content and experiences through digital marketplaces, tokenization, and economic incentive structures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:creator-economy", "labels": [ "Creator Economy", "CreatorEconomy" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Content Distribution Platform", "Content Licensing", "Creator Monetization", "Monetization System", "MSF Taxonomy 2025", "Payment Processing", "Token Economy", "ApplicationLayer", "Blockchain", "Creator Royalty Token", "Decentralized Exchange (DEX)", "Digital Asset Trading", "Digital Asset Workflow", "Digital Goods", "Digital Marketplace", "Digital Wallet", "Metaverse Content Pipeline", "MiddlewareLayer", "NFT Minting", "Royalty Distribution" ] }, { "id": "creator-monetization", "title": "Creator Monetization", "domain": "finance", "domain_name": "Finance", "definition": "Creator Monetization refers to the suite of mechanisms, platforms, and economic models through which independent content creators convert their creative output and audience relationships into sustainable revenue streams. These mechanisms span advertising revenue sharing, subscription and membership tiers, merchandise and physical goods, live events, licensing, and on-chain token-based models including NFTs and creator coins. The field has been shaped by the shift from broadcast media to participatory digital platforms, enabling individuals rather than corporations to capture value from cultural production. Web3 infrastructure is expanding the design space by enabling programmable royalties and direct fan-ownership of creative works.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:creator-monetization", "labels": [ "Creator Monetization", "Creator Monetisation" ], "is_subclass_of": [ "Creator Economy" ], "wikilinks": [] }, { "id": "creator-royalties", "title": "Creator Royalties", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Creator royalties are automatic, programmable payments directed to the original creator of a digital asset \u2014 typically an NFT \u2014 each time that asset is resold on a secondary market. Encoded in smart contract logic or enforced through marketplace policy, royalties provide creators with a fraction of every subsequent transaction price, extending their economic participation beyond the initial sale. The mechanism is foundational to sustainable creator economies on blockchain platforms, though enforcement remains contested between on-chain protocol enforcement and off-chain marketplace compliance. Debate centres on whether royalty obligations can be made non-bypassable at the token-transfer level or rely on voluntary marketplace co-operation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:creator-royalties", "labels": [ "Creator Royalties", "Perpetual Creator Royalties" ], "is_subclass_of": [ "Digital Rights Management" ], "wikilinks": [ "NFT", "Smart Contract", "Creator Royalties" ] }, { "id": "creator-royalty-token", "title": "Creator Royalty Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token granting its holder a share of ongoing from creative works, enabling automated royalty distribution and fractional ownership of intellectual property income streams.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:creator-royalty-token", "labels": [ "Creator Royalty Token", "CreatorRoyaltyToken" ], "is_subclass_of": [ "Token and Asset", "Virtual Economy" ], "wikilinks": [ "Automated Royalty Distribution", "Creator Compensation", "Fan Investment", "Intellectual Property System", "IP Monetization", "IP Rights Metadata", "ISO 24165", "OECD Creative Economy", "OMA3 Media WG", "Ownership Record", "Payment Gateway", "Revenue Distribution Logic", "Revenue Monitoring Service", "Royalty Smart Contract", "Royalty Tracking System", "SmartContractLayer", "AI Agent System", "Blockchain", "BlockchainDomain", "CreativeMediaDomain" ] }, { "id": "credential-definition", "title": "Credential Definition", "domain": "security", "domain_name": "Security", "definition": "A credential definition is an on-ledger or registry artefact, published by a credential issuer, that binds a credential schema to the issuer's cryptographic public keys and signing parameters. It specifies the attributes a credential will contain and the cryptographic material verifiers use to validate signatures and zero-knowledge proofs derived from issued credentials. Central to decentralised-identity ecosystems such as those built on AnonCreds, a credential definition lets a holder prove possession of an issuer-signed credential, selectively disclosing only chosen attributes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:credential-definition", "labels": [ "Credential Definition" ], "is_subclass_of": [ "Verifiable Credential Standard" ], "wikilinks": [] }, { "id": "credential-exchange", "title": "Credential Exchange", "domain": "security", "domain_name": "Security", "definition": "Credential exchange is the protocol-governed flow by which verifiable credentials are issued to, held by and presented from a digital identity wallet to relying parties. It defines how issuers offer credentials, how holders store and selectively disclose them, and how verifiers request and validate proofs. It is a foundational interaction pattern of decentralised and self-sovereign identity systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:credential-exchange", "labels": [ "Credential Exchange" ], "is_subclass_of": [ "Digital Identity Standards" ], "wikilinks": [] }, { "id": "credential-format-standard", "title": "Credential Format Standard", "domain": "security", "domain_name": "Security", "definition": "Technical specifications defining the structure, encoding, and cryptographic verification methods for digital credentials, enabling secure issuance, storage, and verification of identity documents, certifications, and attestations across different platforms and systems through interoperable formats such as W3C Verifiable Credentials and ISO mDL.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:credential-format-standard", "labels": [ "Credential Format Standard" ], "is_subclass_of": [ "Identity Standards" ], "wikilinks": [ "Credential Verification", "Cryptographic Proofs", "DID", "Identity Portability", "ISO mDL", "Issuer Infrastructure", "Trust Interoperability", "W3C Verifiable Credentials", "DID Nostr Identity", "Identity Standards", "metaverse", "Standards Body" ] }, { "id": "credential-issuance", "title": "Credential Issuance", "domain": "security", "domain_name": "Security", "definition": "Credential issuance is the process by which an authoritative entity \u2014 an issuer \u2014 creates, signs, and delivers a structured attestation about a subject's attributes, qualifications, or identity to that subject or to a designated holder. In the W3C Verifiable Credentials model, issuance involves binding claims to a subject's decentralised identifier using the issuer's cryptographic key, producing a tamper-evident credential that the holder can present to verifiers without returning to the issuer. The issuance process encompasses schema selection, claim population, signature generation, and delivery, and may be implemented with varying degrees of issuer privacy, holder binding strength, and revocability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:credential-issuance", "labels": [ "Credential Issuance" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "credential-portability", "title": "Credential Portability", "domain": "security", "domain_name": "Security", "definition": "Credential portability is the property that lets a holder carry verifiable credentials across providers, platforms, and jurisdictions without re-issuance or vendor lock-in. It relies on open standards such as W3C Verifiable Credentials and decentralised identifiers so that issuers, holders, and verifiers interoperate. It is a foundational capability for user-controlled digital identity and trust frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:credential-portability", "labels": [ "Credential Portability" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "credential-presentation", "title": "Credential Presentation", "domain": "security", "domain_name": "Security", "definition": "Credential presentation is the process by which a holder of a verifiable credential shares proof of that credential, or selected claims derived from it, with a verifying party in order to satisfy a request, typically as a signed verifiable presentation rather than the raw credential itself. It follows credential issuance in the standard verifiable-credential lifecycle of issuance, storage, presentation and verification, and may use selective disclosure so the holder reveals only the attributes required rather than the full credential. Credential presentation protocols, such as those built on DIDComm or OpenID for Verifiable Presentations, define how the request, proof and response are exchanged between holder and verifier.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:credential-presentation", "labels": [ "Credential Presentation" ], "is_subclass_of": [ "Credential Issuance" ], "wikilinks": [] }, { "id": "credential-schema", "title": "credential schema", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Credential Schema is a machine-readable, versioned specification \u2014 typically expressed in JSON Schema or a JSON-LD vocabulary \u2014 that defines the mandatory and optional claims, data types, cardinality constraints, and value ranges permissible within a Verifiable Credential. Schemas function as a binding contract between credential issuers, holders, and verifiers: the issuer populates claims in conformance with a declared schema, the holder's identity wallet stores schema metadata alongside the credential, and the verifier resolves and validates the credential's structure against the schema before applying trust policy decisions. Schemas are published at stable, resolvable URIs and are referenced from the credentialSchema property defined in the W3C Verifiable Credentials Data Model; in AnonCreds ecosystems they are written to a Verifiable Data Registry, making them tamper-evident and globally discoverable.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:credential-schema", "labels": [ "Credential Schema" ], "is_subclass_of": [ "Decentralised Identity" ], "wikilinks": [] }, { "id": "credential-storage", "title": "Credential Storage", "domain": "security", "domain_name": "Security", "definition": "Credential storage is the secure persistence of authentication secrets and verifiable credentials\u2014such as passwords, keys, tokens, and signed attestations\u2014so they can be protected at rest and retrieved for verification. It employs encryption, hardware-backed enclaves, and access controls to resist theft and tampering. It is a required capability of any digital identity management system.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:credential-storage", "labels": [ "Credential Storage" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "credential-store", "title": "Credential Store", "domain": "security", "domain_name": "Security", "definition": "A credential store is the component within an identity system that holds user identifiers, secrets, and account attributes used during authentication. Implemented as a directory, database, or secrets vault, it is queried by an identity provider to validate sign-in attempts and issue tokens. Its integrity and confidentiality are critical to the security of the surrounding identity infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:credential-store", "labels": [ "Credential Store", "Credential Vault" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "credential-verification", "title": "Credential Verification", "domain": "security", "domain_name": "Security", "definition": "Credential verification is the process of cryptographically or institutionally confirming the authenticity, integrity, and current validity of a credential \u2014 such as a digital certificate, verifiable credential, or identity assertion \u2014 issued by a trusted authority about a subject, so that a relying party can grant access or trust without real-time contact with the original issuer. In decentralised identity systems it relies on public-key cryptography, digital signatures, and optionally distributed ledgers to enable privacy-preserving, tamper-evident verification. The process must also resolve issuer keys and query revocation registries (via CRL, OCSP, or on-chain status) to ensure invalidated credentials cannot be fraudulently reused. Selective-disclosure techniques, including zero-knowledge proofs and SD-JWTs, allow subjects to prove specific attributes without exposing unnecessary personal data.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:credential-verification", "labels": [ "Credential Verification", "Offline Credential Verification" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "credit-risk", "title": "Credit Risk", "domain": "finance", "domain_name": "Finance", "definition": "The risk of financial loss arising from a borrower or counterparty failing to meet contractual obligations, quantified through probability of default, loss given default and exposure at default, and managed through credit scoring, collateralisation, diversification, provisioning and regulatory capital requirements; the dominant risk category on most bank balance sheets and the core determinant of loan pricing in both traditional and decentralised lending.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:credit-risk", "labels": [ "Credit Risk" ], "is_subclass_of": [ "Risk" ], "wikilinks": [ "Risk", "Risk Management", "Operational Risk", "Lending Protocol" ] }, { "id": "crew-active-dosimeter", "title": "Crew Active Dosimeter", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:crew-active-dosimeter", "labels": [ "Crew Active Dosimeter" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "crew-health-monitoring", "title": "Crew Health Monitoring", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:crew-health-monitoring", "labels": [ "Crew Health Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "crew-ai", "title": "CrewAI", "domain": "ai", "domain_name": "Ai", "definition": "CrewAI is an open-source Python framework for orchestrating teams of autonomous LLM-powered agents that collaborate on complex tasks through defined roles, backstories, goals, and structured processes. It models a crew of specialised agents\u2014each with distinct capabilities and a designated toolset\u2014that delegate work, share context through a shared memory system, and execute sequential, parallel, or hierarchical workflows toward a common objective. With 47,800+ GitHub stars, 27 million downloads, and adoption by 63% of the Fortune 500 as of mid-2026, CrewAI has become one of the fastest-growing multi-agent orchestration frameworks in the agentic AI ecosystem.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:crew-ai", "labels": [ "CrewAI" ], "is_subclass_of": [ "AI Agent System", "Multi-Agent Systems", "CLI Multi-Agent Systems", "Agent Frameworks", "Agentic Workflow" ], "wikilinks": [ "CLI Multi-Agent Systems", "Agentic Workflow", "Large Language Model", "Multi-Agent Systems", "Autonomous Agent", "Tool Use", "Function Calling", "Orchestration", "Agent Memory", "Task Planning", "Retrieval-Augmented Generation", "Model Context Protocol", "AutoGen", "LangGraph", "OpenAI Agents SDK", "ReAct Pattern", "Chain of Thought", "Human-in-the-Loop", "Prompt Engineering", "Prompt Injection" ] }, { "id": "crewed-spacecraft", "title": "Crewed Spacecraft", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:crewed-spacecraft", "labels": [ "Crewed Spacecraft" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "crop-monitoring", "title": "Crop Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:crop-monitoring", "labels": [ "Crop Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "cross-attention", "title": "Cross Attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An attention mechanism where queries come from one sequence whilst keys and values come from a different sequence, enabling information flow between the encoder and decoder in sequence-to-sequence models.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:cross-attention", "labels": [ "Cross Attention", "Cross-Attention", "Cross-Attention Fusion" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "ICLR", "IEEE Transactions on Medical Imaging", "KnoWhere", "Vaswani et al. Transformers", "Computer Vision", "MetaverseDomain", "NVIDIA Omniverse", "PEOPLE" ] }, { "id": "cross-border-authentication", "title": "Cross Border Authentication", "domain": "governance", "domain_name": "Governance", "definition": "Cross border authentication is the verification of a person or entity's identity across national or jurisdictional boundaries, so that credentials issued in one country are trusted and accepted in another. It depends on interoperable identity schemes, mutual-recognition agreements and trust frameworks that reconcile differing legal and technical regimes. The capability enables seamless access to services, regulated transactions and travel without re-establishing identity from scratch in each jurisdiction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:cross-border-authentication", "labels": [ "Cross Border Authentication", "Cross-Border Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "cross-border-compliance", "title": "Cross Border Compliance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multi-jurisdictional regulatory compliance framework governing organisations operating across national boundaries, requiring simultaneous adherence to overlapping and often conflicting legal regimes including data protection law (GDPR, UK GDPR, CCPA, PIPL, DPDPA 2023), AI regu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cross-border-compliance", "labels": [ "Cross Border Compliance" ], "is_subclass_of": [ "Legal and Regulatory", "Compliance Programs", "Regulatory Governance", "Legal Risk Management", "International Law", "Corporate Governance" ], "wikilinks": [ "Adequacy Decisions", "AI Regulation", "Anti-Money Laundering", "Bank Secrecy Act", "Basel Committee", "Basel III Standards", "BC-0479-regulatory-compliance", "BC-0480-kyc-requirements", "BC-0484-markets-in-crypto-assets", "BC-0485-travel-rule", "BC-0486-regulatory-reporting", "BC-0487-compliance-monitoring", "BC-0488-licensing-requirements", "BC-0489-consumer-protection", "Binance", "Binding Corporate Rules", "BitMEX", "Bradford Brussels Effect", "Brummer Minilateralism", "Brussels Effect" ] }, { "id": "cross-border-regulatory-cooperation", "title": "Cross Border Regulatory Cooperation", "domain": "finance", "domain_name": "Finance", "definition": "Cross-border regulatory cooperation is the coordination of financial supervision, rule-making and enforcement across national jurisdictions to manage activities and risks that span borders. It operates through memoranda of understanding, supervisory colleges, information-sharing arrangements, mutual recognition and common standards set by international bodies. By aligning approaches to systemic risk, anti-money-laundering and cross-border payments, it reduces regulatory arbitrage and supports consistent oversight of globally active firms and markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cross-border-regulatory-cooperation", "labels": [ "Cross Border Regulatory Cooperation", "Cross-Border Regulatory Cooperation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "cross-chain-asset-transfer", "title": "Cross Chain Asset Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-chain asset transfer is the cryptographic and protocol-level process of moving digital assets \u2014 cryptocurrencies, fungible tokens, or non-fungible tokens \u2014 from one blockchain network to a distinct, independent blockchain network while preserving the asset's economic properties and enforcing integrity guarantees across the transfer. Because independent blockchains maintain no shared global state, protocols must ensure that an asset locked or burned on the source chain is atomically minted or released on the destination chain, preventing double-spend and maintaining supply conservation. Implementations range from hash time-locked contracts (HTLCs) and lock-and-mint bridge contracts overseen by validator committees to native inter-blockchain communication protocols (IBC) and zero-knowledge proof-based verification bridges that eliminate trusted attestors entirely.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-asset-transfer", "labels": [ "Cross Chain Asset Transfer", "Cross-Chain Asset Transfer", "Cross-Chain Token Transfer", "Cross-Chain Transfer" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "cross-chain-liquidity", "title": "Cross Chain Liquidity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-chain liquidity refers to the availability and free movement of tradable assets across multiple, otherwise isolated blockchain networks. It is achieved through bridges, relayers, liquidity pools and atomic swaps that let value flow between chains without a single custodian. Robust cross-chain liquidity reduces fragmentation, tightens spreads and enables composable decentralised finance across heterogeneous ecosystems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-liquidity", "labels": [ "Cross Chain Liquidity", "Cross-Chain Liquidity" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "cross-domain-authentication", "title": "Cross Domain Authentication", "domain": "security", "domain_name": "Security", "definition": "Cross-domain authentication is the capability for a principal authenticated in one security or administrative domain to prove its identity to services in another domain without re-enrolling separate credentials. It relies on federated trust relationships and standard token exchanges so that an identity provider in one realm is accepted by relying parties in another. This underpins single sign-on across organisations and is essential to federated and distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cross-domain-authentication", "labels": [ "Cross Domain Authentication", "Cross-Domain Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "cross-domain-bridge", "title": "Cross Domain Bridge", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Cross Domain Bridge is an architectural pattern, protocol, or middleware system that enables interoperability and data exchange between heterogeneous collaboration platforms, virtual environments, and organisational systems. Implementations include protocol translators (WebRTC, SIP, H.323), identity federation layers (SAML, OAuth, OpenID Connect), and data transformation services (REST, GraphQL, gRPC). Advanced bridges leverage blockchain for trustless cross-domain transactions and open standards such as OpenXR for spatial interoperability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cross-domain-bridge", "labels": [ "Cross Domain Bridge" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "IEEE P2874", "IETF OAUTH", "OMG", "W3C Federated Identity Working Group", "Blockchain", "OpenXR" ] }, { "id": "cross-metaverse-commerce", "title": "Cross Metaverse Commerce", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Economic activities and transactions that span multiple metaverse platforms and virtual worlds, enabled by interoperability standards, blockchain-based digital assets, and unified digital identity systems that allow users to buy, sell, trade, and transfer value seamlessly across different virtual...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cross-metaverse-commerce", "labels": [ "Cross Metaverse Commerce", "Cross-Metaverse Commerce" ], "is_subclass_of": [ "Platform and Environment", "Virtual Economy" ], "wikilinks": [ "Asset Portability", "Cross-Platform Trading", "Interoperability Standards", "Virtual Market Integration", "Blockchain", "Blockchain Infrastructure", "Digital Wallet", "metaverse", "Virtual Economy" ] }, { "id": "cross-modal-retrieval", "title": "Cross Modal Retrieval", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-modal retrieval is the task of retrieving items in one modality (such as images) using a query expressed in a different modality (such as text), and vice versa. It relies on learning a shared embedding space in which semantically corresponding items across modalities lie close together, so that similarity search can bridge the modality gap. Contrastive vision-language models are the dominant approach, enabling text-to-image search, image captioning retrieval and multimodal recommendation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cross-modal-retrieval", "labels": [ "Cross Modal Retrieval", "Cross-Modal Retrieval" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "cross-platform-digital-twins", "title": "Cross Platform Digital Twins", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital twin systems designed for interoperability across different software platforms and environments, enabling real-time data exchange, collaborative simulation, and unified virtual representations of physical assets or processes that can be accessed and manipulated from multiple compatible sy...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cross-platform-digital-twins", "labels": [ "Cross Platform Digital Twins", "Cross-Platform Digital Twins" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin Technology" ], "wikilinks": [ "API Integration", "Collaborative Simulation", "Interoperability Standards", "Multi-Vendor Ecosystems", "Computer Vision", "Data Integration", "Digital Twin Technology", "metaverse", "Universal Scene Description" ] }, { "id": "cross-border-data-transfer-rule", "title": "Cross-Border Data Transfer Rule", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Regulatory framework governing international movement of personal and sensitive data across jurisdictions, ensuring privacy protection through adequacy assessments and safeguarding mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:cross-border-data-transfer-rule", "labels": [ "Cross-Border Data Transfer Rule" ], "is_subclass_of": [ "Legal and Regulatory", "Privacy Framework" ], "wikilinks": [ "Adequacy Decision Framework", "APEC CBPR", "Binding Corporate Rules", "Data Privacy Governance Framework", "Data Protection Authority Notification", "EU-US Data Privacy Framework", "GDPR", "GDPR Article 45", "International Data Flows", "Legal Basis Determination", "OECD Privacy Framework", "Standard Contractual Clauses", "Transfer Impact Assessment", "User Privacy Protection", "Compliance Verification", "Global Metaverse Operations", "MiddlewareLayer", "Privacy Impact Assessment", "Telecollaboration", "TrustAndGovernanceDomain" ] }, { "id": "cross-border-data-transfer", "title": "Cross-Border Data Transfer", "domain": "governance", "domain_name": "Governance", "definition": "Cross-border data transfer is the movement of personal or regulated data across national jurisdictions, governed by data-protection laws that restrict where and how such data may be processed. Compliance mechanisms include adequacy decisions, standard contractual clauses, binding corporate rules, and localisation requirements. It is a central concern of privacy regimes such as GDPR and various Asia-Pacific frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cross-border-data-transfer", "labels": [ "Cross-Border Data Transfer", "Cross-Border Data Flows", "Cross-Border Data Sharing" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "cross-border-enforcement", "title": "Cross-Border Enforcement", "domain": "governance", "domain_name": "Governance", "definition": "Cross-border enforcement is the process by which legal judgments, regulatory actions, or compliance obligations are recognised and executed across national jurisdictions. It depends on treaties, mutual legal assistance, and cooperation between regulators to act on actors and assets located abroad. It is a persistent challenge for digital and crypto markets that operate without regard to territorial borders.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-border-enforcement", "labels": [ "Cross-Border Enforcement" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "cross-border-identity", "title": "Cross-Border Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-border identity refers to digital identity credentials and verification mechanisms designed to be recognised and trusted across national jurisdictions, allowing an individual or organisation to prove attributes about themselves to a relying party in a different country without relying on a single centralised, national identity authority. It builds on self-sovereign identity principles and interoperable trust frameworks, such as those developed by the Trust Over IP Foundation, to establish mutual recognition of credential formats, issuers and verification methods between jurisdictions. Cross-border identity is a key enabler for international travel, remote work and cross-border financial services that require reliable identity assurance.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:cross-border-identity", "labels": [ "Cross-Border Identity" ], "is_subclass_of": [ "Self Sovereign Identity" ], "wikilinks": [] }, { "id": "cross-border-payment-transparency", "title": "Cross-Border Payment Transparency", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-border payment transparency is the requirement and practice of making the originator, beneficiary, fees, and routing of international payments visible to counterparties and regulators. It supports anti-money-laundering controls and consumer protection by exposing hidden costs and identifying parties to a transfer. In crypto it is advanced through measures such as the FATF Travel Rule and on-chain settlement traceability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-border-payment-transparency", "labels": [ "Cross-Border Payment Transparency" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "cross-border-payments", "title": "cross-border payments", "domain": "finance", "domain_name": "Finance", "definition": "Cross-Border Payments are financial transactions that originate in one country and are settled in another, requiring currency conversion, multi-jurisdiction regulatory compliance, and interoperability between disparate payment systems. Traditional correspondent banking networks rely on SWIFT messaging and chains of intermediary banks, incurring multi-day settlement cycles, layered fees, and significant opacity. Blockchain-based payment rails\u2014including stablecoins, payment channels, and Central Bank Digital Currencies\u2014compress settlement latency and reduce intermediary costs by enabling atomic finality and programmable escrow. The domain sits at the intersection of monetary policy, AML/KYC regulation, ISO 20022 standardisation, and distributed-ledger technology.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:cross-border-payments", "labels": [ "Cross-Border Payments", "Borderless Payments", "Cross Border Payments", "Cross-Border Payment", "Cross-Border Retail Payment", "G20 Cross-Border Payments Initiative", "International Payments System" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "cross-border-remittances", "title": "Cross-Border Remittances", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-border remittances are transfers of money by individuals\u2014typically migrant workers\u2014to recipients in another country, often family members. Traditionally routed through banks and money-transfer operators with high fees and slow settlement, they are increasingly served by stablecoins and crypto rails that lower cost and latency. They represent a major global financial flow and a key driver of crypto payment adoption.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-border-remittances", "labels": [ "Cross-Border Remittances", "Cross-Border Remittance" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "cross-border-settlement", "title": "Cross-Border Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Cross-border settlement is the process of finalising the transfer of funds or assets between counterparties in different jurisdictions, including currency conversion, compliance checks, and the irrevocable discharge of obligations across correspondent banking chains, central bank systems, or emerging blockchain-based rails. It encompasses the full lifecycle from trade initiation to finality across multiple legal, regulatory, and technical environments.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:cross-border-settlement", "labels": [ "Cross-Border Settlement", "Cross-border Tax Settlement" ], "is_subclass_of": [ "Cross-Border Payments" ], "wikilinks": [] }, { "id": "cross-border-transfer", "title": "Cross-Border Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cross-border transfer is the movement of value or funds between parties located in different countries, settled across distinct currency and regulatory regimes. It encompasses bank wires, card networks, money-transfer operators, and increasingly digital-payment and cryptocurrency rails. Speed, cost, foreign-exchange handling, and compliance with multiple jurisdictions are its defining constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cross-border-transfer", "labels": [ "Cross-Border Transfer" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "cross-chain-bridge", "title": "Cross-Chain Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Cross-Chain Bridge is a protocol-level interoperability artefact \u2014 comprising on-chain smart contracts on each connected blockchain plus an off-chain attestation, relay, or proving infrastructure \u2014 that enables the trust-minimised transfer of fungible tokens, non-fungible tokens, arbitrary mess...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cross-chain-bridge", "labels": [ "Cross-Chain Bridge", "Cross-Chain Bridge Protocol", "Cross-Chain Bridge Security", "Cross-Chain Bridging", "CrossChainBridge" ], "is_subclass_of": [ "Network Component", "Blockchain Interoperability", "Interoperability Protocol", "Cross-Chain Application", "Asset Transfer Mechanism", "Smart Contract Protocol" ], "wikilinks": [ "Across Protocol", "Across Protocol Whitepaper 2022", "Asset Transfer Mechanism", "Atomic-Swap-Only Protocol", "Augusto et al. 2024 SoK Bridge Security Privacy", "Axelar", "Axelar Whitepaper 2021", "Belchior et al. 2021 Blockchain Interoperability Survey", "Bhuptani 2021 Connext Trust Taxonomy", "Bridge Contract", "Bridged Stablecoin", "Burn-and-Mint Bridge", "Buterin 2016 Chain Interoperability", "Buterin 2021 Cross-Chain Applications Critique", "Cambridge CCAF Travel Rule 2023", "Centralised Exchange Bridging", "Chain Abstraction", "Chain Abstraction Wallet", "Chain Agnostic Standards Alliance", "Chainalysis 2024 Crypto Crime Report" ] }, { "id": "cross-chain-bridges", "title": "Cross-Chain Bridges", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The category of protocols that move assets and data between distinct blockchains, treated collectively as connective infrastructure between ledgers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cross-chain-bridges", "labels": [ "Cross-Chain Bridges", "Trustless Bridges" ], "is_subclass_of": [ "Blockchain Interoperability" ], "wikilinks": [ "Smart Contract", "Cross Chain Asset Transfer", "Cross-Chain Bridge", "Bridge", "Blockchain Interoperability" ] }, { "id": "cross-chain-communication", "title": "Cross-Chain Communication", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-chain communication is the exchange of messages, data, and asset-transfer instructions between distinct blockchain networks that do not natively share state. It is implemented through protocols, relays, and light-client verification that let one chain trust and act on events from another. It is the foundation of blockchain interoperability and multi-chain applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-communication", "labels": [ "Cross-Chain Communication", "Cross-Chain Function Calls" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "cross-chain-composability", "title": "Cross-Chain Composability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-chain composability is the capacity for smart contracts and protocols deployed on different blockchain networks to interoperate and be combined into higher-order applications, as if operating within a single execution environment. It depends on interoperability infrastructure such as cross-chain bridges and messaging protocols to relay state and asset transfers between chains reliably. Cross-chain composability extends the general property of composability beyond a single chain, enabling multi-chain DeFi strategies and cross-chain governance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:cross-chain-composability", "labels": [ "Cross-Chain Composability" ], "is_subclass_of": [ "Composability" ], "wikilinks": [] }, { "id": "cross-chain-dex", "title": "Cross-Chain DEX", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cross-chain DEX is a decentralised exchange that lets users swap assets native to different blockchains without a centralised custodian. It relies on bridges, atomic swaps, or liquidity-network protocols to coordinate settlement across chains while preserving non-custodial trading. It extends decentralised finance beyond single-chain liquidity silos.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-dex", "labels": [ "Cross-Chain DEX", "Cross-Chain DEX Aggregation" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "cross-chain-governance", "title": "cross-chain governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-Chain Governance is a coordination framework and set of protocols that enable governance proposals, voting outcomes, and policy updates to propagate and be enforced across multiple heterogeneous blockchain networks without requiring a single trusted intermediary. It extends on-chain governance mechanisms \u2014 token-weighted voting, quadratic voting, time-locked execution \u2014 to multi-chain environments using interoperability layers such as IBC (Inter-Blockchain Communication), cross-chain message-passing bridges, or relay networks. Achieving consistent governance state across chains requires solving distributed consensus problems while preserving each chain's sovereignty. The field sits at the intersection of blockchain interoperability, distributed systems coordination, and decentralised decision-making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-governance", "labels": [ "Cross-Chain Governance", "Multi-Chain Governance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "cross-chain-interoperability", "title": "Cross-Chain Interoperability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-Chain Interoperability is the technical capacity for distinct blockchain networks to communicate, transfer assets, and share arbitrary state without relying on a centralised intermediary, achieved through mechanisms such as light-client bridges, relay chains, atomic swaps, and standardised inter-blockchain communication protocols. It addresses the heterogeneous consensus problem\u2014the challenge of enabling two networks with different finality guarantees and trust models to agree on the validity of cross-chain events\u2014through cryptographic proofs, validator sets, or shared security frameworks. Prominent implementations include the Cosmos IBC protocol, Polkadot's Cross-Consensus Message Format (XCM), LayerZero's oracle-relayer model, and zero-knowledge proof bridges that verify source-chain state transitions with minimal on-chain trust. The field is foundational to the composable multi-chain ecosystem in which digital assets, governance rights, and smart contract logic can flow freely across sovereign networks.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:cross-chain-interoperability", "labels": [ "Cross-Chain Interoperability", "Cross-Chain Interoperability Protocol", "Multi-Chain Interoperability" ], "is_subclass_of": [ "Blockchain Interoperability" ], "wikilinks": [] }, { "id": "cross-chain-messaging", "title": "Cross-Chain Messaging", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocols and mechanisms that enable communication and data transfer between different blockchain networks, facilitating interoperability and cross-chain applications without centralised intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-messaging", "labels": [ "Cross-Chain Messaging", "Cross Chain Messaging", "cross-chain-messaging" ], "is_subclass_of": [ "Interoperability Protocol", "Message Passing System" ], "wikilinks": [ "Asset Movement", "Blockchain Proof", "Cross-Chain Function Calls", "Decentralised Computation", "Handler Execution", "Interoperability Protocol", "Message Authentication", "Message Passing System", "Message Queue", "Proof Verification", "Relay Network", "State Synchronisation", "AI Agent System", "BlockchainDomain", "Blockchain Interoperability" ] }, { "id": "cross-chain-nft", "title": "Cross-Chain NFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cross-chain NFT is a non-fungible token that can move or be represented across multiple blockchains while preserving its identity, ownership, and metadata. It uses bridges, burn-and-mint mechanisms, or messaging protocols to transfer the canonical token between networks. It enables NFTs to access liquidity, marketplaces, and applications on chains other than where they were minted.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-nft", "labels": [ "Cross-Chain NFT" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "cross-chain-swap", "title": "Cross-Chain Swap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cross-chain swap is the exchange of a token or asset held on one blockchain network for a different asset held on another network, without relying on a centralised exchange as intermediary. It is typically executed via a cross-chain bridge that locks or burns the source asset and mints or releases an equivalent representation on the destination chain, or via atomic swap protocols that guarantee both legs settle or neither does. Cross-chain swaps are a core primitive for multi-chain liquidity, enabling decentralised exchanges such as Osmosis to route trades across ecosystems.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:cross-chain-swap", "labels": [ "Cross-Chain Swap" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "cross-dao-benchmarking", "title": "Cross-DAO Benchmarking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-DAO benchmarking is the comparative analysis of decentralised autonomous organisations against shared metrics such as treasury size, voter participation, proposal throughput, and contributor activity. It aggregates on-chain governance and financial data to rank or contextualise a DAO's performance relative to peers. It informs governance design, delegate accountability, and treasury strategy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-dao-benchmarking", "labels": [ "Cross-DAO Benchmarking" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "cross-domain-reference-corpus", "title": "Cross-Domain Reference Corpus", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated collection of external hyperlinks spanning blockchain, cryptocurrency, augmented and virtual reality, AI generative tools, and media production resources, assembled as a research reference page for cross-domain topics relevant to the NarrativeGoldmine knowledge graph.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cross-domain-reference-corpus", "labels": [ "Cross-Domain Reference Corpus", "Various Links" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "perplexity" ] }, { "id": "cross-encoder-reranking", "title": "Cross-Encoder Reranking", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Cross-Encoder Reranking is a two-stage information retrieval technique in which a cross-encoder transformer model receives a query and a candidate document concatenated as a single input sequence, performs full bidirectional self-attention across both, and outputs a relevance score used to re-order an initial candidate set retrieved by a faster but less accurate first-stage retriever. It typically yields substantially higher ranking quality than bi-encoder first-stage retrieval at the cost of higher computational latency.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cross-encoder-reranking", "labels": [ "Cross-Encoder Reranking" ], "is_subclass_of": [ "Information Retrieval", "Neural Information Retrieval" ], "wikilinks": [ "Information Retrieval", "Semantic Search", "Embedding Model", "Retrieval-Augmented Generation", "RAG Pipeline", "Transformer", "BERT", "BM25", "Dense Retrieval", "Hybrid Retrieval", "Natural Language Processing", "Question Answering", "Document Retrieval", "Knowledge Distillation", "Cosine Similarity", "Self-Attention", "Approximate Nearest Neighbour Search", "Dense Passage Retrieval", "Enterprise Search", "Neural Information Retrieval" ] }, { "id": "cross-entropy-loss", "title": "Cross-Entropy Loss", "domain": "ai", "domain_name": "Ai", "definition": "Cross-entropy loss is a differentiable scalar objective function that measures the dissimilarity between a model's predicted probability distribution and the true label distribution, computed as the negative log-likelihood of the correct class under the model's output. It is the canonical training objective for classification tasks and language modelling, directly optimising the model to assign maximum probability mass to correct outputs. Minimising cross-entropy is equivalent to maximising the likelihood of the training data under the model's parameterised distribution.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cross-entropy-loss", "labels": [ "Cross-Entropy Loss", "Cross Entropy Loss" ], "is_subclass_of": [ "Loss Function", "Maximum Likelihood Estimation" ], "wikilinks": [ "Loss Function", "Backpropagation", "Gradient Descent", "Supervised Learning", "Language Modeling", "Transformer", "Large Language Models", "Softmax Function", "Neural Network", "Deep Learning", "Natural Language Processing", "Model Training", "KL Divergence", "Information Theory", "Activation Function", "Stochastic Gradient Descent", "Label Smoothing", "Focal Loss", "Direct Preference Optimisation", "Reinforcement Learning from Human Feedback" ] }, { "id": "cross-functional-collaboration", "title": "Cross-Functional Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Cross-functional collaboration is coordinated work between people from different disciplines or organisational units \u2014 such as engineering, design, and operations \u2014 combining diverse expertise toward a shared goal. In distributed and remote settings it depends on digital workplace platforms and virtual collaboration tools to bridge the coordination gaps that would otherwise be closed by physical co-location. It is a recurring requirement for digital workplace platforms and virtual collaboration systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:cross-functional-collaboration", "labels": [ "Cross-Functional Collaboration" ], "is_subclass_of": [ "Virtual Collaboration" ], "wikilinks": [] }, { "id": "cross-modal-conditioning", "title": "Cross-Modal Conditioning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-modal conditioning is a generative modelling technique in which a model producing output in one sensory or representational modality is guided at inference time by a conditioning signal derived from a different modality, typically via cross-attention or adapter mechanisms that inject encoded representations of the conditioning input into the backbone network's intermediate layers. It is the foundational mechanism enabling text-to-image synthesis, text-to-audio generation, audio-driven video synthesis, depth-conditioned inpainting, and other heterogeneous generation tasks where the semantic intent is expressed in one modality and realised in another.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cross-modal-conditioning", "labels": [ "Cross-Modal Conditioning" ], "is_subclass_of": [ "AI Technique", "Attention Mechanism", "Multimodal Learning" ], "wikilinks": [ "Attention Mechanism", "Cross Attention", "Multimodal Learning", "Diffusion Model", "Text-to-Image", "Classifier-Free Guidance", "Embedding", "Transformer Architecture", "Latent Space", "U-Net", "CLIP", "Stable Diffusion", "Foundation Model", "Representation Learning", "Contrastive Learning", "Audio Synthesis", "Video Generation", "Image Generation", "Natural Language Processing", "Computer Vision" ] }, { "id": "cross-platform-asset-exchange", "title": "Cross-Platform Asset Exchange", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Cross-platform asset exchange is the capability to move 3D models, textures, and other digital assets between different metaverse platforms, engines, and tools while preserving fidelity and usability. It depends on shared 3D file formats and import/export pipelines that map assets onto each platform's runtime. It is a prerequisite for interoperable virtual worlds and portable user content.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-asset-exchange", "labels": [ "Cross-Platform Asset Exchange", "Cross-Platform Exchange" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "cross-platform-authentication", "title": "Cross-Platform Authentication", "domain": "security", "domain_name": "Security", "definition": "Cross-platform authentication is the verification of a user's identity in a way that is valid across multiple platforms, devices, or virtual worlds without re-registration. It uses federated identity, single sign-on, and portable credentials so that one verified identity unlocks access everywhere. It is a foundational requirement for interoperable avatars and multiverse experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-authentication", "labels": [ "Cross-Platform Authentication", "Cross-Device Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "cross-platform-compatibility", "title": "Cross-Platform Compatibility", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cross-platform compatibility is the property of a software system, application, or digital asset that allows it to function correctly and consistently across multiple operating systems, hardware architectures, runtime environments, or device categories without requiring separate implementations. Achieving this property typically involves adherence to open standards, abstraction layers, and rigorous testing regimes that surface divergent behaviour across target environments. In immersive and extended-reality contexts it extends to ensuring that 3D content, avatars, and application logic behave equivalently across headsets, smartphones, and desktop viewers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cross-platform-compatibility", "labels": [ "Cross-Platform Compatibility" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "cross-platform-compliance-hub", "title": "Cross-Platform Compliance Hub", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A unified regulatory compliance system that harmonizes and coordinates compliance activities across multiple platforms, jurisdictions, and regulatory frameworks through centralized policy management and audit aggregation.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cross-platform-compliance-hub", "labels": [ "Cross-Platform Compliance Hub" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance", "Regulatory Compliance Framework" ], "wikilinks": [ "Audit Aggregator", "Cross-Platform Auditing", "Data Classification System", "GDPR", "Governance Infrastructure", "ISO 27001", "Legal Framework Database", "Multi-Jurisdictional Policy Store", "Platform Integration API", "Policy Synchronization", "Regulatory Compliance Framework", "Regulatory Harmonization", "Regulatory Mapping Engine", "Reporting Engine", "Risk Assessment Module", "SOC 2", "Unified Compliance Reporting", "DORA", "NIS2", "EU AI Act" ] }, { "id": "cross-platform-content", "title": "Cross-Platform Content", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Cross-platform content is digital media and 3D assets authored once and made usable across multiple platforms, engines, or devices through standardised formats and conversion pipelines. It depends on format parsers and content pipelines that normalise assets to each target's runtime requirements. It enables creators to reach diverse metaverse environments without bespoke re-authoring.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-content", "labels": [ "Cross-Platform Content", "Cross-Platform 3D Content" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "cross-platform-identity", "title": "Cross-Platform Identity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cross-platform identity refers to the capability of linking and managing a user's electronic identity and attributes across multiple distinct systems, platforms, and organisational boundaries, enabling seamless authentication and authorisation across heterogeneous environments through federated identity management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-identity", "labels": [ "Cross-Platform Identity" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [ "Core Technology", "Federated Access", "Single Sign-On", "Trust Relationships", "User Experience", "Blockchain", "Identity Federation", "Identity Management" ] }, { "id": "cross-platform-interoperability", "title": "Cross-Platform Interoperability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The capability for heterogeneous blockchain networks, software platforms, and distributed systems to communicate, exchange data, and transfer value seamlessly without centralized intermediaries. Achieved through standardized protocols, light-client verification, and message-passing frameworks that eliminate trusted third parties.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-interoperability", "labels": [ "Cross-Platform Interoperability" ], "is_subclass_of": [ "Network and Communication", "Next Generation Mobile" ], "wikilinks": [ "DeFi", "ERC-5164", "IEEE blockchain standards", "Next Generation Mobile", "W3C DID", "Blockchain", "Cross-Chain Bridge", "Layer 0", "Light Client", "Relayer" ] }, { "id": "cross-platform-rendering", "title": "Cross-Platform Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-platform rendering is a rendering approach that produces consistent visual output across different operating systems, hardware GPUs and graphics APIs, typically by targeting an abstraction layer rather than a single vendor-specific API. It allows a single rendering codebase to run on Windows, macOS, Linux, mobile and web targets, translating draw calls to the underlying native graphics API such as Vulkan, Metal or DirectX. Standards bodies such as the Khronos Group maintain the graphics APIs and specifications that make cross-platform rendering practical.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:cross-platform-rendering", "labels": [ "Cross-Platform Rendering" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "cross-system-querying", "title": "Cross-System Querying", "domain": "data", "domain_name": "Data", "definition": "Cross-system querying is the ability to issue a single query that retrieves and joins data residing in multiple independent systems or data stores. It is achieved through federation surfaces, linked-data encoding, and query mediators that translate and route requests to heterogeneous sources. It enables unified access to distributed data without centralising or duplicating it.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cross-system-querying", "labels": [ "Cross-System Querying" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "cross-validation", "title": "Cross-Validation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Cross-validation is a statistical resampling technique for evaluating machine learning model performance by partitioning available data into complementary training and validation subsets, training the model on each partition in turn, and averaging the resulting error estimates. It provides a less optimistic and more generalisable estimate of out-of-sample predictive performance than a single train\u2013test split.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cross-validation", "labels": [ "Cross-Validation", "Cross Validation", "Cross-Validation Strategy" ], "is_subclass_of": [ "Model Evaluation", "AI Technique", "Statistical Learning Theory", "Resampling" ], "wikilinks": [ "AI Machine Learning", "Model Evaluation", "Model Selection", "Bias-Variance Tradeoff", "Overfitting", "Hyperparameter Tuning", "Regularisation", "Ensemble Methods", "Statistical Learning Theory", "Supervised Learning", "Deep Learning", "Neural Network", "Convolutional Neural Network", "Applied Machine Learning", "Benign Overfitting", "Annotated Training Data", "Statistical Inference", "Data Governance", "Resampling", "Performance Metric" ] }, { "id": "crowd-simulation", "title": "Crowd Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Crowd simulation models the collective movement and behaviour of large numbers of agents -- pedestrians, vehicles or non-player characters -- navigating shared space while responding to obstacles, goals and one another. It combines agent-based modelling with local avoidance and flocking rules to produce plausible emergent group behaviour at real-time or near-real-time speed. It is used in procedural animation, urban planning tools and game and simulation environments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:crowd-simulation", "labels": [ "Crowd Simulation" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "crowdfunding", "title": "Crowdfunding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Crowdfunding is a financing model in which a project or venture raises capital from a large number of contributors, each providing a relatively small amount, typically through an online platform. Models range from donation and reward-based campaigns to equity and debt offerings, and blockchain variants use token sales to distribute ownership or utility rights. The approach broadens access to early-stage capital while distributing risk across many backers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:crowdfunding", "labels": [ "Crowdfunding" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "crowdsourcing", "title": "Crowdsourcing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Crowdsourcing is the practice of obtaining contributions, labour or judgements from a large distributed group of people, typically through an open call mediated by an online platform. In machine learning it is widely used to collect, label and validate training data by decomposing work into microtasks distributed across many contributors. Effective crowdsourcing combines incentive design with quality-control mechanisms to aggregate noisy individual inputs into reliable results.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:crowdsourcing", "labels": [ "Crowdsourcing" ], "is_subclass_of": [ "Data Collection" ], "wikilinks": [] }, { "id": "cruise-phase", "title": "Cruise Phase", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cruise-phase", "labels": [ "Cruise Phase" ], "is_subclass_of": [ "Mission Phase" ], "wikilinks": [] }, { "id": "crypto-asset-service-provider", "title": "Crypto Asset Service Provider", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A legal person authorised to provide crypto-asset services to third parties on a professional basis \u2014 custody, exchange operation, order execution, placement, transfer services, and advice \u2014 as defined by the EU's Markets in Crypto-Assets Regulation (MiCA); CASPs must obtain authorisation, meet prudential and governance requirements, and comply with anti-money-laundering and travel-rule obligations, gaining passporting rights across the EU single market.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:crypto-asset-service-provider", "labels": [ "Crypto Asset Service Provider", "Crypto-Asset Service Provider" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [ "EU MiCA Regulation", "Cryptocurrency Exchange", "Transfer Of Funds Regulation" ] }, { "id": "crypto-climate-accord", "title": "Crypto Climate Accord", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Crypto Climate Accord is a private-sector initiative launched in 2021 to decarbonise the cryptocurrency and blockchain industry, modelled on the Paris Climate Agreement. It commits signatory companies and projects to achieving net-zero greenhouse-gas emissions from electricity consumption by 2030 and to developing open-source accounting standards and tools for measuring and reporting crypto energy use.", "entityType": "Class", "qualityScore": 0.82, "maturity": "emerging", "iri": "urn:ngm:class:crypto-climate-accord", "labels": [ "Crypto Climate Accord" ], "is_subclass_of": [ "ESG" ], "wikilinks": [] }, { "id": "crypto-regulation", "title": "Crypto Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Crypto regulation is the body of laws and supervisory rules governing the issuance, trading, custody and use of crypto-assets, including authorisation of service providers, disclosure requirements, market conduct rules, client-asset custody standards, and anti-money-laundering obligations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:crypto-regulation", "labels": [ "Crypto Regulation", "Crypto-Asset Regulation", "SEC Crypto Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Regulatory Frameworks", "Investor Protection", "Anti-Money Laundering", "Financial Regulation", "https://www.esma.europa.eu/esmas-activities/digital-finance-and-innovation/markets-crypto-assets-regulation-mica", "https://www.fsb.org/work-of-the-fsb/financial-innovation-and-structural-change/crypto-assets-and-global-stablecoins/" ] }, { "id": "crypto-token", "title": "Crypto Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain-based programmable token representing assets, rights, or utility within a decentralized system, with transferability governed by smart contract logic.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:crypto-token", "labels": [ "Crypto Token", "CryptoToken" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Cryptographic Key", "Governance Voting", "ISO 24165", "Programmable Value", "Reed Smith", "Tokenization System", "AI Agent System", "Blockchain", "BlockchainDomain", "Blockchain Network", "Consensus Mechanism", "Decentralized Exchange", "Digital Ownership", "Loyalty Token", "Metadata Schema", "Middleware Layer", "Non-Fungible Token (NFT)", "Smart Contract", "Stablecoin", "Token Standard" ] }, { "id": "crypto-trading", "title": "Crypto Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Crypto trading is the buying and selling of cryptocurrencies and digital assets for profit, conducted on centralised exchanges, decentralised exchanges, and over-the-counter desks. It spans spot, margin, derivatives, and algorithmic strategies, with prices driven by liquidity, sentiment, and on-chain activity. It is a primary use case and liquidity source for the broader crypto-asset market.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:crypto-trading", "labels": [ "Crypto Trading", "Cryptocurrency Trading" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "crypto-wallet", "title": "Crypto Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A crypto wallet is software or hardware that manages the private keys controlling blockchain assets and enables users to sign transactions, hold tokens, and interact with decentralised applications. Wallets may be custodial or non-custodial, hot or cold, and increasingly serve as the identity and access layer for Web3. They do not store assets themselves but the cryptographic keys that authorise control over on-chain balances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:crypto-wallet", "labels": [ "Crypto Wallet" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "cryptoasset-regulation", "title": "Cryptoasset Regulation", "domain": "finance", "domain_name": "Finance", "definition": "Cryptoasset regulation is the body of laws, supervisory rules, and policy frameworks governing the issuance, custody, trading, and use of digital assets such as cryptocurrencies, stablecoins, and tokenised securities. It applies traditional financial-services objectives, consumer protection, market integrity, financial stability, and the prevention of money laundering and terrorist financing, to crypto-native business models. Landmark regimes include the EU Markets in Crypto-Assets regulation and the UK's phased approach led by the Financial Conduct Authority and HM Treasury, alongside global standards from the Financial Action Task Force such as the travel rule.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:cryptoasset-regulation", "labels": [ "Cryptoasset Regulation" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [] }, { "id": "cryptoasset-regulatory-framework", "title": "Cryptoasset Regulatory Framework", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Crypto Legal from DigiSoc examines the legal, regulatory, and governance landscape surrounding cryptocurrency and digital assets within digital society contexts. It covers UK and EU regulatory positions on crypto-assets as property, stablecoin frameworks, AML/KYC obligations, and the intersection of Web3 economics with metaverse commerce. The field addresses how legacy legal systems must adapt to support low-friction, globally scalable value exchange in virtual social spaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cryptoasset-regulatory-framework", "labels": [ "Cryptoasset Regulatory Framework", "Crypto Legal from DigiSoc", "Cryptoasset Policy" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "Blockchain" ] }, { "id": "cryptocurrency-exchange", "title": "Cryptocurrency Exchange", "domain": "finance", "domain_name": "Finance", "definition": "A cryptocurrency exchange is a platform that enables the trading of digital assets for other cryptocurrencies or for fiat currency. Exchanges match buy and sell orders, provide custody or settlement of funds, and supply liquidity and price discovery for crypto markets. They range from centralised intermediaries holding customer assets to decentralised protocols that execute trades on-chain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cryptocurrency-exchange", "labels": [ "Cryptocurrency Exchange" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [] }, { "id": "cryptocurrency-mining", "title": "Cryptocurrency Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptocurrency Mining is the process by which nodes in a proof-of-work blockchain network compete to solve computationally intensive cryptographic puzzles in order to validate pending transactions and append new blocks to the chain. Successful miners are rewarded with newly minted coins and transaction fees, providing the economic incentive that secures the network. The energy intensity and hardware specialisation of mining have significant environmental and economic implications.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:cryptocurrency-mining", "labels": [ "Cryptocurrency Mining" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "cryptocurrency-regulation", "title": "Cryptocurrency Regulation", "domain": "governance", "domain_name": "Governance", "definition": "The body of laws, supervisory frameworks, and enforcement practice governing the issuance, custody, exchange, and use of cryptocurrencies and related digital assets. It spans securities and commodities classification, anti-money-laundering and counter-terrorist-financing obligations, consumer and investor protection, taxation, stablecoin reserve requirements, and prudential rules for intermediaries, and it varies sharply across jurisdictions \u2014 from comprehensive regimes such as the EU's MiCA to outright prohibitions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:cryptocurrency-regulation", "labels": [ "Cryptocurrency Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Financial Regulation", "Anti-Money Laundering", "Cryptocurrency", "Regulation" ] }, { "id": "cryptocurrency-remuneration", "title": "Cryptocurrency Remuneration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Cryptocurrency remuneration is the practice of paying remote workers, freelancers, or distributed team members in digital assets \u2014 including Bitcoin, Ethereum, or fiat-pegged stablecoins \u2014 rather than traditional currency. It enables borderless, near-instant settlement at low transaction cost, bypassing correspondent banking infrastructure and extending financial access to unbanked populations. Compliance with local payroll tax, AML/KYC, and employment law obligations remains essential.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:cryptocurrency-remuneration", "labels": [ "Cryptocurrency Remuneration", "TELE-253-cryptocurrency-remuneration" ], "is_subclass_of": [ "Protocol and Infrastructure", "Cryptocurrency" ], "wikilinks": [ "TELE-002-telecollaboration", "TELE-250-blockchain-collaboration", "TELE-251-smart-contract-coordination", "Cryptocurrency" ] }, { "id": "cryptocurrency-storage", "title": "Cryptocurrency Storage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptocurrency Storage encompasses the cryptographic key management systems and secure storage solutions for maintaining control over digital assets on blockchain networks. Architectures range from hot wallets with internet connectivity through cold storage hardware devices to multi-signature and threshold-signature schemes. Hierarchical Deterministic (HD) wallets following BIP32/BIP39 standards generate key trees from a single seed phrase, while institutional custody solutions leverage multi-party computation (MPC) and smart-contract-based social recovery for asset governance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cryptocurrency-storage", "labels": [ "Cryptocurrency Storage" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BIP32", "BIP39", "BIP44", "NIST Post-Quantum Cryptography", "Blockchain" ] }, { "id": "cryptocurrency-token", "title": "Cryptocurrency Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A programmable digital asset implemented via smart contracts on a blockchain platform, categorised into utility tokens (access rights), security tokens (equity or debt instruments), governance tokens (protocol voting rights), and non-fungible tokens (NFTs, unique digital ownership). Token behaviour and interoperability are defined by standards such as ERC-20, ERC-721, and ERC-1155.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cryptocurrency-token", "labels": [ "Cryptocurrency Token" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "cryptocurrency-wallet", "title": "Cryptocurrency Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cryptocurrency wallet is software or hardware that manages the cryptographic keys controlling blockchain assets and constructs, signs and broadcasts transactions on a user's behalf. It does not store coins, which exist only as ledger entries, but rather safeguards the private keys that authorise spending and prove ownership. Wallets range from custodial services that hold keys for users to non-custodial and hardware wallets that give users sole control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cryptocurrency-wallet", "labels": [ "Cryptocurrency Wallet" ], "is_subclass_of": [ "Wallet" ], "wikilinks": [] }, { "id": "cryptocurrency", "title": "Cryptocurrency", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptocurrency is a class of bearer-style digital assets whose unit ownership, issuance schedule, and transaction history are jointly secured by Public-Key Cryptography and a permissionless Consensus Mechanism operated by a Peer-to-Peer Network of independent nodes, such that no centr...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cryptocurrency", "labels": [ "Cryptocurrency", "ADA Cryptocurrency", "Cryptocurrency Ecosystem", "Volatile Cryptocurrency" ], "is_subclass_of": [ "Token and Asset", "Network Component", "Digital Asset", "Cryptoasset", "Distributed Ledger Native Token", "Bearer Instrument", "Permissionless Asset" ], "wikilinks": [ "24/7 Settlement", "Ammous 2018 The Bitcoin Standard", "Anchorage Digital", "Arbitrum", "Back 2002 Hashcash", "Bank Deposit", "Bank of England 2023 Digital Pound Discussion Paper", "Bearer Instrument", "Binance", "BIP-32 HD Wallet", "BIP-39 Mnemonic", "BitGo", "Bitcoin Improvement Proposal", "Bittensor", "Bohme Christin Edelman Moore 2015 Bitcoin Economics", "Buterin Hertzog 2017 Casper FFG", "CCAF 2024 Global Cryptoasset Benchmarking Study", "CCAF Cambridge Bitcoin Electricity Consumption Index", "Centralised Exchange", "Chaum 1983 Blind Signatures" ] }, { "id": "cryptoeconomics", "title": "Cryptoeconomics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptoeconomics is the discipline that combines cryptography with economic incentives to design and secure decentralised systems whose participants are assumed to act in their own self-interest. It uses mechanism design and game theory to make honest behaviour the rational choice, so that protocols remain secure and live without a trusted central authority. The field underpins consensus mechanisms, token economies, and the incentive structures of blockchain networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cryptoeconomics", "labels": [ "Cryptoeconomics", "Crypto-Economics" ], "is_subclass_of": [ "Blockchain Economics" ], "wikilinks": [] }, { "id": "cryptographic-accumulator", "title": "Cryptographic Accumulator", "domain": "security", "domain_name": "Security", "definition": "A cryptographic accumulator is a primitive that compresses a large set of elements into a single short value while still permitting compact proofs that a given element is (or is not) a member of the set. Constructions based on RSA groups, bilinear pairings, or Merkle trees allow membership witnesses whose size is independent of the set's cardinality, and dynamic accumulators support efficient addition and removal of elements. Accumulators underpin scalable membership proofs in anonymous credentials, certificate revocation, and stateless blockchain clients.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cryptographic-accumulator", "labels": [ "Cryptographic Accumulator", "Accumulator" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "cryptographic-algorithm", "title": "Cryptographic Algorithm", "domain": "security", "domain_name": "Security", "definition": "A Cryptographic Algorithm is a precisely defined mathematical procedure that transforms data to achieve security properties\u2014confidentiality, integrity, authentication, or non-repudiation\u2014based on computational hardness assumptions. The class encompasses symmetric ciphers, asymmetric (public-key) schemes, hash functions, digital signature algorithms, and zero-knowledge proof systems, each providing different security guarantees and performance characteristics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:cryptographic-algorithm", "labels": [ "Cryptographic Algorithm" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "cryptographic-commitment", "title": "Cryptographic Commitment", "domain": "security", "domain_name": "Security", "definition": "A cryptographic scheme that allows a party to commit to a chosen value while keeping it hidden, with the ability to reveal it later, satisfying the binding property (cannot change the committed value) and the hiding property (the commitment reveals no information about the value). Used in zero-knowledge proofs, atomic swaps, and confidential transactions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:cryptographic-commitment", "labels": [ "Cryptographic Commitment" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "cryptographic-hash-function", "title": "cryptographic hash function", "domain": "security", "domain_name": "Security", "definition": "A cryptographic hash function is a deterministic, one-way algorithm that maps an input of arbitrary length to a fixed-size digest (typically 160\u2013512 bits), satisfying three core security properties: preimage resistance (infeasibility of recovering input from digest), second-preimage resistance (infeasibility of finding a distinct input that maps to the same digest as a known input), and collision resistance (infeasibility of finding any two distinct inputs that share a digest). These properties make hash functions foundational primitives for data integrity verification, digital signatures, message authentication codes, and proof-of-work consensus. Widely deployed algorithms include SHA-256 (Bitcoin), Keccak-256 (Ethereum), BLAKE2, and SHA-3 (NIST FIPS 202); Grover's algorithm on quantum hardware reduces effective security by half, motivating ongoing standardisation of quantum-resistant alternatives by NIST.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:cryptographic-hash-function", "labels": [ "Cryptographic Hash Function", "Cryptographic Hash Functions", "Cryptographic Hashing", "Hash Algorithm", "Hash Computation" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "cryptographic-hash", "title": "Cryptographic Hash", "domain": "security", "domain_name": "Security", "definition": "Deterministic mathematical function transforming arbitrary input data into a fixed-length digest, guaranteeing data integrity verification and tamper detection through collision resistance, pre-image resistance, and avalanche-effect properties.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-hash", "labels": [ "Cryptographic Hash", "CryptographicHash" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "Blockchain Security", "Blockchain Validation", "Cryptographic Hash Functions", "Data Integrity", "Hash Functions", "Merkle Trees", "NCSC", "NIST FIPS PUB 202", "RFC 1321", "Tamper Detection", "AI Agent System", "BlockchainDomain", "Digital Signatures" ] }, { "id": "cryptographic-identity", "title": "Cryptographic Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographic identity is an identity model in which an actor is represented and authenticated through possession of a private key rather than through a centrally issued credential. Control of the corresponding public key \u2014 or an identifier derived from it \u2014 proves the actor's identity by producing verifiable digital signatures. This model underpins blockchain accounts, decentralized identifiers and self-sovereign identity, removing the need for a trusted registry to vouch for who someone is.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-identity", "labels": [ "Cryptographic Identity" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [] }, { "id": "cryptographic-infrastructure", "title": "Cryptographic Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cryptographic infrastructure is the ensemble of hardware, software, protocols, and institutional processes that provision, manage, and operate cryptographic primitives at scale across a computing environment. It encompasses key generation and distribution, certificate authorities, hardware security modules, and the policies governing their lifecycle. As a foundational layer, it underpins secure communications, digital identity, and data integrity across both centralised and decentralised systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:cryptographic-infrastructure", "labels": [ "Cryptographic Infrastructure", "Public Key Infrastructure" ], "is_subclass_of": [ "Cryptographic Algorithm" ], "wikilinks": [] }, { "id": "cryptographic-key-management", "title": "Cryptographic Key Management", "domain": "security", "domain_name": "Security", "definition": "The administration of cryptographic keys throughout their lifecycle, including generation, storage, distribution, rotation, backup, recovery, and destruction, ensuring the security and availability of keying material while preventing unauthorized access or compromise.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-key-management", "labels": [ "Cryptographic Key Management" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "FIPS 140-3", "NIST SP 800-130", "NIST SP 800-57", "AI Agent System", "Cryptographic Keys", "Cryptography", "Digital Signature", "Key Derivation Function", "Private Key", "Random Number Generation", "Security Architecture" ] }, { "id": "cryptographic-key-pair", "title": "Cryptographic Key Pair", "domain": "security", "domain_name": "Security", "definition": "A cryptographic key pair is a mathematically linked pair of keys, a public key and a private key, used in asymmetric cryptography. The private key is kept secret by its owner while the public key may be distributed openly, and operations performed with one key can only be reversed or verified with the other. Key pairs underpin encryption to a recipient, digital signatures that prove authorship, and key agreement, making them foundational to secure communication, authentication and decentralised identity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cryptographic-key-pair", "labels": [ "Cryptographic Key Pair" ], "is_subclass_of": [ "Public Key Cryptography" ], "wikilinks": [] }, { "id": "cryptographic-key", "title": "Cryptographic Key", "domain": "security", "domain_name": "Security", "definition": "A piece of secret or public information that parameterises a cryptographic algorithm, determining how data is encrypted, decrypted or signed. Keys are the inputs that make cryptographic operations specific and reversible only to authorised parties.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cryptographic-key", "labels": [ "Cryptographic Key" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Encryption", "Digital Signature", "Public Key Infrastructure", "Private Key", "Key Management", "Cryptography", "https://csrc.nist.gov/glossary/term/cryptographic_key" ] }, { "id": "cryptographic-keys", "title": "Cryptographic Keys", "domain": "security", "domain_name": "Security", "definition": "Strings of data used in cryptographic algorithms to encrypt, decrypt, sign, or verify data, serving as the secret parameters that transform plaintext to ciphertext and vice versa. Keys may be symmetric (single shared secret) or asymmetric (public-private pairs); their security depends on key length, entropy of generation, and rigorous lifecycle management.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-keys", "labels": [ "Cryptographic Keys", "Viewing Keys" ], "is_subclass_of": [ "Cryptographic Primitive", "Cryptography" ], "wikilinks": [ "AI Agent System", "Asymmetric Encryption", "Cryptographic Key Management", "Cryptography", "Digital Signature", "Key Derivation Function", "Random Number Generation", "Symmetric Encryption" ] }, { "id": "cryptographic-layer", "title": "Cryptographic Layer", "domain": "security", "domain_name": "Security", "definition": "The Cryptographic Layer is the stratum that provides confidentiality, integrity, and authenticity primitives to the layers above. It sits above the Hardware Layer, which supplies entropy and acceleration, and below identity, consensus, and security strata that depend on its guarantees. It contains ciphers, hash functions, signature schemes, and key management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-layer", "labels": [ "Cryptographic Layer" ], "is_subclass_of": [ "Security and Identity", "owl:Thing" ], "wikilinks": [ "Hardware Layer", "Identity Layer", "Consensus Layer", "Public Key Cryptography", "Hash Function", "owl:Thing", "NIST (National Institute of Standards and Technology)" ] }, { "id": "cryptographic-library", "title": "Cryptographic Library", "domain": "security", "domain_name": "Security", "definition": "A cryptographic library is a software package that implements cryptographic primitives and protocols \u2014 encryption, hashing, signing, key exchange, and random-number generation \u2014 behind a programming interface that application developers can use without reimplementing the underlying mathematics. Well-designed libraries such as OpenSSL, libsodium, BoringSSL, and the Rust RustCrypto suite emphasise constant-time implementations, safe defaults, and resistance to misuse. The correctness and side-channel resistance of these libraries is critical, since flaws propagate to every application that depends on them.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cryptographic-library", "labels": [ "Cryptographic Library" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "cryptographic-primitive", "title": "Cryptographic Primitive", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The fundamental atomic building blocks of all cryptographic systems \u2014 including hash functions, symmetric and asymmetric ciphers, digital signatures, and key exchange protocols \u2014 each providing specific, well-defined security guarantees. Primitives are insufficient alone and must be combined in higher-level cryptographic protocols to satisfy multiple security requirements such as confidentiality, integrity, authentication, and non-repudiation. Correct selection and composition of primitives is the foundational concern of applied cryptography.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-primitive", "labels": [ "Cryptographic Primitive", "Cryptographic Primitive (Blockchain)", "Cryptographic Primitives", "CryptographicPrimitive" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "AI Agent System", "Asymmetric Encryption", "Blockchain", "Cryptographic Protocol", "Cryptography", "Digital Signature", "Hash Function", "Symmetric Encryption" ] }, { "id": "cryptographic-privacy-activist", "title": "Cryptographic Privacy Activist", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cypherpunk is an activist who advocates for the widespread use of strong cryptography and privacy-enhancing technologies as tools for social and political change. Emerging from mailing list culture in the early 1990s, cypherpunks believe that individual privacy can only be guaranteed through cryptographic means rather than through legal or institutional protections. Their ethos\u2014'cypherpunks write code'\u2014produced foundational technologies including public-key cryptography applications, anonymous remailers, digital cash, and the conceptual groundwork for Bitcoin and decentralised systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-privacy-activist", "labels": [ "Cryptographic Privacy Activist", "Cypherpunk Movement", "cypherpunk" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "cryptographic-proof-system", "title": "Cryptographic Proof System", "domain": "security", "domain_name": "Security", "definition": "A cryptographic proof system is a formal protocol allowing one party (a prover) to convince another party (a verifier) of the truth of a statement without revealing any information beyond the validity of that statement. These systems provide mathematical guarantees of soundness, completeness, and, in zero-knowledge variants, zero information leakage.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cryptographic-proof-system", "labels": [ "Cryptographic Proof System", "Proof System" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "cryptographic-proof", "title": "Cryptographic Proof", "domain": "security", "domain_name": "Security", "definition": "A cryptographic proof is a mathematical construction that enables one party to demonstrate the truth of a statement or possession of secret knowledge to a verifying party in a computationally sound and tamper-evident manner, without necessarily revealing the underlying information itself. Rooted in complexity theory and interactive proof systems, cryptographic proofs provide infeasibility guarantees: a computationally bounded adversary cannot forge a valid proof for a false statement. The field spans classical constructions such as hash-based commitments and digital signatures, through to advanced non-interactive arguments including zk-SNARKs and STARKs, and underpins security across blockchain, identity, and privacy-preserving computation domains.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:cryptographic-proof", "labels": [ "Cryptographic Proof", "CryptographicProof" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "cryptographic-proofs", "title": "Cryptographic Proofs", "domain": "security", "domain_name": "Security", "definition": "Cryptographic proofs are mathematical constructions that allow one party to convince another of the truth of a statement with cryptographic certainty, often without revealing the underlying data. They include proofs of knowledge, membership proofs, proofs of computation, and zero-knowledge proofs, and rely on primitives such as hash functions, commitments, and elliptic-curve operations. Cryptographic proofs underpin blockchain validity, verifiable computation, privacy-preserving authentication, and data-availability guarantees, letting verifiers trust outcomes they did not themselves compute.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cryptographic-proofs", "labels": [ "Cryptographic Proofs", "CryptographicProofs", "DLEQ Proofs" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "cryptographic-protocol", "title": "Cryptographic Protocol", "domain": "security", "domain_name": "Security", "definition": "Cryptographic Protocol encompasses mathematical frameworks and algorithmic procedures that secure digital systems through cryptographic primitives including hash functions, digital signatures, encryption schemes, zero-knowledge proofs, and commitment protocols, enabling confidentiality, integrity, authentication, and non-repudiation across distributed and blockchain applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:cryptographic-protocol", "labels": [ "Cryptographic Protocol", "Cryptographic-Protocol" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Blockchain-Security", "CMAC", "EdDSA", "Encryption-Scheme", "HMAC", "NIST Post-Quantum Cryptography", "Post-Quantum Cryptography Standards", "Privacy-Preservation", "Schnorr", "blockchain", "Blockchain", "Collaboration", "Consensus-Protocol", "Consensus-Protocol", "Control-Algorithm", "Convergence", "Cryptographic-Protocol", "Cryptography", "Digital-Asset", "Digital-Infrastructure" ] }, { "id": "cryptographic-protocols", "title": "Cryptographic Protocols", "domain": "security", "domain_name": "Security", "definition": "Cryptographic protocols are sequences of operations using cryptographic primitives to achieve security goals such as confidentiality, integrity and authentication.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cryptographic-protocols", "labels": [ "Cryptographic Protocols" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Encryption", "Kerberos", "Network Security", "Cryptography", "https://en.wikipedia.org/wiki/Cryptographic_protocol", "https://csrc.nist.gov/" ] }, { "id": "cryptographic-security", "title": "Cryptographic Security", "domain": "security", "domain_name": "Security", "definition": "Cryptographic Security is the discipline of applying mathematical cryptographic primitives and protocols \u2014 including symmetric encryption, asymmetric public-key cryptography, cryptographic hash functions, digital signatures, message authentication codes, and zero-knowledge proofs \u2014 to enforce confidentiality, integrity, authenticity, and non-repudiation of information and communications. It provides the formal security guarantees upon which trustless distributed systems, secure channels, identity frameworks, and privacy-preserving computation are constructed. The field spans both theoretical hardness assumptions (discrete logarithm, integer factorisation, lattice problems) and practical protocol engineering, encompassing key management, certificate infrastructure, and post-quantum cryptographic migration. In applied contexts it underpins everything from TLS transport security and blockchain transaction authorisation to hardware security modules and secure multi-party computation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:cryptographic-security", "labels": [ "Cryptographic Security" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "cryptographic-signature", "title": "cryptographic signature", "domain": "security", "domain_name": "Security", "definition": "A cryptographic signature is a mathematical scheme that allows a private-key holder to produce an unforgeable, publicly-verifiable proof that a specific message or data item was authorised by them, delivering both authenticity and non-repudiation. The signing algorithm combines a cryptographic hash of the message with the signer's private key to produce a compact signature value; any party holding the corresponding public key can verify the signature without accessing the private key. Dominant schemes include ECDSA (Bitcoin, Ethereum), Ed25519 (Solana, Cosmos, OpenSSH), RSA-PSS (TLS, S/MIME), and Schnorr (BIP-340). Cryptographic signatures are a foundational primitive underpinning transaction authorisation, code signing, verifiable credentials, and authenticated key exchange across virtually all secure digital infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:cryptographic-signature", "labels": [ "Cryptographic Signature" ], "is_subclass_of": [ "Asymmetric Cryptography" ], "wikilinks": [] }, { "id": "cryptographic-signing", "title": "Cryptographic Signing", "domain": "security", "domain_name": "Security", "definition": "Cryptographic signing is the process of producing a verifiable proof of authorship and integrity over a message or document using a private key, such that anyone holding the corresponding public key can confirm the signature without being able to forge it. It underpins authentication, non-repudiation, software-supply-chain integrity, and blockchain transaction authorisation. Signing schemes include RSA, ECDSA, EdDSA, and Schnorr, each defining how a hash of the message is transformed under the signer's secret key into a compact, publicly verifiable signature.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cryptographic-signing", "labels": [ "Cryptographic Signing", "COSE Signing", "Signing Algorithm", "Signing Key" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "cryptographic-sortition", "title": "Cryptographic Sortition", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographic sortition is a technique for randomly and unpredictably selecting a subset of participants from a larger population to perform a role such as block proposal or committee membership, using a verifiable random function so the outcome cannot be predicted or manipulated in advance. Each participant can privately determine whether they were selected and prove it to others without revealing information that would let an adversary target them beforehand. It underpins consensus protocols such as Algorand's, allowing large validator populations to reach agreement with low communication overhead and strong resistance to targeted attacks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-sortition", "labels": [ "Cryptographic Sortition" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "cryptographic-system", "title": "Cryptographic System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An integrated framework of mathematical algorithms, protocols, and mechanisms designed to provide information security properties including confidentiality, integrity, authentication, and non-repudiation in adversarial environments. Blockchain cryptographic systems enable trustless operation through mathematical guarantees rather than trusted intermediaries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cryptographic-system", "labels": [ "Cryptographic System" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity" ], "wikilinks": [ "Cryptographic Hash Function", "FIPS 140-3 Security Requirements", "Homomorphic Encryption", "ISO/IEC 18033 Encryption Algorithms", "NIST Cryptographic Standards", "Threshold Cryptography", "AI Agent System", "Blockchain", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "Digital Signature", "Merkle Tree", "Public Key Cryptography", "Zero-Knowledge Proof" ] }, { "id": "cryptographic-verification", "title": "Cryptographic Verification", "domain": "security", "domain_name": "Security", "definition": "Cryptographic Verification is the process of using cryptographic primitives\u2014such as digital signatures, hash functions, and zero-knowledge proofs\u2014to confirm the authenticity, integrity, and non-repudiation of data, identities, or transactions. It forms the trust foundation for blockchain systems, content authentication, and decentralised identity schemes.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:cryptographic-verification", "labels": [ "Cryptographic Verification", "Cryptographic Verification System" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "cryptography-security-and-privacy", "title": "Cryptography Security and Privacy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cryptography Security and Privacy denotes the cross-disciplinary infrastructure layer combining mathematical primitives (symmetric encryption AES-GCM/ChaCha20-Poly1305/ASCON-128 NIST lightweight finalist February 2023, hash functions SHA-2/SHA-3 Keccak/BLAKE3 throughput 6.8 GB/s single-thread, pa...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cryptography-security-and-privacy", "labels": [ "Cryptography Security and Privacy" ], "is_subclass_of": [ "Security and Identity", "Computer Science", "Information Security", "Applied Mathematics", "Privacy Engineering", "Trusted Computing" ], "wikilinks": [ "AES-GCM", "Anderson Security Engineering 3rd Edition", "Anonymous Communication", "ANSSI France", "ANSSI RGS", "Apple Private Cloud Compute 2024", "Applied Mathematics", "Argon2", "Authentication", "BIP-340 Schnorr Bitcoin Taproot", "BLAKE3", "Boneh & Shoup Graduate Course in Applied Cryptography", "BSI Germany", "BSI TR-02102 Cryptographic Mechanisms", "Bulletproofs Bunz et al 2018", "Cambridge Anderson Murdoch Chip and PIN", "ChaCha20-Poly1305", "CISA Post-Quantum Cryptography Initiative", "CKKS Cheon Kim Kim Song", "Classical Surveillance Systems" ] }, { "id": "cryptography", "title": "Cryptography", "domain": "security", "domain_name": "Security", "definition": "Cryptography is the mathematical science of transforming information through encryption and related primitives, ensuring confidentiality, authenticity, and integrity in digital communications, blockchain systems, and distributed networks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:cryptography", "labels": [ "Cryptography" ], "is_subclass_of": [ "Cryptographic Security", "Security" ], "wikilinks": [ "encryption", "Encryption-Scheme", "ISO/IEC 14888", "ISO/IEC 18033", "NIST", "Post-Quantum Cryptography Standards", "blockchain", "Blockchain", "Collaboration", "Consensus-Protocol", "Control-Algorithm", "Cryptographic-Protocol", "Cryptography", "Digital-Asset", "Digital-Infrastructure", "Digital-Signature", "Digital-Twin", "Hash-Function", "Post-Quantum-Cryptography", "Quantum-Computing" ] }, { "id": "css", "title": "Css", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cascading Style Sheets (CSS) is the standard language for describing the presentation of documents written in markup languages such as HTML, controlling layout, colour, typography and responsive behaviour. It separates content from presentation, with rules that cascade and inherit according to specificity and source order. Maintained as a family of W3C specifications, CSS is a foundational web technology alongside HTML and JavaScript.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:css", "labels": [ "Css", "CSS" ], "is_subclass_of": [ "Web Standard" ], "wikilinks": [] }, { "id": "cubesat", "title": "CubeSat", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A CubeSat is a [[Small Satellite|small-satellite]] form factor built around standardised units, written as **U**. Under the CubeSat Design Specification Revision 14.1, a 1U spacecraft is based on a ten-centimetre cube and may have a mass up to two kilograms; larger configurations combine units into forms such as 2U, 3U, 6U or 12U.[^1] The standardised spacecraft-to-deployer interface is the defining feature. CubeSat is therefore not a synonym for every nanosatellite.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:cubesat", "labels": [ "CubeSat" ], "is_subclass_of": [ "Small Satellite" ], "wikilinks": [ "small-satellite", "payload" ] }, { "id": "cultural-heritage-preservation", "title": "Cultural Heritage Preservation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cultural Heritage Preservation encompasses the methodologies, technologies, and institutional practices used to document, conserve, and transmit tangible and intangible cultural assets to future generations. Digital preservation extends these goals through high-fidelity capture technologies\u2014photogrammetry, LiDAR scanning, and multispectral imaging\u2014combined with long-lived archival formats and distributed storage. Immersive technologies increasingly enable experiential access to heritage sites and artefacts that physical access cannot support. The discipline spans archaeology, architecture, performing arts, indigenous knowledge, and language revitalisation, requiring interdisciplinary collaboration between cultural practitioners and technologists.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:cultural-heritage-preservation", "labels": [ "Cultural Heritage Preservation", "Cultural Heritage Preservation System", "Cultural Heritage Restoration", "Cultural Heritage Tracking", "Heritage Preservation", "Heritage Site Preservation" ], "is_subclass_of": [ "Digital Heritage" ], "wikilinks": [] }, { "id": "cultural-heritage-xr-experience", "title": "Cultural Heritage XR Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An immersive extended reality application designed to preserve, present, and educate users about cultural heritage through interactive 3D reconstructions, AR overlays, and virtual museum experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cultural-heritage-xr-experience", "labels": [ "Cultural Heritage XR Experience" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Cultural Database", "Cultural Education", "Cultural Heritage Preservation System", "Cultural Metadata", "Heritage Documentation", "Heritage Tourism", "ICOM Museum Definition", "Interactive Exhibit", "LiDAR Scanning", "Museum API", "UNESCO World Heritage Convention", "3D Reconstruction", "3D Rendering Engine", "ApplicationLayer", "AR Overlay", "Archaeological Site Reconstruction", "Computer Vision", "CreativeMediaDomain", "Educational Narrative", "Photogrammetry" ] }, { "id": "cultural-heritage", "title": "Cultural Heritage", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The legacy of tangible artefacts (monuments, buildings, artworks, archives) and intangible attributes (language, performance, craft, ritual) inherited from past generations, maintained in the present, and safeguarded for future ones. In digital ecosystems, cultural heritage is increasingly captured, preserved, and re-presented through 3D scanning, digital twins, and immersive experiences, making it a foundational content domain for virtual tourism and the creative industries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:cultural-heritage", "labels": [ "Cultural Heritage" ], "is_subclass_of": [ "Creative Industries" ], "wikilinks": [ "Creative Industries", "Digital Preservation", "Virtual Tourism", "Digital Objects", "Digital Twin", "Immersive Experience" ] }, { "id": "cultural-preservation", "title": "Cultural Preservation", "domain": "governance", "domain_name": "Governance", "definition": "Cultural preservation is the systematic safeguarding, documentation and transmission of tangible and intangible cultural heritage so that it survives for future generations. In digital contexts it involves capturing artefacts as durable metadata-rich records, archiving collective memory, and ensuring long-term accessibility against media decay and format obsolescence. It matters as a governance concern for who controls heritage data and how it remains discoverable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cultural-preservation", "labels": [ "Cultural Preservation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "cultural-provenance-record", "title": "Cultural Provenance Record", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured metadata object that documents the origin, ownership history, authenticity verification, and cultural context of cultural artifacts, artworks, or digital cultural assets to establish legitimacy and preserve heritage lineage.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:cultural-provenance-record", "labels": [ "Cultural Provenance Record" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Authentication Record", "Authenticity Certification", "Blockchain Ledger", "CIDOC-CRM", "Condition Report", "Conservation Database", "Cultural Context", "Cultural Heritage Management System", "Cultural Heritage Tracking", "Heritage Registry", "Museum Collection System", "Museum Information System", "Ownership Chain", "Ownership Transfer", "SPECTRUM Museum Standard", "Artifact Metadata", "Authentication Service", "Blockchain", "CreativeMediaDomain", "Digital Archive" ] }, { "id": "curated-research-link-repository", "title": "Curated Research Link Repository", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated collection of saved hyperlinks to external web resources, serving as a personal or shared reference library for articles, tools, and research material. In knowledge management contexts, web bookmarks act as annotated entry points linking internal knowledge graphs to external sources.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:curated-research-link-repository", "labels": [ "Curated Research Link Repository", "Web Bookmarks" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "current-sensor", "title": "Current Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "Current Sensor - An electrical measurement device (Hall effect, fluxgate, or shunt-based) that detects current flow in motor circuits and power systems, enabling Motor Torque Estimation, Fault Detection, and Energy Monitoring in autonomous robots.", "entityType": "Class", "qualityScore": 0.52, "maturity": "draft", "iri": "urn:ngm:class:current-sensor", "labels": [ "Current Sensor", "Current Sensing" ], "is_subclass_of": [ "Perception and Sensing", "Robotics", "Sensor" ], "wikilinks": [ "Analogue-to-Digital Conversion", "Efficiency Monitoring", "EN 60 068", "Energy Monitoring", "ETSI TS 132 423", "Fault Detection", "ISO 8373:2021", "Motor Control System", "Motor Torque Estimation", "Overload Protection", "Power Management", "Signal Conditioning", "Computer Vision", "Predictive Maintenance", "Robotics", "RoboticsDomain", "Sensor" ] }, { "id": "curriculum-learning", "title": "Curriculum Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A training strategy that presents examples to a model in a meaningful order, typically progressing from easy to difficult, mimicking how humans learn. Curriculum learning improves convergence speed, final performance, and generalisation by structuring the learning progression rather than relying on random example ordering.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:curriculum-learning", "labels": [ "Curriculum Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "IEEE/CVF International Conference on Computer Vision", "IEEE Transactions on Pattern Analysis and Machine Intelligence", "Computer Vision", "MetaverseDomain" ] }, { "id": "cursor", "title": "Cursor", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cursor is an AI-native code editor developed by Anysphere, built on a Visual Studio Code foundation and deeply integrated with large language models to provide inline code completion, multi-file agentic editing, natural-language chat over the codebase, and autonomous background agents. As of 2026, Cursor is the leading independent AI coding assistant by revenue, having reached $2 billion ARR and a $29.3 billion Series D valuation, and is used by over 64% of Fortune 500 companies.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cursor", "labels": [ "Cursor", "Cursor IDE" ], "is_subclass_of": [ "Software Development", "Generative AI", "Agentic AI" ], "wikilinks": [ "GPT", "Software Development", "Generative AI", "Large Language Model", "Transformer Architecture", "Agentic AI", "Agentic Workflow", "Microsoft Copilot", "LLM Agents", "LLM Application Framework", "Tool-Augmented LLM", "Retrieval-Augmented Generation", "Natural Language Processing", "Code Generation", "Model Context Protocol", "https://cursor.com", "https://docs.cursor.com", "AI-GroundedDomain", "ApplicationLayer" ] }, { "id": "curve-finance", "title": "Curve Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Curve Finance is a decentralised exchange protocol operating across Ethereum and multiple EVM-compatible chains, specialising in low-slippage swaps between assets expected to maintain near-parity in value, such as stablecoins and liquid staking tokens. It employs a hybrid invariant automated market maker that blends constant-sum and constant-product behaviour, concentrating liquidity near the peg to dramatically reduce trading costs for correlated assets. The protocol's CRV governance token is distributed to liquidity providers and can be locked in a vote-escrow mechanism (veCRV) that grants voting power over pool incentive allocation, creating a flywheel dynamic known as the Curve Wars. Curve v2 extended the model to volatile asset pairs using a price-repegging invariant, broadening the protocol's scope from stablecoin-only pools to general concentrated-liquidity trading.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:curve-finance", "labels": [ "Curve Finance" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [ "Ethereum", "Automated Market Maker", "Stablecoin", "Decentralised Finance Domain" ] }, { "id": "curve-wars", "title": "Curve Wars", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Curve Wars refers to the competitive dynamic among decentralised finance protocols to accumulate vote-escrowed CRV (veCRV) governance power over Curve Finance, thereby gaining the ability to direct CRV token emissions (gauges) toward liquidity pools in which the protocol has a stake. By controlling gauge weights, protocols attract liquidity providers to their pools by offering higher yields, creating a recursive incentive structure. The conflict intensified with the emergence of vote-aggregation intermediaries such as Convex Finance, which pooled user CRV into a dominant veCRV position and issued its own liquid derivative tokens (cvxCRV, vlCVX) to further abstract the underlying governance asset.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:curve-wars", "labels": [ "Curve Wars" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [ "Vote-Escrow Model", "Convex Finance", "Liquidity Pool", "Liquidity Provision" ] }, { "id": "curve", "title": "Curve", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Curve is a decentralised exchange on Ethereum and other chains optimised for low-slippage swaps between similarly priced assets such as stablecoins. Its CRV token and gauge system govern liquidity incentives.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:curve", "labels": [ "Curve" ], "is_subclass_of": [ "Curve Finance" ], "wikilinks": [ "Automated Market Maker", "Liquidity Pool", "Gauge Voting", "Decentralized Exchange", "Curve Finance" ] }, { "id": "curve25519", "title": "Curve25519", "domain": "security", "domain_name": "Security", "definition": "Curve25519 is a Montgomery-form elliptic curve designed for fast, secure elliptic-curve Diffie-Hellman key exchange at a 128-bit security level. It was engineered to avoid common implementation pitfalls by enabling constant-time, branch-free arithmetic and to sidestep concerns about opaque parameter selection through rigid, transparent design choices. It underpins the X25519 key-agreement function and is widely deployed in modern secure-transport and messaging protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:curve25519", "labels": [ "Curve25519" ], "is_subclass_of": [ "Elliptic Curve Cryptography" ], "wikilinks": [] }, { "id": "custodial-exchange", "title": "Custodial Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A custodial exchange is a cryptocurrency trading venue that holds users' assets and private keys on their behalf, settling trades on its internal ledger rather than on-chain. Users gain convenience, liquidity and familiar account-based access, but cede control of their keys and accept counterparty risk in the operator. It is the centralised counterpart to non-custodial and decentralised trading models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:custodial-exchange", "labels": [ "Custodial Exchange" ], "is_subclass_of": [ "Cryptocurrency Exchange" ], "wikilinks": [] }, { "id": "custodial-wallet", "title": "Custodial Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A custodial wallet is a cryptocurrency wallet in which a third party \u2014 typically an exchange or custodian \u2014 holds and controls the private keys on behalf of the user, who accesses funds through an account rather than direct on-chain key ownership. This model trades the self-sovereignty of non-custodial wallets for convenience, account recovery, and integrated services, while introducing counterparty risk and reliance on the custodian's security and solvency. Custodial arrangements are common at centralised exchanges and increasingly subject to regulatory requirements for asset segregation and proof of reserves.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:custodial-wallet", "labels": [ "Custodial Wallet" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [] }, { "id": "custodian", "title": "Custodian", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A custodian is an entity entrusted with safeguarding assets on behalf of others, holding and securing them while the legal or beneficial owner retains a claim. In digital-asset markets a custodian secures cryptographic keys and underlying holdings for clients and, in the case of fiat-backed stablecoins, holds the reserves that back issued tokens. Custodianship concentrates security and introduces counterparty risk, which is mitigated through controls such as segregation, audits, and proof of reserves.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:custodian", "labels": [ "Custodian" ], "is_subclass_of": [ "Custody" ], "wikilinks": [] }, { "id": "custody-infrastructure", "title": "Custody Infrastructure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Custody infrastructure refers to the integrated technological, operational, and regulatory systems that enable the secure holding, management, and transfer of digital assets on behalf of third parties. It encompasses hardware security modules (HSMs), multi-signature and multi-party computation (MPC) key management schemes, cold and warm storage tiers, policy engines, and compliance workflows that satisfy fiduciary and regulatory obligations. As the institutional-grade equivalent of a traditional custodian bank applied to blockchain-native assets, custody infrastructure underpins every regulated digital asset vehicle\u2014from spot ETFs to tokenised real-world assets. Its design must simultaneously satisfy cryptographic security requirements, operational resilience, regulatory auditability, and the settlement latency demands of professional financial markets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:custody-infrastructure", "labels": [ "Custody Infrastructure" ], "is_subclass_of": [ "Custody" ], "wikilinks": [] }, { "id": "custody-layer", "title": "Custody Layer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Custody Layer is the stratum that governs the safekeeping and authorised control of assets and the keys that command them. It sits above the Cryptographic and Identity strata it depends on and below the settlement and application activity that moves assets. It contains key storage, signing policies, and the authorisation rules for asset control.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:custody-layer", "labels": [ "Custody Layer" ], "is_subclass_of": [ "Token and Asset", "owl:Thing" ], "wikilinks": [ "Cryptographic Layer", "Identity Layer", "Settlement Layer", "Application Layer", "Multi-Signature", "Key Management", "owl:Thing" ] }, { "id": "custody", "title": "Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The holding and safekeeping of assets and the keys that control them, defining who has the authority to move funds on behalf of an owner.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:custody", "labels": [ "Custody", "Collaborative Custody", "Community Custody", "Custody Log", "Custody Risk", "Custody Solution", "Custody Solutions" ], "is_subclass_of": [ "Token and Asset", "Custody Infrastructure" ], "wikilinks": [ "Private Key", "Wallet", "Self-Custody", "Institutional Custody", "Custody Infrastructure" ] }, { "id": "customer-data-platform", "title": "Customer Data Platform", "domain": "data", "domain_name": "Data", "definition": "A Customer Data Platform (CDP) is a packaged software system that ingests first-party customer data from disparate online and offline sources, resolves fragmented identifiers into persistent, unified customer profiles, and exposes those profiles via APIs to downstream marketing, analytics, and personalisation systems. Unlike data warehouses or CRM systems, a CDP is marketer-managed, real-time capable, and purpose-built for identity resolution and audience activation. CDPs enforce consent and privacy preferences at the profile level, making them a key architectural component for compliant, data-driven customer engagement.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:customer-data-platform", "labels": [ "Customer Data Platform" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Data Integration", "Data Management", "Privacy" ] }, { "id": "customer-due-diligence", "title": "Customer Due Diligence", "domain": "finance", "domain_name": "Finance", "definition": "Customer due diligence (CDD) is the set of regulated procedures by which a financial institution or obliged entity identifies and verifies a customer, understands the nature and purpose of the business relationship, and assesses the money-laundering and terrorist-financing risk it poses. It encompasses identity verification, beneficial-ownership identification, screening against sanctions and politically exposed person lists, and ongoing monitoring of transactions and risk profile. CDD is a core obligation of anti-money-laundering regimes, calibrated through a risk-based approach that escalates to enhanced due diligence for higher-risk customers and permits simplified measures for lower-risk ones.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:customer-due-diligence", "labels": [ "Customer Due Diligence" ], "is_subclass_of": [ "Know Your Customer" ], "wikilinks": [] }, { "id": "customer-experience-management", "title": "Customer Experience Management", "domain": "ai", "domain_name": "Ai", "definition": "Customer Experience Management (CEM or CXM) is the systematic discipline of designing, orchestrating, measuring, and continuously improving every interaction a customer has with an organisation across the full lifecycle \u2014 from initial brand awareness through purchase, onboarding, ongoing service, and eventual advocacy or churn. It encompasses the tools, processes, data infrastructure, and organisational culture required to engineer experience quality at scale, relying on journey mapping to document touchpoint sequences, voice-of-customer programmes to capture perception signals, omnichannel platforms to unify channel delivery, and analytics engines to translate interaction data into actionable improvement priorities.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:customer-experience-management", "labels": [ "Customer Experience Management" ], "is_subclass_of": [ "Enterprise Software Platform", "CRM", "User Experience" ], "wikilinks": [ "CRM", "E-Commerce", "Customer Experience", "Customer Data Platform", "Omnichannel", "Journey Mapping", "Voice of Customer", "Sentiment Analysis", "Predictive Personalization", "Behavioral Analytics", "Customer Retention", "Customer Support", "Hyper personalisation", "Natural Language Processing", "Digital Twin of the Customer", "Loyalty Programs", "Service Design", "Data Privacy", "Artificial Intelligence", "Machine Learning" ] }, { "id": "customer-experience", "title": "Customer Experience", "domain": "ai", "domain_name": "Ai", "definition": "Customer experience (CX) is the holistic perception a customer forms of an organisation across every interaction and touchpoint throughout the customer lifecycle, from initial awareness through purchase, post-sale service, and eventual advocacy or churn. It encompasses emotional, cognitive, sensory, and behavioural dimensions of customer engagement and is recognised as a primary competitive differentiator in markets where product parity is high. Organisations systematically measure and engineer CX through journey mapping, voice-of-customer programmes, service design, and data-driven personalisation.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:customer-experience", "labels": [ "Customer Experience", "Customer Experience Enhancement" ], "is_subclass_of": [ "User Experience", "Customer Experience Management" ], "wikilinks": [ "Customer Experience Management", "User Experience", "Behavioral Analytics", "Sentiment Analysis", "Predictive Personalization", "Customer Service Automation", "Hyper personalisation", "Loyalty Programs", "Customer Retention", "Digital Twin of the Customer", "Omnichannel", "Omnichannel Orchestration", "First-Party Data", "Unified Customer Profile", "Customer Support", "Service Design", "Data Privacy", "Artificial Intelligence", "Natural Language Processing", "Large Language Models" ] }, { "id": "customer-lifetime-value", "title": "Customer Lifetime Value", "domain": "data", "domain_name": "Data", "definition": "Customer lifetime value is a predictive metric estimating the total net revenue a business can expect from a customer over the duration of their relationship. It is computed from historical purchase frequency, average order value, and retention rates, often refined with machine-learned models that account for individual customer behaviour. Customer relationship management and customer experience management systems use lifetime value estimates to prioritise retention and personalisation investment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:customer-lifetime-value", "labels": [ "Customer Lifetime Value" ], "is_subclass_of": [ "CRM" ], "wikilinks": [] }, { "id": "customer-retention", "title": "Customer Retention", "domain": "governance", "domain_name": "Governance", "definition": "Customer retention is the set of strategies and mechanisms a business uses to keep existing customers engaged and transacting over time, reducing churn. It commonly relies on loyalty and rewards schemes, personalised omnichannel experiences, and data-driven engagement to maximise customer lifetime value. As an economic mechanism it is typically cheaper than acquisition and central to sustainable revenue.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:customer-retention", "labels": [ "Customer Retention" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "customer-rewards", "title": "Customer Rewards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Loyalty program systems that use blockchain technology and tokenization to create tradeable, interoperable digital rewards, enabling customers to earn, exchange, and redeem tokens across multiple platforms and businesses while providing brands with transparent, fraud-resistant, and cost-effective...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:customer-rewards", "labels": [ "Customer Rewards" ], "is_subclass_of": [ "Platform and Environment", "Loyalty Programs" ], "wikilinks": [ "Cross-Platform Rewards", "Customer Retention", "Token Trading", "Blockchain", "Blockchain Infrastructure", "Digital Wallet", "Loyalty Programs", "metaverse", "Smart Contracts" ] }, { "id": "customer-service-automation", "title": "customer service automation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Customer Service Automation (CSA) is the systematic application of artificial intelligence, rule-based logic, and workflow orchestration to resolve customer enquiries, execute transactions, and manage service interactions without requiring continuous human agent involvement. It integrates conversational AI, intent classification, sentiment analysis, and retrieval-augmented generation with CRM platforms, ticketing systems, and enterprise knowledge bases to deliver contextually accurate, personalised responses at scale. Modern CSA pipelines increasingly use large language models as the generation backbone, enabling coherent multi-turn dialogue capable of handling nuanced, composite, or cross-channel queries. Effectiveness is measured by containment rate, first-contact resolution, and customer satisfaction metrics, with human-in-the-loop escalation pathways ensuring quality and regulatory compliance.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:customer-service-automation", "labels": [ "Customer Service Automation" ], "is_subclass_of": [ "AI Application", "Conversational AI", "AI Agent", "Business Process Automation" ], "wikilinks": [ "Artificial Intelligence", "Conversational AI", "Natural Language Processing", "Large Language Model", "Intent Classification", "Sentiment Analysis", "Retrieval-Augmented Generation", "Knowledge Base", "Chatbot", "Virtual Agent", "Dialogue Management", "Named Entity Recognition", "Interactive Voice Response", "Robotic Process Automation", "Omnichannel", "Customer Relationship Management", "Escalation Management", "Workflow Automation", "Business Process Automation", "AI Agent" ] }, { "id": "customer-support-automation", "title": "Customer Support Automation", "domain": "ai", "domain_name": "Ai", "definition": "Customer Support Automation (CSA-Support) is the application of artificial intelligence, natural language processing, and workflow orchestration to the technical and post-sale support domain, automatically handling customer enquiries, diagnosing faults, routing and resolving service tickets, and delivering self-service resolution pathways through chatbots, virtual agents, and agentic AI systems. It specialises the broader Customer Service Automation domain toward helpdesk operations, IT service management, and product support contexts, integrating with ticketing platforms, knowledge bases, diagnostic APIs, and CRM systems to achieve autonomous first-line resolution of technical and product queries at scale.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:customer-support-automation", "labels": [ "Customer Support Automation" ], "is_subclass_of": [ "Conversational AI", "Customer Service Automation", "AI Agent" ], "wikilinks": [ "Dialogue Systems", "Chatbot", "Natural Language Processing", "Conversational AI", "Customer Service Automation", "Large Language Model", "Intent Classification", "Retrieval-Augmented Generation", "Knowledge Base", "Sentiment Analysis", "Named Entity Recognition", "Virtual Agent", "Dialogue Management", "Robotic Process Automation", "Workflow Automation", "Natural Language Understanding", "Machine Learning", "Transformer Architecture", "Agentic RAG", "Active Learning" ] }, { "id": "customer-support", "title": "Customer Support", "domain": "ai", "domain_name": "Ai", "definition": "Customer Support is the structured operational function and set of AI-augmented processes through which organisations resolve customer queries, technical faults, billing disputes, and product complaints across telephone, email, live chat, self-service portals, and social messaging channels, increasingly powered by Large Language Models, Retrieval-Augmented Generation architectures, and autonomous Agentic Workflow systems that automate first-contact resolution while escalating complex edge cases to human agents with full conversational context.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:customer-support", "labels": [ "Customer Support", "Multilingual Customer Support" ], "is_subclass_of": [ "Customer Service Automation", "Enterprise Workflow" ], "wikilinks": [ "Conversational AI", "Chatbots", "Natural Language Processing", "Sentiment Analysis", "Dialogue System", "Information Retrieval", "Workflow Automation", "Question Answering", "Large Language Models", "Retrieval-Augmented Generation", "CRM", "CRM Integration", "CRM Systems", "Intent Recognition", "Dialogue Management", "Dialogue State Tracking", "Multi-Turn Dialogue", "Knowledge Retrieval", "Speech Recognition", "Reinforcement Learning from Human Feedback" ] }, { "id": "customer-world-model", "title": "Customer World Model", "domain": "economics", "domain_name": "Economics", "definition": "A proprietary, data-driven internal representation of a customer's or merchant's financial behavior and context, used to generate personalized insights and services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:customer-world-model", "labels": [ "Customer World Model" ], "is_subclass_of": [ "World Model" ], "wikilinks": [] }, { "id": "customs-trade-facilitation", "title": "Customs Trade Facilitation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The systematic application of digital infrastructure \u2014 blockchain distributed ledgers, AI-driven risk engines, smart contracts, and interoperable single-window platforms \u2014 to reduce friction in cross-border trade whilst maintaining regulatory compliance for customs authorities and economic operat...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:customs-trade-facilitation", "labels": [ "Customs Trade Facilitation" ], "is_subclass_of": [ "Governance and Regulation", "Supply Chain Traceability", "Regulatory Technology", "Trade Finance", "Digital Customs", "Blockchain Network", "AML KYC Compliance" ], "wikilinks": [ "Advance Cargo Information System", "Alan Turing Institute", "Anti-Fraud Verification", "Authorised Economic Operator", "Automated Duty Calculation", "Carbon Border Adjustment", "Centralised Single Window", "Certificate of Origin", "ComplianceLayer", "Contour", "Coordinated Border Management", "Corda", "Cryptographic Hash Functions", "Customs Declaration", "Data Governance Framework", "Decentralised Identifiers", "Digital Customs", "Distributed Ledger Technology", "Electronic Bill of Lading", "Electronic Signature" ] }, { "id": "cut-mix", "title": "Cut Mix", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A data augmentation technique that creates training examples by cutting and pasting patches between images, with labels mixed proportionally to the patch areas. CutMix improves model robustness and localisation ability by forcing attention to less discriminative regions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:cut-mix", "labels": [ "Cut Mix", "CutMix" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "cvss", "title": "Cvss", "domain": "security", "domain_name": "Security", "definition": "The Common Vulnerability Scoring System (CVSS) is an open, standardised framework for rating the severity of software vulnerabilities on a 0-10 scale. It decomposes severity into Base, Temporal and Environmental metric groups capturing intrinsic exploitability, real-world threat conditions and organisation-specific impact. CVSS provides a vendor-neutral common language so defenders can compare and prioritise remediation consistently across heterogeneous products.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cvss", "labels": [ "Cvss", "CVSS" ], "is_subclass_of": [ "Vulnerability Management" ], "wikilinks": [] }, { "id": "cyber-physical-systems", "title": "Cyber Physical Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cyber Physical Systems (CPS) are engineered systems in which computational elements and physical processes are tightly integrated and coordinated through networked sensing, actuation, and feedback control. They span safety-critical domains including autonomous vehicles, industrial automation, smart grids, medical devices, and aerospace, demanding real-time coordination between embedded computation, heterogeneous communication networks, and physical plant dynamics. CPS extend traditional embedded and control systems by incorporating networked intelligence, adaptive autonomy, and large-scale coordination across geographically distributed components. Their design requires co-engineering of hardware, software, control, and communication layers under stringent timing, reliability, and safety constraints.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:cyber-physical-systems", "labels": [ "Cyber Physical Systems", "Cyber-Physical System", "Cyber-Physical Systems" ], "is_subclass_of": [ "Embedded Systems" ], "wikilinks": [] }, { "id": "cyber-resilience", "title": "Cyber Resilience", "domain": "security", "domain_name": "Security", "definition": "Cyber resilience is the capacity of an organisation or system to anticipate, withstand, recover from, and adapt to adverse cyber events while continuing to deliver its intended outcomes. It extends conventional cybersecurity from breach prevention towards graceful degradation, rapid restoration, and continuous learning under sustained attack. Cyber resilience integrates technical controls, business continuity planning, and governance so that critical functions persist even when individual defences fail.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cyber-resilience", "labels": [ "Cyber Resilience" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "cyber-security-and-cryptography", "title": "Cyber Security and Cryptography", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cyber Security and Cryptography is the integrated discipline governing the confidentiality, integrity, availability, and authenticity of digital information systems through mathematical cryptographic primitives, protocol engineering, and systemic security architecture.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cyber-security-and-cryptography", "labels": [ "Cyber Security and Cryptography" ], "is_subclass_of": [ "Security and Identity", "Information Security", "Applied Mathematics", "Network Security", "Systems Engineering", "Risk Management" ], "wikilinks": [ "Adversarial Machine Learning", "AES-GCM", "Applied Mathematics", "Authentication", "BSI Germany", "CA/Browser Forum", "Castle-and-Moat Architecture", "Certificate Authority", "ChaCha20-Poly1305", "Computational Complexity Theory", "Critical National Infrastructure", "Cryptographic Primitives", "CryptographyDomain", "Data Confidentiality", "Data Integrity", "Defence and Intelligence", "Digital Forensics", "Digital Sovereignty", "ENISA", "Entropy" ] }, { "id": "cyber-security-and-military", "title": "Cyber Security and Military", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cyber Security and Military denotes the doctrinal, organisational, technical and operational fusion of cyberspace as a recognised warfighting domain alongside land, sea, air and space \u2014 codified by NATO at the 2016 Warsaw Summit declaration that cyberspace is \"a domain of operations in which NATO...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:cyber-security-and-military", "labels": [ "Cyber Security and Military" ], "is_subclass_of": [ "Security and Identity", "Cyber Operations", "National Security Capability", "Warfighting Domain", "State-Sponsored Activity", "Defence Capability" ], "wikilinks": [ "Adversary Intelligence Collection", "Allied Coalition Operations", "Attribution Capability", "Battle Damage Assessment", "Buchanan The Hacker and the State 2020", "CISA FBI NSA Volt Typhoon Joint Advisory AA23-144a AA24-038a", "Civilian Cybersecurity", "Classified Compute Infrastructure", "Command and Control Infrastructure", "Computer Network Exploitation", "Computer Network Operations Doctrine", "Computer Networks", "Corporate Information Security", "Counter-Intelligence", "Counter-Proliferation", "Counter-Terrorism", "Critical Infrastructure", "Critical Infrastructure Disruption", "CriticalInfrastructureDomain", "Cyber Deterrence" ] }, { "id": "cybernetics", "title": "Cybernetics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cybernetics is the transdisciplinary science of regulatory systems, feedback mechanisms, and goal-directed behaviour in animals, machines, and organisations, founded by Norbert Wiener in 1948 to provide a unified framework for understanding how systems use information to maintain stability and achieve purposes across biological, mechanical, and social domains. It centres on the study of circular causal processes \u2014 feedback loops \u2014 through which a system compares its actual state against a desired state and acts to reduce the discrepancy, extending naturally to concepts of information, communication, control, and self-organisation. Second-order cybernetics extends this to include the observer as a participant in the system being observed, influencing systems theory, cognitive science, constructivism, and AI.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cybernetics", "labels": [ "Cybernetics" ], "is_subclass_of": [ "Control Theory", "Systems Theory" ], "wikilinks": [ "Control Theory", "Feedback Loop", "Feedback Control", "Information Theory", "Signal Processing", "Adaptive Control", "Reinforcement Learning", "Cognitive Science", "Systems Theory", "Complex Adaptive Systems", "Emergence", "Machine Learning", "Human Robot Interaction", "Brain Computer Interfaces", "Self-Organisation", "AI Alignment", "Autonomous Systems", "Robotics", "Self-Organised Criticality", "Homeostasis" ] }, { "id": "cybersecurity-framework", "title": "Cybersecurity Framework", "domain": "governance", "domain_name": "Governance", "definition": "A cybersecurity framework is a structured set of standards, guidelines and practices that organisations use to identify, protect against, detect, respond to and recover from cyber threats. Frameworks such as the NIST Cybersecurity Framework provide a common taxonomy and maturity model for managing security risk in a repeatable, auditable way. They align technical controls with governance, risk management and regulatory obligations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:cybersecurity-framework", "labels": [ "Cybersecurity Framework" ], "is_subclass_of": [ "Security Framework" ], "wikilinks": [] }, { "id": "cybersecurity-policy", "title": "Cybersecurity Policy", "domain": "governance", "domain_name": "Governance", "definition": "A cybersecurity policy is a formal set of rules, roles, and expectations that govern how an organisation or jurisdiction protects information systems, data, and networks from threats. It translates risk appetite and legal obligations into actionable standards covering access, incident handling, data protection, and acceptable use. Cybersecurity policy operates at organisational level as internal governance and at national level as regulation and strategy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cybersecurity-policy", "labels": [ "Cybersecurity Policy" ], "is_subclass_of": [ "Policy" ], "wikilinks": [] }, { "id": "cybersecurity-risk-management", "title": "Cybersecurity Risk Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured discipline for identifying, assessing, prioritising, and treating risks to information systems, data assets, and digital infrastructure arising from threats such as malicious actors, system vulnerabilities, and operational failures. Cybersecurity risk management integrates threat modelling, vulnerability assessment, control selection, and residual risk acceptance into a repeatable governance cycle aligned with organisational risk appetite. Frameworks such as NIST CSF, ISO 27005, and FAIR provide structured methodologies. It bridges technical security practice with executive governance and regulatory compliance.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:cybersecurity-risk-management", "labels": [ "Cybersecurity Risk Management" ], "is_subclass_of": [ "Artificial Intelligence", "Cybersecurity Framework", "Risk Management", "Information Security Management" ], "wikilinks": [ "Governance", "Security", "Risk Management", "Information Security Management", "Threat Modelling", "Vulnerability Assessment", "Continuous Monitoring", "Security Framework", "Risk Assessment", "Incident Response", "Compliance Monitoring", "AI Safety", "Supply Chain Security", "Penetration Testing", "NIST Cybersecurity Framework", "ISO 27001", "Resilience", "Accountability", "Security Operations", "Zero Trust Architecture" ] }, { "id": "cybersecurity-standard", "title": "Cybersecurity Standard", "domain": "robotics", "domain_name": "Robotics", "definition": "Cybersecurity standard addresses digital security threats to robot systems through standardised practices for vulnerability assessment, secure communications, authentication, and incident response.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:cybersecurity-standard", "labels": [ "Cybersecurity Standard" ], "is_subclass_of": [ "Safety and Standards", "Robot Standard", "Security Standards" ], "wikilinks": [ "Audit Logging", "Authentication Mechanism", "Communication Protocols", "Compliance Assurance", "Cryptographic Infrastructure", "Device Hardening", "Encryption Protocol", "IEC 62443", "ISO/SAE 21434", "Malicious Control Prevention", "Network Security", "Protected Data Transmission", "Secure Communication Channel", "Secure Teleoperation", "Security Standards", "Threat Detection", "Vulnerability Assessment", "Access Control", "AI Agent System", "Control Algorithm" ] }, { "id": "cybersecurity", "title": "Cybersecurity", "domain": "security", "domain_name": "Security", "definition": "Cybersecurity is the discipline concerned with protecting computer systems, networks, and data from unauthorised access, damage, and attack. In the AI governance context it addresses adversarial threats to machine learning models, data poisoning, model extraction, and the use of AI techniques for automated threat detection, vulnerability analysis, and intrusion detection. It overlaps strongly with privacy, encryption, access control, and authentication.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:cybersecurity", "labels": [ "Cybersecurity", "Cybersecurity Analytics" ], "is_subclass_of": [ "Information Security" ], "wikilinks": [ "owl:Thing" ] }, { "id": "cycloidal-drive", "title": "Cycloidal Drive", "domain": "robotics", "domain_name": "Robotics", "definition": "A cycloidal drive is a mechanical speed-reduction gearbox in which an eccentrically mounted cycloidal disc engages a ring of pins to transmit motion at a high reduction ratio. It offers low backlash, high torque density and good shock resistance, making it a favoured actuator component in robot joints and precision positioning systems. Compared with planetary gearing it tolerates impact loads and distributes force across many contact points.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:cycloidal-drive", "labels": [ "Cycloidal Drive" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "cylindrical-robot", "title": "Cylindrical Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Cylindrical robot employs one rotary joint (azimuth) and two prismatic joints (radial and vertical) that produce a cylindrical workspace, enabling efficient reach over rectangular work envelopes.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:cylindrical-robot", "labels": [ "Cylindrical Robot" ], "is_subclass_of": [ "Robot Type", "Actuation and Control", "Industrial Robot", "Coordinate Robots", "Fixed-Base Manipulator" ], "wikilinks": [ "Articulated Robots", "Assembly Automation", "Azimuth Drive", "Base Support", "Component Assembly", "Control Systems", "Coordinate Robots", "Depalletising", "End-Effector Mount", "Fixed-Base Manipulator", "Material Handling", "Palletising", "Pick and Place", "Prismatic Actuators", "Radial Actuator", "Rotary Joint", "Vertical Actuator", "Wrist", "Force Control", "Industrial Robot" ] }, { "id": "cypher-query-language", "title": "Cypher Query Language", "domain": "data", "domain_name": "Data", "definition": "Cypher is a declarative graph query language originally developed by Neo4j for querying property graph databases, using an ASCII-art syntax to express patterns of nodes and relationships intuitively. It allows users to describe graph patterns using parentheses for nodes and arrows for relationships, making queries readable and expressive without requiring deep knowledge of graph traversal algorithms. Cypher has become the basis of the openCypher project and GQL ISO standard, establishing it as a cross-vendor lingua franca for graph databases.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:cypher-query-language", "labels": [ "Cypher Query Language" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "dai", "title": "DAI", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DAI is a decentralised stablecoin issued by the MakerDAO protocol on Ethereum that aims to hold a value close to one US dollar through collateral backing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dai", "labels": [ "DAI", "DAI Stablecoin" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [ "MakerDAO", "Smart Contract", "Ethereum", "Stablecoin" ] }, { "id": "dall-e-3", "title": "DALL-E 3", "domain": "ai", "domain_name": "Ai", "definition": "DALL-E 3 is a text-to-image generation model developed by OpenAI that produces images from natural language descriptions. It improves prompt adherence over earlier versions by reformulating user prompts with a language model.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dall-e-3", "labels": [ "DALL-E 3", "DALL-E" ], "is_subclass_of": [ "Text-to-Image Generation", "Deep Generative Model", "Generative Model", "Generative AI" ], "wikilinks": [ "Diffusion Model", "Language Model", "Image Generation", "OpenAI", "Generative AI", "Text-to-Image Generation", "CLIP", "Latent Diffusion", "Variational Autoencoder", "Vision Transformer", "Contrastive Learning", "Generative Adversarial Network", "Stable Diffusion Image Model", "Midjourney Text-to-Image Service", "Deep Generative Model", "Generative Model", "CLIP Encoder", "Latent Diffusion Model Training", "Generative AI Engineering", "Creative Tools" ] }, { "id": "dao-analytics", "title": "DAO Analytics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Specialised data analysis and business intelligence infrastructure providing quantitative insight into Decentralised Autonomous Organisation performance across governance participation, treasury health, delegate accountability, proposal lifecycle dynamics, sybil resistance, and ecosystem-wide...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:dao-analytics", "labels": [ "DAO Analytics" ], "is_subclass_of": [ "Governance and Regulation", "Analytics Platforms", "Blockchain Analytics", "Governance Tooling", "Data-Driven Governance", "Decentralised Autonomous Organisation" ], "wikilinks": [ "Aave", "Aave Governance", "ABI Repository", "Analytics Platforms", "Apache Parquet", "Apache Spark", "Apache Trino", "Aragon", "Arbitrum", "Arbitrum DAO", "Base", "Block Explorer", "Blockchain Analytics", "BNB Chain", "Boardroom", "Centralised Corporate Governance Reporting", "Chainlink", "Closed-Source Analytics", "Community Understanding", "Compound" ] }, { "id": "dao-governance-for-telecollaboration", "title": "DAO Governance for Telecollaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The application of decentralised autonomous organisation (DAO) governance mechanisms\u2014token-weighted voting, proposal systems, treasury management\u2014to coordinate geographically distributed teams through on-chain decision-making, enabling democratic, transparent collaboration without centralised ma...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:dao-governance-for-telecollaboration", "labels": [ "DAO Governance for Telecollaboration" ], "is_subclass_of": [ "Protocol and Infrastructure", "Blockchain Technology" ], "wikilinks": [ "Blockchain Technology", "DecentralisedAutonomousOrganisation", "TELE-002-telecollaboration", "TELE-250-blockchain-collaboration", "TELE-251-smart-contract-coordination" ] }, { "id": "dao-governance", "title": "DAO Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DAO governance refers to the decision-making structures and voting mechanisms within Decentralised Autonomous Organisations, enabling token holders to collectively govern protocol parameters, treasury allocations, and organisational direction through on-chain and off-chain processes. Governance rules are encoded as smart contracts that automatically execute approved decisions, supporting a spectrum of voting models including token-weighted, quadratic, delegated, conviction, and reputation-based approaches.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:dao-governance", "labels": [ "DAO Governance", "DAO Governance Framework", "DAO Governance Protocol" ], "is_subclass_of": [ "Community Governance" ], "wikilinks": [ "Decentralised Decision-Making", "Blockchain" ] }, { "id": "dao-legal-structures", "title": "DAO Legal Structures", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DAO Legal Structures are the corpus of statutory wrappers, jurisdictional incorporation regimes, and bespoke entity forms \u2014 spanning US state limited-liability companies, US state unincorporated nonprofit associations, offshore foundation companies, Swiss associations and foundations, Liechtenste...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:dao-legal-structures", "labels": [ "DAO Legal Structures", "DAO Legal Frameworks" ], "is_subclass_of": [ "Governance and Regulation", "Legal Entity Wrappers", "Governance Structures", "Corporate Forms", "Liability Shield Mechanisms", "Regulatory Compliance Frameworks" ], "wikilinks": [ "a16z crypto Legal", "Aave Companies", "Algorithmic Management Doctrine", "Allen & Overy Shearman", "Anti-Money Laundering Compliance", "Aragon", "Articles of Organization", "Astor v MakerDAO 2024", "B\u00fcrgi N\u00e4geli Rechtsanw\u00e4lte", "Bank Account Access", "Beneficial Ownership Disclosure", "Beneficial Ownership Registers", "BVI Business Companies Act 2004", "BVI Limited Company", "Bylaws", "Cambridge Centre for Alternative Finance (CCAF)", "Cayman Enterprise City", "Cayman Foundation Companies Act 2017", "Cayman Foundation Company", "CCAF 2024 Global Cryptoasset Benchmarking Study" ] }, { "id": "dao-tooling", "title": "DAO Tooling", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DAO Tooling is the integrated software stack of governance platforms, treasury custodians, execution oracles, deliberation forums, identity primitives, analytics layers, and compensation rails that operationalises decentralised autonomous organisations as functioning socio-technical institutions ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:dao-tooling", "labels": [ "DAO Tooling" ], "is_subclass_of": [ "Governance and Regulation", "Governance Infrastructure", "Blockchain Infrastructure", "Organisational Software", "Coordination Technology", "Web3 Stack" ], "wikilinks": [ "0xPARC", "1Hive", "a16z", "a16z State of Crypto 2024", "Aave", "Aave Companies", "Aave DAO", "Aavegotchi", "Aavenomics", "Admin", "Agentic AI", "Aggelos Kiayias", "Agora", "ai16z", "ai16z DAO", "AIP-1", "Albert Hirschman", "Aleo", "Alexei Zamyatin", "Allo Protocol" ] }, { "id": "dao", "title": "DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised Autonomous Organisation operating through transparent code-based rules and distributed governance rather than hierarchical management, enabling community coordination at scale via smart contracts, token-based voting, and on-chain treasury management.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:dao", "labels": [ "DAO", "DAO Operations", "DAO Structure", "DAOs", "Data DAO", "Decentralized Autonomous Organization" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "Blockchain-Based Records", "Consensus Voting", "DAOs", "Distributed Consensus", "AI Agent System", "BlockchainDomain", "Blockchain Infrastructure", "Smart Contracts" ] }, { "id": "daogovernance", "title": "DAOGovernance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The set of on-chain and off-chain processes by which a decentralised autonomous organisation makes and enforces collective decisions. It covers proposal submission, voting and execution of approved actions through smart contracts.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:daogovernance", "labels": [ "DAOGovernance" ], "is_subclass_of": [ "Decentralized Autonomous Organization" ], "wikilinks": [ "Smart Contract", "Governance", "Quadratic Voting", "Decentralized Governance", "Token", "Decentralized Autonomous Organization", "https://ethereum.org/en/dao/" ] }, { "id": "dapp", "title": "DApp", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised application whose backend logic runs on a blockchain or other decentralised network rather than on centrally controlled servers. It typically combines smart contracts with a conventional user interface.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dapp", "labels": [ "DApp" ], "is_subclass_of": [ "Decentralized Application" ], "wikilinks": [ "Smart Contract", "Web3", "Decentralized Application" ] }, { "id": "dbpedia", "title": "DBpedia", "domain": "data", "domain_name": "Data", "definition": "DBpedia is a community project that extracts structured information from Wikipedia and publishes it as an interlinked, openly licensed knowledge graph in RDF. It exposes millions of entities through a SPARQL endpoint and dereferenceable URIs, forming a central hub of the Linked Open Data cloud. It is widely used as background knowledge for knowledge-graph construction, entity linking and semantic search.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dbpedia", "labels": [ "DBpedia" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "dc-servo-motor", "title": "DC Servo Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "DC servo motor combines a direct-current electric motor with integrated or external feedback control electronics to enable precise position, velocity, or torque regulation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:dc-servo-motor", "labels": [ "DC Servo Motor", "Brushless DC Motor" ], "is_subclass_of": [ "Actuation and Control", "Electric Motor", "Servo Motor" ], "wikilinks": [ "Bearings", "Control Electronics", "Cooling System", "Current Driver", "DC Motor", "DC Power Supply", "Dynamic Control", "Feedback Sensor", "Load Adaptation", "Mechanical Load", "Motor Shaft", "Position Encoder", "Precise Positioning", "Robotic Joint", "Robotic Manipulation", "Servo Amplifier", "Trajectory Tracking", "AI Agent System", "Electric Motor", "Motion Control" ] }, { "id": "dds-middleware", "title": "DDS Middleware", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Data Distribution Service (DDS) Middleware is an OMG-standardised publish-subscribe communication middleware designed for real-time, scalable, and decentralised data exchange in safety-critical and high-performance distributed systems. It defines a Data-Centric Publish-Subscribe (DCPS) model in which participants discover each other automatically and exchange typed data via a global data space, governed by a rich set of Quality of Service (QoS) policies covering reliability, latency, deadline, liveliness, and durability. DDS is the foundational communication layer for ROS 2, NATO's STANAG 4910 tactical networks, and aerospace control systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dds-middleware", "labels": [ "DDS Middleware" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [] }, { "id": "dds", "title": "DDS", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "DDS, the Data Distribution Service, is an Object Management Group standard for a publish-subscribe middleware that distributes real-time data between distributed system components.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dds", "labels": [ "DDS", "Cyclone DDS", "OMG DDS Specification", "OMG DDS Standard" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [ "Network Protocol", "Robot Operating System", "Distributed Systems", "Middleware" ] }, { "id": "ddos-mitigation", "title": "DDoS Mitigation", "domain": "security", "domain_name": "Security", "definition": "DDoS mitigation is the set of techniques and infrastructure used to detect and absorb distributed denial-of-service traffic so that a targeted service remains available to legitimate users. Approaches include traffic scrubbing, rate limiting, anycast-based load distribution, and edge filtering at content delivery networks positioned close to end users. Effective mitigation depends on capacity headroom large enough to absorb attack volumes while distinguishing malicious traffic from genuine demand spikes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:ddos-mitigation", "labels": [ "DDoS Mitigation" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "dex", "title": "DEX", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The common abbreviation for a decentralised exchange, a smart-contract trading venue where users swap assets without surrendering custody to a central operator.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dex", "labels": [ "DEX", "Perpetual DEX" ], "is_subclass_of": [ "Decentralised Exchange" ], "wikilinks": [ "Smart Contract", "Liquidity Pool", "Automated Market Maker", "Order Book", "Decentralised Exchange" ] }, { "id": "dicom", "title": "DICOM", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "DICOM (Digital Imaging and Communications in Medicine) is the international standard for storing, transmitting, and managing medical images and associated metadata. It defines both a file format that binds pixel data to rich patient, study, and acquisition attributes, and network services for exchanging images between modalities, archives, and viewing workstations. DICOM enables interoperability across radiology, cardiology, and other imaging-intensive specialities, and serves as the canonical data substrate for medical imaging artificial intelligence.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dicom", "labels": [ "DICOM" ], "is_subclass_of": [ "Interoperability Standard", "Healthcare Data Standard" ], "wikilinks": [ "Interoperability Standard", "Medical Imaging", "Medical Imaging AI", "Healthcare AI", "Computer Vision", "Deep Learning", "Data Protection", "Interoperability Framework", "HL7 FHIR", "Digital Health", "Electronic Health Record", "Natural Language Processing", "Federated Learning", "Data Governance", "Neural Network", "Convolutional Neural Network", "Standards Body", "Interoperability", "REST API", "Clinical Decision Support" ] }, { "id": "did-controller", "title": "DID Controller", "domain": "security", "domain_name": "Security", "definition": "The entity\u2014person, organisation, or autonomous system\u2014authorised to make changes to a DID document, as defined by the W3C DID Core specification. The DID controller proves its authority by controlling the cryptographic keys or verification methods designated in the DID document, and may be the DID subject itself (self-sovereign control), a guardian acting for a subject such as a child or an IoT device, or a set of parties sharing control. It is unrelated to the PID controller of control engineering, with which it shares only a surface string.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:did-controller", "labels": [ "DID Controller" ], "is_subclass_of": [ "Entity" ], "wikilinks": [ "Decentralized Identifiers", "DID Document", "Cryptographic Key" ] }, { "id": "did-document", "title": "did document", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A DID Document is a JSON-LD resource resolved from a Decentralised Identifier (DID) that expresses the DID subject's cryptographic public keys, authentication methods, assertion methods, key agreement protocols, capability delegation mechanisms, and service endpoints. Defined by the W3C DID Core specification, a DID Document is the machine-readable artefact that enables any verifier to discover how to authenticate with or communicate securely to a DID subject without relying on a centralised identity authority. DID Documents are anchored to a Verifiable Data Registry \u2014 such as a distributed ledger, blockchain, or peer-to-peer network \u2014 via a DID method-specific read operation, making them a foundational component of self-sovereign identity and verifiable credential ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:did-document", "labels": [ "DID Document" ], "is_subclass_of": [ "Decentralised Identifier" ], "wikilinks": [] }, { "id": "did-method", "title": "DID Method", "domain": "security", "domain_name": "Security", "definition": "A DID Method is a specification that defines how a particular type of Decentralised Identifier (DID) is created, resolved, updated, and deactivated, encoding the syntax rules and CRUD semantics for identifiers anchored to a specific verifiable data registry, ledger, or peer-to-peer network. Each method is identified by a unique method name embedded in the DID string (e.g. did:web, did:key, did:ion), and must provide a compliant DID Resolver that retrieves the corresponding DID Document containing public keys and service endpoints. DID Methods are standardised through the W3C DID Core specification and registered in the DIF DID Method Registry, enabling interoperable self-sovereign identity across heterogeneous systems. The diversity of methods reflects trade-offs between decentralisation guarantees, key management complexity, ledger dependency, and resolver performance.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:did-method", "labels": [ "DID Method", "DID Method Driver", "W3C DID Method", "did:indy Method Specification" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [ "Decentralized Identity", "Digital Identity", "W3C" ] }, { "id": "did-nostr-identity", "title": "DID Nostr Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A W3C DID Core|W3C Decentralised Identifier (did:nostr:) binding a VisionClaw Agentic Container|VisionClaw agent to its BIP-340 Schnorr Keypair|BIP-340 x-only public key, enabling cryptographic proof of identity, self-sovereign key management, and inter-agent trust without a c...", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:did-nostr-identity", "labels": [ "DID Nostr Identity" ], "is_subclass_of": [ "Network Component", "Distributed Identity" ], "wikilinks": [ "ADR-013", "AgenticSystemsDomain", "BIP-340", "BIP-340 Cryptography", "BIP-340 Schnorr Keypair", "BIP-340 Schnorr Keypair", "Blockchain Identity", "Credential Issuance", "Cryptographic Identity", "CryptographyDomain", "Decentralised Trust", "DID Resolution", "Ethereum Name Service", "IdentityDomain", "Message Signing", "Nostr NIP-01", "RFC 8610", "Schnorr Signatures", "Self-Sovereign Key Management", "Smart Contract Verification" ] }, { "id": "did-resolution", "title": "DID Resolution", "domain": "security", "domain_name": "Security", "definition": "DID Resolution is the process of dereferencing a Decentralised Identifier (DID) to retrieve its associated DID Document, which contains cryptographic public keys, authentication methods, and service endpoints. Defined by the W3C DID Core specification, resolution is performed by a DID Resolver that applies the read operation of the relevant DID Method \u2014 such as did:web, did:key, or did:ion \u2014 against its underlying verifiable data registry. The resolution process returns both a DID Document and Resolution Metadata describing the outcome, enabling verifiers to authenticate subjects, verify Verifiable Credentials, and establish secure communication channels in decentralised identity systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:did-resolution", "labels": [ "DID Resolution", "DID Resolution Protocol" ], "is_subclass_of": [ "Decentralized Identifier" ], "wikilinks": [ "Decentralized Identifier", "Identity Verification System", "Identity Management" ] }, { "id": "did-resolver", "title": "DID Resolver", "domain": "security", "domain_name": "Security", "definition": "A DID resolver is a component that takes a Decentralized Identifier (DID) and returns its associated DID document containing public keys, verification methods, and service endpoints. It implements the W3C DID Resolution specification, dispatching to method-specific drivers (e.g. did:web, did:ion, did:key) to locate and verify the document. Resolvers are the lookup layer that makes DIDs actionable for authentication and credential exchange.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:did-resolver", "labels": [ "DID Resolver" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "did", "title": "DID", "domain": "security", "domain_name": "Security", "definition": "A decentralised identifier, a type of globally unique identifier that enables verifiable, self-sovereign digital identity without reliance on a central registry.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:did", "labels": [ "DID", "Decentralised Identifiers" ], "is_subclass_of": [ "Self Sovereign Identity" ], "wikilinks": [ "Public Key", "Cryptography", "Verifiable Credentials", "Self Sovereign Identity", "Decentralized Identifiers" ] }, { "id": "didcomm-v2", "title": "DIDComm v2", "domain": "security", "domain_name": "Security", "definition": "DIDComm v2 is a transport-agnostic, end-to-end encrypted messaging protocol that lets two or more parties communicate securely using their Decentralized Identifiers. Messages are signed and encrypted with keys discovered from DID documents, providing confidentiality, authenticity, and mutual authentication independent of any central server or transport. It is the secure communication layer underpinning self-sovereign identity interactions such as verifiable credential exchange.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:didcomm-v2", "labels": [ "DIDComm v2" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "didcomm", "title": "DIDComm", "domain": "security", "domain_name": "Security", "definition": "DIDComm (Decentralised Identifier Communication) is a secure, private messaging protocol built on top of the W3C Decentralised Identifiers specification, enabling peer-to-peer, end-to-end encrypted communication between parties whose identities are anchored to DIDs. The protocol specifies how messages are packaged, signed, and encrypted using keys derived from DID documents, without reliance on any central server or directory. DIDComm messages are transport-agnostic, operating over HTTP, Bluetooth, NFC, or any delivery mechanism, making the protocol suitable for both online and offline identity interactions. It forms the communication backbone of the self-sovereign identity ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:didcomm", "labels": [ "DIDComm", "DIDComm Messaging" ], "is_subclass_of": [ "Decentralized Identity (DID)" ], "wikilinks": [] }, { "id": "dlss", "title": "DLSS", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "DLSS (Deep Learning Super Sampling) is NVIDIA's family of AI-based rendering techniques that use neural networks running on dedicated tensor hardware to reconstruct high-resolution, high-frame-rate images from lower-resolution rendered inputs. By upscaling, accumulating temporal information, and generating intermediate frames, DLSS delivers image quality approaching or exceeding native rendering at a fraction of the GPU cost, enabling demanding effects such as real-time ray tracing to run smoothly. It exemplifies the integration of learned models into the real-time graphics pipeline.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dlss", "labels": [ "DLSS" ], "is_subclass_of": [ "Rendering Technology" ], "wikilinks": [] }, { "id": "dlt", "title": "DLT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed Ledger Technology (DLT) is a class of systems in which transaction records are replicated, shared, and synchronised across multiple nodes without a single central authority. Blockchains are one form of DLT, but the term also covers directed-acyclic-graph and other non-chained ledger structures. DLT underpins many digital-asset and central-bank-digital-currency designs because it provides tamper-evidence, auditability, and resilience.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dlt", "labels": [ "DLT" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "dns", "title": "DNS", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Domain Name System (DNS) is a hierarchical, distributed naming system and directory service that translates human-readable domain names into IP addresses and other resource records required for locating internet services. Operating as a global, federated database partitioned into zones, DNS uses a delegation tree rooted at thirteen authoritative root-server clusters and propagates queries recursively through top-level domain, second-level domain, and sub-domain name servers. It is defined in RFC 1034 and RFC 1035 and extended by DNSSEC (RFC 4033-4035) to provide cryptographic integrity verification of responses, mitigating cache poisoning, spoofing, and man-in-the-middle attacks against the resolution chain.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:dns", "labels": [ "DNS", "DNS Resolution", "DNS Service" ], "is_subclass_of": [ "Application Layer" ], "wikilinks": [ "Network Protocol", "HTTP", "Communication Protocols", "Application Layer" ] }, { "id": "do-178-c", "title": "DO-178C", "domain": "standards", "domain_name": "Standards", "definition": "DO-178C (Software Considerations in Airborne Systems and Equipment Certification) is the primary international standard governing the development and certification of airborne software, published by RTCA in 2011 as the successor to DO-178B. It defines five software levels (A through E) based on failure severity, each requiring progressively more rigorous development assurance activities including requirements traceability, structural coverage testing, and independence in reviews. Compliance with DO-178C is required by aviation regulatory authorities (FAA, EASA) for software installed in certified aircraft.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:do-178-c", "labels": [ "DO-178C", "DO-178C Avionics" ], "is_subclass_of": [ "Safety Standard" ], "wikilinks": [] }, { "id": "dscsa-compliance", "title": "DSCSA Compliance", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "DSCSA Compliance refers to meeting the requirements of the U.S. Drug Supply Chain Security Act, which mandates electronic, interoperable tracing of prescription drugs at the unit (package) level throughout the supply chain. It requires serialisation, product identifiers, transaction history exchange, and verification of suspect or illegitimate product. Full enforcement of unit-level traceability obligates manufacturers, distributors, and dispensers to maintain auditable end-to-end records.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dscsa-compliance", "labels": [ "DSCSA Compliance", "FDA DSCSA" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "dspy", "title": "DSPy", "domain": "ai", "domain_name": "Ai", "definition": "DSPy is an open-source framework for programming language models in which developers declare the structure of a task using typed signatures and composable modules, and an optimiser automatically generates and tunes the prompts and few-shot examples needed to maximise a defined metric. By treating prompts as learnable parameters rather than hand-written strings, DSPy shifts language-model application development from manual prompt engineering toward systematic, metric-driven compilation of pipelines. It targets reliable, portable multi-stage LLM programs such as retrieval-augmented and agentic systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dspy", "labels": [ "DSPy" ], "is_subclass_of": [ "LLM Application Framework", "Automatic Prompt Optimisation", "AI Framework" ], "wikilinks": [ "LLM Application Framework", "Prompt Engineering", "Automatic Prompt Optimisation", "LLM Orchestration", "Chain-of-Thought Prompting", "LLM Agents", "Large Language Models", "Retrieval Augmented Generation", "Few-Shot Prompting", "In-Context Learning", "Machine Learning", "Bayesian Optimisation", "Natural Language Processing", "Transformer Architecture", "Training and Fine Tuning", "Instruction Following", "Model Optimisation and Performance", "Hallucination", "Evaluation Benchmarks and Leaderboards", "Reinforcement Learning" ] }, { "id": "dtcc", "title": "DTCC", "domain": "finance", "domain_name": "Finance", "definition": "DTCC, the Depository Trust and Clearing Corporation, is a United States company that provides clearing, settlement, and recordkeeping services for securities markets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dtcc", "labels": [ "DTCC", "DTCC Digital Securities Initiative" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [ "Traditional Finance", "Asset Tokenisation", "Financial Infrastructure" ] }, { "id": "dvc", "title": "DVC", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "DVC (Data Version Control) is an open-source, Git-integrated data and model versioning tool that applies version control semantics to machine learning datasets, trained model artefacts, and reproducible ML pipelines. It stores large binary artefacts in external object storage while committing hash-based pointer files to Git, enabling deterministic experiment replay, collaborative data science, and automated CI/CD-driven ML pipelines.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dvc", "labels": [ "DVC" ], "is_subclass_of": [ "AI Infrastructure", "MLOps", "Data Versioning", "Machine Learning Discipline" ], "wikilinks": [ "Checkpoints", "MLOps", "Data Versioning", "Experiment Tracking", "MLflow", "Model Registry", "Git", "Data Pipeline", "Continuous Integration", "Machine Learning Discipline", "AI Infrastructure", "Weights and Biases", "Model Versioning", "Reproducibility", "Artifact Metadata", "Amazon S3", "Google Cloud Storage", "Azure Blob Storage", "Hyperparameter Tuning", "CI/CD Pipeline" ] }, { "id": "dwpose", "title": "DWPose", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "DWPose is a whole-body 2D human pose estimation model developed by IDEA-Research that detects 133 keypoints spanning body, hands, face, and feet using a two-stage teacher-student knowledge distillation scheme. Its skeletal conditioning maps are widely used to drive spatial-conditioning systems such as ControlNet in diffusion-based image and video generation pipelines.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dwpose", "labels": [ "DWPose" ], "is_subclass_of": [ "Computer Vision", "Pose Estimation", "Knowledge Distillation", "Keypoint Detection" ], "wikilinks": [ "Pose Estimation", "ControlNet and Similar Spatial Conditioning Systems", "Diffusion Model", "Computer Vision", "OpenPose", "RTMPose", "MMPose", "Knowledge Distillation", "COCO WholeBody", "Stable Diffusion", "Image Generation", "Video Generation", "Human Body Model", "Keypoint Detection", "Skeleton Map", "Convolutional Neural Network", "Transformer Architecture", "Character Animation", "Gesture Recognition", "Digital Human Technology" ] }, { "id": "dagger-ci-pipeline-engine", "title": "Dagger CI Pipeline Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Dagger is a programmable CI/CD engine that unifies pipeline definition and execution within typed, composable functions authored in general-purpose languages such as Go, Python, or TypeScript. Unlike traditional Dockerfile-and-shell-script approaches, Dagger caches the result of every function call at fine granularity, achieves CI/local parity by running identically on developer machines and cloud runners, and exposes pipelines as strongly-typed APIs discoverable via `dagger functions`.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:dagger-ci-pipeline-engine", "labels": [ "Dagger CI Pipeline Engine", "Dagger" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Blockchain" ] }, { "id": "dama-dmbok", "title": "Dama Dmbok", "domain": "data", "domain_name": "Data", "definition": "The DAMA Data Management Body of Knowledge (DAMA-DMBOK) is a reference framework published by DAMA International that codifies the principles, functions, and best practices of enterprise data management. It organises the discipline into knowledge areas such as data governance, data quality, metadata, master data, and data architecture, arranged around a central data governance function. DAMA-DMBOK provides a common vocabulary and standard structure for establishing and assessing data management programmes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:dama-dmbok", "labels": [ "Dama Dmbok", "DAMA DMBOK", "DAMA-DMBOK" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "danksharding", "title": "Danksharding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Danksharding is a data-availability scaling design for Ethereum that scales the blockchain by providing large amounts of cheap data space for rollups rather than sharding execution. It uses a unified fee market and a merged block-building process in which a single proposer commits to a block containing many data blobs, the availability of which is verified through data-availability sampling and erasure coding. Danksharding is the long-term target architecture that proto-danksharding incrementally builds towards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:danksharding", "labels": [ "Danksharding" ], "is_subclass_of": [ "Sharding" ], "wikilinks": [] }, { "id": "dao-treasury", "title": "Dao Treasury", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A DAO treasury is the pool of on-chain assets collectively owned and governed by a decentralised autonomous organisation, typically held in smart contracts and controlled through member governance. It funds operations, grants and incentives, with disbursements authorised by token-weighted votes and executed by multisignature or programmatic controls. Treasury management balances runway, diversification and accountability without a central custodian.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:dao-treasury", "labels": [ "Dao Treasury", "DAO Treasury" ], "is_subclass_of": [ "DAO" ], "wikilinks": [] }, { "id": "dark-energy", "title": "Dark Energy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dark-energy", "labels": [ "Dark Energy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "dark-matter", "title": "Dark Matter", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dark-matter", "labels": [ "Dark Matter" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-access-interface", "title": "Data Access Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Data Access Interface is a formalised contract\u2014such as a REST API, GraphQL endpoint, or SPARQL query service\u2014that mediates structured access to data stores, registries, or knowledge graphs. It enforces authentication, authorisation, and schema validation, decoupling data consumers from underlying storage implementations. In metaverse and spatial computing contexts, data access interfaces expose asset repositories, user identity records, and scene graphs to applications and AI agents.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-access-interface", "labels": [ "Data Access Interface" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-acquisition", "title": "Data Acquisition", "domain": "robotics", "domain_name": "Robotics", "definition": "Data acquisition is the process of gathering raw signals and measurements from sensors and the environment, then conditioning and recording them for downstream use in perception, learning, and control. In robotics it spans sampling sensor streams, time-synchronising heterogeneous sources, and logging structured datasets for training and analysis. High-quality acquisition is foundational because the fidelity of perception and learned policies depends on the data captured.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-acquisition", "labels": [ "Data Acquisition" ], "is_subclass_of": [ "Data Collection" ], "wikilinks": [] }, { "id": "data-aggregation", "title": "Data Aggregation", "domain": "data", "domain_name": "Data", "definition": "Data aggregation is the systematic process of collecting, combining, and summarising records from multiple heterogeneous sources into a unified, reduced representation suitable for analysis, reporting, or further processing. It encompasses both batch and streaming paradigms, applying operations such as grouping, counting, summing, averaging, and deduplication to transform raw, high-volume data into structured, lower-dimensionality outputs. Aggregation underpins analytical workflows from simple dashboards to complex federated query engines, acting as a bridge between raw data capture and actionable intelligence. It is distinct from raw data replication in that it deliberately reduces detail while preserving statistically significant structure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:data-aggregation", "labels": [ "Data Aggregation", "Data Aggregation Layer", "Financial Data Aggregation", "Multi-Source Data Aggregation" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [ "Data Integration", "Community Detection", "Knowledge Graph" ] }, { "id": "data-analysis-service", "title": "Data Analysis Service", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-analysis-service", "labels": [ "Data Analysis Service" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-analysis", "title": "Data Analysis", "domain": "data", "domain_name": "Data", "definition": "Data analysis is the systematic process of inspecting, cleaning, transforming, and modelling data to extract useful information, support decision-making, and test hypotheses. It encompasses descriptive summarisation, exploratory analysis to surface structure and anomalies, inferential statistics to generalise from samples, and predictive modelling. Data analysis spans manual statistical work through to automated analytics pipelines, and it is the disciplinary core from which data science, business intelligence, and machine-learning workflows draw. Rigorous analysis attends to data quality, sampling bias, and the validity of inferential assumptions.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:data-analysis", "labels": [ "Data Analysis", "Data Analysis Agents" ], "is_subclass_of": [ "Process" ], "wikilinks": [] }, { "id": "data-analytics", "title": "Data Analytics", "domain": "data", "domain_name": "Data", "definition": "Data Analytics is the systematic computational examination of raw data sets to uncover patterns, correlations, and actionable insights that support organisational decision-making. It spans four analytical tiers: descriptive analytics (summarising historical state), diagnostic analytics (identifying causes of past events), predictive analytics (forecasting future outcomes via statistical and machine learning models), and prescriptive analytics (recommending optimal actions). Analytics pipelines ingest data from heterogeneous sources, apply transformation and enrichment steps, execute statistical or ML algorithms, and surface results through visualisations, dashboards, and automated alerts. As a discipline, data analytics integrates elements of statistics, computer science, domain knowledge, and data engineering to create measurable business and scientific value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-analytics", "labels": [ "Data Analytics", "Analytics", "Big Data Analytics", "ESG Data Analytics", "Historical Analytics" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "data-annotation", "title": "Data Annotation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Data Annotation is the process of labeling or tagging raw data (images, text, audio, video) with structured, meaningful labels that provide ground truth for supervised machine learning models, enabling algorithms to learn from human-validated examples.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:data-annotation", "labels": [ "Data Annotation", "Annotation", "Cost-Effective Annotation" ], "is_subclass_of": [ "AI Technique", "Machine Learning Pipeline" ], "wikilinks": [ "Cohen's Kappa", "GDPR", "HIPAA", "Snorkel", "Machine Learning Pipeline", "Training Data" ] }, { "id": "data-anonymization-pipeline", "title": "Data Anonymization Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An automated, multi-stage process that systematically removes, masks, or generalizes personally identifiable information (PII) from datasets to protect individual privacy while preserving data utility for analysis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:data-anonymization-pipeline", "labels": [ "Data Anonymization Pipeline" ], "is_subclass_of": [ "Governance and Safety", "Data Protection" ], "wikilinks": [ "Data Classification", "De-identification Module", "ENISA Anonymization Guide", "GDPR Compliance", "Generalization Engine", "ISO 20889", "Perturbation Function", "PII Detection", "PII Detector", "Privacy Engineering", "Privacy Policy", "Risk Assessor", "Suppression Filter", "Access Control", "Blockchain", "Compliance Framework", "Data Governance", "Data Layer", "Data Protection", "Differential Privacy" ] }, { "id": "data-architecture", "title": "Data Architecture", "domain": "data", "domain_name": "Data", "definition": "Data architecture is the discipline of designing the structures, standards, models and integration patterns that govern how an organisation collects, stores, transforms, moves and consumes data. It defines logical and physical data models, storage technologies, data flows, master data strategies and the policies that ensure consistency, quality and security across systems. As a component of enterprise architecture, it provides the blueprint that aligns data assets with business and analytical requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-architecture", "labels": [ "Data Architecture" ], "is_subclass_of": [ "Enterprise Architecture" ], "wikilinks": [] }, { "id": "data-archiving", "title": "Data Archiving", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-archiving", "labels": [ "Data Archiving" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-augmentation-strategies", "title": "Data Augmentation Strategies", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Techniques that create modified versions of training examples to increase dataset diversity and model robustness. Data augmentation strategies apply transformations that preserve label semantics whilst introducing variation, improving generalisation and reducing overfitting.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:data-augmentation-strategies", "labels": [ "Data Augmentation Strategies" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Apple Machine Learning Research", "arXiv", "IEEE/CVF International Conference on Computer Vision", "Nature", "Computer Vision", "MetaverseDomain" ] }, { "id": "data-augmentation", "title": "Data Augmentation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Data augmentation is a machine learning technique that expands a training dataset by applying label-preserving transformations or synthesising new examples from existing data. Typical methods include geometric and photometric image transforms, noise injection, and generative model sampling such as GAN-produced samples, diffusion-model synthesis, and mixing-based strategies such as Mixup and CutMix. It improves model generalisation and robustness, mitigating overfitting when labelled data is scarce. Modern automated augmentation pipelines such as AutoAugment and RandAugment use reinforcement learning or random search to discover optimal policies, while 2024-2026 diffusion-based approaches such as DiffuseMix enable label-preserving generation of high-fidelity training examples that improve performance on imbalanced and low-resource benchmarks.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:data-augmentation", "labels": [ "Data Augmentation", "Monolingual Data Augmentation" ], "is_subclass_of": [ "Machine Learning Technique", "Regularisation", "Machine Learning" ], "wikilinks": [ "Machine Learning", "Deep Learning", "Generative Adversarial Networks", "Variational Autoencoder", "Diffusion Model", "Computer Vision", "Natural Language Processing", "Regularisation", "Overfitting", "Transfer Learning", "Supervised Learning", "Semi-Supervised Learning", "Few-Shot Learning", "Convolutional Neural Networks", "Transformer Architecture", "Object Detection", "Image Classification", "Training Pipeline", "Synthetic Data", "Noise Injection" ] }, { "id": "data-availability-sampling", "title": "Data Availability Sampling", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Data availability sampling (DAS) is a technique that lets resource-limited clients gain high statistical confidence that all the data behind a block has been published, without downloading the whole block. Each client requests a few random fragments of an erasure-coded dataset; if enough randomly chosen fragments are returned across many clients, the full data can be reconstructed, so withholding it becomes detectable. DAS is a foundational primitive for scalable, modular blockchains and rollup-centric architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:data-availability-sampling", "labels": [ "Data Availability Sampling" ], "is_subclass_of": [ "Data Availability" ], "wikilinks": [] }, { "id": "data-availability", "title": "Data Availability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Data availability is the property of a distributed system\u2014most critically blockchain networks and rollup scaling architectures\u2014that guarantees all data necessary to verify and reconstruct network state is published and retrievable by any participant. The core data availability problem arises when a block producer publishes a block header without releasing the underlying transaction data, making it impossible for validators or light clients to verify the block's correctness without downloading all data. Modern solutions combine erasure coding (expanding data such that any sufficient subset allows full reconstruction) with data availability sampling (DAS), enabling light nodes to probabilistically confirm full publication by checking only a small random subset of encoded chunks. Data availability is a foundational primitive in modular blockchain architectures, separating the data publication concern from execution, consensus, and settlement layers.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:data-availability", "labels": [ "Data Availability", "Data Availability Layer", "Data-Availability", "Ethereum Data Availability" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "data-breach-notification", "title": "Data Breach Notification", "domain": "governance", "domain_name": "Governance", "definition": "Data breach notification is the legal and operational obligation to inform supervisory authorities and, where there is a high risk, affected individuals after a breach of personal data. Frameworks such as the UK and EU GDPR require controllers to report qualifying breaches to the regulator without undue delay, typically within seventy-two hours of becoming aware, and to document the breach regardless of whether it is reportable. The duty turns an internal security incident into a regulated disclosure event, with timing, content and assessment of risk all prescribed by law.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-breach-notification", "labels": [ "Data Breach Notification" ], "is_subclass_of": [ "Data Protection Law" ], "wikilinks": [] }, { "id": "data-breach", "title": "Data Breach", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Data Breach is a security incident in which sensitive, protected, or confidential data is accessed, disclosed, copied, transmitted, or destroyed by an unauthorised actor, whether through external attack, insider threat, or accidental exposure. In AI contexts, data breaches can compromise training datasets, model weights, inference outputs, or user interaction logs, triggering GDPR notification obligations (within 72 hours to supervisory authorities), regulatory penalties, and reputational damage. AI systems increase both the attack surface (by aggregating and processing large personal-data collections) and the potential for novel breach vectors such as model-inversion attacks and membership-inference attacks that reconstruct or identify individuals from model outputs without direct database access.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:data-breach", "labels": [ "Data Breach" ], "is_subclass_of": [ "AI Governance and Ethics", "Risk" ], "wikilinks": [ "Artificial Intelligence", "Risk", "Cybersecurity", "GDPR", "Privacy", "Membership Inference", "Model Inversion", "Data Protection", "Access Control", "Audit Trail", "Regulatory Compliance", "Risk Management", "Risk Assessment", "Machine Learning", "Threat Intelligence", "Incident Response", "Encryption", "Differential Privacy", "Adversarial Attack", "Ransomware" ] }, { "id": "data-brokers", "title": "Data Brokers", "domain": "governance", "domain_name": "Governance", "definition": "Data brokers are firms that collect, aggregate, enrich and resell personal and behavioural data about individuals, typically without a direct relationship with those individuals. They assemble profiles from public records, online tracking and commercial sources for marketing, scoring and risk uses. They are a focal point of privacy regulation and surveillance concern because their trade can enable pervasive monitoring and profiling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-brokers", "labels": [ "Data Brokers", "Data Broker" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-catalog", "title": "Data Catalog", "domain": "data", "domain_name": "Data", "definition": "A data catalog is a centralised, searchable inventory of an organisation's data assets enriched with metadata, descriptions, ownership and usage context. It enables discovery, governance and self-service analytics by indexing datasets, schemas and their relationships, often integrating glossaries and lineage. As a component of metadata management and data fabric architectures it is essential for finding and trusting data at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-catalog", "labels": [ "Data Catalog", "AWS Glue Data Catalog" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-catalogue", "title": "Data Catalogue", "domain": "data", "domain_name": "Data", "definition": "A data catalogue is a centralised, searchable inventory of an organisation's data assets enriched with metadata, descriptions, lineage and usage information. It helps users discover, understand, trust and govern data by linking technical metadata with business context and policies. Modern catalogues often automate metadata harvesting and use collaboration features to capture institutional knowledge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-catalogue", "labels": [ "Data Catalogue" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-center-infrastructure", "title": "Data Center Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical and digital systems, including power, cooling, and networking, required to house and operate large-scale computing and AI workloads.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-center-infrastructure", "labels": [ "Data Center Infrastructure" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "data-center-moratorium", "title": "Data Center Moratorium", "domain": "governance", "domain_name": "Governance", "definition": "Local or state legislative pauses on new data center permits, driven by concerns over energy consumption, noise, and community impact.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-center-moratorium", "labels": [ "Data Center Moratorium" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-center-power-demand", "title": "Data Center Power Demand", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The increasing energy consumption of data centers driven by AI workloads, impacting grid capacity and energy policy.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-center-power-demand", "labels": [ "Data Center Power Demand" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "data-center-sustainability", "title": "Data Center Sustainability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The practice of designing and operating data centers to minimize environmental impact through efficient energy use, water management, and carbon reduction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-center-sustainability", "labels": [ "Data Center Sustainability" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "data-centers", "title": "Data Centers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Facilities that house the critical components of an organization's information technology infrastructure, including servers, storage, and networking equipment, essential for large-scale AI computation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-centers", "labels": [ "Data Centers" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "data-centre", "title": "Data Centre", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A data centre is a dedicated facility that houses computing, storage and networking infrastructure together with the power, cooling and physical security needed to operate them reliably. It provides the consolidated environment for hosting servers, cloud services and high-performance workloads such as AI training. Its energy and power demands make efficiency, measured through metrics like PUE, a central operational and sustainability concern.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:data-centre", "labels": [ "Data Centre", "AI Data Centre", "Data Center", "Data Centre Infrastructure", "Data Centre Operations", "Data Centres" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "data-classification", "title": "Data Classification", "domain": "governance", "domain_name": "Governance", "definition": "Data classification is the process of organising data into categories based on its sensitivity, value and regulatory obligations, so that appropriate handling, protection and access controls can be applied consistently. Typical schemes assign labels such as public, internal, confidential and restricted, which then drive encryption, retention, sharing and disposal rules. It is a foundational activity in data governance and information security, enabling organisations to prioritise protection of their most sensitive information and to demonstrate regulatory compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-classification", "labels": [ "Data Classification" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-cleaning", "title": "Data Cleaning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data Cleaning is a artificial intelligence concept and a type of Data Preprocessing. that enables Data Analysis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:data-cleaning", "labels": [ "Data Cleaning" ], "is_subclass_of": [ "AI Technique", "Data Preprocessing" ], "wikilinks": [ "Data Analysis", "Data Preprocessing", "ISO 8601" ] }, { "id": "data-collection", "title": "Data Collection", "domain": "data", "domain_name": "Data", "definition": "Data collection is the systematic process of gathering raw observations, measurements, or records from primary sources \u2014 including sensors, user interactions, instruments, surveys, and web scraping \u2014 in a form suitable for storage, processing, and analysis. As a foundational stage of the data lifecycle, it determines the completeness, representativeness, and quality of all downstream analytical products. In machine learning contexts, data collection encompasses sourcing, labelling, and curating training datasets that govern model capability and bias characteristics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:data-collection", "labels": [ "Data Collection", "Data Collection Pipeline", "Data Collection System", "Failure Data Collection", "Mass Data Collection", "Real-World Data Collection", "Supplier Data Collection", "Survey-Based Data Collection", "User Data Collection" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-completeness", "title": "Data Completeness", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-completeness", "labels": [ "Data Completeness" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-compression", "title": "Data Compression", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data Compression in AI encompasses techniques for reducing the size of datasets, models, and computational representations while preserving essential information and predictive performance. Key methods include model quantisation (reducing numerical precision of weights), pruning (removing redundant parameters), knowledge distillation (training compact student models from large teachers), and neural compression via autoencoders. These techniques are critical for deploying AI on resource-constrained edge devices and for reducing storage, bandwidth, and energy costs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-compression", "labels": [ "Data Compression", "Compression", "Compression Algorithm", "Meshopt Compression" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Edge AI", "Model Compression", "Quantization", "Knowledge Distillation" ] }, { "id": "data-confidentiality", "title": "Data Confidentiality", "domain": "security", "domain_name": "Security", "definition": "The property that data is accessible only to those authorised to view it, protecting information from disclosure to unauthorised parties. It is one of the core objectives of information security.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:data-confidentiality", "labels": [ "Data Confidentiality", "Confidentiality" ], "is_subclass_of": [ "Information Security" ], "wikilinks": [ "Cryptography", "Privacy", "Information Security" ] }, { "id": "data-consistency", "title": "Data Consistency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data consistency is the guarantee that data remains valid, coherent and in agreement across copies, transactions and nodes of a system. In distributed systems it spans a spectrum from strong consistency, where all readers observe the latest write, to eventual consistency, where replicas converge over time. It is a core correctness property traded off against availability and latency, and it underpins fault-tolerant data layers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-consistency", "labels": [ "Data Consistency", "Enterprise Data Consistency" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "data-contracts", "title": "Data Contracts", "domain": "data", "domain_name": "Data", "definition": "A data contract is a formal, versioned agreement between data producers and consumers that specifies the schema, semantics, quality guarantees and service-level expectations of a dataset or stream. It makes data interfaces explicit and enforceable, enabling automated validation and breaking-change detection in data pipelines. Data contracts are a metadata-management practice that improves reliability and trust across decentralised data ownership.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-contracts", "labels": [ "Data Contracts", "Data Contract" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-curation", "title": "Data Curation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data curation is the process of collecting, filtering, cleaning, annotating and organising raw data into a coherent, high-quality dataset suitable for training or evaluating machine learning models. It includes deduplication, removal of low-quality or harmful content, balancing of class or source distributions, and documentation of provenance and licensing. Rigorous data curation has a substantial effect on downstream model quality and is increasingly recognised as being as important as model architecture in large-scale AI systems.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:data-curation", "labels": [ "Data Curation" ], "is_subclass_of": [ "Data Preprocessing" ], "wikilinks": [] }, { "id": "data-deduplication", "title": "Data Deduplication", "domain": "data", "domain_name": "Data", "definition": "Data deduplication is the process of detecting and eliminating redundant copies of data so that only unique instances are retained or referenced. In storage it reduces capacity and bandwidth needs through chunk- or block-level matching, while in data preparation it removes duplicate records to improve quality. It is foundational for clean training data and for identity resolution where records must be matched and merged.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-deduplication", "labels": [ "Data Deduplication" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-dictionary", "title": "Data Dictionary", "domain": "data", "domain_name": "Data", "definition": "A data dictionary is a centralised, structured repository that describes the data elements of an information system, recording each element's name, definition, data type, format, allowable values, relationships and ownership. It serves as an authoritative reference for the meaning and structure of data, distinct from the data itself, enabling consistent understanding across teams and systems. Data dictionaries underpin data governance, integration and quality assurance by making the semantics of stored data explicit.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-dictionary", "labels": [ "Data Dictionary" ], "is_subclass_of": [ "Metadata Management" ], "wikilinks": [] }, { "id": "data-discovery", "title": "data discovery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data Discovery is the iterative process of locating, profiling, cataloguing, and contextualising data assets distributed across an organisation's storage systems, databases, data lakes, SaaS applications, and streaming pipelines to make them findable, understandable, and trustworthy for analytics, governance, and compliance purposes. It encompasses automated metadata extraction, schema inference, data profiling (statistical characterisation), lineage tracing (upstream/downstream dependencies), and classification (sensitivity tagging). Data Discovery is foundational to implementing Data Governance frameworks and enabling self-service analytics in data mesh and data fabric architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-discovery", "labels": [ "Data Discovery", "Spatial Data Discovery" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-distribution-service", "title": "Data Distribution Service", "domain": "robotics", "domain_name": "Robotics", "definition": "The Data Distribution Service is a middleware standard for real-time, data-centric publish-subscribe communication between distributed system components. It defines a global data space in which publishers and subscribers exchange typed data samples without direct knowledge of one another, governed by configurable quality-of-service policies for reliability, latency, and durability. It is widely used in robotics, autonomous vehicles, and other systems requiring deterministic machine-to-machine messaging.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-distribution-service", "labels": [ "Data Distribution Service" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [] }, { "id": "data-drift", "title": "Data Drift", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data drift is the change over time in the statistical distribution of the input data fed to a deployed machine learning model relative to the distribution it was trained on. Unlike concept drift, which alters the relationship between inputs and targets, data drift (also called covariate or feature drift) shifts the marginal distribution of the features themselves and can silently degrade model accuracy even when the learned mapping remains valid. Detecting and responding to data drift is a central concern of model monitoring and MLOps, typically driving alerts, scheduled retraining, or fallback policies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-drift", "labels": [ "Data Drift" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [] }, { "id": "data-efficient-learning", "title": "Data Efficient Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data Efficient Learning encompasses machine learning techniques that achieve strong generalisation from limited labelled training examples, including transfer learning, few-shot learning, semi-supervised learning, data augmentation, and self-supervised pretraining. These methods address real-world constraints where large annotated datasets are impractical to obtain, making AI deployable in specialised or resource-constrained domains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:data-efficient-learning", "labels": [ "Data Efficient Learning", "Data-Efficient Learning", "Sample Efficient Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "owl:Thing" ] }, { "id": "data-engineering", "title": "Data Engineering", "domain": "data", "domain_name": "Data", "definition": "Data Engineering is the discipline concerned with designing, building and operating the systems that collect, store, transform and serve data at scale. It covers data pipelines, storage architectures, batch and streaming processing, data modelling and the orchestration and monitoring of workflows. Its purpose is to make reliable, well-structured data available for analytics, reporting and machine learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-engineering", "labels": [ "Data Engineering" ], "is_subclass_of": [ "Data Management", "Infrastructure Domain" ], "wikilinks": [ "Data Pipeline", "ETL", "Data Warehouse", "Distributed Systems Domain", "Machine Learning Domain", "Business Intelligence", "Data Science", "Stream Processing", "Infrastructure Domain" ] }, { "id": "data-ensemble", "title": "Data Ensemble", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-ensemble", "labels": [ "Data Ensemble" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-exchange-format", "title": "Data Exchange Format", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A data exchange format is an agreed structure and encoding for representing data so that it can be reliably transmitted, stored, and interpreted between independent systems. It defines syntax, data types, and often a schema, allowing a producer and a consumer that share no internal code to exchange information without loss of meaning. Common examples include JSON, XML, CSV, and binary formats such as Protocol Buffers, each trading off readability, compactness, and parsing speed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:data-exchange-format", "labels": [ "Data Exchange Format" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "data-exchange", "title": "Data Exchange", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data Exchange refers to the structured transfer of data between systems, organisations, or parties using agreed-upon formats, protocols, and governance frameworks, enabling interoperability without requiring identical internal architectures. It encompasses both technical standards (APIs, file formats, serialisation protocols) and organisational arrangements (data sharing agreements, trust frameworks, data marketplaces) that govern how data flows across boundaries. Modern data exchange platforms provide cataloguing, consent management, lineage tracking, and value exchange mechanisms alongside raw data transfer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-exchange", "labels": [ "Data Exchange" ], "is_subclass_of": [ "Infrastructure", "Data Interoperability" ], "wikilinks": [] }, { "id": "data-fabric-architecture", "title": "Data Fabric Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An integrated data-management architecture that provides unified access, governance, security, and orchestration across distributed and heterogeneous data sources through active metadata management, automated data integration, and policy-driven controls.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:data-fabric-architecture", "labels": [ "Data Fabric Architecture" ], "is_subclass_of": [ "Data Management", "Data Management" ], "wikilinks": [ "Access Control Layer", "Cross-Domain Governance", "Data Catalog", "Data Integration Service", "Data Lineage Tracking", "Data Schema", "Distributed Storage", "FAIR DO", "Federated Queries", "Gartner Data Fabric Research", "Self-Service Analytics", "Semantic Layer", "Unified Data Access", "W3C Data Fabric BP", "API Gateway", "Blockchain", "Computation And Intelligence Domain", "Data Layer", "Data Management", "Data Virtualization" ] }, { "id": "data-fabric", "title": "Data Fabric", "domain": "data", "domain_name": "Data", "definition": "A data fabric is an architectural approach that provides a unified, metadata-driven layer for accessing, integrating and governing data across heterogeneous and distributed sources. It uses active metadata, knowledge graphs and automation to connect data without forcing physical consolidation. By abstracting underlying systems it delivers consistent discovery, access and governance, reducing the friction of fragmented data estates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-fabric", "labels": [ "Data Fabric" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-federation", "title": "Data Federation", "domain": "data", "domain_name": "Data", "definition": "Data federation is a data integration technique that presents multiple distributed and heterogeneous data sources as a single virtual database queryable in place, without physically moving or copying the data. A federation engine decomposes queries, pushes work to source systems and combines results on the fly. It enables real-time unified access and is a core mechanism behind data virtualization and integration interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-federation", "labels": [ "Data Federation" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-flow-diagram", "title": "Data Flow Diagram", "domain": "security", "domain_name": "Security", "definition": "A data flow diagram (DFD) is a visual representation of how data moves through a system, showing processes, data stores, external entities, and the trust boundaries between them. In security practice it is the standard input artefact for threat modelling, since attack surfaces and applicable threats are identified by walking each flow and boundary crossing in the diagram. It is a structural tool rather than a behavioural one, deliberately omitting control flow and timing to keep the focus on where data moves and where it is trusted less.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:data-flow-diagram", "labels": [ "Data Flow Diagram" ], "is_subclass_of": [ "Threat Modelling" ], "wikilinks": [] }, { "id": "data-format-standard", "title": "Data Format Standard", "domain": "standards", "domain_name": "Standards", "definition": "Technical specifications defining how data is structured, encoded, and exchanged across metaverse platforms and 3D applications, including standards like glTF for efficient 3D asset transmission and Universal Scene Description (USD) for complex scene composition and collaboration.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:data-format-standard", "labels": [ "Data Format Standard", "Data Format Specification" ], "is_subclass_of": [ "Technical Standards" ], "wikilinks": [ "3D Asset Interoperability", "Content Portability", "Cross-Platform Exchange", "glTF", "Implementation Libraries", "Khronos Group", "MaterialX", "Metaverse Standards Forum", "USD", "Validation Tools", "Computer Vision", "metaverse", "Standards Body", "Technical Standards" ] }, { "id": "data-format", "title": "Data Format", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Data Format is a formal structural and encoding specification that governs how raw bytes are organised, typed, and interpreted when data is stored, transmitted, or exchanged between computational systems. Formats encode decisions about byte ordering, field delimitation, schema evolution, compression, and type systems, ranging from binary serialisation protocols (Protocol Buffers, Apache Avro, MessagePack) to human-readable interchange formats (JSON, XML, YAML, TOML) and domain-specific schemas (DICOM for medical imaging, glTF for 3D assets, FITS for astronomical data). The choice of data format critically determines interoperability scope, parsing overhead, storage efficiency, schema evolution capability, and compatibility with downstream processing pipelines and analytics engines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-format", "labels": [ "Data Format" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "data-governance-framework", "title": "Data Governance Framework", "domain": "governance", "domain_name": "Governance", "definition": "A Data Governance Framework is a structured set of policies, processes, roles, and standards that define how an organisation acquires, manages, protects, and disposes of data assets throughout their lifecycle. It establishes accountability structures, data quality standards, metadata management practices, and compliance mechanisms aligned to regulatory obligations such as GDPR. Effective frameworks balance centralised oversight with federated data ownership through data mesh and stewardship models. They underpin trustworthy AI deployments by ensuring training data provenance and ongoing monitoring of data quality and lineage.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:data-governance-framework", "labels": [ "Data Governance Framework" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-governance", "title": "Data Governance", "domain": "governance", "domain_name": "Governance", "definition": "Data Governance is a system of policies, processes, roles, and standards that establish accountability and control over an organisation's data assets throughout their lifecycle. It defines who may take what actions with which data, under what circumstances, and using what methods, ensuring data quality, consistency, security, and regulatory compliance. Effective data governance bridges technical data management practices with organisational strategy, aligning data stewardship with business objectives and legal obligations. It is foundational to trustworthy analytics, AI model training, and interoperability across enterprise systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:data-governance", "labels": [ "Data Governance", "DataGovernance" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-infrastructure", "title": "Data Infrastructure", "domain": "data", "domain_name": "Data", "definition": "Data infrastructure is the integrated set of systems, platforms, and pipelines that ingest, store, process, govern, and serve data across an organisation. It encompasses storage layers such as data warehouses and data lakes, processing and integration pipelines, cataloguing and governance, and the compute fabric that supports analytics and machine learning. Data infrastructure provides the reliable, scalable foundation on which data products and decision-making depend.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-infrastructure", "labels": [ "Data Infrastructure" ], "is_subclass_of": [ "Infrastructure", "Data Management (Infrastructure)" ], "wikilinks": [] }, { "id": "data-ingestion", "title": "Data Ingestion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Data ingestion is the process of acquiring data from heterogeneous sources and moving it into a target store or processing system for downstream use. It covers batch and streaming acquisition, format normalisation, validation and routing, and forms the entry stage of data pipelines. In spatial-computing contexts ingestion handles sensor streams and captured geometry before reconstruction and analysis.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-ingestion", "labels": [ "Data Ingestion", "Data Ingestion Pipeline" ], "is_subclass_of": [ "Data Pipeline" ], "wikilinks": [] }, { "id": "data-integration-interface", "title": "Data Integration Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A standardized set of protocols, rules, and formats for unifying and mediating data flows across heterogeneous platforms, enabling seamless data exchange and interoperability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:data-integration-interface", "labels": [ "Data Integration Interface" ], "is_subclass_of": [ "Data Management", "Data Management" ], "wikilinks": [ "Cross-Platform Data Exchange", "Data Adapter", "Data Federation", "DataManagementDomain", "Data Schema", "ETSI GR ARF 010", "InteroperabilityDomain", "ISO 23247", "Message Broker", "Protocol Translator", "Real-Time Synchronization", "Schema Mapper", "Service Discovery", "API Gateway", "Blockchain", "Communication Protocol", "Data Governance", "Data Layer", "Data Management", "Interoperability Framework" ] }, { "id": "data-integration", "title": "Data Integration", "domain": "data", "domain_name": "Data", "definition": "Data Integration is the set of processes, architectures, and tools that combine data from multiple heterogeneous sources \u2014 including relational databases, APIs, event streams, file systems, and third-party services \u2014 into a unified, consistent, and queryable representation suitable for analytics, machine learning, or operational workloads. It encompasses extract-transform-load (ETL) and extract-load-transform (ELT) pipelines, schema harmonisation, semantic mapping, identity resolution, and real-time federation patterns. Modern data integration extends beyond batch movement to include change-data capture (CDC), streaming integration via message brokers, and virtual federation through query engines, enabling organisations to maintain a single source of truth across distributed data estates. In AI and spatial-computing contexts, data integration connects sensor telemetry, digital-twin feeds, user-behaviour streams, and knowledge-graph stores into coherent data products that power downstream inference and immersive experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-integration", "labels": [ "Data Integration", "Data Integration Service", "Health Data Integration" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-integrity-verification", "title": "Data Integrity Verification", "domain": "security", "domain_name": "Security", "definition": "Data integrity verification is the process of confirming that a piece of data has not been altered, corrupted, or tampered with since it was created or last authorised, typically by comparing a cryptographic digest computed over the data against a previously recorded or independently trusted value. Techniques include cryptographic hash functions, checksums, and Merkle tree proofs, which allow large datasets to be verified efficiently by checking a small root value rather than the full contents. It underpins trust in distributed systems, blockchains, and any pipeline where data passes through untrusted intermediaries.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:data-integrity-verification", "labels": [ "Data Integrity Verification" ], "is_subclass_of": [ "Data Integrity" ], "wikilinks": [] }, { "id": "data-integrity", "title": "data integrity", "domain": "data", "domain_name": "Data", "definition": "Data integrity is the property guaranteeing that data remains accurate, complete, consistent, and unaltered throughout its entire lifecycle\u2014spanning creation, storage, transmission, and processing\u2014except through authorised operations. It is enforced through a layered combination of technical controls including cryptographic hash functions, digital signatures, Merkle trees, access control mechanisms, and ACID-compliant transactions, as well as procedural controls such as change-control workflows, audit trails, and immutable logging. Violations\u2014whether from storage errors, transmission corruption, software bugs, or deliberate tampering\u2014can propagate silently and have cascading consequences in any system relying on the data for decision-making or compliance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:data-integrity", "labels": [ "Data Integrity", "DataIntegrity" ], "is_subclass_of": [ "Data Quality" ], "wikilinks": [] }, { "id": "data-interoperability", "title": "Data Interoperability", "domain": "data", "domain_name": "Data", "definition": "Data Interoperability is the capability of heterogeneous systems, applications, and data sources to exchange, interpret, and act upon shared data without loss of meaning across organisational and technical boundaries. It encompasses syntactic interoperability (shared formats and wire protocols), semantic interoperability (common vocabularies, ontologies, and data models), and pragmatic interoperability (agreed processes, policies, and trust frameworks). Achieving full data interoperability requires alignment across API contracts, schema registries, identity frameworks, and data governance regimes. It is a foundational property enabling federated analytics, cross-domain knowledge graphs, and open ecosystem integration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-interoperability", "labels": [ "Data Interoperability" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-inventory", "title": "Data Inventory", "domain": "governance", "domain_name": "Governance", "definition": "A data inventory is a structured, maintained record of the categories, locations, and flows of data - particularly personal data - that an organisation collects, processes, and stores. It underpins compliance with data protection regimes such as CCPA and GDPR by allowing an organisation to demonstrate what data it holds, why it holds it, and where it resides when responding to access, deletion, or portability requests. Maintaining an accurate data inventory is a prerequisite for exercising rights such as the right to be forgotten.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:data-inventory", "labels": [ "Data Inventory" ], "is_subclass_of": [ "Data Catalog" ], "wikilinks": [] }, { "id": "data-labelling", "title": "Data Labelling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Data labelling is the process of annotating raw data, such as images, text, audio or sensor readings, with the target outputs or categories a supervised model is expected to predict. It produces the ground-truth signal that links inputs to desired outputs and largely determines the achievable accuracy of trained models. Labelling combines human annotators, guidelines, tooling and quality control, increasingly augmented by model-assisted and active-learning workflows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-labelling", "labels": [ "Data Labelling" ], "is_subclass_of": [ "Data Annotation" ], "wikilinks": [] }, { "id": "data-lake", "title": "Data Lake", "domain": "data", "domain_name": "Data", "definition": "A centralized repository that stores structured, semi-structured, and unstructured data at any scale in its native format, deferring schema enforcement to query time (schema-on-read). Data lakes enable big-data analytics, machine learning pipelines, and exploratory analysis without upfront data modelling, serving as the foundational ingestion layer for modern data architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:data-lake", "labels": [ "Data Lake", "Data Lakes" ], "is_subclass_of": [ "Data Storage", "Artificial Intelligence" ], "wikilinks": [ "Big Data", "Data Engineering", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "data-layer", "title": "Data Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data Layer is the architectural tier within a layered, n-tier, hexagonal, or modular system responsible for the durable persistence, transactional integrity, indexed retrieval, replication, and abstraction of stateful data \u2014 encompassing both the enterprise-architecture sense (the data access lay...", "entityType": "Class", "qualityScore": 0.53, "maturity": "established", "iri": "urn:ngm:class:data-layer", "labels": [ "Data Layer", "Blockchain Data Layer", "DataLayer" ], "is_subclass_of": [ "Data Management", "Network and Communication", "Architectural Layer", "System Tier", "Software Layer" ], "wikilinks": [ "ACID Properties", "Active Record", "ANSI SQL", "Architectural Layer", "Avail", "Backup Strategy", "Bloom Filter", "Business Logic Layer", "Cache Layer", "Celestia", "Celestia Specifications", "CIDR Architecture", "Clean Architecture", "Connection Pool", "Consensus Layer", "CQRS", "Cypher", "Data Availability", "Data Availability Sampling", "Data Consistency" ] }, { "id": "data-lifecycle", "title": "Data Lifecycle", "domain": "data", "domain_name": "Data", "definition": "The sequence of stages that data passes through from initial creation or capture to eventual archival or destruction, typically comprising creation, storage, use, sharing, archiving, and disposal, with each stage carrying distinct security, quality, and regulatory obligations that organisations manage through lifecycle policies, retention schedules, and classification-driven controls.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:data-lifecycle", "labels": [ "Data Lifecycle" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Data Management", "Data Governance", "Data Classification" ] }, { "id": "data-lineage-tracking", "title": "Data Lineage Tracking", "domain": "data", "domain_name": "Data", "definition": "Data lineage tracking is the automated capture and maintenance of provenance metadata as data is created, transformed and moved across pipelines and platforms. It instruments transformations and queries to build a continuously updated lineage graph rather than a static map. Often expressed using provenance vocabularies such as PROV-O, it underpins reproducibility, trust and governance in data fabric architectures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-lineage-tracking", "labels": [ "Data Lineage Tracking", "Lineage Tracker", "Model Lineage Tracking" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-lineage", "title": "Data Lineage", "domain": "data", "domain_name": "Data", "definition": "Data lineage is the documented record of data's origins, movements, transformations and consumption as it flows through systems and pipelines. It maps how a data element is derived end to end, supporting impact analysis, debugging, audit and regulatory compliance. As a pillar of data governance and metadata management it makes data trustworthy by exposing its provenance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-lineage", "labels": [ "Data Lineage", "AI Training Data Lineage", "Lineage Metadata" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-link-layer", "title": "Data Link Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The data link layer is the second layer of the OSI model, responsible for node-to-node data transfer across a single physical link and for framing raw bits from the physical layer into structured frames. It provides addressing through hardware (MAC) addresses, error detection, and media access control that arbitrates shared transmission media. By presenting a reliable link to the network layer above, it abstracts away the imperfections of the underlying physical medium.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:data-link-layer", "labels": [ "Data Link Layer" ], "is_subclass_of": [ "OSI Model" ], "wikilinks": [] }, { "id": "data-localisation", "title": "Data Localisation", "domain": "governance", "domain_name": "Governance", "definition": "Data localisation is the legal or regulatory requirement that certain categories of data be stored, processed, or retained within the geographic borders of a particular country or jurisdiction. Governments impose such rules to assert data sovereignty, protect citizens' personal information, support law-enforcement access, or shield strategic sectors, and the requirements range from mandating in-country copies to outright bans on cross-border transfer. Compliance forces organisations to architect regional data residency, partition storage, and reconcile conflicting national regimes, materially shaping cloud architecture and international data flows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-localisation", "labels": [ "Data Localisation" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-loss-prevention", "title": "Data Loss Prevention", "domain": "security", "domain_name": "Security", "definition": "Data loss prevention (DLP) is a set of security controls and technologies that detect and block the unauthorised exfiltration, leakage or misuse of sensitive data. DLP systems classify content, monitor data in use, in motion and at rest, and enforce policies at endpoints, networks and cloud services. It is a core data-protection capability supporting compliance with privacy and confidentiality requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-loss-prevention", "labels": [ "Data Loss Prevention" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "data-management-platform", "title": "Data Management Platform", "domain": "data", "domain_name": "Data", "definition": "A data management platform is a centralised system for collecting, organising, segmenting, and activating large volumes of data drawn from multiple sources so that it can be used for analytics, audience targeting, and decision-making. Historically associated with advertising, where it aggregated and anonymised audience data for campaign targeting, the term also denotes broader enterprise platforms that unify data pipelines, storage, and governance. It is closely related to, but distinct from, customer data platforms and data warehouses.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-management-platform", "labels": [ "Data Management Platform" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "data-management-system", "title": "Data Management System", "domain": "data", "domain_name": "Data", "definition": "A data management system is software that stores, organises, secures and provides controlled access to data throughout its lifecycle. It encompasses capabilities for ingestion, modelling, querying, integrity enforcement and governance, with database management systems and graph databases as common specialisations. It is the foundational platform on which metadata standards and downstream applications depend.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-management-system", "labels": [ "Data Management System" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-management", "title": "Data Management", "domain": "data", "domain_name": "Data", "definition": "Data Management is the disciplined set of practices, architectures, and technologies for acquiring, validating, storing, organising, protecting, and retiring data across its full lifecycle. It encompasses data modelling, quality assurance, metadata governance, access control, and lineage tracking to ensure data assets remain accurate, discoverable, and trustworthy. In AI and distributed-systems contexts it additionally covers ingestion pipelines, versioned feature stores, reproducible training datasets, and consent frameworks. Effective data management underpins regulatory compliance, interoperability, and the reliability of analytical and machine-learning workloads.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-management", "labels": [ "Data Management", "Data Management (Infrastructure)", "Health Data Management" ], "is_subclass_of": [], "wikilinks": [ "Blockchain", "Metaverse Technology" ] }, { "id": "data-mapping", "title": "Data Mapping", "domain": "governance", "domain_name": "Governance", "definition": "Data mapping is the process of defining correspondences between fields, elements, or schemas of a source data structure and those of a target, specifying how each attribute is matched, transformed, or combined during movement between systems. It is foundational to data integration, migration, and interoperability, guiding extract-transform-load pipelines and message translation so that semantically equivalent information aligns across heterogeneous models. In privacy and governance contexts, data mapping also denotes the cataloguing of where personal data resides and how it flows, which is a prerequisite for lineage tracking and regulatory compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-mapping", "labels": [ "Data Mapping" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [] }, { "id": "data-marketplace", "title": "Data Marketplace", "domain": "data", "domain_name": "Data", "definition": "A data marketplace is a platform on which organisations and individuals discover, buy, sell or exchange datasets and data access rights under defined licensing and pricing terms. It provides discovery, valuation, provenance and settlement mechanisms so that data can be treated as a tradeable asset rather than a siloed internal resource. Data marketplaces range from centralised commercial exchanges to decentralised, blockchain-mediated platforms that use smart contracts for access control and payment.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:data-marketplace", "labels": [ "Data Marketplace" ], "is_subclass_of": [ "Marketplace" ], "wikilinks": [] }, { "id": "data-masking", "title": "Data Masking", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-masking", "labels": [ "Data Masking" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-mesh", "title": "Data Mesh", "domain": "data", "domain_name": "Data", "definition": "Data Mesh is a socio-technical paradigm for large-scale analytical data architecture, coined by Zhamak Dehghani (2020), that decentralises data ownership to domain-aligned teams who treat their analytical datasets as first-class products accessible via standardised interfaces. It rests on four principles: domain-oriented decentralised data ownership, data as a product, self-serve data infrastructure as a platform, and federated computational governance. Unlike monolithic data lakes or warehouses governed by a central data engineering team, a Data Mesh distributes accountability so that the teams who generate data also maintain its quality, documentation, and SLAs as discoverable data products. The model draws on domain-driven design and microservices thinking applied to analytical rather than operational data flows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:data-mesh", "labels": [ "Data Mesh" ], "is_subclass_of": [ "Data Governance Framework" ], "wikilinks": [] }, { "id": "data-minimisation", "title": "Data Minimisation", "domain": "data", "domain_name": "Data", "definition": "Data Minimisation is a privacy principle and GDPR requirement (Article 5(1)(c)) mandating that personal data collection and processing be limited to what is adequate, relevant, and necessary for specified purposes, reducing privacy risks by avoiding accumulation of excessive data that could be misused, breached, or enable function creep.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-minimisation", "labels": [ "Data Minimisation", "GDPR Data Minimisation" ], "is_subclass_of": [ "Privacy and Data Governance" ], "wikilinks": [ "GDPR Article 25", "GDPR Article 5(1)(c)", "ISO 29100", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "data-mining", "title": "Data Mining", "domain": "data", "domain_name": "Data", "definition": "Data mining is the computational process of discovering non-obvious patterns, correlations, anomalies and predictive structure in large datasets, sitting at the core of the knowledge discovery in databases (KDD) pipeline. It combines techniques from machine learning, statistics and database systems \u2014 including clustering, classification, association-rule learning, regression and anomaly detection \u2014 to turn raw operational data into actionable models and insights across science, commerce and government.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:data-mining", "labels": [ "Data Mining" ], "is_subclass_of": [ "Knowledge Discovery" ], "wikilinks": [ "Knowledge Discovery", "Clustering", "Text Mining", "Big Data", "Machine Learning" ] }, { "id": "data-model", "title": "Data Model", "domain": "data", "domain_name": "Data", "definition": "A data model is an abstract specification of how data elements are structured, related and constrained within a domain or system. It defines entities, attributes, relationships and integrity rules at conceptual, logical or physical levels of abstraction. Data models are prerequisites for metadata standards and for designing storage, exchange and querying of data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-model", "labels": [ "Data Model" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-modelling", "title": "Data Modelling", "domain": "data", "domain_name": "Data", "definition": "Data modelling is the process of defining the structure, relationships, constraints, and semantics of data to be stored, processed, or exchanged in an information system. It produces formal artefacts \u2014 conceptual, logical, and physical models \u2014 that guide database design, API contracts, and data integration. Effective data modelling ensures data consistency, reduces redundancy, and aligns technical storage structures with the business domain concepts they represent.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:data-modelling", "labels": [ "Data Modelling" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-models", "title": "Data Models", "domain": "data", "domain_name": "Data", "definition": "Data models are the collected structural specifications that define how information is represented, related and constrained within a system or framework. In digital-twin and layered service architectures they describe the shape of state, telemetry and interactions exchanged between components. Well-defined data models are a prerequisite for interoperability, validation and consistent behaviour across service layers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-models", "labels": [ "Data Models" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-monetisation", "title": "Data Monetisation", "domain": "governance", "domain_name": "Governance", "definition": "Data monetisation is the practice of generating economic value from data assets, either directly by selling data and data-derived products, or indirectly by using data to improve decisions, products and operations. It spans approaches from licensing raw or aggregated datasets through data marketplaces to embedding analytics in services that command a premium. Responsible data monetisation must be balanced against data privacy, consent and governance obligations, and it raises questions about data ownership and the ethics of treating personal information as a tradeable resource.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:data-monetisation", "labels": [ "Data Monetisation" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-observability", "title": "Data Observability", "domain": "data", "domain_name": "Data", "definition": "Data observability is the practice of continuously monitoring the health and reliability of data and data pipelines to detect, diagnose and resolve issues before they affect consumers. It tracks pillars such as freshness, volume, schema, distribution and lineage, often using automated anomaly detection. As an extension of metadata management it brings software-style monitoring discipline to data systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-observability", "labels": [ "Data Observability" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-parallelism", "title": "Data Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Data parallelism is a parallel computing strategy in which the same operation is applied simultaneously to different partitions of a dataset distributed across multiple processing units. In machine learning it is the dominant approach to scaling model training: the model is replicated on every worker, each worker computes gradients on a distinct mini-batch shard, and gradients are aggregated through collective communication such as all-reduce before the synchronised parameter update. This contrasts with model parallelism, which splits a single model across devices, and underpins large-scale training on GPU and TPU clusters.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:data-parallelism", "labels": [ "Data Parallelism" ], "is_subclass_of": [ "Parallel Processing", "Distributed Training", "Distributed AI Training" ], "wikilinks": [] }, { "id": "data-partitioning", "title": "Data Partitioning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data partitioning is the practice of dividing a dataset into smaller, independently manageable subsets distributed across storage nodes or processing units to improve scalability, performance, and availability. Partitioning strategies include horizontal splitting by key range or hash, vertical splitting by column, and functional splitting by domain, each balancing query locality against load distribution. Effective partitioning underpins distributed databases and large-scale data systems by enabling parallel processing while minimising cross-partition coordination.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-partitioning", "labels": [ "Data Partitioning" ], "is_subclass_of": [ "Sharding" ], "wikilinks": [] }, { "id": "data-persistence", "title": "data persistence", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data Persistence is the property of data surviving beyond the lifetime of the process, session, or physical medium that created it, encompassing all mechanisms and design patterns \u2014 relational databases, distributed file systems, append-only event logs, object stores, and key-value stores \u2014 that guarantee this durability. In transactional systems, persistence is formalised through the ACID properties (Atomicity, Consistency, Isolation, Durability), where the Durability guarantee ensures committed transactions survive system failure by means of write-ahead logging and fsync operations. Beyond single-node databases, distributed systems trade aspects of consistency for availability and partition tolerance as described in the CAP theorem and the PACELC model, producing a spectrum of persistence strategies from strongly consistent relational systems to eventually consistent distributed NoSQL stores. The selection of a persistence strategy profoundly shapes a system's fault tolerance, latency, throughput, and operational complexity.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:data-persistence", "labels": [ "Data Persistence" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-pipeline", "title": "Data Pipeline", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An automated, end-to-end sequence of connected processing stages that orchestrates data ingestion, transformation, validation, and delivery, enforcing quality assurance and error handling at each stage to produce reliable, actionable analytical outputs for downstream consumers such as machine learning systems and business intelligence platforms.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:data-pipeline", "labels": [ "Data Pipeline", "DataPipeline" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "appliesTransformation", "AutomatedPipeline", "Availability", "Data Delivery", "Data Ingestion", "DataPreprocessing", "DataQuality", "Data Reliability", "Data Transformation", "DataTransformation", "Data Validation", "DistributedStorage", "dt:coordinatedBy", "dt:feeds", "dt:storedOn", "dt:trackedOn", "dt:validatedBy", "Error Handling", "ETL Patterns", "ingestsFrom" ] }, { "id": "data-poisoning", "title": "Data Poisoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A training-time adversarial attack where malicious actors inject, modify, or manipulate training data to compromise model integrity, causing targeted misclassifications, backdoor triggers, or general performance degradation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-poisoning", "labels": [ "Data Poisoning", "Training Data Poisoning" ], "is_subclass_of": [ "AI Technique", "Adversarial Attack" ], "wikilinks": [ "ball2020metaverse", "ISO/IEC 23894:2023", "MITRE ATLAS", "NCSC", "neural networks", "NIST AI Risk Management Framework", "Wouters2022", "computer vision", "Computer Vision", "Diffusion Models", "Fooocus", "Knowledge Graphing", "latent space", "MetaverseDomain", "Transformers", "Unreal Engine" ] }, { "id": "data-portability", "title": "Data Portability", "domain": "data", "domain_name": "Data", "definition": "Data portability is the ability for users to obtain, move and reuse their personal or application data across services in a structured, machine-readable format without lock-in. It is both a legal right under regimes such as GDPR and a technical property enabled by open formats and interoperable storage. In decentralised-web designs it lets individuals carry data between providers, shifting control from platforms to users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-portability", "labels": [ "Data Portability", "Financial Data Portability", "Health Data Portability", "Patient Data Portability" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-preprocessing", "title": "Data Preprocessing", "domain": "ai", "domain_name": "Ai", "definition": "Data preprocessing is the stage of a machine learning workflow that transforms raw data into a clean, consistent form suitable for modelling. It encompasses cleaning, normalisation, encoding, imputation and feature engineering to remove noise and align scales and types. The quality of preprocessing strongly determines downstream model accuracy and is a prerequisite for reliable training.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:data-preprocessing", "labels": [ "Data Preprocessing", "Corpus Preprocessing" ], "is_subclass_of": [ "Machine Learning Technique", "Machine Learning Pipeline", "Data Pipeline", "Data Science Workflow" ], "wikilinks": [] }, { "id": "data-privacy", "title": "Data Privacy", "domain": "governance", "domain_name": "Governance", "definition": "Data Privacy is the governance and engineering discipline that ensures individuals retain meaningful control over their personal information through legal frameworks, technical safeguards, and organisational policies governing the appropriate collection, processing, storage, and sharing of personal data. It spans both the regulatory compliance dimension\u2014expressed in instruments such as GDPR, CCPA, PIPEDA, and sector-specific regulations\u2014and the engineering discipline of privacy-by-design that minimises data exposure through techniques such as anonymisation, pseudonymisation, differential privacy, federated learning, and consent management. As AI training practices, surveillance capitalism, and cross-border data flows intensify the stakes of personal information handling, data privacy functions as a core organisational risk management and trust-building domain. The field requires integration across legal, technical, and organisational layers to be effective, and is increasingly operationalised through dedicated Privacy-Enhancing Technologies (PETs).", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:data-privacy", "labels": [ "Data Privacy", "Metadata Privacy", "Payment Data Privacy" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "data-processing-hardware", "title": "Data Processing Hardware", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The computing infrastructure including GPUs, CPUs, specialized accelerators, and edge computing devices that power metaverse applications, virtual reality experiences, and immersive environments, providing the massive processing capabilities required for real-time 3D rendering, AI inference, and ...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:data-processing-hardware", "labels": [ "Data Processing Hardware" ], "is_subclass_of": [ "Computing and Cloud", "Computing Infrastructure" ], "wikilinks": [ "AI Processing", "Cooling Systems", "High-Bandwidth Memory", "Immersive Computing", "Power Infrastructure", "Sensor Input", "Computing Infrastructure", "metaverse", "Real-Time Rendering" ] }, { "id": "data-processing-state", "title": "Data Processing State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-processing-state", "labels": [ "Data Processing State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-processing", "title": "Data Processing", "domain": "data", "domain_name": "Data", "definition": "Data Processing is the systematic application of operations \u2014 collection, validation, transformation, aggregation, enrichment, and storage \u2014 that convert raw, unstructured, or heterogeneous input data into organised, queryable, and semantically coherent representations suitable for analysis, machine learning, or real-time decision-making. It encompasses both batch and streaming paradigms, spanning ETL/ELT pipelines, in-memory computation frameworks, and edge preprocessing workflows that must balance throughput, latency, fault-tolerance, and data quality. Architecturally, data processing sits between raw data ingestion and higher-order analytical or AI workloads, acting as the foundational substrate for knowledge extraction, model training, and operational intelligence. Mature implementations employ declarative query languages, distributed execution engines, schema registries, and lineage tracking to ensure reproducibility and governance across the full data lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-processing", "labels": [ "Data Processing", "Local Data Processing", "Raw Data Processing" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "data-product-version", "title": "Data Product Version", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-product-version", "labels": [ "Data Product Version" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-profile", "title": "Data Profile", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:data-profile", "labels": [ "Data Profile" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "data-protection-impact-assessment", "title": "Data Protection Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A Data Protection Impact Assessment is a structured process for identifying, evaluating and mitigating the risks that a planned data-processing activity poses to the rights and freedoms of individuals. It documents the nature, scope, context and purposes of processing, assesses necessity and proportionality, and records measures that reduce identified risks. Under the General Data Protection Regulation it is mandatory where processing is likely to result in high risk, such as large-scale profiling or use of sensitive data. It is a core accountability instrument linking privacy-by-design to demonstrable compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-protection-impact-assessment", "labels": [ "Data Protection Impact Assessment" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "data-protection-law", "title": "Data Protection Law", "domain": "governance", "domain_name": "Governance", "definition": "The body of law governing how personal data may be collected, processed, stored and shared, and the rights afforded to individuals over their personal information.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-protection-law", "labels": [ "Data Protection Law" ], "is_subclass_of": [ "Data Protection" ], "wikilinks": [ "Data Protection", "Privacy", "Data Privacy", "GDPR", "CCPA" ] }, { "id": "data-protection-officer", "title": "Data Protection Officer", "domain": "governance", "domain_name": "Governance", "definition": "A Data Protection Officer (DPO) is an organisational role responsible for overseeing an entity's data-protection strategy and monitoring compliance with applicable privacy law such as the GDPR. The DPO advises on data-protection obligations, conducts and reviews privacy impact assessments, serves as the contact point for supervisory authorities and data subjects, and operates with independence from operational management. Designation of a DPO is mandatory under the GDPR for public authorities and for controllers or processors whose core activities involve large-scale or sensitive personal-data processing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-protection-officer", "labels": [ "Data Protection Officer" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-protection-regulation", "title": "Data Protection Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Data Protection Regulation comprises the body of legally binding instruments \u2014 including the EU General Data Protection Regulation (GDPR, 2018), the UK Data Protection Act 2018, the California Consumer Privacy Act (CCPA), and equivalent national or regional laws \u2014 that govern the collection, processing, storage, transfer, and erasure of personal data by organisations. These frameworks establish fundamental principles such as lawfulness, purpose limitation, data minimisation, and accountability; confer enforceable rights on data subjects (access, rectification, erasure, portability, and objection); and require proportionate technical and organisational safeguards, including impact assessments for high-risk processing. The regulation of personal data intersects directly with AI system design, automated decision-making constraints, consent management infrastructure, and cross-border data flows in globally distributed architectures.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:data-protection-regulation", "labels": [ "Data Protection Regulation", "General Data Protection Regulation" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-protection", "title": "Data Protection", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive set of processes and technologies that safeguard personal and system data in virtual environments through encryption, access control, privacy preservation, and regulatory compliance mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:data-protection", "labels": [ "Data Protection", "Biometric Data Protection", "Data Protection Framework", "Data Protection Obligations", "Data Protection Rights", "Data Protection by Default" ], "is_subclass_of": [ "Governance and Safety", "Security Architecture" ], "wikilinks": [ "Audit System", "Authentication", "Authorization", "Compliance Management", "Data Loss Prevention", "Data Privacy", "Data Sovereignty", "Encryption Service", "ETSI GR ARF 010", "GDPR", "GDPR Compliance", "ISO 27701", "Privacy Engineering", "Privacy Policy Engine", "Regulatory Requirements", "Security Policy", "User Trust", "Access Control System", "Application Layer", "Blockchain" ] }, { "id": "data-provenance", "title": "Data Provenance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A continuous process of recording and tracking the origin, lineage, and transformation history of data objects, enabling traceability, validation of data quality, and verification of authenticity throughout the data lifecycle.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:data-provenance", "labels": [ "Data Provenance", "IoT Data Provenance", "Training Data Provenance" ], "is_subclass_of": [ "Data Management", "Data Management" ], "wikilinks": [ "Attribution", "Compliance Audit", "Data Quality Assessment", "ETSI GR ARF 010", "Event Logging", "Lineage Tracker", "Metadata", "Provenance Recorder", "Reproducibility", "Signature Validator", "Timestamp Authority", "Timestamp Service", "W3C PROV-O", "Audit Trail", "Blockchain", "Data Governance", "Data Layer", "Data Management", "Digital Signature", "Identity Management" ] }, { "id": "data-quality-management", "title": "Data Quality Management", "domain": "data", "domain_name": "Data", "definition": "Data quality management is the discipline of measuring, monitoring, and improving the accuracy, completeness, consistency, and timeliness of data across its lifecycle. It combines profiling, validation, cleansing, and continuous monitoring with governance policies that define quality expectations. Reliable data quality is a precondition for trustworthy analytics, machine learning, and regulatory reporting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-quality-management", "labels": [ "Data Quality Management" ], "is_subclass_of": [ "Data Quality" ], "wikilinks": [] }, { "id": "data-quality", "title": "Data Quality", "domain": "data", "domain_name": "Data", "definition": "Data Quality is the degree to which a dataset is fit for its intended purpose, evaluated across multiple dimensions including accuracy, completeness, consistency, timeliness, uniqueness, and validity. It encompasses the processes, standards, and metrics used to assess, monitor, cleanse, and continuously improve the reliability and trustworthiness of data assets. High data quality is a foundational prerequisite for sound analytics, machine learning model performance, regulatory compliance, and effective decision-making across all data-driven disciplines. Its absence propagates errors through downstream systems, undermining the value of even sophisticated analytical and AI pipelines.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:data-quality", "labels": [ "Data Quality", "Data Quality Assessment", "Data Quality Assurance" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Data Governance", "Master Data Management", "Data Integration", "Data Management" ] }, { "id": "data-redundancy", "title": "Data Redundancy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data redundancy is the deliberate storage of duplicate or recoverable copies of data so that information survives hardware failure, corruption or loss. Implemented through replication, mirroring, RAID and erasure or forward-error-correction codes, it trades extra capacity for durability and availability. It is a foundational reliability technique in storage systems and distributed infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-redundancy", "labels": [ "Data Redundancy" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "data-registry", "title": "Data Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Data Registry is a centralised or federated catalogue that maintains authoritative records of data assets, their schemas, provenance, ownership, and access policies. In digital and metaverse infrastructure, registries track digital assets, avatar identities, spatial anchors, and ontological terms, enabling discovery, governance, and interoperability across distributed systems. Blockchain-anchored registries provide tamper-evident provenance and decentralised control.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-registry", "labels": [ "Data Registry" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-replication", "title": "Data Replication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The process of copying and maintaining data across multiple nodes, servers, or locations in distributed systems to ensure consistency, availability, and fault tolerance, using consensus algorithms like Paxos, Raft, and Byzantine Fault Tolerant protocols to coordinate state across decentralised networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:data-replication", "labels": [ "Data Replication" ], "is_subclass_of": [ "Data Management", "Distributed Systems" ], "wikilinks": [ "Consistency Guarantees", "Data Availability", "Storage Systems", "Blockchain", "Consensus Algorithm", "Distributed Systems", "Fault Tolerance", "metaverse", "Network Infrastructure" ] }, { "id": "data-residency", "title": "Data Residency", "domain": "governance", "domain_name": "Governance", "definition": "The physical or geographic location in which an organisation's data is stored and processed, and the practice of controlling that location to satisfy regulatory, contractual, tax, or policy requirements; data residency underpins localisation mandates and sovereignty claims by determining which jurisdiction's laws, courts, and government access powers apply to the data.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:data-residency", "labels": [ "Data Residency" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [ "Data Sovereignty", "Data Localisation", "Data Governance" ] }, { "id": "data-schema", "title": "Data Schema", "domain": "data", "domain_name": "Data", "definition": "A Data Schema is a formal specification that defines the structure, types, constraints, and semantics of data within a system or exchange protocol. It enumerates fields, their data types, cardinality rules, allowed values, and relationships between entities, enabling machines to validate conformance and enabling humans to understand data contracts. Data schemas are expressed in formalisms such as JSON Schema, XML Schema Definition, RDF SHACL, or OWL, and are versioned to manage evolution over time. They underpin interoperability between systems, database design, API contracts, and linked-data publishing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-schema", "labels": [ "Data Schema" ], "is_subclass_of": [ "Schema Definition" ], "wikilinks": [] }, { "id": "data-science", "title": "Data Science", "domain": "data", "domain_name": "Data", "definition": "Data science is an interdisciplinary field that combines statistics, programming, and domain knowledge to extract insight and build predictive models from data. It spans the full lifecycle from data acquisition and cleaning through exploratory analysis, modelling, and communication of results. Data science underpins evidence-based decision-making and supplies the analytical foundation for applied machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-science", "labels": [ "Data Science" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "data-security", "title": "Data Security", "domain": "security", "domain_name": "Security", "definition": "Data security comprises the policies, controls, and technologies that protect digital information from unauthorised access, corruption, or theft throughout its lifecycle. It spans encryption at rest and in transit, access control mechanisms, identity verification, and governance frameworks that together ensure confidentiality, integrity, and availability of data assets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:data-security", "labels": [ "Data Security", "Data Security Law" ], "is_subclass_of": [ "Information Security" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-serialization", "title": "Data Serialization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data Serialization is the process of converting structured in-memory data objects \u2014 including primitive types, collections, and complex graphs \u2014 into a byte-sequence or character-stream representation that can be stored persistently, transmitted across a network boundary, or reconstructed (deserialised) into an equivalent in-memory representation on a different machine or at a different time, potentially running different software. Serialization formats vary along axes of human readability (JSON, YAML, XML versus binary), schema enforcement (Protocol Buffers, Apache Avro with mandatory schemas versus schemaless JSON), compactness, and cross-language support, making format selection a critical engineering decision that affects system interoperability, performance, and evolvability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-serialization", "labels": [ "Data Serialization", "Data Serialisation" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "data-sharing", "title": "Data Sharing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The controlled exchange of datasets, data streams, or derived information between organisations, systems, or individuals under agreed governance, access control, and privacy constraints. Data sharing in distributed infrastructure contexts relies on interoperability standards, consent management frameworks, and data governance policies to balance openness with security and sovereignty.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:data-sharing", "labels": [ "Data Sharing", "Clinical Trial Data Sharing", "Cross-Organisational Data Sharing" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "data-silo", "title": "Data Silo", "domain": "data", "domain_name": "Data", "definition": "A data silo is an isolated repository of data controlled by one team, system, or department and not readily accessible to the rest of an organisation. Silos arise from fragmented tooling, organisational boundaries, and incompatible formats, producing duplicated, inconsistent, and underused data. They are the principal obstacle that data integration, governance, and interoperability efforts seek to dismantle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-silo", "labels": [ "Data Silo", "Data Silos" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-sovereignty", "title": "data sovereignty", "domain": "governance", "domain_name": "Governance", "definition": "Data sovereignty is the legal and political principle that digital data is subject to the laws, governance structures, and enforcement jurisdiction of the nation, region, or community in which it originates or is processed. It encompasses requirements such as data localisation (mandating that data remain physically within specified geographic boundaries), restrictions on cross-border data flows, and the right of governments or communities to compel access for regulatory, security, or cultural-preservation purposes. Data sovereignty directly shapes cloud architecture decisions, AI training dataset curation, federated system design, and multinational data-sharing agreements. It intersects with \u2014 yet is conceptually distinct from \u2014 privacy rights, data protection law, and cybersecurity, being principally concerned with jurisdictional control rather than individual rights.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:data-sovereignty", "labels": [ "Data Sovereignty" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-standards", "title": "Data Standards", "domain": "data", "domain_name": "Data", "definition": "Data Standards are formally agreed specifications, schemas, formats, vocabularies, and protocols that define how data is structured, encoded, exchanged, and interpreted across systems, organisations, and domains. They enable interoperability between heterogeneous systems by establishing shared semantics and syntactic conventions, reducing integration costs and data quality errors. Governed by bodies such as ISO, W3C, IETF, and NIST, data standards span a spectrum from de facto industry conventions (e.g., JSON, CSV) to de jure normative specifications (e.g., SQL:2023, OWL2, FHIR). They are foundational to data governance, metadata management, and the reliable operation of distributed data ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:data-standards", "labels": [ "Data Standards" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "data-stewardship", "title": "Data Stewardship", "domain": "governance", "domain_name": "Governance", "definition": "Data stewardship is the operational accountability for managing an organisation's data assets across their lifecycle, ensuring they are accurate, well-documented, secure, and used in line with policy. Data stewards act as the human layer of data governance, owning definitions, resolving quality issues, and curating metadata and lineage. It bridges governance policy and day-to-day data practice, enabling trustworthy, reusable data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-stewardship", "labels": [ "Data Stewardship" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "data-storage-layer", "title": "Data Storage Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software layer managing persistent storage, retrieval, and lifecycle of digital assets, metadata, world state, user data, and transactional records in metaverse systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:data-storage-layer", "labels": [ "Data Storage Layer" ], "is_subclass_of": [ "Data Management", "Data Layer" ], "wikilinks": [ "Asset Persistence", "Backup Systems", "Blockchain Storage", "Cache Layer", "CDN Storage", "Content Distribution", "Data Indexing", "Encryption Service", "MSF Taxonomy 2025", "Object Storage Service", "Replication Service", "User Profile Storage", "World State Management", "Blockchain", "Data Analytics", "Data Layer", "Database System", "InfrastructureDomain", "Metaverse Stack", "Network Infrastructure" ] }, { "id": "data-storage", "title": "Data Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data storage encompasses the systems, technologies, and architectures used to capture, retain, and retrieve digital information for ongoing and future use. It includes file, block, and object storage paradigms alongside the hardware and software infrastructure ensuring data persistence, accessibility, availability, and protection against loss or corruption.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:data-storage", "labels": [ "Data Storage" ], "is_subclass_of": [ "Data Management", "Infrastructure Component" ], "wikilinks": [ "Backup and Recovery", "Core Technology", "Data Persistence", "Data Redundancy", "Storage Architecture", "Blockchain", "Information Retrieval", "Infrastructure Component" ] }, { "id": "data-structure", "title": "Data Structure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A data structure is an organizational scheme for efficiently storing, accessing, and manipulating data, encompassing arrays, trees, graphs, hash tables, and tensors that underpin algorithmic computation and machine learning systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:data-structure", "labels": [ "Data Structure", "Authenticated Data Structure", "Data Structures", "DataStructure", "Graph Data Structure", "Tree Data Structure" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Graph Neural Networks", "IEEE Micro", "Memory Optimization", "Sparse Matrices", "Tensor Operations" ] }, { "id": "data-subject-rights", "title": "Data Subject Rights", "domain": "governance", "domain_name": "Governance", "definition": "Data subject rights are the legal entitlements granted to individuals over personal data that organisations hold about them, including rights of access, rectification, erasure, restriction, portability and objection. Codified in regimes such as the GDPR, these rights give people control over how their data is collected, used and shared, and impose corresponding obligations on data controllers and processors. Exercising and honouring them is a core requirement of modern data protection and privacy compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-subject-rights", "labels": [ "Data Subject Rights" ], "is_subclass_of": [ "Data Protection" ], "wikilinks": [] }, { "id": "data-synchronization", "title": "Data Synchronization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data synchronization is the process of keeping data consistent across multiple devices, replicas or systems by propagating changes and reconciling conflicts. It may be one-way or bidirectional, real-time or batch, and relies on change tracking, versioning and conflict-resolution strategies. It is essential wherever distributed copies of data must converge, from device sync to collective-intelligence systems aggregating many participants.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-synchronization", "labels": [ "Data Synchronization", "Real-Time Data Synchronization" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "data-transformation", "title": "Data Transformation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data transformation is the process of converting data from one structure, format or representation into another to make it suitable for storage, integration or analysis. It includes cleansing, type conversion, normalisation, aggregation, enrichment, schema mapping and serialisation, and is typically expressed as declarative or programmatic steps within a pipeline. Transformation reconciles heterogeneous sources, enforces quality and conformance rules, and shapes raw inputs into the canonical forms required by downstream systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-transformation", "labels": [ "Data Transformation" ], "is_subclass_of": [ "Data Processing" ], "wikilinks": [] }, { "id": "data-type", "title": "Data Type", "domain": "data", "domain_name": "Data", "definition": "A data type is a classification that specifies which values a piece of data may hold and which operations may be performed on it, such as integer, string, boolean, or array. Type declarations allow compilers, interpreters, and schema validators to catch category errors before or during execution, and inform how data is stored and serialised. Data types are the atomic building blocks from which composite structures such as records and schemas are constructed.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:data-type", "labels": [ "Data Type" ], "is_subclass_of": [ "Type System" ], "wikilinks": [] }, { "id": "data-validation", "title": "Data Validation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data validation is the process of checking that data conforms to defined rules, formats, ranges and constraints before it is processed, stored or transmitted. It detects errors, inconsistencies and anomalies at ingestion and transformation boundaries, ensuring that only well-formed and trustworthy data enters downstream systems. Validation is enforced through schema checks, type and range assertions, referential integrity rules and business-logic constraints.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:data-validation", "labels": [ "Data Validation" ], "is_subclass_of": [ "Infrastructure", "Data Management" ], "wikilinks": [] }, { "id": "data-versioning", "title": "Data Versioning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Data Versioning is the systematic practice of tracking, storing, and managing changes to datasets, model artefacts, and derived data products over time, providing reproducibility of experiments, auditability of data lineage, and governance of the full ML data supply chain. It applies software version-control semantics \u2014 branching, tagging, diffing, commit history, and rollback \u2014 to the large binary assets that underpin machine learning: raw datasets, processed feature stores, trained model weights, and pipeline configuration. Tools such as DVC (Data Version Control), lakeFS, and Apache Iceberg implement data versioning at different granularities, from file-level pointers through table-level snapshots to full environment branching over petabyte-scale data lakes. Data Versioning is a required foundation for reproducible science, regulatory compliance under frameworks such as the EU AI Act, and the operationalisation of mature MLOps practice.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:data-versioning", "labels": [ "Data Versioning", "Dataset Versioning" ], "is_subclass_of": [ "AI Infrastructure", "Machine Learning Pipeline", "Data Management", "Artificial Intelligence" ], "wikilinks": [ "Data Engineering", "DVC", "MLOps", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Blockchain", "Machine Learning Pipeline", "Experiment Tracking", "Feature Store", "Model Registry", "Data Pipeline", "Data Management", "Version Control", "Artifact Metadata", "Feature Engineering", "Model Training", "Reproducibility", "Data Lineage", "Data Governance", "Workflow Automation" ] }, { "id": "data-virtualization", "title": "Data Virtualization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A data integration approach that creates a virtual abstraction layer enabling users and applications to access, query, and integrate data from multiple disparate sources as a single unified system without physically moving or replicating the underlying data, supporting real-time access and reduci...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:data-virtualization", "labels": [ "Data Virtualization" ], "is_subclass_of": [ "Data Management", "Data Integration" ], "wikilinks": [ "Data Connectors", "Data Federation", "Query Engine", "Real-Time Data Access", "Unified Data View", "Blockchain", "Data Integration", "Metadata Management", "metaverse" ] }, { "id": "data-visualisation", "title": "Data Visualisation", "domain": "data", "domain_name": "Data", "definition": "Data Visualisation is the systematic practice of encoding data attributes into perceptual channels \u2014 position, colour, size, shape, and motion \u2014 to enable rapid human comprehension of patterns, trends, anomalies, and relational structures within datasets. It encompasses both static and interactive representations, spanning two-dimensional charts, network graphs, geospatial maps, and immersive three-dimensional scenes rendered in augmented or virtual reality environments. Effective data visualisation integrates principles from information design, cognitive psychology, and computer graphics, requiring rendering infrastructure, data pipelines, and interaction paradigms suited to the display medium. In the context of real-time analytics and spatial computing, visualisation systems must handle streaming data, high-dimensionality reduction, and multi-user collaborative viewing at interactive frame rates.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:data-visualisation", "labels": [ "Data Visualisation", "Data Visualization" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "data-warehouse", "title": "Data Warehouse", "domain": "data", "domain_name": "Data", "definition": "A data warehouse is a centralised analytical repository that integrates cleansed, structured data from multiple operational sources into a subject-oriented, historical model optimised for querying and reporting. It supports business intelligence through schemas such as star and snowflake and columnar storage for fast aggregation. It is a core data-engineering asset enabling consistent enterprise analytics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:data-warehouse", "labels": [ "Data Warehouse" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "data-warehousing", "title": "Data Warehousing", "domain": "data", "domain_name": "Data", "definition": "Data warehousing is the practice of consolidating data from multiple operational and external sources into a central, integrated, subject-oriented repository optimised for query and analysis rather than transaction processing. A data warehouse stores historical, cleansed and conformed data structured for reporting and decision support, typically populated through extract-transform-load pipelines and queried using online analytical processing. It provides the persistent analytical substrate on which business intelligence, dashboards and downstream analytics are built.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:data-warehousing", "labels": [ "Data Warehousing" ], "is_subclass_of": [ "Data Engineering" ], "wikilinks": [] }, { "id": "data", "title": "Data", "domain": "data", "domain_name": "Data", "definition": "Data is the recorded representation of facts, observations or measurements in a form suitable for storage, processing and communication. It is the raw material from which information and knowledge are derived through interpretation and analysis, and may be structured, semi-structured or unstructured. Across computing and analytics, data is captured, modelled, stored, transformed and governed throughout a lifecycle that determines its usefulness and trustworthiness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:data", "labels": [ "Data" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "database-management-system", "title": "Database Management System", "domain": "data", "domain_name": "Data", "definition": "A database management system (DBMS) is software that defines, stores, retrieves, secures and manages structured data while enforcing integrity, concurrency and durability. It mediates all access to the underlying database through a query interface and transaction manager, supporting models such as relational, document, key-value and graph. It is the foundational platform for persistent data services including ontology repositories and taxonomy registries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:database-management-system", "labels": [ "Database Management System" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "database-query", "title": "Database Query", "domain": "data", "domain_name": "Data", "definition": "A structured request submitted to a database management system to retrieve, insert, update, or delete data according to defined criteria. Queries are expressed in formal query languages such as SQL for relational databases or SPARQL for RDF triple stores, and underpin knowledge graph retrieval, analytics pipelines, and real-time application data access.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:database-query", "labels": [ "Database Query" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "database-schema", "title": "Database Schema", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A database schema is the formal, declarative description of the logical structure of a database, specifying its tables, columns, data types, primary and foreign keys, integrity constraints, views, indexes, and relationships. It constitutes the logical layer of a three-level ANSI/SPARC architecture that separates the conceptual organisation of data from physical storage details and from application-level views, and is enforced at runtime by the Database Management System. Schemas evolve through controlled migration scripts \u2014 governed by tools such as Flyway and Liquibase \u2014 and contrast with ontologies in being prescriptive, closed-world storage structures rather than open, inference-supporting conceptual models for reasoning.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:database-schema", "labels": [ "Database Schema" ], "is_subclass_of": [ "Data Model", "Relational Database" ], "wikilinks": [ "Data Model", "Relational Database", "Database Management System", "SQL", "Database Query", "Ontology", "Knowledge Representation", "OWL Class Hierarchy", "Data Integration", "Data Governance", "Metadata", "Data Warehouse", "Data Interoperability", "NoSQL Database", "Object-Relational Mapping", "Data Migration", "Normalisation", "Referential Integrity", "Relational Algebra", "ACID Transactions" ] }, { "id": "database-system", "title": "Database System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A database system (DBMS - Database Management System) is software that enables users to define, create, maintain, and control access to structured collections of data. It encompasses relational databases using SQL for structured table-based data and NoSQL databases supporting flexible schemas for document, key-value, graph, and wide-column data models, providing mechanisms for concurrent access, data integrity, and persistent storage.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:database-system", "labels": [ "Database System", "Schemaless Database" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Data Persistence", "Query Processing", "Transaction Management", "Blockchain", "ETSI Domain: Data Management", "InfrastructureDomain", "Technology Domain" ] }, { "id": "database-systems", "title": "Database Systems", "domain": "data", "domain_name": "Data", "definition": "Database systems are organised software stacks \u2014 comprising a storage engine, query processor, transaction manager, and access-control layer \u2014 that persistently store, retrieve, and manipulate structured or semi-structured data. They enforce ACID or BASE consistency guarantees, coordinate concurrent access via locking or multi-version concurrency control (MVCC), and expose declarative query languages such as SQL or graph-query dialects. Modern database systems span relational, document, key-value, columnar, time-series, and graph data models, each optimised for distinct access patterns and workload characteristics.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:database-systems", "labels": [ "Database Systems" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Graph Databases", "Microservices", "Distributed Systems", "https://en.wikipedia.org/wiki/Database", "https://www.postgresql.org/docs/" ] }, { "id": "database", "title": "Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A database is an organised collection of structured data managed by a database management system that supports persistent storage, efficient retrieval, concurrent access and integrity guarantees. Databases expose query interfaces, enforce schemas or schemaless models, and provide transactional or eventual-consistency semantics depending on their design. They form the durable state layer beneath most data-intensive applications and services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:database", "labels": [ "Database" ], "is_subclass_of": [ "Infrastructure", "Data Management" ], "wikilinks": [] }, { "id": "datalog-knowledge-graph-query-language", "title": "Datalog Knowledge Graph Query Language", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Queries, in the Logseq context, are structured Datalog or advanced query expressions embedded in pages using the `#+BEGIN_QUERY` / `#+END_QUERY` syntax. They dynamically retrieve and display blocks or pages matching specified conditions\u2014such as filtering private pages by the absence of a `#Public` tag\u2014without modifying underlying data. In the broader NarrativeGoldmine ontology, Queries represent the retrieval-layer mechanism that surfaces knowledge from the graph at read time.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:datalog-knowledge-graph-query-language", "labels": [ "Datalog Knowledge Graph Query Language", "Queries" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "dataset-curation", "title": "Dataset Curation", "domain": "data", "domain_name": "Data", "definition": "Dataset curation is the deliberate selection, cleaning, documentation, and maintenance of data collections used to train and evaluate machine learning models. It encompasses sourcing and licensing, de-duplication, filtering of low-quality or harmful content, labelling and label auditing, balancing for coverage and representation, and versioning with provenance records. Curation quality is a first-order determinant of model behaviour: the same architecture trained on better-curated data is routinely more capable, safer, and easier to evaluate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dataset-curation", "labels": [ "Dataset Curation" ], "is_subclass_of": [ "Data Curation" ], "wikilinks": [ "Data Curation", "Data Management", "Dataset", "Data Quality", "Data Labelling" ] }, { "id": "dataset", "title": "Dataset", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Dataset is a structured collection of data records sharing a common schema, gathered for a specific purpose such as training machine learning models, conducting research, or supporting analytics. Datasets are characterised by their size, modality (text, image, tabular, audio, video, graph, etc.), provenance, and licensing terms, all of which affect their fitness for use. Data quality, curation methodology, and bias documentation are critical attributes that determine the reliability of downstream AI systems. The movement from model-centric to data-centric AI has elevated dataset engineering to a first-class research and engineering discipline, with dataset documentation frameworks such as Datasheets for Datasets, Dataset Nutrition Labels, and the Croissant machine-readable format formalising the metadata contract between dataset creators and downstream consumers.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:dataset", "labels": [ "Dataset" ], "is_subclass_of": [ "AI Infrastructure", "Data Architecture", "Computational Modelling" ], "wikilinks": [ "Machine Learning", "Data Quality", "Data Governance", "Data Curation", "Data Annotation", "Data Labelling", "Benchmark Dataset", "Training Data", "Supervised Learning", "Unsupervised Learning", "Data Augmentation", "Data Lineage", "Data Catalogue", "Data Collection", "Data Cleaning", "Bias", "Fairness", "Model Evaluation", "Overfitting", "Transfer Learning" ] }, { "id": "datasheets-for-datasets", "title": "Datasheets for Datasets", "domain": "governance", "domain_name": "Governance", "definition": "A documentation framework proposed by Gebru et al. in which every machine learning dataset is accompanied by a structured datasheet recording its motivation, composition, collection process, preprocessing, recommended uses, distribution, and maintenance. Modelled on the datasheets that accompany electronic components, the practice surfaces provenance, consent, and bias considerations at the data layer, enabling informed dataset selection, reproducibility, and accountability across the machine learning lifecycle.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:datasheets-for-datasets", "labels": [ "Datasheets for Datasets" ], "is_subclass_of": [ "AI Documentation Standards" ], "wikilinks": [ "AI Documentation Standards", "Model Cards", "Dataset", "Data Governance" ] }, { "id": "datum-transformation", "title": "Datum Transformation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:datum-transformation", "labels": [ "Datum Transformation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "david-chaum", "title": "David Chaum", "domain": "security", "domain_name": "Security", "definition": "American computer scientist and cryptographer who pioneered digital cash, blind signatures and anonymous communication, and is widely regarded as a founder of the cypherpunk and privacy technology fields.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:david-chaum", "labels": [ "David Chaum" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Public-Key Cryptography", "Privacy", "Cryptocurrency", "Consensus", "Cryptography" ] }, { "id": "de-ber-ta", "title": "De BER Ta", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Decoding-enhanced BERT with Disentangled Attention: an improved BERT architecture that uses disentangled attention (separating content and position) and an enhanced mask decoder for better performance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:de-ber-ta", "labels": [ "De BER Ta" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "de-fi-protocol", "title": "De Fi Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain-based financial application providing decentralized financial services via smart contracts.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:de-fi-protocol", "labels": [ "De Fi Protocol", "DeFi Protocol", "DeFi Protocol Security", "DeFiProtocol" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "Asset Swaps", "Automated Market Makers", "Borrowing", "Governance Tokens", "Lending", "Trading", "AI Agent System", "BlockchainDomain", "Blockchain Infrastructure", "Smart Contract", "Virtual Economy" ] }, { "id": "de-fi-services", "title": "De Fi Services", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Decentralised Finance (DeFi) Services are financial products and protocols\u2014including lending, borrowing, automated market-making, and derivatives\u2014implemented as smart contracts on programmable blockchains without centralised intermediaries. DeFi Services enable permissionless, non-custodial access to financial primitives and are a key component of virtual economies within metaverse and spatial computing platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:de-fi-services", "labels": [ "De Fi Services", "DeFi Services" ], "is_subclass_of": [ "Platform and Environment", "Digital Asset Ecosystem" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "defi-infrastructure", "title": "DeFi Infrastructure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DeFi infrastructure is the layer of foundational protocols and services \u2014 price oracles, stablecoins, bridges, and indexing services \u2014 that decentralised finance applications depend on but do not themselves implement. It sits beneath end-user-facing DeFi protocols such as lending markets and exchanges, supplying reliable price feeds and settlement assets like USDC that those protocols compose into their own functionality. Its reliability and decentralisation materially affect the security of everything built on top of it.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:defi-infrastructure", "labels": [ "DeFi Infrastructure" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "defi-protocol", "title": "DeFi Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A DeFi Protocol is a set of immutable or upgradeable smart contracts deployed on a public blockchain that implements a specific financial primitive\u2014such as lending, decentralised exchange, derivatives, or yield aggregation\u2014in a permissionless and non-custodial manner. Protocol logic encodes all rules governing asset custody, interest-rate models, fee distribution, and liquidation mechanics directly in on-chain code that executes deterministically without intermediaries. Most DeFi protocols expose a composable interface through standardised token standards so that outputs (liquidity tokens, receipt tokens, or yield-bearing positions) can be consumed as inputs by other protocols. Governance over protocol parameters is typically delegated to token holders via an on-chain voting system, making protocol evolution itself a decentralised process.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:defi-protocol", "labels": [ "DeFi Protocol" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "de-fi", "title": "defi", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DeFi (Decentralised Finance) is the ecosystem of open, permissionless financial protocols and applications deployed as Smart Contracts on public blockchains \u2014 primarily Ethereum \u2014 that replicate and extend traditional financial services such as lending, borrowing, trading, yield generation, and derivatives without centralised intermediaries such as banks, brokers, or clearinghouses. Core DeFi primitives include Automated Market Makers (AMMs), over-collateralised lending protocols, stablecoins, flash loans, and yield aggregators, all coordinated through token-based incentive mechanisms and governed by Decentralised Autonomous Organisations (DAOs). DeFi protocols are composable \u2014 outputs of one protocol become inputs to another, enabling complex financial strategies assembled from protocol primitives. The open, programmable architecture distinguishes DeFi from traditional finance and from centralised cryptocurrency exchanges, making it a distinct paradigm within the broader blockchain economy.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:de-fi", "labels": [ "DeFi", "Confidential DeFi", "DeFi Ecosystem", "DeFi Integration", "Decentralised Finance", "Decentralized Finance", "Defi" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "dead-reckoning", "title": "Dead Reckoning", "domain": "robotics", "domain_name": "Robotics", "definition": "Dead reckoning is a navigation technique that estimates an entity's current position by applying its known speed, heading, and elapsed time to a previously determined position. It relies on relative motion measurements from sensors such as inertial measurement units and wheel encoders rather than external position fixes. Because errors accumulate over time as drift, dead reckoning is typically fused with absolute references like GPS to maintain accuracy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:dead-reckoning", "labels": [ "Dead Reckoning" ], "is_subclass_of": [ "Localisation" ], "wikilinks": [] }, { "id": "deanonymisation", "title": "Deanonymisation", "domain": "security", "domain_name": "Security", "definition": "Deanonymisation is the adversarial process of re-identifying individuals within data that was intended to be anonymous or pseudonymous, reversing the protective transformation that anonymisation applied. It works by correlating a supposedly de-identified record against auxiliary information \u2014 public datasets, quasi-identifiers such as postcode, age, and gender, behavioural fingerprints, or linkage across leaked corpora \u2014 until a unique individual is singled out. Landmark demonstrations against released medical, mobility, and streaming datasets have shown that removing direct identifiers is rarely sufficient. As the antonym of anonymisation, deanonymisation is both a privacy threat to defend against and an analytical technique used to test whether a dataset's protection actually holds.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:deanonymisation", "labels": [ "Deanonymisation" ], "is_subclass_of": [ "Adversarial Attack" ], "wikilinks": [ "Anonymisation", "Pseudonymity", "Privacy" ] }, { "id": "death-of-the-internet", "title": "Death of the Internet", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Death of the Internet is a concept cluster in the ics-society and Media Theory domains capturing the convergent degradation of the public World Wide Web as a space for authentic human knowledge-exchange, discovery, and democratic discourse.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:death-of-the-internet", "labels": [ "Death of the Internet" ], "is_subclass_of": [ "AI Governance and Ethics", "Platform Decay", "Information Integrity", "Digital Society Surveillance", "AI Ethics", "Media Theory", "Surveillance Capitalism" ], "wikilinks": [ "ActivityPub", "Advertising Economics", "Age Verification", "AI Training Data", "AIGovernanceDomain", "AlgorithmLayer", "Algorithmic Capture", "API Rate Limiting", "Attention Manipulation", "Authentic Content", "Authentic Internet", "Background Tokens", "Blockchain Attestation", "Bot Traffic", "C2PA", "Content Credentials", "ContentLayer", "Content Provenance", "Dark Forest Theory", "Data Provenance Initiative 2024" ] }, { "id": "debugging", "title": "Debugging", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Debugging is the systematic process of locating, diagnosing and resolving defects in software so that it behaves as intended. It typically involves reproducing the fault, observing program state through breakpoints, logging and runtime inspection, forming hypotheses about the cause, and verifying a fix. Debugging spans interactive use of debuggers, analysis of stack traces and logs, and reasoning about concurrency and integration boundaries. It is a core software development activity complementary to, but distinct from, automated testing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:debugging", "labels": [ "Debugging", "Software Debugging" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "decarbonisation", "title": "Decarbonisation", "domain": "governance", "domain_name": "Governance", "definition": "Decarbonisation is the systematic reduction of carbon dioxide and other greenhouse-gas emissions from an activity, organisation or economy, ultimately toward net zero. It combines energy efficiency, electrification, renewable energy procurement and removal of residual emissions, often guided by science-based targets. For energy-intensive systems such as blockchains it is the pathway to credible carbon-neutral operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decarbonisation", "labels": [ "Decarbonisation", "Decarbonisation Strategy", "Supplier Decarbonisation" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "decentraland", "title": "Decentraland", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentraland is a blockchain-based virtual world in which users own parcels of virtual land and in-world assets as non-fungible tokens on the Ethereum network, governed by a decentralised autonomous organisation through on-chain voting with governance tokens.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentraland", "labels": [ "Decentraland" ], "is_subclass_of": [ "Metaverse Platform" ], "wikilinks": [ "Ethereum", "Smart Contract", "Asset Tokenization", "Social VR", "Decentralized Autonomous Organization", "Metaverse Platform" ] }, { "id": "decentralisation", "title": "Decentralisation", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Decentralisation is the architectural and governance principle of distributing authority, control, computation, or data storage across multiple independent nodes or actors rather than concentrating it in a single entity. In technical systems it manifests as distributed ledgers, peer-to-peer networks, and federated protocols that eliminate single points of failure and censorship. In governance it refers to transferring decision-making power from central authorities to local or community actors. The degree of decentralisation exists on a spectrum and involves trade-offs between efficiency, security, scalability, and resilience.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decentralisation", "labels": [ "Decentralisation", "Decentralisation Analysis" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "decentralised-agent-coordination-initiative", "title": "Decentralised Agent Coordination Initiative", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An open collaborative initiative advancing decentralised agent frameworks that integrate Nostr relays for communication, Bitcoin Lightning for micropayments, distributed identity, and open data connectors to enable trustless, economically-incentivised multi-agent coordination without central intermediaries.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-agent-coordination-initiative", "labels": [ "Decentralised Agent Coordination Initiative", "Agentic Alliance", "Decentralised Agent Coordination" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Agents", "Bitcoin", "Blockchain", "Distributed Computing", "Distributed Identity", "Melvin Carvalho", "Nostr protocol", "Projects" ] }, { "id": "decentralised-agentic-infrastructure-stack", "title": "Decentralised Agentic Infrastructure Stack", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A convergent set of nascent and rapidly maturing technologies\u2014including decentralised identity, linked data, nostr key-pair agents, lightning-network micropayments, and knowledge graph tooling\u2014that together enable a new paradigm of interoperable, agent-driven digital infrastructure. The stack combines protocol-level primitives with knowledge management and collaborative AI to support autonomous, verifiable workflows.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-agentic-infrastructure-stack", "labels": [ "Decentralised Agentic Infrastructure Stack", "Emerging tech stack" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "decentralised-ai", "title": "Decentralised Ai", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised AI refers to artificial intelligence systems whose training, inference, data provenance, or governance are distributed across many independent participants rather than controlled by a single centralised entity. It commonly combines machine-learning techniques such as federated learning with blockchain or peer-to-peer infrastructure to coordinate compute, verify contributions, and align incentives through crypto-economic mechanisms. The goal is to reduce single points of control and failure, preserve data sovereignty, and enable open marketplaces for models, data, and compute.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-ai", "labels": [ "Decentralised Ai", "Decentralised AI" ], "is_subclass_of": [ "Blockchain", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "decentralised-application", "title": "Decentralised Application", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised application (dApp) is software whose backend logic runs on a decentralised peer-to-peer network, typically as smart contracts on a blockchain, rather than on centrally controlled servers. It combines on-chain contracts for trustless state and logic with off-chain front ends and storage. dApps are a defining component of Web3, enabling permissionless, censorship-resistant services such as exchanges, lending and marketplaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-application", "labels": [ "Decentralised Application", "Decentralised Application Ecosystems", "Decentralised Applications" ], "is_subclass_of": [ "Distributed System" ], "wikilinks": [] }, { "id": "decentralised-authentication", "title": "Decentralised Authentication", "domain": "security", "domain_name": "Security", "definition": "Decentralised authentication verifies the identity of a party without relying on a single central identity provider, instead using cryptographic keys, decentralised identifiers and verifiable credentials controlled by the user. The holder proves control of an identifier and presents credentials that a verifier checks against a distributed trust registry rather than a federated login service. This shifts control of identity from platforms to individuals while preserving cryptographic assurance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-authentication", "labels": [ "Decentralised Authentication" ], "is_subclass_of": [ "Authentication Protocol" ], "wikilinks": [] }, { "id": "decentralised-autonomous-organisation", "title": "decentralised autonomous organisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Decentralised Autonomous Organisation (DAO) is a blockchain-native organisational structure in which governance rules, treasury management, and operational logic are encoded as smart contracts executing on a public distributed ledger, enabling token-holder communities to propose, deliberate, and vote on decisions without recourse to centralised management or traditional corporate hierarchy. DAOs achieve censorship-resistant, transparent coordination by replacing trusted intermediaries with deterministic on-chain execution, typically combining governance tokens that confer weighted voting rights with treasury contracts that custodise and disburse collective assets. As an emergent governance primitive, DAOs span DeFi protocols, grant programmes, investment collectives, open-source software foundations, and decentralised media organisations, presenting novel challenges around legal personality, voter apathy, and Sybil resistance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-autonomous-organisation", "labels": [ "Decentralised Autonomous Organisation", "Decentralised Autonomous Organisations", "DecentralisedAutonomousOrganisation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "decentralised-coordination", "title": "Decentralised Coordination", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised coordination refers to mechanisms by which multiple autonomous agents\u2014human or machine\u2014achieve collective outcomes without a central authority directing behaviour, relying instead on shared protocols, incentive structures, consensus rules, or emergent social norms. In blockchain contexts it manifests as DAO governance, token-based voting, and smart-contract-enforced agreements that replace managerial hierarchy with algorithmic rule.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-coordination", "labels": [ "Decentralised Coordination" ], "is_subclass_of": [ "Autonomous Governance" ], "wikilinks": [] }, { "id": "decentralised-creative-metaverse-framework", "title": "Decentralised Creative Metaverse Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A conceptual and technical framework for a decentralised, AI-agent-driven metaverse ecosystem enabling global creative collaboration, autonomous task execution, and value exchange. The architecture integrates Nostr for identity and communication, Bitcoin and Lightning Network for payments, and USD/Omniverse for 3D asset manipulation, with agentic actors managing task negotiation, content delivery, and digital-object provenance across interconnected virtual spaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-creative-metaverse-framework", "labels": [ "Decentralised Creative Metaverse Framework", "Agentic Metaverse for Global Creatives" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Digital Twin" ] }, { "id": "decentralised-decision-making", "title": "Decentralised Decision-Making", "domain": "governance", "domain_name": "Governance", "definition": "Decentralised Decision-Making is a governance paradigm in which authority to make binding choices is distributed across multiple autonomous actors \u2014 individuals, nodes, or smart contracts \u2014 rather than concentrated in a single hierarchical centre, typically relying on voting mechanisms, consensus protocols, or market-based coordination to aggregate preferences and resolve conflicts. In blockchain and DAO contexts, decentralised decision-making is operationalised through on-chain proposal-and-vote cycles, token-weighted or quadratic voting, and cryptographically enforced execution of outcomes via smart contracts. The paradigm spans organisational theory, political science, distributed systems, and mechanism design, with implementations ranging from holocratic corporations to fully autonomous protocol governance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decentralised-decision-making", "labels": [ "Decentralised Decision-Making", "Decentralised Decision Making", "Decentralized Decision-Making" ], "is_subclass_of": [ "Decentralized Governance" ], "wikilinks": [] }, { "id": "decentralised-exchange", "title": "Decentralised Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A trading venue that allows users to swap digital assets directly through smart contracts, without a central operator holding custody of funds or matching orders off-chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralised-exchange", "labels": [ "Decentralised Exchange" ], "is_subclass_of": [ "Exchange Mechanism" ], "wikilinks": [ "Smart Contract", "Liquidity Pool", "Permissionless Trading", "Automated Market Maker", "Order Book", "Exchange Mechanism" ] }, { "id": "decentralised-finance", "title": "decentralised finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised Finance (DeFi) encompasses financial services and instruments\u2014including lending, borrowing, trading, derivatives, and yield generation\u2014implemented as permissionless, non-custodial smart contracts on public blockchains, eliminating the need for traditional financial intermediaries such as banks, brokers, clearinghouses, and custodians. Protocol logic is encoded directly in on-chain code that executes deterministically and transparently, allowing any party with an internet connection to inspect the rules governing their assets. DeFi protocols achieve composability by adhering to shared token standards and interacting through standardised interfaces, enabling complex multi-step financial strategies assembled from interoperable building blocks. The sector operates without central points of control, exposing users to smart-contract risk, oracle manipulation, and governance attacks in place of traditional counterparty risk.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:decentralised-finance", "labels": [ "Decentralised Finance", "Decentralised Finance Protocol", "DecentralisedFinance" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "decentralised-governance", "title": "Decentralised Governance", "domain": "governance", "domain_name": "Governance", "definition": "A model of collective decision-making in which authority is distributed across participants rather than concentrated in a central body, often implemented through on-chain voting, token-weighted proposals, and smart contracts that execute decisions automatically once thresholds are met.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralised-governance", "labels": [ "Decentralised Governance", "Decentralised Grant Programme" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [ "Smart Contract", "Transparency", "Decentralised Autonomous Organisation", "Decentralisation", "DAO", "Governance Framework" ] }, { "id": "decentralised-identifier", "title": "Decentralised Identifier", "domain": "security", "domain_name": "Security", "definition": "A type of globally unique identifier, defined by a W3C standard, that is controlled by its subject and resolves to a DID document containing public keys and service endpoints, without depending on a central registry or issuing authority.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralised-identifier", "labels": [ "Decentralised Identifier", "DID Decentralised Identifier", "Decentralised Identifier Architecture", "W3C Decentralised Identifier" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [ "Public-Key Cryptography", "Verifiable Credentials", "Decentralised Identity", "Identity Management", "Digital Identity" ] }, { "id": "decentralised-identifiers", "title": "Decentralised Identifiers", "domain": "security", "domain_name": "Security", "definition": "A W3C standard for globally unique identifiers that are created, owned and controlled by their subject without reliance on a central registration authority, enabling verifiable, self-sovereign digital identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralised-identifiers", "labels": [ "Decentralised Identifiers", "W3C Decentralised Identifiers" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [ "Cryptography", "Verifiable Credentials", "W3C", "Identity", "Digital Identity" ] }, { "id": "decentralised-identity", "title": "decentralised identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised Identity is a model for digital identity management in which individuals, organisations, and devices create, own, and control their own cryptographic identifiers and verifiable credentials without dependence on centralised identity providers or registries. Identifiers are anchored to a verifiable data registry \u2014 typically a blockchain or distributed ledger \u2014 as Decentralised Identifiers (DIDs) standardised by the W3C DID Core 1.0 specification, which defines a URI scheme resolving to a DID Document containing public keys and service endpoints. Credentials attesting to attributes of the DID subject are issued by trusted parties as W3C Verifiable Credentials, stored in a user-controlled digital wallet, and selectively disclosed to verifiers using zero-knowledge proofs or selective-disclosure mechanisms. The model operationalises Self-Sovereign Identity principles, giving subjects full autonomy over their identity data across systems and jurisdictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decentralised-identity", "labels": [ "Decentralised Identity", "DecentralisedIdentity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "decentralised-network", "title": "Decentralised Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Decentralised Network is a network architecture in which control, data and decision-making are distributed across many independent nodes rather than concentrated in a central authority or server. Nodes communicate peer-to-peer, share responsibility for routing, storage and consensus, and the network continues to operate even when individual nodes fail or leave. This topology improves resilience, censorship resistance and fault tolerance at the cost of greater coordination complexity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:decentralised-network", "labels": [ "Decentralised Network", "Decentralized Network" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "decentralised-storage", "title": "Decentralised Storage", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Storage systems that distribute data across many independent nodes rather than a single central provider, using content addressing and cryptographic verification to ensure data integrity, availability, and censorship resistance without a trusted intermediary.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralised-storage", "labels": [ "Decentralised Storage" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Cryptographic Hash", "Provenance", "IPFS", "Filecoin", "Distributed Systems" ] }, { "id": "decentralised-trust", "title": "Decentralised Trust", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralised trust is an architectural property of a system in which confidence in the correctness of data, identity claims, or transaction outcomes is derived from cryptographic proofs, distributed consensus, and transparent protocol rules rather than from reliance on any single central authority or intermediary. Rather than placing faith in a bank, government registry, or certificate authority, participants in a decentralised trust model verify claims independently using public-key cryptography, Merkle proofs, or zero-knowledge proofs enforced by a distributed network. The result is a system where trust is a computable property of the protocol rather than a social or legal delegation to an institution.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decentralised-trust", "labels": [ "Decentralised Trust" ], "is_subclass_of": [ "Trust" ], "wikilinks": [] }, { "id": "decentralised-web", "title": "Decentralised Web", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Decentralised Web (dWeb) is the architectural counter-movement to platform-consolidated Web 2.0, comprising the family of protocols, data models, identity primitives and economic mechanisms that re-establish the original distributed character of the internet by replacing location-addressed client...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:decentralised-web", "labels": [ "Decentralised Web", "Decentralised Social Web", "Decentralised Web Stack" ], "is_subclass_of": [ "Network and Communication", "Sociotechnical System", "Web Architecture", "Distributed System", "Peer To Peer Network", "Open Protocol" ], "wikilinks": [ "Aave Companies", "Account Portability", "ActivityPub", "Aggelos Kiayias", "Anna's Archive", "Archive Permanence", "Argent", "Arweave", "AT Protocol", "Aztec", "Aztec Protocol", "BBC R&D", "BBC R&D 2019 Reimagining Public Service Media Distributed Network", "Benet 2014 IPFS White Paper arXiv 1407.3561", "Berners-Lee Hendler Lassila 2001 Semantic Web Scientific American", "Berners-Lee Shadbolt 2012 Open Government Data CACM", "Bitcoin Lightning Network", "Bitcraft", "Blockchain-only Web3", "Bluesky" ] }, { "id": "decentralised-file-storage", "title": "Decentralised file storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Decentralised file storage denotes a class of infrastructure protocols and networks that disaggregate object/blob persistence across geographically distributed, mutually untrusting node operators using content-addressing, cryptographic accountability, and (in most implementations) crypto-economic...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:decentralised-file-storage", "labels": [ "Decentralised file storage" ], "is_subclass_of": [ "Data Management", "Content-Addressed Storage", "Distributed Storage", "Peer-to-Peer Network", "Web3 Infrastructure", "Permissionless Protocol" ], "wikilinks": [ "AI Dataset Distribution", "ar.io Permanode Network Documentation", "Arweave", "Ateniese et al 2007 Provable Data Possession", "Bandwidth", "Benet 2014 IPFS Whitepaper", "Bitswap Protocol", "CAR File", "CAR File Specifications", "Centralised File Server", "CID", "Cloud Object Storage", "Content-Addressed Storage", "Content Delivery Network", "Content Identifier", "Cryptographic Hash Function", "Data Sovereignty", "Decentralized Application", "Decentralized Identity", "Decentralized Web" ] }, { "id": "decentralization-layer", "title": "Decentralization Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Structural layer that distributes data and control across nodes to reduce central dependence and increase trust through P2P networking, blockchain, and distributed consensus mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:decentralization-layer", "labels": [ "Decentralization Layer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Cryptographic Primitives", "Data Sovereignty", "Distributed Hash Table", "Distributed Storage", "MSF Taxonomy 2025", "P2P Network", "Trust Distribution", "Blockchain", "Censorship Resistance", "Consensus Protocol", "Fault Tolerance", "InfrastructureDomain", "Middleware Layer", "Network Infrastructure" ] }, { "id": "decentralization", "title": "Decentralization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The architectural property of distributing control, data storage, and decision-making across many independent network participants rather than concentrating them in a single authority. In blockchain systems, decentralisation is achieved through open participation, distributed consensus, and cryptographic enforcement of rules, providing censorship resistance, fault tolerance, and permissionless access.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decentralization", "labels": [ "Decentralization", "Decentralisation" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure", "Telecollaboration" ] }, { "id": "decentralized-application", "title": "Decentralized Application", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralized application (dApp) is a software application whose backend logic runs on a blockchain or peer-to-peer network rather than on centralised servers controlled by a single entity, ensuring that no single party can unilaterally modify, censor, or shut down the application. The on-chain components \u2014 typically smart contracts \u2014 enforce business logic and state transitions transparently, while user-facing frontends may remain conventional web or mobile interfaces that communicate with the blockchain via wallet connectors. dApps inherit the censorship resistance and trustlessness of their underlying blockchain while exposing usability challenges related to transaction costs, latency, and key management.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:decentralized-application", "labels": [ "Decentralized Application" ], "is_subclass_of": [ "Blockchain Application" ], "wikilinks": [] }, { "id": "decentralized-authentication", "title": "Decentralized Authentication", "domain": "security", "domain_name": "Security", "definition": "Decentralised Authentication is an approach to verifying identity claims without relying on a single central authority or identity provider. It uses cryptographic credentials, decentralised identifiers and distributed trust mechanisms so that users control their own authentication material and prove claims directly to relying parties. The model reduces single points of failure, limits data centralisation and aligns with self-sovereign identity principles.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-authentication", "labels": [ "Decentralized Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "decentralized-autonomous-organization", "title": "Decentralized Autonomous Organization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An organisational entity operating through Smart Contracts, Distributed Governance, and community voting mechanisms without centralised authority or hierarchical control, enabling transparent resource allocation, democratic decision-making, and distributed treasury management at global scale.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-autonomous-organization", "labels": [ "Decentralized Autonomous Organization", "Decentralised Autonomous Organisation", "Decentralized Autonomous Organisation", "Decentralized Organization", "DecentralizedAutonomousOrganization" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "BC-0142-smart-contract", "BC-0201-decentralization", "BC-0462-on-chain-voting", "BC-0463-governance-token", "BC-0464-treasury-management", "BC-0465-proposal-system", "Distributed Governance", "AI Agent System", "BlockchainDomain", "Smart Contracts", "Virtual Economy" ] }, { "id": "decentralized-control", "title": "Decentralized Control", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Decentralized control is a control paradigm in which decision-making authority is distributed across many independent agents or nodes rather than concentrated in a single central controller, with global behaviour emerging from local rules and limited peer-to-peer communication. It trades the predictability of centralised control for robustness to single-point failure, scalability to large numbers of agents, and resilience under communication or node loss. Decentralized control is applied both in political and organisational governance and, in engineering, in domains such as swarm robotics where large numbers of simple agents must coordinate without a central coordinator.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-control", "labels": [ "Decentralized Control" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "decentralized-exchange-dex", "title": "Decentralized Exchange (DEX)", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Peer-to-peer marketplace enabling direct token swaps and digital asset trading through smart contracts without centralized intermediaries or custodial control.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:decentralized-exchange-dex", "labels": [ "Decentralized Exchange (DEX)" ], "is_subclass_of": [ "DeFi and Economics", "Smart Contract" ], "wikilinks": [ "Cryptographic Signature", "DeFi WG", "Decentralized Trading", "ISO 24165", "Price Oracle", "Token Swapping", "Trading Interface", "Automated Market Maker", "Blockchain", "Blockchain Oracle", "Consensus Mechanism", "Digital Wallet", "Liquidity Pool", "Liquidity Provision", "MiddlewareLayer", "Price Discovery", "Smart Contract", "Token Standard", "VirtualEconomyDomain" ] }, { "id": "decentralized-exchange", "title": "Decentralized Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Decentralized Exchange (DEX) is a peer-to-peer cryptoasset trading protocol implemented as a set of Smart Contract state machines on one or more public blockchains that enables non-custodial atomic swaps between digital assets without a centralised matching engine, custodial wallet, or oper...", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:decentralized-exchange", "labels": [ "Decentralized Exchange", "DecentralizedExchange" ], "is_subclass_of": [ "DeFi and Economics", "Marketplace", "Financial Protocol", "Blockchain Application", "Decentralised Finance Primitive", "Trading Venue" ], "wikilinks": [ "0x Protocol", "1inch", "1inch Fusion", "Aave", "Adams 2018 Uniswap V1", "Adams Zinsmeister Robinson 2020 Uniswap V2 Whitepaper", "Adams Zinsmeister Salem Keefer Robinson 2021 Uniswap V3 Core", "Aevo", "Aggregator", "Angeris Chitra 2020 Improved Price Oracles CFMM", "Angeris Kao Chiang Noyes Chitra 2019 Analysis of Uniswap Markets", "Arbitrum", "Asset Tokenisation", "Avalanche", "Balancer Protocol", "Bancor Network", "Base", "Batch Auction Matching", "Binance", "BIS Aramonte Doerr Huang Schrimpf 2022 DeFi Decentralisation Illusion" ] }, { "id": "decentralized-finance-de-fi", "title": "Decentralized Finance (DeFi)", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralized Finance (DeFi) is an open, permissionless financial ecosystem built on public blockchain networks\u2014principally ereum but increasingly on Layer 2 Networks (Arbitrum, Optimism, Base) and rival Layer 1s (Solana, Avalanche, BNB Chain)\u2014that replicates and extends traditional finan...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:decentralized-finance-de-fi", "labels": [ "Decentralized Finance (DeFi)" ], "is_subclass_of": [ "DeFi and Economics", "Token Economy", "Blockchain", "Financial Technology", "Smart Contracts", "Cryptographic Protocols" ], "wikilinks": [ "Algorithmic Stablecoin", "Atomic Composability", "CeFi", "Central Bank", "Chainlink", "Collateral", "Commercial Banking", "Concentrated Liquidity", "Cryptographic Oracle", "Cryptographic Proofs", "CryptographicProtocols", "Cryptographic Protocols", "Decentralised Exchange", "EIP-1559", "ERC-20 Token Standard", "EVM", "FATF Guidance on Virtual Assets", "Flash Loan Arbitrage", "Flash Loans", "Gas Fee Mechanisms" ] }, { "id": "decentralized-finance", "title": "Decentralized Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralized finance (DeFi) is an ecosystem of financial services built on public blockchains using smart contracts to provide lending, trading, derivatives and asset management without traditional intermediaries. Protocols are composable and permissionless, letting users transact directly from self-custodied wallets. DeFi introduces novel mechanisms such as automated market makers and liquidity pools, along with risks like impermanent loss and smart-contract exploits.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-finance", "labels": [ "Decentralized Finance", "Decentralised Finance" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "decentralized-governance", "title": "Decentralized Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralized Governance is a class of collective decision-making systems in which authority, rule-setting, and enforcement mechanisms are distributed across a network of participants rather than concentrated in a single central actor, implemented through formal on-chain voting protocols (token-w...", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:decentralized-governance", "labels": [ "Decentralized Governance", "Decentralised Governance" ], "is_subclass_of": [ "Governance and Regulation", "Governance", "Distributed Systems", "Institutional Economics", "Collective Action", "Common Pool Resources" ], "wikilinks": [ "Benevolent Dictatorship", "BIP Process", "Blockchain Consensus", "Centralised Governance", "Collective Action", "Common Pool Resources", "Community Ownership", "Community Treasury", "Compound Finance", "Compound GovernorBravo", "Corporate Board Governance", "Cosmos", "Cosmos Governance Module", "Cryptographic Proofs", "Cryptographic Signatures", "DAOs", "DeFi Protocols", "Decentralized Finance", "Delegation System", "Discourse Forums" ] }, { "id": "decentralized-identifier", "title": "Decentralized Identifier", "domain": "security", "domain_name": "Security", "definition": "A Decentralised Identifier (DID) is a globally unique, persistent, cryptographically verifiable identifier that a subject creates and controls without reliance on any centralised registry, authority, or intermediary. Defined by the W3C DID Core specification, a DID is a URI of the form did:method:identifier that resolves to a DID Document containing public keys, authentication mechanisms, and service endpoints. DIDs enable self-sovereign identity by decoupling identifier ownership from third-party identity providers, allowing verifiable credential exchange and privacy-preserving authentication across heterogeneous systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:decentralized-identifier", "labels": [ "Decentralized Identifier" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [ "DID Resolution", "Identity Management", "Centralized Identity Provider", "Digital Identity" ] }, { "id": "decentralized-identifiers", "title": "Decentralized Identifiers", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A W3C-standardised URI scheme (DID Core v1.0 Recommendation 19 July 2022, v1.1 Working Draft progressing through 2024-2025) defining globally unique cryptographically verifiable identifiers of the form did:: controlled directly by their subjects without dependency on c...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:decentralized-identifiers", "labels": [ "Decentralized Identifiers", "BC-0457-decentralized-identifiers" ], "is_subclass_of": [ "Cryptographic Primitive", "Token and Asset", "Identifier", "URI Scheme", "Self-Sovereign Identifier", "Cryptographic Identifier" ], "wikilinks": [ "Alan Turing Institute", "Allen 2016 Path to Self-Sovereign Identity", "Animo Solutions", "AnonCreds", "Aries-Cloudagent-Python (ACA-Py)", "Aries Framework JavaScript", "Aries Framework .NET", "Avast", "BankID", "BBS+ Signatures", "BC Wallet", "BlockChain Hyperledger Belfast", "Bloom", "Bluesky AT-Protocol", "British Columbia Government", "Cabinet Office GDS", "CBOR-LD", "Centralized Identity Provider", "cheqd", "Civic" ] }, { "id": "decentralized-identity-did", "title": "Decentralized Identity (DID)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A W3C standard for self-sovereign digital identities that are globally unique, cryptographically verifiable, and controlled by the identity subject without requiring centralized authorities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:decentralized-identity-did", "labels": [ "Decentralized Identity (DID)" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain Technology", "Cross-Domain Identity", "Decentralized Authentication", "DID Document", "DID Method", "DID Resolver", "DID URI", "JSON-LD", "Privacy-Preserving Identity", "Self-Sovereign Identity (SSI)", "W3C DID Core Specification", "W3C DID Specification", "Cryptographic Keys", "DID Nostr Identity", "Distributed Ledger", "Identity Management System", "MiddlewareLayer", "Public Key Infrastructure", "TrustAndGovernanceDomain", "Verifiable Credential (VC)" ] }, { "id": "decentralized-identity-foundation", "title": "decentralized identity foundation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Decentralized Identity Foundation (DIF) is an open industry consortium founded in 2017 that develops interoperable specifications, protocols, and reference implementations for decentralised digital identity systems. It produces foundational work items including the Presentation Exchange specification, the Decentralized Web Node (DWN) protocol for identity-linked personal data storage, the DIDComm messaging protocol, and the Sidetree protocol family for scalable DID method implementations. DIF operates as a coordination layer complementing W3C standards bodies, bridging specification gaps between DID methods, verifiable credential formats, and identity wallet implementations. Its outputs are adopted in major identity frameworks including the European Union Digital Identity (EUDI) Wallet Architecture Reference Framework.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralized-identity-foundation", "labels": [ "Decentralized Identity Foundation", "Decentralised Identity Foundation" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "decentralized-identity", "title": "Decentralized Identity", "domain": "security", "domain_name": "Security", "definition": "Decentralized Identity is an approach to digital identity management in which individuals, organisations, and devices hold cryptographic control over their own identifiers and credentials without dependence on any centralised identity provider. Built upon the W3C Decentralized Identifiers (DID) v1.0 Recommendation and the W3C Verifiable Credentials Data Model, it establishes a tripartite trust triangle of issuers, holders, and verifiers where credential authenticity is established through public-key cryptography anchored to a verifiable data registry \u2014 such as a distributed ledger or DNS \u2014 rather than through a privileged intermediary. This architecture realises self-sovereign identity principles: the subject generates and controls their key material, selectively discloses attributes using mechanisms such as BBS+ signatures or SD-JWT, and satisfies verifiers without exposing credentials to central surveillance or requiring real-time queries to the issuer.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-identity", "labels": [ "Decentralized Identity", "Decentralized Identity Ecosystem", "DecentralizedIdentity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "decentralized-key-storage", "title": "Decentralized Key Storage", "domain": "security", "domain_name": "Security", "definition": "Decentralized key storage is the practice of distributing a cryptographic key across multiple independent holders or locations so that no single party ever possesses the whole key, and no single point of compromise or failure can expose or destroy it. Typically realised with Shamir secret sharing or threshold cryptography, the key is split into shares such that a defined quorum reconstructs or jointly uses it while any smaller subset reveals nothing. This is distinct from decentralised data storage \u2014 which spreads arbitrary files across a network such as IPFS \u2014 because the object being protected is the secret itself and the security goal is quorum-controlled reconstruction rather than content availability.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decentralized-key-storage", "labels": [ "Decentralized Key Storage" ], "is_subclass_of": [ "Key Management" ], "wikilinks": [ "Key Management", "Shamir Secret Sharing", "Threshold Cryptography" ] }, { "id": "decentralized-storage", "title": "Decentralized Storage", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed storage infrastructure that distributes data across peer-to-peer networks rather than centralised data centres, enabling data persistence, redundancy, and access without single points of failure, often incentivised by cryptographic token mechanisms.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-storage", "labels": [ "Decentralized Storage", "DecentralizedStorage" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Arweave", "BlockchainHistory", "ContentDistribution", "Data Availability", "Decentralised Storage", "DigitalArt", "DistributedFileSystem", "dt:archives", "dt:enables", "dt:hosts", "dt:preserves", "dt:stores", "Filecoin", "IEEE", "incentivizes", "IPFS", "NFTMetadata", "PersistentStorage", "providesRedundancy", "storesData" ] }, { "id": "decentralized-swarm-control", "title": "Decentralized Swarm Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Decentralised swarm control distributes decision-making across all robot agents, where each robot computes commands based solely on local sensor information and direct communication with neighbouring agents.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:decentralized-swarm-control", "labels": [ "Decentralized Swarm Control" ], "is_subclass_of": [ "Robot Type", "Swarm Robotics", "Swarm Control", "Robotics" ], "wikilinks": [ "Adaptive Swarm Behaviour", "Agent Autonomy", "Asynchronous Coordination", "Autonomous Vehicle Platooning", "Biologically Inspired Robotics", "Collective Construction", "Cooperative Behaviour", "Distributed Algorithm", "Distributed Control Theory", "Emergent Behaviour", "Emergent Task Performance", "Foraging Robots", "Graph Theory", "Inter-Agent Communication", "Local Decision Logic", "Local Sensing", "Multi-Agent Systems", "Neighbourhood Communication", "Neural Networks", "Scalability" ] }, { "id": "decentralized-trading", "title": "Decentralized Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Decentralized trading is the exchange of crypto-assets directly between parties through smart contracts on a blockchain, without a centralized intermediary holding custody of funds or matching orders. Trades settle peer-to-contract against liquidity pools or order books governed by deterministic on-chain logic, giving traders self-custody and permissionless access. It is the activity layer enabled by decentralized exchanges and automated market makers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decentralized-trading", "labels": [ "Decentralized Trading" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "decision-engine", "title": "Decision Engine", "domain": "ai", "domain_name": "Ai", "definition": "A decision engine is a software component that encapsulates decision logic\u2014rules, models, policies, or heuristics\u2014and evaluates inputs against that logic to produce actionable outputs such as approvals, classifications, recommendations, or routing choices. Decision engines decouple business logic from application code, enabling non-developers to modify decision policies without code deployments, and support auditability by providing traceable reasoning paths for each decision. They range from rule-based expert systems to ML model inference services to hybrid architectures combining both. The emergence of large language model reasoning cores has extended the decision engine concept to unstructured input spaces, enabling complex multi-step decisioning with natural language explanation traces.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:decision-engine", "labels": [ "Decision Engine" ], "is_subclass_of": [ "Inference Engine", "Automated Reasoning", "Expert Systems", "Decision Intelligence" ], "wikilinks": [ "Automated Reasoning", "Policy Engine", "Inference Engine", "Decision Support", "Automated Planning", "Informed decision-making", "Decision Transparency", "Orchestration", "Expert Systems", "Machine Learning", "Explainable AI", "Large Language Models", "Knowledge Representation", "Business Rules", "Automated Decision Making", "Constraint Satisfaction", "Workflow Automation", "Risk Scoring", "Recommendation Engine", "Audit Trail" ] }, { "id": "decision-making", "title": "Decision Making", "domain": "ai", "domain_name": "Ai", "definition": "Decision making is the cognitive and computational process of selecting a course of action from a set of available alternatives by evaluating options against objectives, constraints, and uncertainty. It spans normative models of rational choice (expected utility, Bayesian reasoning), descriptive behavioural accounts that document systematic biases, and computational frameworks such as Markov Decision Processes and reinforcement learning that formalise sequential choice under uncertainty. In artificial intelligence, decision making underpins agent planning, autonomous systems, and large-scale optimisation, bridging probability theory, control theory, and game theory into operational policies.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:decision-making", "labels": [ "Decision Making", "Decision Making System", "Ethical Decision-Making", "Investor Decision-Making" ], "is_subclass_of": [ "AI Research Area", "Decision Theory", "Operations Research", "Cognitive Science" ], "wikilinks": [ "Reinforcement Learning", "Behavioural Economics", "owl:Thing", "Utility Theory", "Bayesian Inference", "Markov Decision Process", "Multi-Criteria Decision Analysis", "Uncertainty Quantification", "Objective Function", "Knowledge Representation", "Autonomous Agent", "Planning", "Policy Optimisation", "Probability Theory", "Optimisation", "Game Theory", "Machine Learning", "Bounded Rationality", "Governance", "AI Ethics" ] }, { "id": "decision-rights", "title": "Decision Rights", "domain": "governance", "domain_name": "Governance", "definition": "Decision rights are the formally allocated authorities that determine who may make, approve, or veto a given class of decisions within an organisation or governance system. They specify the mapping between roles or stakeholders and the scope of choices they control, forming the backbone of accountability. Clear decision rights reduce ambiguity, prevent conflicting actions, and enable auditable governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:decision-rights", "labels": [ "Decision Rights" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "decision-support", "title": "Decision Support", "domain": "ai", "domain_name": "Ai", "definition": "Decision Support refers to the class of information systems, analytical frameworks, and AI-augmented tools designed to improve the quality, speed, and consistency of human or automated decision-making by surfacing relevant data, models, and recommendations at the point of choice. Modern decision support systems integrate structured data from enterprise systems with unstructured signals from documents, sensor streams, and language models to present synthesised options with associated confidence levels and risk profiles. They range from simple rule-based dashboards to sophisticated agentic pipelines that autonomously gather evidence, run simulations, and present ranked courses of action. The field spans domains including clinical medicine, financial trading, military command, supply chain management, and public policy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:decision-support", "labels": [ "Decision Support", "Decision Support System", "Decision Support Systems", "Investor Decision Support" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [] }, { "id": "decision-theory", "title": "Decision Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Decision theory is the formal study of how a rational agent should choose among actions whose outcomes are uncertain, combining probabilities over states of the world with utilities over outcomes to select actions that maximise expected utility. Its normative branch prescribes optimal choice under axioms of rationality, while its descriptive branch studies how agents actually decide. It provides the foundational framework for rational action in artificial intelligence, economics, and operations research.", "entityType": "Class", "qualityScore": 0.9, "maturity": "mature", "iri": "urn:ngm:class:decision-theory", "labels": [ "Decision Theory" ], "is_subclass_of": [ "Artificial Intelligence", "Decision Making", "Operations Research", "Probability Theory" ], "wikilinks": [ "Probability Theory", "Bayesian Inference", "Utility Theory", "Game Theory", "Operations Research", "Decision Making", "Markov Decision Process", "Reinforcement Learning", "AI Safety Research", "Risk Management", "Artificial Intelligence", "Optimisation", "Behavioural Economics", "Bounded Rationality", "Uncertainty Quantification", "Cognitive Science", "Rational Agent", "Planning", "Knowledge Representation", "Objective Function" ] }, { "id": "decision-transparency", "title": "Decision Transparency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Decision transparency is the property of an AI system whereby the basis, logic, and contributing factors of its outputs are made accessible and comprehensible to relevant stakeholders. It requires that decision processes be traceable \u2014 from input data through model architecture to final output \u2014 and that explanations be appropriate to the audience, whether technical developers, domain experts, or affected individuals. Decision transparency is a foundational prerequisite for meaningful accountability, contestability, and regulatory compliance in AI deployments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:decision-transparency", "labels": [ "Decision Transparency" ], "is_subclass_of": [ "AI Governance and Ethics", "Responsible AI", "AI Ethics" ], "wikilinks": [ "Artificial Intelligence", "Blockchain", "Digital Twin", "Explainable AI", "Interpretability", "Audit Trail", "Accountability", "Algorithmic Accountability", "AI Trustworthiness", "Responsible AI", "AI Governance", "Regulatory Compliance", "AI Ethics", "AI Fairness", "Human Oversight", "Model Interpretability", "Algorithmic Transparency Index", "AI Model Card", "Stakeholder Engagement in AI", "EU AI Act" ] }, { "id": "decision-tree", "title": "Decision Tree", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A decision tree is a supervised learning model that predicts an outcome by recursively partitioning the feature space into regions, represented as a tree of decision nodes and leaf nodes. Each internal node tests a feature against a threshold and routes an instance down a branch, while leaves assign a class label or numeric value. Decision trees are valued for their interpretability and form the building blocks of ensemble methods such as random forests and gradient boosting.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:decision-tree", "labels": [ "Decision Tree" ], "is_subclass_of": [ "Supervised Learning", "AI Technique", "Interpretable Model", "Explainable AI", "Machine Learning" ], "wikilinks": [ "Supervised Learning", "Classification", "Regression", "Ensemble Methods", "Interpretability", "Explainability", "Random Forest", "Gradient Boosted Trees", "Overfitting", "Pruning", "Bias-Variance Tradeoff", "Feature Importance", "Gini Coefficient", "Information Gain", "Decision Transparency", "Explainable AI", "Machine Learning", "Training Data", "Evaluation Metric", "Cross Validation" ] }, { "id": "declination", "title": "Declination", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:declination", "labels": [ "Declination" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "decoder-network", "title": "Decoder Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A decoder network is the component of an encoder-decoder architecture that maps a compact latent representation back to a target output such as a reconstructed input, an image or a sequence of tokens. It learns to invert the compression performed by the encoder, reconstructing high-dimensional structure from low-dimensional codes. Decoders are central to autoencoders, variational autoencoders, sequence-to-sequence models and generative systems.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:decoder-network", "labels": [ "Decoder Network" ], "is_subclass_of": [ "Neural Network", "Encoder-Decoder Architecture" ], "wikilinks": [ "Autoencoder", "Encoder", "Variational Autoencoder", "Latent Space", "Neural Network", "Convolutional Neural Network", "Transformer", "Representation Learning", "Generative Model", "Image Generation", "Deep Learning", "Generative Adversarial Network", "Diffusion Model", "U-Net", "Attention Mechanism", "Backpropagation", "Loss Function", "Recurrent Neural Network", "Natural Language Processing", "Sequence-to-Sequence Model" ] }, { "id": "decoder", "title": "Decoder", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The component in an encoder-decoder architecture that generates the output sequence autoregressively, using masked self-attention, cross-attention to encoder outputs, and feed-forward layers.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:decoder", "labels": [ "Decoder", "Action Decoder", "Mask Decoder", "Multimodal Decoder", "X-Decoder" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "deduplication", "title": "Deduplication", "domain": "data", "domain_name": "Data", "definition": "Deduplication is a data-management technique that eliminates redundant copies of identical data by storing a single instance and referencing it wherever the same content recurs. Implementations typically hash data chunks and compare digests, so identical blocks resolve to the same stored object. It reduces storage footprint, backup windows, and network transfer in content-addressed and backup systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:deduplication", "labels": [ "Deduplication", "Content Deduplication" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "deep-generative-model", "title": "Deep Generative Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Deep Generative Model is a class of deep neural network trained to learn and approximate the underlying probability distribution of a dataset so that novel, statistically plausible samples can be drawn from it. The principal families \u2014 Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), normalising flows, and diffusion models \u2014 differ in how they parameterise and optimise the generative distribution, offering distinct trade-offs among sample fidelity, mode coverage, training stability, and latent-space interpretability. These models underpin modern generative AI capabilities across images, audio, video, text, and structured scientific data such as molecular graphs.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:deep-generative-model", "labels": [ "Deep Generative Model", "Generative Video Model" ], "is_subclass_of": [ "Generative AI", "Deep Learning" ], "wikilinks": [ "Generative AI", "Deep Learning", "Generative Adversarial Network", "Variational Autoencoder", "Diffusion Model", "Normalising Flow", "Autoregressive Model", "Neural Network", "Latent Space", "Training Data", "Probabilistic Inference", "Image Generation", "Synthetic Data", "Text-to-Image", "Data Augmentation", "Drug Discovery", "Stochastic Gradient Descent", "Backpropagation", "Attention Mechanism", "Encoder-Decoder Architecture" ] }, { "id": "deep-knowledge-tracing", "title": "Deep Knowledge Tracing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A machine-learning approach to modelling student knowledge, introduced by Piech et al. in 2015, that feeds a learner's full sequence of exercise interactions into a recurrent neural network which maintains a latent state of mastery and predicts the probability of answering the next item correctly, outperforming Bayesian Knowledge Tracing by learning inter-skill structure and forgetting dynamics directly from data at the cost of interpretability.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:deep-knowledge-tracing", "labels": [ "Deep Knowledge Tracing" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [ "Machine Learning", "Recurrent Neural Network", "Bayesian Knowledge Tracing", "Adaptive Learning" ] }, { "id": "deep-learning-framework", "title": "Deep Learning Framework", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A software library that provides the building blocks for defining, training and deploying deep neural networks, including tensor operations, automatic differentiation and hardware acceleration.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:deep-learning-framework", "labels": [ "Deep Learning Framework", "Deep Learning Frameworks" ], "is_subclass_of": [ "Deep Learning", "Machine Learning Software", "Deep Learning Domain", "Machine Learning" ], "wikilinks": [ "Automatic Differentiation", "Hardware Acceleration", "Backpropagation", "PyTorch", "Deep Learning Domain", "TensorFlow", "JAX", "Neural Network", "Computation Graph", "Tensor Operation", "Gradient Descent", "Stochastic Gradient Descent", "Convolutional Neural Network", "Transformer", "GPU Compute", "Distributed Training", "Transfer Learning", "Model Inference", "Mixed Precision Training", "Python" ] }, { "id": "deep-learning", "title": "Deep Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Deep Learning is a subset of Machine Learning based on Artificial Neural Networks with multiple layers (depth) that learn hierarchical representations of data through Backpropagation and Gradient Descent.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:deep-learning", "labels": [ "Deep Learning", "Deep Learning Model", "Deep Learning System", "Deep Learning Training", "Deep Learning for Robotics", "DeepLearning" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline", "Artificial Intelligence" ], "wikilinks": [ "AlgorithmicLayer", "Artificial Neural Networks", "Blockchain Validation", "DataManagementDomain", "Decentralized AI", "Geoffrey Hinton", "GPU Compute", "Hidden Layer", "Hyperparameter Tuning", "ISO/IEC 22989:2022", "NIST AI Standards", "Optimizer", "Self-Supervised Learning", "Yann LeCun", "Yoshua Bengio", "Activation Function", "AI-GroundedDomain", "Artificial Intelligence", "Backpropagation", "ComputationAndIntelligenceDomain" ] }, { "id": "deep-neural-network", "title": "Deep Neural Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A deep neural network is an artificial neural network with multiple hidden layers between its input and output, enabling it to learn hierarchical representations of data. Each layer applies a learnable linear transformation followed by a non-linear activation, and the network is trained by gradient descent with backpropagation to minimise a loss function. Depth lets the model compose simple features into increasingly abstract ones, which underlies modern deep learning across vision, language, and audio.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:deep-neural-network", "labels": [ "Deep Neural Network" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "deep-reinforcement-learning", "title": "Deep Reinforcement Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Deep Reinforcement Learning (DRL) is a machine learning paradigm that combines deep neural networks with reinforcement learning, enabling agents to learn optimal policies for sequential decision-making tasks by interacting with an environment, receiving scalar reward signals, and updating neural network parameters through gradient-based optimisation. Grounded in the Markov Decision Process framework, DRL applies the Bellman equation recursively to approximate value functions or directly optimise policies using methods such as DQN, PPO, SAC, and MuZero, achieving superhuman performance in games, robotics control, chip design, and language model alignment.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:deep-reinforcement-learning", "labels": [ "Deep Reinforcement Learning" ], "is_subclass_of": [ "Deep Learning", "Reinforcement Learning", "Artificial Intelligence" ], "wikilinks": [ "Deep Learning", "Reinforcement Learning", "Neural Network", "Markov Decision Process", "Bellman Equation", "Backpropagation", "Policy Gradient Methods", "Q-Learning", "Temporal Difference Learning", "Proximal Policy Optimisation", "Reward Signal", "Value Function", "Action Space", "Environment Model", "Simulation Environment", "Experience Replay", "GPU Acceleration", "Convolutional Neural Network", "Agent", "Agentic AI" ] }, { "id": "deep-research", "title": "Deep Research", "domain": "ai", "domain_name": "Ai", "definition": "Deep research is an agentic AI capability in which a language model autonomously plans and executes multi-step investigations, browsing the web, reading sources, and synthesising a cited report on a topic. The agent decomposes a query into sub-questions, iteratively gathers evidence across many pages, and reconciles findings rather than answering from parametric memory alone. It trades latency for depth, breadth of sources, and verifiable citations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:deep-research", "labels": [ "Deep Research" ], "is_subclass_of": [ "AI Application", "Agents", "Agentic Workflow" ], "wikilinks": [] }, { "id": "deep-think", "title": "Deep Think", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An advanced AI model developed by Google, characterized by high performance on reasoning benchmarks and the inclusion of autonomous agents for complex tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:deep-think", "labels": [ "Deep Think" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "deep-dao", "title": "DeepDAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "DeepDAO is a leading analytics and data aggregation platform dedicated to tracking and measuring the organisational health, financial activity, governance participation, and membership dynamics of decentralised autonomous organisations across multiple blockchain networks. It provides standardised metrics \u2014 total assets under management, number of token holders, proposal counts, voter participation rates, and delegate rankings \u2014 that enable comparative analysis of DAO ecosystems and inform investment, research, and governance design decisions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:deep-dao", "labels": [ "DeepDAO" ], "is_subclass_of": [ "DAO Analytics" ], "wikilinks": [] }, { "id": "deep-mind", "title": "DeepMind", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "DeepMind is an artificial intelligence research laboratory owned by Google (Alphabet), headquartered in London, known for pioneering work in reinforcement learning, deep learning, and scientific AI applications including AlphaGo, AlphaFold, Gemini, and Gato.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:deep-mind", "labels": [ "DeepMind" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Reinforcement Learning", "Deep Learning", "Machine Learning", "Google", "Artificial Intelligence", "https://deepmind.google", "https://deepmind.google/research/" ] }, { "id": "deep-speed", "title": "DeepSpeed", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "DeepSpeed is an open-source deep learning optimisation library developed by Microsoft Research that enables training and inference of extremely large neural network models through ZeRO (Zero Redundancy Optimizer) memory partitioning, pipeline parallelism, and mixed-precision arithmetic. It reduces the per-device memory footprint of model states by partitioning optimiser states, gradients, and parameters across data-parallel devices.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:deep-speed", "labels": [ "DeepSpeed" ], "is_subclass_of": [ "Distributed Training", "Distributed AI Training", "Distributed Computing", "Machine Learning", "High Performance Computing" ], "wikilinks": [] }, { "id": "deepfake-detection", "title": "Deepfake Detection", "domain": "ai", "domain_name": "Ai", "definition": "Deepfake detection is the application of machine learning and signal processing techniques to identify synthetic or manipulated media\u2014including face-swapped video, voice-cloned audio, and AI-generated images\u2014by analysing artefacts, inconsistencies, and statistical signatures that distinguish fabricated from authentic content. It operates as a countermeasure to generative adversarial networks and related synthesis methods.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:deepfake-detection", "labels": [ "Deepfake Detection" ], "is_subclass_of": [ "Computer Vision", "Machine Learning", "Media Forensics", "Signal Processing", "Deep Learning" ], "wikilinks": [] }, { "id": "deepfakes-and-fraudulent-content", "title": "Deepfakes and fraudulent content", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Deepfakes and fraudulent content denote synthetic audio, image, video and text artefacts produced by deep generative models (notably GAN-based face-swap pipelines DeepFaceLab/Faceswap/StyleGAN3, diffusion-based face-conditioning ReActor/Roop/InstantID/PhotoMaker/IP-Adapter-Face, lip-sync video ge...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:deepfakes-and-fraudulent-content", "labels": [ "Deepfakes and fraudulent content" ], "is_subclass_of": [ "AI Application", "AI Safety", "Synthetic Media", "Online Harm", "Information Integrity Threat", "Identity Fraud", "Generative AI Misuse" ], "wikilinks": [ "Audiovisual Forgery", "Authentic UGC", "Business Email Compromise", "C2PA", "C2PA 2024 Specification 2.0", "CAI Coalition for Content Provenance and Authenticity", "CGI VFX", "China Deep Synthesis Provisions", "Consensual Avatar", "Content Authenticity", "ContentAuthenticityDomain", "ContentLayer", "CyberSecurityDomain", "D-ID", "Dathathri et al. 2024 SynthID Text Nature", "DeepFaceLab", "Deloitte Deepfake Banking Fraud Forecast 2024", "Disclosed Synthetic Media", "Disinformation Campaign", "Distribution Platform" ] }, { "id": "deepfakes", "title": "Deepfakes", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI-generated or manipulated synthetic media content that convincingly alters a person's appearance, voice, or actions using deep learning techniques such as GANs, autoencoders, and voice synthesis models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:deepfakes", "labels": [ "Deepfakes", "Deepfake" ], "is_subclass_of": [ "AI Application", "Synthetic Media", "AI-Generated Content", "Generative AI", "Computer Vision Application" ], "wikilinks": [ "AI-Generated Content", "Audio Manipulation", "Audio Processing", "Computational Infrastructure", "Content Creation", "Entertainment Production", "Face Recognition", "Face Swapping", "Gesture Synthesis", "Identity Deception", "ISO 29100", "Media Manipulation", "Misinformation", "Reed Smith", "Synthetic Media", "Synthetic Video Generation", "Training Dataset", "Voice Cloning", "Autonomous Robot", "Computer Vision" ] }, { "id": "defect-detection", "title": "Defect Detection", "domain": "robotics", "domain_name": "Robotics", "definition": "Defect Detection is the automated identification of flaws, anomalies, or deviations from specification in products, materials, or processes, typically performed using machine vision and machine-learning models on images or sensor data. It localises and classifies defects such as cracks, scratches, contamination, dimensional errors, or assembly faults, often in real time on a production line. Modern systems combine high-resolution imaging, deep-learning object detection and anomaly detection, and feedback to robotic or process controls. Defect detection underpins quality assurance and industrial inspection, reducing waste and ensuring consistency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:defect-detection", "labels": [ "Defect Detection" ], "is_subclass_of": [ "Machine Vision" ], "wikilinks": [] }, { "id": "defence-in-depth", "title": "Defence In Depth", "domain": "security", "domain_name": "Security", "definition": "Defence in Depth is a cybersecurity strategy that layers multiple independent security controls such that the failure of any single control does not expose the system to compromise. Originating from military doctrine, it applies redundancy and diversity across physical, technical, and administrative security dimensions. The strategy reduces the probability of successful attack by requiring adversaries to defeat multiple independent barriers sequentially.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:defence-in-depth", "labels": [ "Defence In Depth", "Defence in Depth" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "defense-in-depth", "title": "Defense In Depth", "domain": "security", "domain_name": "Security", "definition": "Defense in depth is a security strategy that layers multiple, independent controls across an environment so that the failure or bypass of any single control does not lead to compromise. Adapted from military doctrine, it spans physical, network, host, application, and data layers, combining preventive, detective, and responsive measures to slow attackers, increase the cost of intrusion, and provide redundancy. It assumes no control is infallible and complements modern paradigms such as zero trust by ensuring that defences are distributed rather than concentrated at a single perimeter.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:defense-in-depth", "labels": [ "Defense In Depth", "Defence in Depth", "Defense in Depth" ], "is_subclass_of": [ "Security Architecture" ], "wikilinks": [] }, { "id": "deferred-rendering", "title": "Deferred Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Deferred rendering is a real-time shading technique that separates geometry processing from lighting by first rasterising scene attributes into a set of screen-space buffers, then computing lighting in a second pass over those buffers. By storing per-pixel position, normal, albedo, and material data in a geometry buffer, lighting cost becomes independent of scene complexity and scales with the number of lights instead. This decoupling makes it efficient to render scenes with many dynamic light sources.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:deferred-rendering", "labels": [ "Deferred Rendering" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "definitions-and-frameworks-for-metaverse", "title": "Definitions and frameworks for Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "This page surveys competing definitions and analytical frameworks used to characterise the metaverse, drawing from industry reports, academic literature, and standards bodies. It compares platform-centric, social-first, and open-interoperability framings; examines primitives such as persistence, embodiment, real-time 3D graphics, and economic actors; and contextualises failed early attempts (Decentraland, Sandbox) against more enduring platforms (Roblox). The analysis informs design principles for trust, value transfer, and AI-mediated governance in spatial social systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:definitions-and-frameworks-for-metaverse", "labels": [ "Definitions and frameworks for Metaverse" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Ontology" ], "wikilinks": [ "Aoki2003", "chen2011leet", "cole2013call", "Cook1977; @Kleinke1986; @Fagel2010", "gadekallu2022blockchain", "glas2013battlefields", "heiphetz2010training; @aldrich2005learning", "iser1993fictive", "Kendon1967", "Kleinke1986; @Nguyen2009", "kraus2022facebook", "Otsuka2005", "park2022metaverse", "serapis2008coming", "siyaev2021towards", "taylor2009play", "Apple", "Blockchain", "Computer Vision", "Metaverse Ontology" ] }, { "id": "deflationary-token", "title": "Deflationary Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Deflationary Token is a cryptocurrency or blockchain token whose total circulating supply decreases over time through one or more programmatic burning mechanisms\u2014such as transaction fee burns, scheduled buyback-and-burn events, or protocol-level token destruction\u2014creating supply-side scarcity intended to exert upward price pressure and incentivise long-term holding.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:deflationary-token", "labels": [ "Deflationary Token", "Deflationary Mechanism" ], "is_subclass_of": [ "Token and Asset", "Blockchain Entity", "Economic Mechanism", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain", "Virtual Economy" ] }, { "id": "deforestation-monitoring", "title": "Deforestation Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:deforestation-monitoring", "labels": [ "Deforestation Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "delay-tolerant-space-networking", "title": "Delay-tolerant Space Networking", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:delay-tolerant-space-networking", "labels": [ "Delay-tolerant Space Networking" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "delegate-democracy", "title": "Delegate Democracy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Delegate Democracy (also known as liquid democracy, delegative democracy, or proxy democracy) is a hybrid political and organisational governance model \u2014 formalised in modern computer-mediated form by Bryan Ford in his 2002 paper Delegative Democracy (Yale, later Swiss EPFL) and intellectuall...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:delegate-democracy", "labels": [ "Delegate Democracy", "Delegative Democracy" ], "is_subclass_of": [ "Governance and Regulation", "Governance Model", "Hybrid Democracy", "Collective Decision Making", "Social Choice Mechanism", "DAO Governance" ], "wikilinks": [ "Aave", "Adhocracy", "Agora Forum", "ai16z", "ai16z DAO", "Aragon Court", "Arbitrum", "Arrow 1951 Social Choice and Individual Values", "Behrens Kistner Nitsche Swierczek 2014 Principles of LiquidFeedback", "Better Reykjavik", "Brandt Conitzer Endriss Lang Procaccia 2016 Handbook of Computational Social Choice", "Bryan Ford", "Buterin 2021 Moving Beyond Coin Voting Governance", "Buterin Hitzig Weyl 2018 Quadratic Funding", "Buterin Weyl Ohlhaver 2022 Decentralized Society Soulbound", "Cambridge Centre for the Future of Democracy 2020 Global Satisfaction", "Carroll 1884 Principles of Parliamentary Representation", "Carroll Proportional Representation", "CCAF 2024 Global Cryptoasset Benchmarking Study", "Christoff Grossi 2017 Liquid Democracy TARK" ] }, { "id": "delegated-authorisation", "title": "Delegated Authorisation", "domain": "security", "domain_name": "Security", "definition": "Delegated authorisation is the pattern in which a resource owner grants a third-party application limited access to their resources without sharing their credentials. An authorisation server issues scoped, often time-limited tokens that the application presents to the resource server, so access can be granted, constrained, and revoked independently of the owner's password. OAuth 2.0 is the dominant framework that implements this pattern across web and API ecosystems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:delegated-authorisation", "labels": [ "Delegated Authorisation" ], "is_subclass_of": [ "OAuth 2.0" ], "wikilinks": [] }, { "id": "delegated-proof-of-stake", "title": "Delegated Proof of Stake", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Proof-of-Stake variant in which token holders delegate block production rights to a fixed number of elected delegates (witnesses or block producers) via a token-weighted democratic voting mechanism. DPoS achieves high throughput and fast block times by concentrating validation among a small elected set, with delegates subject to removal by voters if they misbehave or underperform.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:delegated-proof-of-stake", "labels": [ "Delegated Proof of Stake", "Delegated Proof Of Stake", "Delegated Staking", "DelegatedProofOfStake" ], "is_subclass_of": [ "Protocol and Consensus", "Proof of Stake" ], "wikilinks": [ "AI Agent System", "Blockchain", "Proof of Stake", "Virtual Economy" ] }, { "id": "delegation-registry", "title": "Delegation Registry", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A delegation registry is an on-chain contract that records mappings from a token holder's address to a delegate authorised to act or vote on their behalf. In DAO governance it lets holders assign their voting power to a representative without transferring assets, and the registry serves as the canonical source consulted when tallying votes. It enables liquid democracy patterns and separation of custody from voting authority.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:delegation-registry", "labels": [ "Delegation Registry" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "delegation-system", "title": "Delegation System", "domain": "governance", "domain_name": "Governance", "definition": "A delegation system is the governance mechanism by which stakeholders entrust their decision-making or voting power to chosen representatives. In decentralized governance it supports liquid democracy, where delegated authority can be reassigned or revoked at any time, scaling participation without requiring every holder to vote directly. It balances broad legitimacy with the expertise and availability of active delegates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:delegation-system", "labels": [ "Delegation System", "Delegation Mechanism", "Delegation Mechanisms" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "delegation", "title": "Delegation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Delegation in blockchain systems is the act of assigning one's stake or voting rights to another participant who acts on one's behalf in consensus or governance, without transferring ownership of the underlying assets. In proof-of-stake networks token holders delegate stake to validators, sharing in rewards and slashing risk while the validator performs block production. In on-chain governance, delegation lets holders entrust their votes to representatives, a pattern formalised as liquid democracy. Delegation lowers the participation barrier for ordinary holders and concentrates operational responsibility with capable operators, while introducing trust and centralisation trade-offs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:delegation", "labels": [ "Delegation" ], "is_subclass_of": [ "Proof of Stake" ], "wikilinks": [] }, { "id": "deliberate-demonstration", "title": "Deliberate Demonstration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A paradigm for AI learning where users intentionally teach the system by providing explicit examples, instructions, or demonstrations of desired tasks and behaviors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:deliberate-demonstration", "labels": [ "Deliberate Demonstration" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [] }, { "id": "delivery-versus-payment", "title": "Delivery-Versus-Payment", "domain": "finance", "domain_name": "Finance", "definition": "Delivery-Versus-Payment (DvP) is a securities settlement mechanism that links the transfer of a financial instrument to the simultaneous transfer of payment, ensuring that delivery occurs if and only if payment occurs, thereby eliminating principal risk in securities transactions. It is the standard settlement model mandated or strongly encouraged by financial market infrastructures globally.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:delivery-versus-payment", "labels": [ "Delivery-Versus-Payment", "Delivery Versus Payment", "Delivery versus Payment" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [] }, { "id": "delta-differential-one-way-ranging", "title": "Delta Differential One-way Ranging", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:delta-differential-one-way-ranging", "labels": [ "Delta Differential One-way Ranging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "delta-robot", "title": "Delta Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Delta robot employs parallel kinematics where three or more kinematic chains connect actuators to an end-effector platform, constraining motion through parallelogram linkages to pure translation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:delta-robot", "labels": [ "Delta Robot" ], "is_subclass_of": [ "Robot Type", "Parallel Robot", "Industrial Robot", "Parallel Robots" ], "wikilinks": [ "Actuator Motors", "Coordinated Control", "Electronics Assembly", "End-Effector Platform", "Food Processing", "High-Performance Actuators", "High-Speed Picking", "Inverse Kinematics Computation", "Kinematic Chains", "Parallel Robots", "Parallelogram Linkages", "Pick and Place", "Rapid Sorting", "Small Parts Handling", "Stiff Support Structure", "Synchronisation", "Ultra-Fast Placement", "End-Effector", "Industrial Robot", "Machine Learning" ] }, { "id": "delta-v", "title": "Delta-v", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:delta-v", "labels": [ "Delta-v" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "demand-forecasting", "title": "Demand Forecasting", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Demand forecasting is the analytical process of estimating future customer demand for products or services over a specified time horizon using historical sales data, statistical models, and increasingly machine learning algorithms. Accurate demand forecasts drive inventory replenishment, capacity planning, production scheduling, and financial budgeting decisions across retail, manufacturing, logistics, and energy sectors.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:demand-forecasting", "labels": [ "Demand Forecasting" ], "is_subclass_of": [ "Predictive Analytics" ], "wikilinks": [] }, { "id": "demand-planning", "title": "Demand Planning", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Demand planning is the supply-chain process of estimating future customer demand and aligning inventory, production, and procurement to meet it efficiently. It consolidates statistical forecasts, market intelligence, and business constraints into an actionable plan that minimises stockouts and excess holding. Modern demand planning increasingly relies on time-series forecasting and machine-learning models fed by historical sales and external signals.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:demand-planning", "labels": [ "Demand Planning" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "demand-response-mining", "title": "Demand Response Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Demand response mining is the practice of operating cryptocurrency miners as a flexible, interruptible electrical load that ramps up or down in response to grid conditions and price signals. Miners absorb surplus or stranded generation and shut off during peak demand, helping balance the grid while monetising otherwise curtailed energy. It is studied as both a grid-stabilisation tool and a mitigation for the environmental criticism of proof-of-work mining.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:demand-response-mining", "labels": [ "Demand Response Mining" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "demand-response", "title": "Demand Response", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Demand response is the deliberate modification of electricity consumption patterns by end users in response to signals from grid operators, energy markets, or automated control systems, with the aim of balancing supply and demand in real time. It enables consumers\u2014from large industrial facilities to domestic smart appliances\u2014to reduce or shift load during peak periods or grid stress events in exchange for financial incentives or reduced tariffs. Demand response programmes are a cornerstone of smart grid operation, deferring or avoiding costly investment in peak generation capacity. Increasingly, AI-driven automation and IoT sensor networks enable fine-grained, real-time demand response at residential scale without requiring manual intervention from end users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:demand-response", "labels": [ "Demand Response", "Demand Response Protocol", "Demand Response Services", "Grid Demand Response" ], "is_subclass_of": [ "Smart Grid" ], "wikilinks": [] }, { "id": "demand-side-response", "title": "Demand-Side Response", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Demand-side response is the adjustment of electricity consumption by end users in reaction to grid signals, time-of-use tariffs, or scarcity events, rather than altering generation. Smart-home and building automation systems shift or curtail loads such as heating, EV charging, and appliances to flatten peaks and exploit cheap or low-carbon periods. It improves grid reliability and lets consumers reduce energy cost and carbon footprint.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:demand-side-response", "labels": [ "Demand-Side Response" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "democratic-engagement", "title": "Democratic Engagement", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Democratic Engagement refers to the mechanisms, platforms, and practices through which citizens and communities participate in governance decisions, policy formation, and collective sense-making, including via digital and virtual channels. In metaverse and telecollaboration contexts, democratic engagement encompasses online deliberation tools, DAO-based governance, participatory spatial environments, and AI-assisted facilititation of broad-based civic participation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:democratic-engagement", "labels": [ "Democratic Engagement" ], "is_subclass_of": [ "Governance and Safety", "Metaverse and Telecollaboration" ], "wikilinks": [ "owl:Thing", "Telecollaboration" ] }, { "id": "democratic-governance", "title": "Democratic Governance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The exercise of collective decision-making through participatory, accountable, and transparent processes in which stakeholders deliberate and vote on rules and resource allocation; in digital and blockchain contexts it encompasses DAO voting, token-weighted and one-person-one-vote mechanisms, and community oversight of platforms and protocols.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:democratic-governance", "labels": [ "Democratic Governance", "Democratic Governance System", "Democratic Legitimacy", "Democratic Oversight of AI" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "democratic-participation", "title": "Democratic Participation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The mechanisms and systems enabling citizens to engage in collective decision-making processes within virtual environments, digital governance platforms, and metaverse communities, encompassing voting, deliberation, and civic engagement through immersive technologies and decentralized governance ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:democratic-participation", "labels": [ "Democratic Participation" ], "is_subclass_of": [ "Governance and Safety", "Digital Governance" ], "wikilinks": [ "Civic Engagement", "Community Decision Making", "Virtual Voting", "Blockchain", "Digital Governance", "Governance Framework", "Identity Verification", "metaverse", "Voting Systems" ] }, { "id": "democratic-values", "title": "Democratic Values", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Principles and practices of democratic governance\u2014including pluralism, participatory decision-making, transparent governance, electoral integrity, free formation of political will, and protection of civic space\u2014which AI systems and digital platforms should respect and strengthen rather than undermine. As codified in the OECD AI Principles 2024 and the EU AI Act, these values impose specific constraints on AI deployment in electoral, civic, and public-interest contexts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:democratic-values", "labels": [ "Democratic Values" ], "is_subclass_of": [ "Legal and Regulatory", "Policy Enforcement" ], "wikilinks": [ "Council of Europe", "OECD AI Principles 2024", "UNESCO", "Venice Commission", "EU AI Act", "MetaverseDomain", "Telecollaboration" ] }, { "id": "demographic-parity", "title": "Demographic Parity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Demographic parity is a group-fairness criterion requiring that a model's positive prediction rate be equal across protected groups, independent of the true label. Also called statistical parity, it is satisfied when the probability of a favourable decision does not depend on membership of a protected attribute such as gender or ethnicity. It is one of several formal, often mutually incompatible, definitions of algorithmic fairness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:demographic-parity", "labels": [ "Demographic Parity" ], "is_subclass_of": [ "Algorithmic Fairness" ], "wikilinks": [] }, { "id": "demonstration-content-curation-tag", "title": "Demonstration Content Curation Tag", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A placeholder tagging concept used within the knowledge graph to label pages or artefacts intended for demonstration purposes. It serves as a lightweight organisational marker allowing authors to filter, surface, or present selected graph nodes during product demonstrations or stakeholder presentations without altering the underlying ontological structure.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:demonstration-content-curation-tag", "labels": [ "Demonstration Content Curation Tag", "new tag for demos" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "OpenAI" ] }, { "id": "denavit-hartenberg-parameters", "title": "Denavit-Hartenberg Parameters", "domain": "robotics", "domain_name": "Robotics", "definition": "Denavit-Hartenberg (DH) parameters are a standardised four-parameter convention for describing the relative geometry between consecutive links of a robotic manipulator. Each joint is characterised by link length, link twist, link offset, and joint angle, yielding a homogeneous transformation matrix per joint. Chaining these matrices gives a compact, systematic model of the arm's kinematics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:denavit-hartenberg-parameters", "labels": [ "Denavit-Hartenberg Parameters", "Denavit-Hartenberg Convention" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "denoising-score-matching", "title": "Denoising Score Matching", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Denoising score matching is a training objective for learning the score function, the gradient of the log probability density, of a data distribution. Rather than estimating the score directly, it perturbs data with known Gaussian noise and trains a model to predict the noise, which is equivalent to estimating the score of the noise-perturbed distribution. This objective avoids the intractable normalising constant of energy-based models and underpins score-based generative models and diffusion models, where the learned score guides iterative sampling from noise back to data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:denoising-score-matching", "labels": [ "Denoising Score Matching" ], "is_subclass_of": [ "Generative Model", "AI Technique" ], "wikilinks": [] }, { "id": "dense-passage-retrieval", "title": "Dense Passage Retrieval", "domain": "ai", "domain_name": "Ai", "definition": "Dense Passage Retrieval (DPR) is an information retrieval approach in which both queries and document passages are encoded into dense continuous vector representations using dual-encoder neural networks, enabling similarity search via dot-product or cosine distance rather than sparse lexical matching. It substantially outperforms traditional BM25 retrieval on semantic matching tasks and forms the retriever component of open-domain question-answering and retrieval-augmented generation systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dense-passage-retrieval", "labels": [ "Dense Passage Retrieval" ], "is_subclass_of": [ "Information Retrieval", "Neural Information Retrieval", "Semantic Search", "Bi-Encoder Retrieval", "REALM", "DrQA", "ORQA", "ColBERT", "SPLADE", "BGE-M3", "ANCE", "E5 Embedding", "Hybrid Retrieval" ], "wikilinks": [] }, { "id": "dense-retrieval", "title": "Dense Retrieval", "domain": "ai", "domain_name": "Ai", "definition": "Dense retrieval is an information-retrieval method that encodes queries and documents into dense vector embeddings and ranks results by vector similarity rather than lexical term overlap. A learned bi-encoder maps text into a shared semantic space so that conceptually related items are close even without shared keywords. It underpins semantic search and retrieval-augmented generation, often paired with approximate nearest-neighbour indexes for scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dense-retrieval", "labels": [ "Dense Retrieval", "Dense Retriever" ], "is_subclass_of": [ "Information Retrieval", "Semantic Search", "Embedding Search", "Neural Information Retrieval" ], "wikilinks": [] }, { "id": "density-estimation", "title": "Density Estimation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Density estimation is the statistical problem of inferring the underlying probability density function of a random variable from a finite set of observed samples. Parametric approaches assume a fixed functional form whose parameters are fitted by maximum likelihood, while non-parametric approaches such as kernel density estimation and histograms make minimal distributional assumptions. Modern deep density estimation uses normalizing flows, autoregressive models, and variational methods to model complex high-dimensional distributions, making it foundational to generative modelling, anomaly detection, and unsupervised learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:density-estimation", "labels": [ "Density Estimation" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "deorbiting", "title": "Deorbiting", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:deorbiting", "labels": [ "Deorbiting" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "dependency-graph", "title": "Dependency Graph", "domain": "data", "domain_name": "Data", "definition": "A dependency graph is a directed graph in which nodes represent build artefacts, assets or tasks and edges represent a requires-before relationship between them. Traversing the graph in topological order determines a valid build or execution sequence and reveals cycles that would otherwise deadlock a pipeline. Asset and content pipelines rely on dependency graphs to determine what must be rebuilt when a source file changes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:dependency-graph", "labels": [ "Dependency Graph" ], "is_subclass_of": [ "Graph Theory" ], "wikilinks": [] }, { "id": "deployer", "title": "Deployer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Deployer, as defined in EU AI Act Article 3(4), is a natural or legal person, public authority, agency, or other body that uses an AI system under its authority in a professional context. Deployers bear obligations for human oversight, input data monitoring, logging, and fundamental rights impact assessments for high-risk AI systems. They are distinct from providers (who develop or place AI systems on the market) and incur provider-level obligations if they substantially modify an AI system.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:deployer", "labels": [ "Deployer" ], "is_subclass_of": [ "Software Engineering", "Digital Infrastructure" ], "wikilinks": [ "Autonomous Robot", "EU AI Act", "MetaverseDomain" ] }, { "id": "deposit-return-scheme", "title": "Deposit Return Scheme", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "A deposit return scheme is a circular-economy mechanism in which consumers pay a small refundable deposit on a beverage container at purchase and reclaim it on returning the empty for recycling or reuse. The financial incentive drives high collection rates and clean material streams, reducing litter and supporting closed-loop recycling. It is a recognised instrument of reverse logistics and extended producer responsibility.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:deposit-return-scheme", "labels": [ "Deposit Return Scheme" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "depth-buffer", "title": "Depth Buffer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A depth buffer (also called a z-buffer) is a per-pixel array maintained during rasterised rendering that stores the depth of the nearest surface drawn at each screen position. As fragments are generated, their interpolated depth is compared against the stored value, and only fragments closer to the camera overwrite the colour and depth, resolving visibility automatically. The depth buffer is the standard hidden-surface-removal mechanism in real-time graphics pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:depth-buffer", "labels": [ "Depth Buffer" ], "is_subclass_of": [ "Rendering Pipeline" ], "wikilinks": [] }, { "id": "depth-camera", "title": "Depth Camera", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A depth camera is a sensor that captures, for each pixel, the distance from the camera to objects in the scene, producing a depth map or 3D point cloud rather than only colour intensity. Common operating principles include structured light, time-of-flight, and stereo disparity, often combined with a colour stream to yield RGB-D data. Depth cameras are foundational sensors for spatial perception, enabling reconstruction, mapping, and interaction with three-dimensional environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:depth-camera", "labels": [ "Depth Camera" ], "is_subclass_of": [ "Camera" ], "wikilinks": [] }, { "id": "depth-estimation", "title": "Depth Estimation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Depth Estimation is the computer vision task of recovering per-pixel distance from a camera (or a virtual viewpoint) to the surfaces of a 3D scene, producing a depth map D(u,v) \u2208 \u211d\u207a aligned to an image I(u,v) that encodes scene geometry needed for 3D reconstruction, robotic perception, augmented-...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:depth-estimation", "labels": [ "Depth Estimation", "Depth Map Estimation", "DepthFM", "Stereo Depth Estimation" ], "is_subclass_of": [ "AI Technique", "3D Perception", "Computer Vision", "Dense Prediction", "Geometric Vision", "Scene Understanding" ], "wikilinks": [ "3D Perception", "3DReconstructionDomain", "AlgorithmLayer", "AR Occlusion", "ARCore", "ARCore Depth API Documentation", "ARKit", "ARKit Depth API Documentation", "Autonomous Driving", "Bhat et al. 2023 ZoeDepth", "Block Matching", "Bokeh Rendering", "Bundle Adjustment", "Camera Calibration", "Camera Extrinsics", "Camera Intrinsics", "Chang & Chen 2018 PSMNet", "ComputerVisionDomain", "Convolutional Neural Networks", "Cost Volume" ] }, { "id": "depth-map", "title": "Depth Map", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A depth map is a per-pixel image in which each value encodes the distance from the camera to the corresponding point in the scene rather than its colour. It provides the 2.5D geometric structure needed to reconstruct surfaces, segment foreground from background and place virtual content in spatial computing. Depth maps are produced by stereo matching, structured light, time-of-flight sensing or learned monocular estimation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:depth-map", "labels": [ "Depth Map" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "depth-sensing", "title": "Depth Sensing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Depth Sensing is the acquisition and processing of per-pixel or per-point distance information from a sensor to surfaces in a scene, yielding depth maps, range images, or 3D point clouds. Hardware modalities include structured light projection, time-of-flight (ToF) imaging, passive and active stereo vision, LiDAR scanning, and monocular depth estimation via machine learning. Depth sensing is a foundational enabling technology for spatial computing, robotic navigation, autonomous vehicles, augmented and mixed reality occlusion, gesture recognition, and 3D scene reconstruction, providing the geometric substrate upon which higher-level understanding is built.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:depth-sensing", "labels": [ "Depth Sensing" ], "is_subclass_of": [ "Sensor Fusion" ], "wikilinks": [ "Sensor Input", "owl:Thing" ] }, { "id": "depth-sensor", "title": "Depth Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A Depth Sensor is a hardware device or sensor modality that measures the distance between the sensor and objects in a scene, producing per-pixel or per-point depth information as its primary output. Operating principles include structured light projection, time-of-flight measurement, stereo vision correlation, and LiDAR pulse ranging. Depth sensors are fundamental components in robotics, augmented and mixed reality, autonomous vehicles, and industrial inspection, providing the three-dimensional scene understanding that colour cameras alone cannot supply. The output is typically represented as a depth map, disparity map, or three-dimensional [[Point Cloud]], forming the input to downstream perception and scene reconstruction pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:depth-sensor", "labels": [ "Depth Sensor" ], "is_subclass_of": [ "Exteroceptive Sensor" ], "wikilinks": [] }, { "id": "depth-first-search", "title": "Depth-First Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Depth-first search (DFS) is a graph and tree traversal algorithm that explores as far as possible along each branch before backtracking, following one path to its end before considering alternatives. It is naturally expressed through recursion or an explicit last-in-first-out stack and runs in time linear in the number of vertices and edges. DFS underpins many algorithms including topological sorting, cycle detection, finding connected components and solving maze and constraint problems through systematic backtracking.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:depth-first-search", "labels": [ "Depth-First Search" ], "is_subclass_of": [ "Graph Algorithms", "Search Algorithm", "Graph Search", "Algorithm" ], "wikilinks": [] }, { "id": "deregulation", "title": "Deregulation", "domain": "governance", "domain_name": "Governance", "definition": "Deregulation is the process of removing, reducing, or simplifying government rules and controls that constrain how firms operate in a market, typically to increase competition, lower barriers to entry, and improve economic efficiency. It is the deliberate counterpart to regulation and is often pursued in sectors such as telecommunications, finance, energy, and transport. Deregulation does not mean the absence of oversight but rather a shift toward lighter-touch or market-based governance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:deregulation", "labels": [ "Deregulation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "derivative-control", "title": "Derivative Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Derivative control responds to the rate of change of system error, providing damping that opposes oscillations and improves transient response.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:derivative-control", "labels": [ "Derivative Control" ], "is_subclass_of": [ "Actuation and Control", "Feedback Control", "Control Actions", "PID Control Component" ], "wikilinks": [ "Control Actions", "Control Systems", "Damping Action", "Derivative Gain", "Error Rate Measurement", "Error Signal", "Faster Convergence", "Gain Tuning", "Integral Control", "Kalman Filters", "Learning-Based Controllers", "Low-Pass Filtering", "Model Predictive Control", "Oscillation Damping", "PID Control", "PID Control Component", "Reduced Overshoot", "Robotic Joint", "Stability Improvement", "Time Derivative" ] }, { "id": "derivatives-trading", "title": "Derivatives Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Derivatives trading is the buying and selling of financial contracts whose value is derived from an underlying asset, index or rate, such as futures, options, swaps and perpetual contracts. Traders use these instruments to hedge exposure, gain leveraged directional exposure or speculate on price movements without holding the underlying. In decentralised finance, derivatives are implemented through smart contracts that handle margin, settlement and liquidation on-chain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:derivatives-trading", "labels": [ "Derivatives Trading" ], "is_subclass_of": [ "Digital Asset Trading" ], "wikilinks": [] }, { "id": "description-logic", "title": "Description Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A family of formal knowledge representation languages used to describe concepts, roles, and individuals with well-defined model-theoretic semantics, providing decidable fragments of first-order logic tailored to structured knowledge representation and automated reasoning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:description-logic", "labels": [ "Description Logic", "Description Logics", "Resource Description", "SHOIQ Description Logic" ], "is_subclass_of": [ "AI Research Area", "First-Order Logic", "Knowledge Representation", "Formal Logic", "Modal Logic" ], "wikilinks": [ "Set Theory", "OWL", "Knowledge Graph", "Semantic Web", "Ontology" ] }, { "id": "deserialisation", "title": "Deserialisation", "domain": "data", "domain_name": "Data", "definition": "The process of reconstructing in-memory data structures or objects from a serialised byte stream or textual encoding such as JSON, XML, or Protocol Buffers, reversing serialisation so that transmitted or persisted state can be used by a running program. Deserialisation must validate structure, types, and bounds of untrusted input, since naive object reconstruction is a well-known source of remote-code-execution vulnerabilities.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:deserialisation", "labels": [ "Deserialisation" ], "is_subclass_of": [ "Data Format Standard" ], "wikilinks": [ "Serialisation", "Data Format Standard", "Data Exchange", "Protobuf" ] }, { "id": "design-pattern", "title": "Design Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A named, reusable solution template for a recurring design problem in software, describing the participating roles, their collaborations, and the trade-offs of applying the solution in a given context. Catalogued most famously by the Gang of Four as creational, structural, and behavioural patterns, design patterns form a shared vocabulary that lets engineers communicate architecture concisely, guide refactoring towards proven structures, and evaluate components against well-understood alternatives.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:design-pattern", "labels": [ "Design Pattern" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Software Engineering", "Software Architecture", "Refactoring" ] }, { "id": "design-software", "title": "Design Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Design software comprises applications used to create, edit, visualise, and publish visual, spatial, or interactive artefacts across disciplines including graphic design, user interface and user experience design, 3D modelling, architectural drafting, motion graphics, and industrial product design. These tools span 2D vector and raster editors, parametric and direct-modelling CAD environments, sculpting and procedural-generation suites, prototyping platforms, and real-time 3D authoring environments that feed directly into game engines and extended reality runtimes. Modern design software increasingly integrates generative AI assistance for layout suggestion, texture synthesis, mesh generation, and design-variant exploration. The category sits at the intersection of creative practice, software engineering, and spatial computing infrastructure, making it a foundational enabler of digital content pipelines across entertainment, architecture, manufacturing, and immersive media.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:design-software", "labels": [ "Design Software", "Traditional Design Software" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "design-systems", "title": "Design Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A design system is a single source of truth that combines reusable UI components, design tokens, patterns, and documentation governing how a product looks and behaves. It unifies designers and engineers around a shared component library and usage guidelines, ensuring consistency and accelerating delivery across teams and platforms. It encompasses both the visual language and the coded implementation that enforces it.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:design-systems", "labels": [ "Design Systems", "Design System" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "design-thinking", "title": "Design Thinking", "domain": "governance", "domain_name": "Governance", "definition": "Design Thinking is a human-centred, iterative problem-solving methodology that prioritises deep empathy with end users, rapid ideation, and prototype-driven experimentation to arrive at innovative solutions. Originating in industrial design and formalised at Stanford's d.school, it proceeds through five non-linear phases: empathise, define, ideate, prototype, and test. The approach deliberately suspends assumptions and defers judgement during divergent thinking stages, producing a rich solution space before converging on testable prototypes. Design Thinking is now applied widely in product development, service design, organisational strategy, and technology innovation contexts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:design-thinking", "labels": [ "Design Thinking" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "destination-chain-execution", "title": "Destination Chain Execution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Destination chain execution is the phase of a cross-chain operation in which a message or transaction validated from a source chain is finally executed on the target blockchain. After relayers and verification prove the source event, the destination chain mints, releases, or calls a contract to complete the bridged action. Correct execution requires replay protection, message ordering guarantees, and trust in the verification layer.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:destination-chain-execution", "labels": [ "Destination Chain Execution" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "destination-marketing", "title": "Destination Marketing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The strategic promotion of physical or virtual locations using immersive technologies, AR/VR experiences, and digital twin representations to attract visitors, investors, and residents by showcasing destinations through interactive 3D visualizations and metaverse presence.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:destination-marketing", "labels": [ "Destination Marketing" ], "is_subclass_of": [ "Content and Assets", "Digital Marketing" ], "wikilinks": [ "3D Content Creation", "Immersive Advertising", "Location Promotion", "VR Experiences", "Blockchain", "Digital Marketing", "Digital Twin", "metaverse", "Virtual Tourism" ] }, { "id": "detailed-balance", "title": "Detailed Balance", "domain": "ai", "domain_name": "Ai", "definition": "Detailed balance is a condition on a Markov chain stating that, in equilibrium, the probability flux between any two states is equal in both directions. When a transition kernel satisfies detailed balance with respect to a target distribution, that distribution is a stationary distribution of the chain. It is the central design principle behind most Markov chain Monte Carlo samplers, including Metropolis-Hastings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:detailed-balance", "labels": [ "Detailed Balance" ], "is_subclass_of": [ "Markov Chain Monte Carlo", "Reversibility Condition", "Markov Chain Property", "Stochastic Process Condition", "Equilibrium Condition" ], "wikilinks": [] }, { "id": "determinism", "title": "Determinism", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Determinism, in computing and distributed systems, is the property by which a computational process produces identical outputs given identical inputs and initial state, regardless of when, where, or how many times it is executed. Deterministic systems are essential for reproducibility of scientific experiments, predictability of embedded and safety-critical control systems, and correctness of distributed consensus protocols where all participating nodes must reach the same conclusion from the same inputs. In the context of blockchain and smart contracts, determinism is a hard requirement because non-deterministic execution would cause different nodes to compute different state transitions, breaking consensus.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:determinism", "labels": [ "Determinism" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "deterministic-execution", "title": "Deterministic Execution", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Deterministic Execution guarantees that a computation produces identical outputs and state transitions whenever it is run on the same inputs in the same order, irrespective of host, timing, or scheduling. It is a prerequisite for state machine replication and blockchain smart contracts, where independent nodes must reach byte-identical results to agree on shared state. Achieving it demands eliminating sources of nondeterminism such as wall-clock time, unordered concurrency, and floating-point divergence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:deterministic-execution", "labels": [ "Deterministic Execution" ], "is_subclass_of": [ "State Machine Replication" ], "wikilinks": [] }, { "id": "deterministic-finality", "title": "Deterministic Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction finality achieved through explicit protocol mechanisms in BFT-based consensus systems, providing absolute mathematical guarantee that finalized blocks cannot be reverted once a supermajority of validators has committed to them.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:deterministic-finality", "labels": [ "Deterministic Finality", "Consensus Mechanism" ], "is_subclass_of": [ "Transaction Finality", "Blockchain" ], "wikilinks": [ "Absolute Transaction Guarantee", "Validator", "Validator Set", "AI Agent System", "Blockchain", "Byzantine Fault Tolerance", "Consensus Mechanism", "Probabilistic Finality", "Transaction Finality", "Virtual Economy" ] }, { "id": "deterministic-networking", "title": "Deterministic Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Deterministic networking is a class of networking techniques that provide bounded latency, low jitter and negligible packet loss for time-critical traffic over shared packet-switched infrastructure, typically through reserved bandwidth, scheduled transmission windows and time synchronisation across network nodes. It underpins standards such as IEEE Time-Sensitive Networking and IETF DetNet, which extend best-effort Ethernet and IP networks with guarantees suitable for industrial control, robotics and audio-video applications. Deterministic behaviour is achieved by combining traffic shaping, redundancy and precise clock synchronisation rather than relying on statistical over-provisioning alone.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:deterministic-networking", "labels": [ "Deterministic Networking" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "deterministic-scheduling", "title": "Deterministic Scheduling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Deterministic scheduling is a task-scheduling approach in which the order and timing of task execution are fully predictable given the same inputs, so that worst-case latency and execution order can be guaranteed in advance. It is a defining requirement of real-time computing and real-time operating systems, which must bound task response times to meet hard or soft deadlines. Deterministic scheduling contrasts with best-effort scheduling, which optimises average throughput without timing guarantees.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:deterministic-scheduling", "labels": [ "Deterministic Scheduling" ], "is_subclass_of": [ "Real-Time Computing" ], "wikilinks": [] }, { "id": "deterministic-serialisation", "title": "Deterministic Serialisation", "domain": "data", "domain_name": "Data", "definition": "Deterministic serialisation is the encoding of structured data into bytes such that semantically identical inputs always yield exactly the same byte sequence. It fixes ambiguities like map-key ordering, number formatting, and whitespace so that the output is canonical and reproducible. This property is essential for hashing, digital signatures, and content addressing, where any byte difference changes the resulting digest.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:deterministic-serialisation", "labels": [ "Deterministic Serialisation" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "devops", "title": "DevOps", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "DevOps is a sociotechnical discipline that unifies software development (Dev) and IT operations (Ops) through cultural practices, shared toolchains, and automation pipelines to shorten the system development lifecycle and deliver high-quality software continuously. It operationalises collaboration between development and operations teams by eliminating organisational silos, embracing infrastructure as code, and instrumenting every stage of delivery with feedback loops spanning continuous integration, continuous delivery, monitoring, and incident response. The discipline extends Agile principles beyond code authorship to encompass deployment, reliability engineering, and production observability as first-class engineering concerns.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:devops", "labels": [ "DevOps" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "dev-sec-ops", "title": "DevSecOps", "domain": "ai", "domain_name": "Ai", "definition": "DevSecOps is a software engineering practice that integrates security controls, testing, and policy enforcement directly into the continuous integration and delivery pipeline, making security a shared responsibility across development, operations, and security teams. It applies the principle of shift-left security, automating vulnerability scanning, dependency auditing, and policy-as-code checks at every stage from code commit through production deployment.", "entityType": "Class", "qualityScore": 0.95, "maturity": "established", "iri": "urn:ngm:class:dev-sec-ops", "labels": [ "DevSecOps" ], "is_subclass_of": [ "Agile Software Development", "Software Engineering", "Cybersecurity", "Continuous Delivery" ], "wikilinks": [] }, { "id": "development-platform", "title": "Development Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Development Platform is an integrated suite of tools, APIs, runtimes, and infrastructure services that enables engineers and creators to build, test, and deploy applications. In spatial computing contexts, development platforms provide SDKs for XR hardware, scene-graph editors, physics engines, and asset pipelines supporting OpenXR and Universal Scene Description (USD). They abstract hardware heterogeneity and supply governance tooling for collaborative multi-stakeholder development.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:development-platform", "labels": [ "Development Platform" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "device-drivers", "title": "Device Drivers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A device driver is a software component that mediates between an operating system and a specific hardware device, translating generic OS calls into the device's command and register protocol. Drivers expose a uniform interface so applications can use peripherals without knowing their internal details, and they handle interrupts, buffering, and power state. They are the foundation of the hardware abstraction layer that makes hardware portable to software.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:device-drivers", "labels": [ "Device Drivers", "Device Driver" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "device-identity", "title": "Device Identity", "domain": "security", "domain_name": "Security", "definition": "Device identity is a unique, verifiable credential bound to a physical device that allows it to authenticate itself to a network, service, or other devices independently of any human user. It is typically established through a hardware root of trust, such as a TPM or secure element, that stores a cryptographic key pair generated or provisioned at manufacture. In IoT and sensor network deployments, device identity is foundational to zero-trust access control, allowing gateways and platforms to verify which physical device is sending data before admitting it to the network. Without it, spoofed or cloned devices can inject false data or gain unauthorised network access.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:device-identity", "labels": [ "Device Identity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "dexterous-grasping", "title": "Dexterous Grasping", "domain": "robotics", "domain_name": "Robotics", "definition": "Dexterous grasping is the capability of a robotic hand with multiple independently actuated fingers to acquire and reposition objects using coordinated, in-hand manipulation rather than a simple two-jaw pinch. It requires fine control over contact forces and finger placement, often informed by tactile sensing to detect slip and adjust grip in real time without relying solely on vision. Reinforcement learning has become a dominant approach for training dexterous grasping policies, since the high dimensionality of multi-fingered hand kinematics and contact dynamics is difficult to model and plan for analytically. It is a prerequisite for robots handling irregular, deformable, or fragile objects that a rigid parallel gripper cannot reliably grip.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:dexterous-grasping", "labels": [ "Dexterous Grasping" ], "is_subclass_of": [ "Robotic Grasping" ], "wikilinks": [] }, { "id": "dexterous-manipulation", "title": "Dexterous Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Dexterous manipulation is the capability of robotic systems to grasp, reorient, assemble, and interact with objects using multi-fingered hands or compliant end-effectors in ways that require fine motor control, contact-rich reasoning, and real-time adaptation to object geometry and physical properties. It encompasses grasp planning, in-hand manipulation, and tactile feedback integration to replicate or exceed human-hand dexterity in unstructured environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:dexterous-manipulation", "labels": [ "Dexterous Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "di-lo-co", "title": "DiLoCo", "domain": "ai", "domain_name": "Ai", "definition": "DiLoCo (Distributed Low-Communication training) is a method for training large language models across loosely connected, geographically distributed compute clusters with minimal inter-node communication. Workers perform many local optimisation steps before periodically synchronising via an outer optimiser, drastically reducing the bandwidth and latency demands of conventional data-parallel training. It enables collaborative model training over the public internet rather than within a single tightly coupled datacentre.", "entityType": "Class", "qualityScore": 0.95, "maturity": "active-research", "iri": "urn:ngm:class:di-lo-co", "labels": [ "DiLoCo" ], "is_subclass_of": [ "Distributed Computing", "Distributed AI Training", "Federated Learning", "Distributed Training" ], "wikilinks": [] }, { "id": "diagrams-as-code", "title": "Diagrams as Code", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Diagrams as Code (DaC) is a software-engineering discipline and tooling category in which technical diagrams (architecture, sequence, state, deployment, entity-relationship, Gantt, wireframe, timing, dataflow) are authored as plain-text source artefacts in a constrained, machine-readable grammar ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:diagrams-as-code", "labels": [ "Diagrams as Code" ], "is_subclass_of": [ "Software Engineering", "Documentation As Code", "Declarative Specification", "Domain-Specific Language", "Text-Based Authoring", "Version-Controlled Artefact" ], "wikilinks": [ "ADR", "ADR", "Antora", "Architectural Decision Records", "ASCIIflow", "Atlassian", "AWS", "Azure", "Bass Clements Kazman 2021 Software Architecture in Practice 4th", "Belouadi Lauscher Eger 2024 AutomaTikZ ICLR", "Brandes & K\u00f6pf 2001 Fast and Simple Horizontal Coordinate Assignment", "Brown 2018 C4 Model for Visualising Software Architecture", "Brown 2020 Software Architecture for Developers Volume 2", "Build Pipeline", "BuildSystemLayer", "C4 Model", "C4 Model", "ChartGPT", "Class Diagram", "Confluence" ] }, { "id": "dialogue-management", "title": "Dialogue Management", "domain": "ai", "domain_name": "Ai", "definition": "Dialogue management is the component of a conversational system that tracks the evolving state of an interaction and decides the system's next action at each turn. It maintains context across utterances, integrates recognised user intents and slots, and selects responses or operations that move the conversation toward the user's goal. It is the control layer that connects language understanding to language generation in a dialogue agent.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:dialogue-management", "labels": [ "Dialogue Management" ], "is_subclass_of": [ "Conversational AI" ], "wikilinks": [] }, { "id": "dialogue-state-tracking", "title": "Dialogue State Tracking", "domain": "ai", "domain_name": "Ai", "definition": "A component of conversational systems that maintains a structured representation of the user's goals and the context of a conversation across turns. It updates the dialogue state as new utterances are processed.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dialogue-state-tracking", "labels": [ "Dialogue State Tracking" ], "is_subclass_of": [ "Dialogue System", "Dialogue Systems", "Natural Language Processing", "Task-Oriented Dialogue Systems", "Conversational AI" ], "wikilinks": [ "Natural Language Processing", "Conversational AI", "Dialogue System" ] }, { "id": "dialogue-system", "title": "Dialogue System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Dialogue System (conversational AI system) is an AI application that engages in natural language conversations with users through text or speech, managing multi-turn interactions, maintaining conversational context, and executing task-oriented or open-domain dialogues. Modern dialogue systems employ transformer-based language models, dialogue state tracking, and reinforcement learning to power virtual assistants, customer service chatbots, and conversational interfaces.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:dialogue-system", "labels": [ "Dialogue System", "NPC Dialogue System", "Retrieval-Based Dialogue System" ], "is_subclass_of": [ "AI Application", "Natural Language Processing" ], "wikilinks": [ "cashu", "Chatbot", "MetaverseDomain", "Natural Language Processing", "Question Answering", "Telecollaboration" ] }, { "id": "dialogue-systems", "title": "Dialogue Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Dialogue systems are software systems that converse with users in natural language across one or more turns. They include task-oriented assistants, slot-filling pipelines, and open-domain conversational agents, managing intent recognition, dialogue state tracking, response selection and natural language generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dialogue-systems", "labels": [ "Dialogue Systems" ], "is_subclass_of": [ "Conversational AI", "Natural Language Processing", "Human Computer Interaction" ], "wikilinks": [ "Natural Language Processing", "Customer Support Automation", "Chatbot", "Tool-Augmented Reasoning", "Conversational AI" ] }, { "id": "differentiability", "title": "Differentiability", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Differentiability is the mathematical property of a function having a well-defined derivative at every point in its domain, allowing gradients to be computed via calculus. In machine learning it is a prerequisite for gradient-based optimisation: activation and cost functions must be differentiable, or approximately so, for backpropagation to compute parameter updates. Non-differentiable operations require relaxations, subgradients, or surrogate approximations to remain trainable.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:differentiability", "labels": [ "Differentiability" ], "is_subclass_of": [ "Calculus" ], "wikilinks": [] }, { "id": "differentiable-architecture", "title": "differentiable architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Differentiable Architecture refers to a neural network design paradigm, most prominently realised in Differentiable Architecture Search (DARTS), in which discrete structural choices \u2014 such as which operation to place at each edge of a candidate graph or how to connect layers \u2014 are relaxed to continuous mixture weights over a predefined set of operations. This relaxation renders the architecture selection problem differentiable, allowing the architecture parameters to be optimised jointly with network weights via gradient descent on a validation loss. Once optimised, the continuous mixture is discretised to yield a final architecture, dramatically reducing the computational cost of neural architecture search compared to evolutionary or reinforcement learning methods.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:differentiable-architecture", "labels": [ "Differentiable Architecture", "DARTS Differentiable NAS" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "differentiable-function", "title": "Differentiable Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A differentiable function is one that has a well-defined derivative at every point in its domain, meaning it can be locally approximated by a linear map. Neural network components must be differentiable, or approximated by a differentiable surrogate, for gradient-based optimisation to compute usable error signals. Automatic differentiation and backpropagation both depend on the composed network being a differentiable function of its parameters.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:differentiable-function", "labels": [ "Differentiable Function" ], "is_subclass_of": [ "Calculus" ], "wikilinks": [] }, { "id": "differentiable-programming", "title": "Differentiable Programming", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Differentiable programming is a programming paradigm in which entire programs, not just isolated functions, are constructed so that gradients can be computed automatically through them via automatic differentiation. It treats control flow, loops and composed functions as differentiable building blocks, enabling gradient-based optimisation of arbitrary computational pipelines rather than only fixed neural network layers. It underlies techniques such as normalising flows, where a chain of invertible, differentiable transformations must be optimised end-to-end.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:differentiable-programming", "labels": [ "Differentiable Programming" ], "is_subclass_of": [ "Programming Paradigm" ], "wikilinks": [] }, { "id": "differentiable-rendering", "title": "differentiable rendering", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Differentiable rendering is a class of algorithms that reformulate the image-formation pipeline so that pixel values are differentiable with respect to scene parameters \u2014 including geometry, surface materials, lighting, and camera pose. By enabling backpropagation through the rendering process, these methods support gradient-based optimisation for inverse rendering, scene reconstruction from 2D observations, and end-to-end training of 3D-aware neural models. The field bridges classical computer graphics with deep learning, treating rendering as a learnable, differentiable module within a larger neural architecture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:differentiable-rendering", "labels": [ "Differentiable Rendering", "Differentiable Renderer", "DifferentiableRendering" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "Computer Graphics", "Deep Learning", "Inverse Rendering" ], "wikilinks": [] }, { "id": "differential-drive-robot", "title": "Differential Drive Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Differential drive robot uses two independently controlled wheels on opposite sides to enable both forward/backward locomotion and in-place rotation, forming the most widely deployed Mobile Robot architecture.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:differential-drive-robot", "labels": [ "Differential Drive Robot" ], "is_subclass_of": [ "Robot Type", "Wheeled Robot", "Nonholonomic Robot" ], "wikilinks": [ "Arc Trajectory Following", "Autonomous Vehicles", "Caster Wheel", "Differential Drive Controller", "Dubins Curves", "Forward Locomotion", "In-Place Rotation", "Independent Wheel Motors", "Inertial Measurement Units", "Kinematic Model", "Left Drive Motor", "Localisation", "Mobile Manipulation", "Motor Drivers", "Nonholonomic Robot", "Odometry Sensors", "Point Turning", "Power Battery", "Reeds-Shepp Paths", "Right Drive Motor" ] }, { "id": "differential-dynamic-programming", "title": "Differential Dynamic Programming", "domain": "robotics", "domain_name": "Robotics", "definition": "Differential Dynamic Programming (DDP) is a trajectory-optimisation algorithm that solves optimal-control problems by iteratively improving a control sequence using second-order local approximations of the dynamics and cost along the current trajectory. It performs a backward pass computing value-function derivatives and feedback gains, followed by a forward pass that applies the improved controls, converging quadratically near a solution. Together with its Gauss-Newton variant iLQR, DDP is widely used in robotics and model-predictive control for generating smooth, dynamically feasible motions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:differential-dynamic-programming", "labels": [ "Differential Dynamic Programming" ], "is_subclass_of": [ "Optimal Control" ], "wikilinks": [] }, { "id": "differential-equations", "title": "Differential Equations", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Equations relating a function to its derivatives, used to model how quantities change with respect to one or more independent variables; foundational to physics, engineering, biology, and modern machine-learning architectures such as neural ODEs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:differential-equations", "labels": [ "Differential Equations", "Differential Equation" ], "is_subclass_of": [ "AI Research Area", "owl:Thing", "Mathematics", "Applied Mathematics", "Scientific Computing" ], "wikilinks": [ "Linear Algebra", "Numerical Methods", "Dynamical Systems Theory", "Complex Systems", "owl:Thing" ] }, { "id": "differential-gnss", "title": "Differential GNSS", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:differential-gnss", "labels": [ "Differential GNSS" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "differential-interferometric-sar", "title": "Differential Interferometric SAR", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:differential-interferometric-sar", "labels": [ "Differential Interferometric SAR" ], "is_subclass_of": [ "Interferometric Synthetic Aperture Radar" ], "wikilinks": [] }, { "id": "differential-kinematics", "title": "Differential Kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Differential kinematics maps velocities between Robot Joint space and task-space (Cartesian) coordinates using the Jacobian matrix, enabling velocity-level analysis and control of robot manipulators.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:differential-kinematics", "labels": [ "Differential Kinematics" ], "is_subclass_of": [ "Actuation and Control", "Robot Kinematics", "Motion Mathematics" ], "wikilinks": [ "End-Effector Velocity Control", "Force/Torque Transformation", "Geometric Relationships", "Jacobian Matrix", "Joint Configuration", "Learning-Based Control", "Model Predictive Control", "Motion Mathematics", "Numerical Inverse Kinematics", "PID", "Position Kinematics", "Singularity Analysis", "Singularity Avoidance", "Singularity Condition", "Singularity Detection", "Velocity Inputs", "Velocity Transformation", "Force Control", "Forward Kinematics", "Impedance Control" ] }, { "id": "differential-privacy", "title": "Differential Privacy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Differential Privacy is a mathematical framework providing provable privacy guarantees by adding carefully calibrated noise to data queries or model outputs, ensuring that the presence or absence of any single individual's data has negligible impact on analysis results. The epsilon (\u03b5) parameter quantifies the privacy budget, with smaller values indicating stronger guarantees.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:differential-privacy", "labels": [ "Differential Privacy" ], "is_subclass_of": [ "AI Technique", "Privacy-Enhancing Technologies", "AI Safety Technique" ], "wikilinks": [ "Apple Differential Privacy", "Dwork et al. (2006)", "U.S. Census 2020", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin" ] }, { "id": "difficulty-adjustment", "title": "Difficulty Adjustment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Difficulty Adjustment is a Proof-of-Work consensus mechanism that periodically recalibrates the cryptographic puzzle difficulty so that blocks are produced at a statistically stable rate (e.g., approximately every 10 minutes in Bitcoin). It ensures network stability and security as total hash rate fluctuates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:difficulty-adjustment", "labels": [ "Difficulty Adjustment", "Difficulty Adjustment Algorithm", "Dynamic Difficulty Adjustment" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "difficulty-target", "title": "Difficulty Target", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A 256-bit threshold value in proof-of-work blockchains that a block's hash must be numerically less than or equal to for the block to be considered valid. The difficulty target is periodically recalculated by the Difficulty Adjustment algorithm to maintain a consistent average inter-block time regardless of changes in aggregate hash power. It is the primary mechanism by which mining difficulty is quantified and enforced across all network participants.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:difficulty-target", "labels": [ "Difficulty Target" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "difficulty", "title": "Difficulty", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Difficulty is a dynamic parameter in proof-of-work blockchain systems that expresses the computational effort required to find a valid block hash\u2014specifically, the number of leading zeros (or equivalent target threshold) that a candidate block hash must satisfy for the block to be accepted by the network. The difficulty value is adjusted periodically by the protocol based on the observed rate of block production relative to the target interval, ensuring that new blocks are produced at a predictable rate regardless of fluctuations in aggregate network hash rate. Difficulty is the primary mechanism by which proof-of-work blockchains self-regulate their monetary policy and security budget.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:difficulty", "labels": [ "Difficulty", "Network Difficulty" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "diffie-hellman-key-exchange", "title": "Diffie-Hellman Key Exchange", "domain": "security", "domain_name": "Security", "definition": "Diffie-Hellman Key Exchange is a cryptographic method by which two parties establish a shared secret over an insecure channel without ever transmitting the secret itself. Each party combines its private value with the other party's public value such that both arrive at the same key, which an eavesdropper cannot feasibly compute. Its security rests on the difficulty of the discrete logarithm problem, with elliptic-curve variants offering equivalent strength at smaller key sizes. It underpins forward-secret session establishment in protocols such as Transport Layer Security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:diffie-hellman-key-exchange", "labels": [ "Diffie-Hellman Key Exchange" ], "is_subclass_of": [ "Key Exchange" ], "wikilinks": [] }, { "id": "diffusion-model", "title": "Diffusion Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Diffusion Model is a class of probabilistic generative model that learns to synthesise data by reversing a learned forward diffusion process in which training examples are progressively corrupted with Gaussian noise across a fixed Markov chain of timesteps. At inference time the model iteratively denoises a sample drawn from pure noise, guided by a parametrised score function or noise-prediction network, until a high-fidelity output is recovered. Architecturally, the denoising backbone is typically a U-Net or Vision Transformer conditioned on timestep embeddings and optional guidance signals such as text or class labels. Diffusion models achieve state-of-the-art quality on image, audio, video, and molecular generation tasks and underpin production systems including Stable Diffusion, DALL-E 3, Sora, and AudioLDM.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:diffusion-model", "labels": [ "Diffusion Model", "Denoising Diffusion", "Diffusion Model Customisation", "Pretrained Diffusion Model" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture", "Probabilistic Generative Model", "Deep Learning" ], "wikilinks": [] }, { "id": "diffusion-models", "title": "Diffusion Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Diffusion Models are a class of deep generative models that learn the data distribution p_data(x) by reversing a fixed forward Markov noising process q(x_t | x_{t-1}) that progressively corrupts data x_0 with Gaussian noise across T timesteps until x_T \u2248 N(0,I), then training a neural denoiser \u03b5_...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:diffusion-models", "labels": [ "Diffusion Models" ], "is_subclass_of": [ "AI Model Architecture", "Generative Model", "Deep Generative Model", "Neural Network", "Score-Based Model", "Latent Variable Model", "Markov Chain Model" ], "wikilinks": [ "3D Generation", "Abramson et al. 2024 AlphaFold 3", "Albergo & Vanden-Eijnden 2023 Stochastic Interpolants", "AlgorithmLayer", "Audio Synthesis", "Autoregressive Model", "Blattmann et al. 2023 Stable Video Diffusion", "Classifier-Free Guidance", "CLIP Encoder", "ControlNet", "Creative Tools", "Croitoru et al. 2023 Diffusion Models Vision Survey", "CVPR", "Deep Generative Model", "Denoising Score Matching", "Dhariwal & Nichol 2021 Diffusion Beats GANs", "Differentiable Architecture", "Diffusion Transformer", "Drug Discovery", "Energy-Based Model" ] }, { "id": "diffusion-policy", "title": "Diffusion Policy", "domain": "robotics", "domain_name": "Robotics", "definition": "Diffusion Policy is a class of robot learning algorithms that represent robot action sequences as the output of a conditional denoising diffusion process, treating action prediction as iterative noise removal conditioned on sensor observations rather than as direct regression or classification. By leveraging the expressiveness of diffusion models to capture multi-modal action distributions, Diffusion Policy can represent one-to-many mappings from observation to action \u2014 a critical capability for dexterous manipulation tasks where multiple valid action trajectories exist. The approach, introduced by Chi et al. (2023), achieves state-of-the-art performance on imitation learning benchmarks and generalises across diverse manipulation settings.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:diffusion-policy", "labels": [ "Diffusion Policy" ], "is_subclass_of": [ "Robot Learning" ], "wikilinks": [] }, { "id": "diffusion-transformer", "title": "Diffusion Transformer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Diffusion Transformer (DiT) is a generative model architecture that replaces the convolutional U-Net backbone traditionally used in diffusion models with a scalable transformer architecture operating in a compressed latent space. DiT conditions the denoising process on class labels or text embeddings injected via adaptive layer normalisation or cross-attention, and processes image or video patches as sequences of tokens, enabling the model to leverage the scaling laws well-established for language transformers. Introduced by Peebles and Xie (2023), DiT demonstrated that transformer-based denoisers match or surpass U-Net performance while scaling predictably with model size and compute, forming the basis for state-of-the-art image and video generation systems including Stable Diffusion 3, FLUX, and OpenAI's Sora.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:diffusion-transformer", "labels": [ "Diffusion Transformer" ], "is_subclass_of": [ "Generative Model", "Diffusion Models", "Latent Diffusion", "Transformer Architecture" ], "wikilinks": [] }, { "id": "diffusion-of-innovations-theory", "title": "Diffusion of Innovations Theory", "domain": "governance", "domain_name": "Governance", "definition": "Diffusion of Innovations Theory, formulated by Everett Rogers, explains how, why, and at what rate new ideas and technologies spread through a social system over time. It segments adopters into innovators, early adopters, early majority, late majority, and laggards, and identifies perceived attributes such as relative advantage, compatibility, and trialability that govern uptake. It provides the analytical lens for understanding and forecasting technology adoption curves.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:diffusion-of-innovations-theory", "labels": [ "Diffusion of Innovations Theory" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "digital-actor-creation", "title": "Digital Actor Creation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of designing and generating photorealistic or stylized virtual human characters using 3D modeling, motion capture, AI synthesis, and deep learning techniques for use in entertainment, virtual production, metaverse experiences, and interactive media applications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-actor-creation", "labels": [ "Digital Actor Creation" ], "is_subclass_of": [ "AI Application", "Digital Human Technology" ], "wikilinks": [ "AI Synthesis", "Digital Twins", "Synthetic Media", "Virtual Performances", "3D Modeling", "DID Nostr Identity", "Digital Human Technology", "metaverse", "Motion Capture" ] }, { "id": "digital-archive", "title": "Digital Archive", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A systematic collection and preservation system for digital assets, cultural artifacts, documents, and media using distributed storage, metadata standards, and immersive access technologies to ensure long-term accessibility and discovery within virtual environments and knowledge management systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:digital-archive", "labels": [ "Digital Archive" ], "is_subclass_of": [ "Content and Assets", "Digital Preservation" ], "wikilinks": [ "Cultural Heritage Access", "Historical Research", "Knowledge Preservation", "Access Control", "Blockchain", "Digital Preservation", "Metadata Standards", "metaverse", "Storage Infrastructure" ] }, { "id": "digital-art-application", "title": "Digital Art Application", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software platforms and tools enabling artists to create, manipulate, and distribute visual art using digital technologies, including generative AI systems, 3D modeling software, and NFT minting platforms that facilitate artistic expression and commerce in virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-art-application", "labels": [ "Digital Art Application" ], "is_subclass_of": [ "Platform and Environment", "Creative Software" ], "wikilinks": [ "Digital Art Creation", "Graphics Processing", "Storage Systems", "Virtual Exhibitions", "Computer Vision", "Creative Software", "metaverse", "NFT Minting", "User Interface" ] }, { "id": "digital-art", "title": "Digital Art", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Digital art encompasses visual, interactive, and generative artworks created, stored, and distributed using digital technologies as primary medium or tool. It spans a wide range of practices\u2014from pixel painting and vector illustration to algorithmically generated pieces, interactive installations, and AI-synthesised imagery. The form has been transformed by NFT infrastructure, which introduced scarcity and provenance guarantees to inherently copyable digital objects. Digital art sits at the intersection of technological capability and aesthetic intent, continuously expanding as new computational tools emerge.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:digital-art", "labels": [ "Digital Art", "DigitalArt" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "digital-asset-custody", "title": "Digital Asset Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital asset custody is the safekeeping and administration of cryptographic private keys that control crypto-assets on behalf of their owners. Custodians use controls such as hardware security modules, multi-party computation, and cold storage to protect keys against theft, loss, and unauthorised use while meeting regulatory and audit requirements. It is the trust and security backbone enabling institutional participation in digital-asset markets.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:digital-asset-custody", "labels": [ "Digital Asset Custody", "Asset Custody", "Digital Asset Custody Innovation" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "digital-asset-ecosystem", "title": "Digital Asset Ecosystem", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The interconnected network of platforms, protocols, services, and participants that collectively enable the creation, storage, exchange, and utilization of blockchain-based digital assets including cryptocurrencies, tokens, NFTs, and tokenized real-world assets across decentralized and centralized infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-ecosystem", "labels": [ "Digital Asset Ecosystem" ], "is_subclass_of": [ "Platform and Environment", "Digital Economy" ], "wikilinks": [ "Exchange Platforms", "Wallet Infrastructure", "Asset Tokenization", "Blockchain", "Blockchain Network", "DeFi Services", "Digital Economy", "Digital Ownership", "metaverse" ] }, { "id": "digital-asset-governance", "title": "Digital Asset Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital asset governance is the set of rules, processes, and authority structures by which tokenised assets and the protocols managing them are controlled, upgraded, and held accountable across their lifecycle. It spans on-chain governance via governance tokens and DAOs, custody and key-management policy, token-standard conformance, and alignment with external regulatory and compliance regimes. By defining who may change parameters, mint or burn supply, and adjudicate disputes, it determines the legitimacy, security, and resilience of digital-asset systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:digital-asset-governance", "labels": [ "Digital Asset Governance" ], "is_subclass_of": [ "Decentralized Governance" ], "wikilinks": [] }, { "id": "digital-asset-infrastructure", "title": "Digital Asset Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The foundational technology stack comprising blockchain networks, node infrastructure, custody systems, key management solutions, and integration APIs that enable secure creation, storage, transfer, and management of digital assets across institutional and retail applications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-infrastructure", "labels": [ "Digital Asset Infrastructure" ], "is_subclass_of": [ "Computing and Cloud", "Digital Infrastructure" ], "wikilinks": [ "Asset Custody", "Cryptographic Systems", "Key Management", "Node Network", "Storage Solutions", "Blockchain", "Digital Infrastructure", "metaverse", "Transaction Processing" ] }, { "id": "digital-asset-lending", "title": "Digital Asset Lending", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Financial services enabling cryptocurrency holders to lend their digital assets to borrowers through centralized platforms or decentralized protocols, earning interest yields while providing liquidity for trading, leverage, and other financial activities secured by collateralized positions.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-lending", "labels": [ "Digital Asset Lending" ], "is_subclass_of": [ "Platform and Environment", "DeFi Services" ], "wikilinks": [ "Collateral Management", "Leverage Trading", "Liquidation Systems", "Yield Generation", "Blockchain", "DeFi Services", "Liquidity Provision", "metaverse", "Smart Contracts" ] }, { "id": "digital-asset-management", "title": "Digital Asset Management", "domain": "finance", "domain_name": "Finance", "definition": "The professional oversight and administration of cryptocurrency portfolios, tokenized assets, and blockchain-based investments through systematic strategies, risk management frameworks, and fiduciary practices to optimise returns and preserve capital for individuals and institutions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-management", "labels": [ "Digital Asset Management" ], "is_subclass_of": [ "Investment Management" ], "wikilinks": [ "Analytics Tools", "Asset Allocation", "Compliance Systems", "Custody Solutions", "Portfolio Optimization", "Blockchain", "Investment Management", "metaverse", "Risk Management" ] }, { "id": "digital-asset-market", "title": "Digital Asset Market", "domain": "finance", "domain_name": "Finance", "definition": "A marketplace \u2014 centralised or decentralised \u2014 where digital assets including cryptocurrencies, NFTs, and tokenised securities are bought, sold, and exchanged. Digital asset markets encompass order-book exchanges, automated market makers, NFT marketplaces, and OTC desks, each governed by distinct liquidity, pricing, and regulatory mechanisms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-asset-market", "labels": [ "Digital Asset Market", "Digital Asset Marketplace" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-asset-regulation", "title": "Digital Asset Regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital asset regulation is the body of legal and supervisory rules governing the issuance, trading, custody, and use of cryptographic assets such as cryptocurrencies, tokens, and stablecoins. It spans securities law, anti-money-laundering requirements, consumer and investor protection, and bespoke frameworks tailored to blockchain-based instruments. Because asset classifications and jurisdictional approaches differ widely, the field is characterised by evolving guidance, enforcement actions, and efforts toward harmonised cross-border standards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:digital-asset-regulation", "labels": [ "Digital Asset Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "digital-asset-risks", "title": "Digital Asset Risks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Comprehensive analysis of risks associated with digital assets, particularly Bitcoin, encompassing technical vulnerabilities, regulatory challenges, geopolitical concerns, financial stability threats, and systemic implications for global monetary systems. Covers attack vectors such as 51% attacks, selfish mining, network partitioning, exchange failures, and regulatory suppression, alongside macro-level risks including volatility, liquidity crises, and stablecoin depegging.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-asset-risks", "labels": [ "Digital Asset Risks" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Agustin Carstens", "Alameda Research", "AML", "anderson2002free", "apostolaki2016hijacking; @apostolaki2017hijacking; @johnson2014game; @stinner2022proof", "apostolaki2017hijacking", "Balaji Srinivasan", "Bank of International Settlement", "Basel Committee on Banking Supervision", "Biden", "BIS", "BlackRock", "Black Swan", "booth2022bitcoin", "budish2018economic", "carlsten2016instability", "CBDC", "CBOE", "central bank digital currencies", "CFTC" ] }, { "id": "digital-asset-standards", "title": "Digital Asset Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Formal specifications governing how digital assets are represented, identified, exchanged, and verified across platforms, including token standards such as ERC-721 and ERC-1155, metadata schemas, and interoperability profiles that keep virtual items portable and provably owned across metaverse environments and blockchain networks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-standards", "labels": [ "Digital Asset Standards", "Digital Asset Standard" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-asset-trading", "title": "Digital Asset Trading", "domain": "finance", "domain_name": "Finance", "definition": "The buying, selling, and exchange of cryptocurrencies, tokens, and other blockchain-based assets through centralised exchanges, decentralised protocols, and over-the-counter markets using various order types, trading strategies, and execution mechanisms to achieve price discovery and liquidity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-asset-trading", "labels": [ "Digital Asset Trading" ], "is_subclass_of": [ "Financial Trading" ], "wikilinks": [ "Exchange Platform", "Market Making", "Order Matching", "Settlement Systems", "Blockchain", "Financial Trading", "Liquidity Provision", "metaverse", "Price Discovery" ] }, { "id": "digital-asset-workflow", "title": "Digital Asset Workflow", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Asset Workflow is a type of Spatial Computing in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-asset-workflow", "labels": [ "Digital Asset Workflow" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Approval Workflow", "Asset Monetization", "Content Creation Pipeline", "Content Distribution", "Content Management System", "Distribution System", "SMPTE ST 2128", "Version Control", "Asset Archive", "Asset Registry", "Blockchain", "Blockchain Infrastructure", "CreativeMediaDomain", "Creator Economy", "DataLayer", "Digital Goods", "Digital Rights Management", "Metadata Management", "MiddlewareLayer", "NFT Minting" ] }, { "id": "digital-asset", "title": "Digital Asset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital asset is any electronically stored item of value that carries ownership rights and can be transferred, traded, or programmatically controlled \u2014 encompassing cryptocurrencies, tokenised securities, non-fungible tokens, stablecoins, and programmable financial instruments. Digital assets achieve verifiable scarcity and ownership through cryptographic proofs, with blockchain-based variants recorded immutably on distributed ledgers and governed by smart contracts. The class spans both on-chain native assets (e.g., BTC, ETH) and real-world asset tokenisations (RWAs), where legal property rights are encoded into blockchain representations enabling fractional ownership, automated compliance, and atomic settlement. Regulatory frameworks globally now classify digital assets across securities, commodities, payment instruments, and utility categories, shaping issuance, custody, and trading obligations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-asset", "labels": [ "Digital Asset", "Digital Assets", "Digital-Asset", "DigitalAsset", "Onyx Digital Assets" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "digital-avatar", "title": "Digital Avatar", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Digital Avatar is a persistent digital representation of a user or entity within virtual environments, exhibiting embodied presence, visual customisation, and behavioural agency. Digital avatars serve as identity anchors across metaverse platforms and enable social presence, interactive expression, and blockchain-verified ownership of associated assets.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:digital-avatar", "labels": [ "Digital Avatar", "DigitalAvatar" ], "is_subclass_of": [ "Content and Assets", "Interaction Technology" ], "wikilinks": [ "Behavioural Agency", "DigitalFashion", "dt:animatedBy", "dt:authenticatedBy", "dt:enhancedBy", "dt:ownedVia", "dt:tradedOn", "Embodied Presence", "hasAppearance", "Metaverse Standards Forum", "NFT", "NFTMarketplace", "performsAction", "representsUser", "Virtual Environments", "Visual Customisation", "wearsItem", "AvatarCustomization", "DID Nostr Identity", "DigitalIdentity" ] }, { "id": "digital-certificate", "title": "Digital Certificate", "domain": "security", "domain_name": "Security", "definition": "A cryptographic credential issued by a Certificate Authority that binds a public key to an identified entity, authenticates users, and secures transactions across networks, metaverse platforms, and blockchain systems through public key infrastructure and verifiable attestations.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:digital-certificate", "labels": [ "Digital Certificate", "Digital Certificates", "DigitalCertificate" ], "is_subclass_of": [ "Digital Security" ], "wikilinks": [ "Certificate Authority", "Secure Communication", "Trust Establishment", "Cryptographic Keys", "DID Nostr Identity", "Digital Security", "Identity Verification", "metaverse", "Public Key Infrastructure" ] }, { "id": "digital-cinema", "title": "Digital Cinema", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Digital cinema is the production, distribution, and theatrical projection of motion pictures using digital files rather than photochemical film. It is governed by the DCI specification, which defines the Digital Cinema Package (DCP) format, JPEG 2000 image compression, encryption, and colour standards for cinema-grade exhibition. It enables consistent high-resolution playback, secure content delivery, and the broader shift to file-based proprietary video pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-cinema", "labels": [ "Digital Cinema" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "digital-citizens-assembly", "title": "Digital Citizens' Assembly", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Citizens' Assembly is a type of Metaverse governance and safeguarding in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-citizens-assembly", "labels": [ "Digital Citizens' Assembly", "Citizen Assembly" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Civic Engagement Platform", "Collective Decision-Making", "Decision Recording System", "Deliberation Agent", "Democratic Deliberation", "Democratic Governance System", "Distributed Voting", "Multi-Agent Coordination", "Policy Synthesis Engine", "Secure Communication", "Sensor Input", "UN Habitat Digital Civics", "Voting Mechanism", "Consensus Protocol", "Identity Verification", "MiddlewareLayer", "Participant Management System", "Participatory Policy Making", "Transparent Governance", "VirtualSocietyDomain" ] }, { "id": "digital-citizenship", "title": "Digital Citizenship", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Citizenship is a type of Virtual Society in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-citizenship", "labels": [ "Digital Citizenship" ], "is_subclass_of": [ "Governance and Safety", "Virtual Society" ], "wikilinks": [ "Access Controls", "Access to Services", "Civic Duties", "Community Membership", "Community Voting", "GDPR", "IEEE Digital Identity Standards", "Privacy Protection", "Social Interaction", "UN Digital Rights Framework", "ApplicationLayer", "Civic Participation", "Community Governance Model", "Digital Constitution", "Digital Rights", "Governance Token", "Identity Management", "Identity Verification", "Legal Framework", "Metaverse Platform" ] }, { "id": "digital-collectible", "title": "Digital Collectible", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital collectible is a uniquely identifiable digital item whose scarcity, ownership and authenticity are recorded on a blockchain, allowing it to be owned, displayed and traded much like a physical collectible. Typically issued as a non-fungible token or an inscription, each collectible carries provenance and metadata that distinguish it from copies of the same media. Digital collectibles span art, trading cards, in-game items and membership artefacts, and they form a core consumer use case for blockchain-based digital assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:digital-collectible", "labels": [ "Digital Collectible" ], "is_subclass_of": [ "Non-Fungible Token" ], "wikilinks": [] }, { "id": "digital-commerce", "title": "Digital Commerce", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Commerce encompasses the exchange of goods, services, and digital assets through internet-connected and virtual platforms, including metaverse storefronts, NFT marketplaces, and cryptocurrency-denominated payment rails. It extends conventional e-commerce by incorporating programmable smart contracts for trustless settlement, tokenised ownership of virtual goods, and AI-mediated personalisation. In metaverse contexts, digital commerce enables low-friction cross-border value transfer without leaving the virtual environment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-commerce", "labels": [ "Digital Commerce" ], "is_subclass_of": [ "Platform and Environment", "Digital Asset Ecosystem" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-commons", "title": "Digital Commons", "domain": "governance", "domain_name": "Governance", "definition": "The digital commons are shared, non-rivalrous information resources that a community collectively produces, governs, and stewards under open licences and self-defined rules, rather than through private enclosure or state provision. Examples include open-source software, open data, free knowledge repositories, and open scientific outputs, often coordinated via peer production and commons-based governance. As a governance concept it bridges classical commons theory with digital public goods, examining how communities sustain shared resources, prevent enclosure, and align incentives in distributed and blockchain-supported settings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-commons", "labels": [ "Digital Commons" ], "is_subclass_of": [ "Public Goods" ], "wikilinks": [] }, { "id": "digital-constitution", "title": "Digital Constitution", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Constitution is a type of Virtual Society in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-constitution", "labels": [ "Digital Constitution" ], "is_subclass_of": [ "Governance and Safety", "Virtual Society" ], "wikilinks": [ "Amendment Procedure", "Aragon Constitutional Framework", "Bill of Rights", "Checks and Balances", "Constitution DAO", "Constitutional Rights", "Dispute Resolution Process", "Enforcement Mechanism", "ISO 37001 Governance", "Judicial System", "Legitimate Authority", "Rights Protection", "UN Digital Rights Framework", "ApplicationLayer", "Blockchain", "Blockchain Infrastructure", "Community Governance Model", "Consensus Mechanism", "Decentralized Autonomous Organization", "Democratic Governance" ] }, { "id": "digital-content-creation", "title": "Digital Content Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of authoring, editing, and producing digital assets\u2014including 3D models, textures, audio, video, and interactive experiences\u2014using software toolchains. In spatial computing and metaverse contexts, digital content creation encompasses generative AI assistance, physically-based material authoring, and export pipelines targeting real-time rendering engines.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:digital-content-creation", "labels": [ "Digital Content Creation", "Digital Content Creation Tools" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-content-layer", "title": "Digital Content Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An architectural abstraction within metaverse and spatial computing systems that organizes and manages digital assets, 3D objects, interactive media, and user-generated content as discrete layers that can be rendered, composed, and manipulated independently within virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-content-layer", "labels": [ "Digital Content Layer" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Architecture" ], "wikilinks": [ "Content Composition", "Dynamic Rendering", "Layer Management", "Asset Management", "Computer Vision", "metaverse", "Metaverse Architecture", "Rendering Engine", "Spatial Mapping" ] }, { "id": "digital-content-overlay", "title": "Digital Content Overlay", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Content Overlay is a type of Augmented Reality in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-content-overlay", "labels": [ "Digital Content Overlay" ], "is_subclass_of": [ "Display and Rendering", "Augmented Reality" ], "wikilinks": [ "Camera Systems", "Information Display", "Interactive Guidance", "Sensor Input", "SLAM Technology", "Spatial Annotation", "Augmented Reality", "Display Hardware", "metaverse" ] }, { "id": "digital-content-provenance-marking", "title": "Digital Content Provenance Marking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Techniques for embedding imperceptible or visible signals into digital content\u2014such as text, images, audio, or AI-generated outputs\u2014to assert provenance, ownership, or authenticity. AI watermarking approaches include statistical token-distribution biasing (for large language models) and frequency-domain embedding (for images), enabling detection of machine-generated content and supporting intellectual property protection and content provenance verification.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-content-provenance-marking", "labels": [ "Digital Content Provenance Marking", "Watermarks" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "digital-content", "title": "Digital Content", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Content encompasses any information or media encoded in digital form, including text, images, audio, video, 3D models, and interactive experiences. Within spatial computing platforms, digital content is the primary artefact that users create, exchange, and monetise, often represented as NFTs or other digital assets with embedded provenance.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:digital-content", "labels": [ "Digital Content" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-curation-platform", "title": "Digital Curation Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Curation Platform is a type of Digital Asset Management in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-curation-platform", "labels": [ "Digital Curation Platform" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Collection Management", "Content Discovery", "Content Repository", "Cultural Heritage Preservation", "ISO 21127", "Metadata Manager", "Preservation Engine", "Preservation Policy", "Public Access", "UNESCO Digital Heritage", "Versioning System", "Access Control", "Application Layer", "Authentication Service", "Blockchain", "ComputationAndIntelligenceDomain", "CreativeMediaDomain", "Data Layer", "Digital Asset Management", "Long-Term Archival" ] }, { "id": "digital-currency", "title": "Digital Currency", "domain": "finance", "domain_name": "Finance", "definition": "Any form of monetary value that exists purely in electronic format, encompassing central bank digital currencies, cryptocurrencies, stablecoins, and virtual currencies used within metaverse economies for transactions, payments, and value exchange without physical representation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-currency", "labels": [ "Digital Currency" ], "is_subclass_of": [ "Financial Technology" ], "wikilinks": [ "Programmable Money", "Security Infrastructure", "Transaction Network", "Virtual Transactions", "Blockchain", "Digital Payments", "Digital Wallet", "Financial Technology", "metaverse" ] }, { "id": "digital-democracy", "title": "Digital Democracy", "domain": "governance", "domain_name": "Governance", "definition": "Digital democracy is the use of digital technologies to broaden and deepen citizen participation in political decision-making, including online deliberation, e-petitions, participatory budgeting, and novel voting mechanisms. It seeks to make governance more transparent, inclusive, and responsive by lowering the barriers to engagement. It encompasses both incremental e-government tools and experimental collective-choice systems such as quadratic voting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:digital-democracy", "labels": [ "Digital Democracy" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "digital-divide", "title": "Digital Divide", "domain": "governance", "domain_name": "Governance", "definition": "The digital divide describes the structured inequality between individuals, communities, and nations in their ability to access, afford, use, and benefit from information and communication technologies, including broadband connectivity, computing devices, and the digital literacy skills required for effective participation. The divide operates across multiple axes \u2014 geographic (urban versus rural), socioeconomic (income and education), demographic (age, gender, disability), and geopolitical (Global North versus Global South) \u2014 and is self-reinforcing because limited digital access restricts access to the economic and educational resources needed to close the gap. Policy interventions target three layers: physical infrastructure deployment, affordability of devices and data, and cultivation of digital skills and trust through digital literacy programmes. As digital systems mediate healthcare, education, employment, public services, and civic participation, the digital divide functions as a meta-inequality that amplifies pre-existing social disparities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-divide", "labels": [ "Digital Divide", "North-South Divide" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [ "Telecommunications", "Digital Transformation", "Privacy", "Digital Economy" ] }, { "id": "digital-dualism", "title": "Digital Dualism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Dualism is the conceptual framework positing a fundamental, hierarchical separation between digital/virtual and physical/offline experience domains, treating them as mutually exclusive rather than mutually constitutive. Critiqued by Nathan Jurgenson (2011), who proposed an augmented reality perspective recognising that digital and physical are increasingly entangled: digital interactions produce physical consequences, and physical activities generate digital traces. The critique is particularly relevant to metaverse and telecollaboration design, where dualist assumptions lead to poorly integrated hybrid experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-dualism", "labels": [ "Digital Dualism" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Blended Reality", "Digital Embodiment", "Hybrid Space", "Phygital Experience", "Augmented Reality", "Metaverse", "Telecollaboration" ] }, { "id": "digital-economy", "title": "Digital Economy", "domain": "finance", "domain_name": "Finance", "definition": "The interconnected system of economic activities, transactions, and value creation occurring through digital platforms, blockchain networks, and virtual environments, encompassing cryptocurrency markets, NFT trading, virtual real estate, and the tokenised exchange of goods and services in the metaverse.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-economy", "labels": [ "Digital Economy" ], "is_subclass_of": [ "Economic Systems" ], "wikilinks": [ "Digital Value Exchange", "Blockchain", "Blockchain Infrastructure", "Digital Payments", "Economic Systems", "metaverse", "Smart Contracts", "Token Economics", "Virtual Commerce" ] }, { "id": "digital-elevation-model", "title": "Digital Elevation Model", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:digital-elevation-model", "labels": [ "Digital Elevation Model" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "digital-entertainment", "title": "Digital Entertainment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Entertainment encompasses interactive and passive media experiences delivered through digital platforms, including video games, streaming services, virtual concerts, and immersive XR content. Within the metaverse context, digital entertainment drives user engagement and economic activity, often integrating blockchain-based ownership of digital assets and in-world economies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-entertainment", "labels": [ "Digital Entertainment" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-ethics", "title": "Digital Ethics", "domain": "governance", "domain_name": "Governance", "definition": "The branch of applied ethics that examines how digital technologies \u2014 data collection, algorithms, platforms, connected devices, and AI systems \u2014 affect human values such as autonomy, privacy, fairness, dignity, and wellbeing, and that develops principles, design methods, and governance instruments to align technology development with those values. It underpins frameworks including the EU HLEG Ethics Guidelines for Trustworthy AI and value-based engineering standards such as IEEE 7000.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:digital-ethics", "labels": [ "Digital Ethics" ], "is_subclass_of": [ "Ethics" ], "wikilinks": [ "Ethics", "AI Ethics", "Value-Sensitive Design", "IEEE 7000" ] }, { "id": "digital-euro", "title": "Digital Euro", "domain": "finance", "domain_name": "Finance", "definition": "The digital euro is a prospective retail central bank digital currency (CBDC) to be issued by the European Central Bank, providing eurozone households and businesses with a risk-free digital form of public money usable for everyday payments alongside cash, without requiring a bank account at a commercial institution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:digital-euro", "labels": [ "Digital Euro" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [] }, { "id": "digital-evidence-chain-of-custody", "title": "Digital Evidence Chain of Custody", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Digital Evidence Chain of Custody is a type of Legal Framework in the infrastructure domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-evidence-chain-of-custody", "labels": [ "Digital Evidence Chain of Custody" ], "is_subclass_of": [ "Legal and Regulatory", "Legal Framework" ], "wikilinks": [ "Audit Log", "Blockchain Ledger", "Evidence Collection Protocol", "Evidence Integrity Verification", "Forensic Investigation", "ISO 27037", "Legal Admissibility", "Secure Storage", "Tamper Detection", "Timestamp Authority", "Access Control", "Blockchain", "Cryptographic Hash", "DataLayer", "Digital Forensics Framework", "Digital Signature", "Identity Verification", "Legal Framework", "MiddlewareLayer", "Non-Repudiation" ] }, { "id": "digital-experience", "title": "Digital Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Experience encompasses the totality of interactions a user has with digital products, services, and environments \u2014 spanning web, mobile, and immersive (XR) surfaces. It integrates user interface design, content delivery, personalisation, and spatial computing to shape how users perceive and engage with digital platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-experience", "labels": [ "Digital Experience" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Sensor Input", "owl:Thing" ] }, { "id": "digital-fabrication", "title": "Digital Fabrication", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital Fabrication, in the AI context, integrates artificial intelligence with additive manufacturing, CNC machining, and robotic production technologies to optimise fabrication processes. AI enables generative design for novel geometries, predictive maintenance via sensor analytics, real-time quality control through computer vision, and adaptive toolpath planning using reinforcement learning. Machine learning models predict material behaviour, detect defects, and enable mass customisation through digital twin simulation before physical production.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-fabrication", "labels": [ "Digital Fabrication" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Generative Design", "Process Optimization", "Autonomous Robot", "Computer Vision", "Digital Twin", "Robotics" ] }, { "id": "digital-forensics-framework", "title": "Digital Forensics Framework", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured modology and toolset for identifying, preserving, analysing, and documenting digital evidence from computing systems, networks, and virtual environments to support cybersecurity investigations, legal proceedings, and incident response within metaverse and blockchain contexts.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-forensics-framework", "labels": [ "Digital Forensics Framework" ], "is_subclass_of": [ "Governance and Safety", "Cybersecurity" ], "wikilinks": [ "Chain of Custody", "Data Integrity", "Evidence Collection", "Forensic Tools", "Incident Investigation", "Legal Documentation", "NIST Framework", "Blockchain", "Cybersecurity", "metaverse" ] }, { "id": "digital-forensics", "title": "Digital Forensics", "domain": "security", "domain_name": "Security", "definition": "Digital forensics is the scientific discipline concerned with the identification, preservation, extraction, analysis, and presentation of digital evidence from computing systems, networks, storage media, and connected devices in a manner that is legally admissible and reproducible. It applies structured methodologies \u2014 including write-blocked acquisition, cryptographic hashing for evidence integrity, and chain-of-custody documentation \u2014 to support criminal investigations, civil litigation, incident response, and regulatory compliance inquiries. Sub-disciplines include network forensics, mobile forensics, memory forensics, and cloud forensics, each requiring specialist tools and legal frameworks adapted to the peculiarities of the evidence medium. Practitioner outputs must survive rigorous judicial scrutiny, compelling both technical rigour and defensible, auditable process documentation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-forensics", "labels": [ "Digital Forensics" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "digital-gaming", "title": "Digital Gaming", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Gaming encompasses interactive software experiences delivered across platforms including PCs, consoles, and mobile devices, encompassing game design, virtual economies, player interaction systems, and increasingly blockchain-based asset ownership. It represents a convergence of entertainment, spatial computing, and digital economy mechanisms within persistent or session-based virtual environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-gaming", "labels": [ "Digital Gaming" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-gold", "title": "Digital Gold", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital Gold is a conceptual and investment category that characterises a cryptocurrency\u2014most commonly Bitcoin\u2014as a bearer instrument that performs the monetary functions traditionally attributed to gold: a scarce, durable, fungible, portable, and divisible store of value that preserves purchasing power across time and political regimes without dependence on any issuing authority. The analogy rests on Bitcoin's algorithmically enforced supply cap of 21 million coins, its proof-of-work scarcity mechanism, and its censorship resistance, which proponents argue replicate or improve upon gold's monetary properties in a digital-native form. The digital gold thesis underpins institutional Bitcoin investment strategies and shapes regulatory and macroeconomic discourse about the role of cryptographic assets in global monetary systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:digital-gold", "labels": [ "Digital Gold" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [] }, { "id": "digital-goods-registry", "title": "Digital Goods Registry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Goods Registry is a type of Spatial Computing in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-goods-registry", "labels": [ "Digital Goods Registry" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Asset Catalog", "Asset Discovery", "ETSI ARF 010", "Metadata Store", "OMA3 Media WG", "Ownership Records", "Ownership Transfer", "Provenance Tracker", "API Gateway", "Authentication Service", "Blockchain", "Blockchain Infrastructure", "Cross-Platform Interoperability", "Data Layer", "Data Storage", "Identity Provider", "Marketplace Integration", "Middleware Layer", "Provenance Verification", "Search Index" ] }, { "id": "digital-goods", "title": "Digital Goods", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual items and assets that can be owned, transferred, traded, or used within metaverse environments, typically with provable scarcity and verifiable ownership recorded on distributed ledgers; they span avatar wearables, virtual land, in-game items, and creative works monetised through creator economies and virtual commerce.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-goods", "labels": [ "Digital Goods", "DigitalGoods" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Creator Monetization", "Metadata", "Metaverse 101", "Ownership Token", "Usage Rights", "User Ownership", "ApplicationLayer", "Asset Registry", "Asset Trading", "Blockchain", "Blockchain Infrastructure", "Creator Economy", "Digital Asset", "Digital Rights Management", "Digital Wallet", "MiddlewareLayer", "NFT Standards", "Smart Contracts", "Virtual Commerce", "Virtual Economy" ] }, { "id": "digital-governance", "title": "Digital Governance", "domain": "governance", "domain_name": "Governance", "definition": "Digital Governance is the set of policies, institutional frameworks, regulatory instruments, and accountability mechanisms through which societies, organisations, and governments manage the development, deployment, and societal impact of digital technologies and data systems. It encompasses rule-making for the internet, data protection, platform regulation, algorithmic accountability, and cybersecurity policy. Digital Governance operates at multiple levels \u2014 international, national, and organisational \u2014 and draws on both technical standards and legal norms to align technological capability with public values. As digital infrastructure becomes critical to economic and social life, effective digital governance balances innovation incentives with risk mitigation, rights protection, and equitable access.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-governance", "labels": [ "Digital Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-health", "title": "Digital Health", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital health is the use of information and communications technology, sensors, data, and artificial intelligence to deliver, manage, and improve healthcare and wellbeing. It spans telemedicine, remote patient monitoring, electronic health records, wearable biosensors, clinical decision support, and AI-assisted diagnostics. The field aims to widen access, personalise care, and improve outcomes while raising acute requirements for patient privacy, data security, and regulatory oversight.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-health", "labels": [ "Digital Health" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [] }, { "id": "digital-heritage", "title": "Digital Heritage", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The digitisation, preservation, and interactive presentation of cultural artefacts, historical sites, and intangible heritage through spatial computing technologies such as 3D reconstruction, AR, and VR. Digital heritage enables remote access, scholarly analysis, and public engagement with cultural memory that may otherwise be physically inaccessible or at risk of loss.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-heritage", "labels": [ "Digital Heritage", "Heritage Digitisation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-human-avatar-representation", "title": "Digital Human Avatar Representation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The combined domain encompassing the creation, animation, and management of digital representations of human beings in virtual environments, including photorealistic digital humans, stylised avatars, and fictional characters. This domain integrates motion capture, AI-driven face and body generation, lip-synchronisation, and avatar portability standards to produce consistent, expressive agents across metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-human-avatar-representation", "labels": [ "Digital Human Avatar Representation", "Humans, Avatars , Character" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "digital-human-technology", "title": "Digital Human Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A class of technologies for creating photorealistic or stylised computer-generated human representations, encompassing 3D body and face modelling, motion capture-driven animation, procedural skin and cloth simulation, and real-time rendering pipelines. Digital human technology enables believable avatars, virtual actors, and AI-driven conversational agents in spatial computing, entertainment, and training applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-human-technology", "labels": [ "Digital Human Technology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "DID Nostr Identity", "owl:Thing" ] }, { "id": "digital-humans", "title": "Digital Humans", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital humans are photorealistic, AI-driven virtual beings that replicate human appearance, movement, voice, and conversational behaviour, rendered in 3D and designed for real-time interaction within immersive and spatial computing environments. They integrate natural language processing, motion capture, generative AI, and real-time rendering to simulate lifelike presence, serving roles in customer service, education, healthcare, and entertainment. As an emerging category they represent the convergence of embodied AI with avatar systems and synthetic media.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:digital-humans", "labels": [ "Digital Humans", "Digital Human" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "DID Nostr Identity", "MetaverseDomain" ] }, { "id": "digital-identity-framework", "title": "Digital Identity Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Digital Identity Framework is a type of Infrastructure in the infrastructure domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-identity-framework", "labels": [ "Digital Identity Framework" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Cryptographic Systems", "eIDAS 2.0", "Identity Policies", "ISO/IEC 24760", "Policy Frameworks", "Privacy Controls", "Privacy Protection", "Secure Authentication", "Trust Mechanisms", "Authentication Standards", "Cross-Platform Identity", "DataLayer", "DID Nostr Identity", "Digital Identity Management", "Governance Framework", "MiddlewareLayer", "Standardization Bodies", "TrustAndGovernanceDomain", "Trust Architecture" ] }, { "id": "digital-identity-management", "title": "Digital Identity Management", "domain": "security", "domain_name": "Security", "definition": "The systems, processes, and technologies for creating, maintaining, and verifying digital representations of individuals and entities across virtual environments, incorporating self-sovereign identity principles, decentralized identifiers, and verifiable credentials for secure cross-platform authentication.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:digital-identity-management", "labels": [ "Digital Identity Management" ], "is_subclass_of": [ "Identity Systems" ], "wikilinks": [ "Credential Storage", "eIDAS 2.0", "Identity Providers", "ISO 27001", "Metaverse Standards Forum", "Privacy Control", "User Authentication", "W3C DID Core 1.0", "Cross-Platform Identity", "Cryptographic Keys", "DID Nostr Identity", "Identity Systems", "metaverse", "Verifiable Credentials" ] }, { "id": "digital-identity-standards", "title": "Digital Identity Standards", "domain": "security", "domain_name": "Security", "definition": "The technical specifications, protocols, and frameworks established by standards bodies such as W3C and ISO that define interoperable formats for digital identifiers, verifiable credentials, and authentication mechanisms enabling secure identity management across metaverse platforms and decentralized systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:digital-identity-standards", "labels": [ "Digital Identity Standards" ], "is_subclass_of": [ "Security and Identity", "Technical Standards" ], "wikilinks": [ "Conformance Testing", "Credential Exchange", "eIDAS 2.0", "Identity Interoperability", "ISO 27001", "Metaverse Standards Forum", "Protocol Specifications", "Standards Bodies", "Trust Frameworks", "W3C DID Core 1.0", "DID Nostr Identity", "metaverse", "Technical Standards", "Verifiable Credentials" ] }, { "id": "digital-identity-verification", "title": "Digital Identity Verification", "domain": "security", "domain_name": "Security", "definition": "Digital Identity Verification is the remote, automated process of confirming that a person's claimed identity corresponds to a real individual and that the claimant is who they say they are, typically through document analysis, biometric matching, liveness detection, and database cross-referencing. It underpins Know Your Customer (KYC) compliance, onboarding workflows, and access control in digital services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-identity-verification", "labels": [ "Digital Identity Verification" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "digital-identity-wallet", "title": "Digital Identity Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Digital Identity Wallet is a software application \u2014 typically running on a smartphone, secure element, or cloud-hosted enclave \u2014 that stores, manages, and selectively presents cryptographically signed digital credentials (verifiable credentials, mobile driving licences, electronic identity atte...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-identity-wallet", "labels": [ "Digital Identity Wallet", "EU Digital Identity Wallet", "European Digital Identity Wallet", "Identity Wallet" ], "is_subclass_of": [ "Network Component", "Token and Asset", "Identity Management System", "Cryptographic Key Container", "Mobile Application", "Personal Data Store", "Self-Sovereign Identity Component" ], "wikilinks": [ "AgID IT Wallet Specifications 2024", "Age Verification", "BBS+ Signature", "BBS Signature Scheme IETF CFRG", "Biometric Authentication", "Biometric Authentication Subsystem", "Bluetooth Low Energy", "Boneh Boyen Shacham 2004 Short Group Signatures", "Camenisch & Lysyanskaya 2002 CL Signatures", "Centralized Identity Provider", "Christopher Allen 2016 Path to Self-Sovereign Identity", "ClientLayer", "Credential Portability", "Credential Schema Registry", "Cross-Border Identity Verification", "Cryptographic Key Container", "Cryptographic Library", "CryptographyDomain", "Decentralized Identity Foundation", "Deep Linking" ] }, { "id": "digital-identity", "title": "Digital Identity", "domain": "security", "domain_name": "Security", "definition": "A digital representation of an entity encompassing personally identifiable information, behavioral data, credentials, and authentication attributes that enables individuals and organizations to establish presence, ownership, and trust within virtual environments, blockchain networks, and metaverse platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-identity", "labels": [ "Digital Identity", "Cross-Border Digital Identity", "DigitalIdentity" ], "is_subclass_of": [ "Identity" ], "wikilinks": [ "Authentication", "Authorization", "Credential Storage", "Digital Presence", "Privacy Controls", "DID Nostr Identity", "Identity", "Identity Verification", "metaverse" ] }, { "id": "digital-inclusion", "title": "Digital Inclusion", "domain": "governance", "domain_name": "Governance", "definition": "Digital Inclusion is the set of policies and practices that ensure all individuals and communities, especially the disadvantaged, have meaningful access to and the ability to use digital technologies. It addresses affordable connectivity, accessible devices and services, relevant skills and trustworthy content so that no group is left behind as services move online. Digital Inclusion is the constructive response to the digital divide and a precondition for equitable participation in modern civic and economic life.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-inclusion", "labels": [ "Digital Inclusion" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "digital-infrastructure", "title": "Digital Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The foundational computing, networking, and storage systems that underpin large-scale digital services and telecollaboration platforms, including hyperscale data centres with GPU clusters, content delivery networks, software-defined networking, and edge computing nodes. Digital infrastructure is increasingly defined by cloud-native patterns\u2014containerisation, orchestration, and infrastructure-as-code\u2014enabling programmatic resource allocation and resilient, globally distributed deployments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:digital-infrastructure", "labels": [ "Digital Infrastructure", "Digital-Infrastructure", "Member State Digital Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "CNCF", "IEEE", "IETF", "Linux Foundation", "Open Compute Project", "Blockchain" ] }, { "id": "digital-jurisdiction", "title": "Digital Jurisdiction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Jurisdiction is a type of Metaverse governance and safeguarding in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:digital-jurisdiction", "labels": [ "Digital Jurisdiction" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Conflict Resolution Mechanism", "Cross-Border Enforcement", "Dispute Resolution", "Enforcement Mechanism", "International Jurisdiction and the Internet Working Group", "Legal Authority", "Legal Entity", "Legal System", "Multi-Jurisdictional Coordination", "Regulatory Authority", "Sovereignty Model", "UNCITRAL Model Law on Electronic Commerce", "ApplicationLayer", "Blockchain", "Digital Identity", "Governance Framework", "Governance Token", "Jurisdictional Boundary", "Platform Governance", "Regulatory Framework" ] }, { "id": "digital-marketing", "title": "Digital Marketing", "domain": "data", "domain_name": "Data", "definition": "Digital Marketing is the practice of promoting products, services, and brands through internet-connected and data-driven channels \u2014 including search engines, social media platforms, email, content networks, and programmatic advertising systems \u2014 to reach and convert target audiences. It encompasses both inbound strategies such as search engine optimisation and content marketing, and outbound strategies such as paid search, display advertising, and email campaigns. Modern digital marketing is distinguished by its measurability, real-time feedback loops, and ability to personalise messages at scale through machine learning, behavioural analytics, and customer data platforms. It intersects with artificial intelligence for predictive targeting, with data infrastructure for audience segmentation and attribution, and with emerging channels such as augmented reality and conversational interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:digital-marketing", "labels": [ "Digital Marketing" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-marketplace", "title": "Digital Marketplace", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A virtual platform enabling the discovery, purchase, sale, and exchange of digital assets, NFTs, virtual goods, and services within metaverse environments, utilising blockchain technology for transparent transactions, ownership verification, and decentralised commerce.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-marketplace", "labels": [ "Digital Marketplace" ], "is_subclass_of": [ "Platform and Environment", "E-Commerce" ], "wikilinks": [ "Digital Wallets", "NFT Trading", "Virtual Asset Exchange", "Blockchain", "Blockchain Infrastructure", "Creator Economy", "E-Commerce", "metaverse", "Smart Contracts" ] }, { "id": "digital-markets-act", "title": "Digital Markets Act", "domain": "governance", "domain_name": "Governance", "definition": "The Digital Markets Act (DMA) is a European Union regulation that imposes ex ante obligations and prohibitions on large online platforms designated as gatekeepers to ensure contestable and fair digital markets. Rather than waiting for case-by-case competition enforcement, it sets binding rules on designated core platform services covering interoperability, data use, self-preferencing, default settings, and access for business users. Gatekeepers must comply with a list of do's and don'ts, with the European Commission as sole enforcer empowered to levy substantial fines and structural remedies for non-compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-markets-act", "labels": [ "Digital Markets Act" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "digital-model", "title": "Digital Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Digital Model is a structured computational representation of a real-world or conceptual entity, capturing geometry, behaviour, and semantic attributes for use in simulation, visualisation, or analysis. Digital models underpin virtual environments, digital twins, and spatial computing applications by providing machine-readable, interoperable representations of physical or abstract objects.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-model", "labels": [ "Digital Model" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-modeling", "title": "Digital Modeling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Modelling is the process of creating mathematical or geometric representations of physical objects, environments, or systems in a computer, encompassing polygonal mesh construction, NURBS surfaces, procedural generation, and scan-based reconstruction. Digital models serve as the foundation for real-time rendering, simulation, digital twin creation, and asset delivery in spatial computing platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-modeling", "labels": [ "Digital Modeling" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-money", "title": "Digital Money", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Any form of monetary value existing purely in electronic format, encompassing e-money, central bank digital currencies, cryptocurrencies, and stablecoins, that can be stored, transferred, and transacted electronically across payment networks, metaverse economies, and blockchain systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-money", "labels": [ "Digital Money" ], "is_subclass_of": [ "Platform and Environment", "Financial Instruments" ], "wikilinks": [ "Electronic Payments", "Instant Settlement", "Payment Network", "Programmable Money", "Security Infrastructure", "Blockchain", "Digital Wallet", "Financial Instruments", "metaverse" ] }, { "id": "digital-objects", "title": "Digital Objects", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Digital Objects is an umbrella ontology concept covering three closely-related but historically distinct senses of \"an addressable, transferable, identifiable unit of digital content\": (1) Digital Object Architecture (DOA), the formal information-infrastructure paradigm proposed by Robert E.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-objects", "labels": [ "Digital Objects", "Digital Object" ], "is_subclass_of": [ "Data Management", "Information Resource", "Digital Artefact", "Networked Resource", "Identified Entity" ], "wikilinks": [ "ARK", "Arweave", "Authentication", "Bitcoin Ordinals", "Buterin Hitzig Weyl 2022 Decentralized Society Soulbound Tokens", "C2PA", "C2PA Technical Specification 2.0", "Catlow et al. 2017 Artists Rethinking the Blockchain", "Chainalysis 2024 NFT Market Report", "Citation", "Cryptographic Binding", "Cultural Heritage", "Database Row", "De Filippi & Wright 2018 Blockchain and the Law", "Decentralised Identifier", "Decentralised Identity", "Digital Art", "Digital Artefact", "Digital Object Architecture", "Digital Objects (NFT)" ] }, { "id": "digital-onboarding", "title": "Digital Onboarding", "domain": "security", "domain_name": "Security", "definition": "Digital onboarding is the remote, self-service process by which an organisation enrols a new customer or user, capturing and verifying their identity entirely through digital channels rather than in person. It typically combines document capture, biometric checks, liveness detection and database lookups to satisfy regulatory know-your-customer obligations while minimising friction. The aim is to convert a prospect into a verified, account-holding user quickly and securely.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-onboarding", "labels": [ "Digital Onboarding" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [] }, { "id": "digital-ontology-repository", "title": "Digital Ontology Repository", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Ontology Repository is a type of Metadata Repository in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-ontology-repository", "labels": [ "Digital Ontology Repository" ], "is_subclass_of": [ "Standards and Interoperability", "Metadata Repository" ], "wikilinks": [ "Cross-Domain Integration", "Database Management System", "ISO/IEC 11179", "MSF Register WG", "Ontology Reuse", "Ontology Storage System", "Query Interface", "Semantic Interoperability", "Semantic Reasoning Engine", "URI Resolution Service", "Validation Engine", "Version Control", "W3C", "Access Control System", "Authentication Service", "ComputationAndIntelligenceDomain", "Computer Vision", "DataLayer", "Knowledge Sharing", "Metadata Registry" ] }, { "id": "digital-ownership", "title": "Digital Ownership", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The cryptographically verified right to possess, control, and transfer digital assets including NFTs, virtual real estate, in-game items, and tokenised content, established through blockchain technology and smart contracts that provide immutable proof of authenticity and provenance within metaverse economies and broader digital commerce contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-ownership", "labels": [ "Digital Ownership", "DigitalOwnership" ], "is_subclass_of": [ "Property Rights" ], "wikilinks": [ "Asset Transfer", "Creator Royalties", "Blockchain", "Blockchain Network", "Digital Wallet", "metaverse", "Property Rights", "Provenance Tracking", "Smart Contracts" ] }, { "id": "digital-payment-system", "title": "Digital Payment System", "domain": "finance", "domain_name": "Finance", "definition": "The integrated infrastructure of protocols, platforms, and financial instruments enabling secure monetary transactions, encompassing cryptocurrency payments, stablecoin transfers, fiat gateways, smart contract-based payment automation, and regulated payment rails for both physical and virtual commerce.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-payment-system", "labels": [ "Digital Payment System", "Digital Payment Infrastructure" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [ "Cross-Border Payments", "Payment Gateway", "Virtual Transactions", "Blockchain", "Blockchain Network", "Digital Wallet", "Financial Infrastructure", "metaverse", "Micropayments" ] }, { "id": "digital-payments", "title": "Digital Payments", "domain": "finance", "domain_name": "Finance", "definition": "Electronic monetary transactions executed through digital channels including blockchain networks, mobile wallets, and online platforms, enabling the transfer of value for goods, services, and assets within both traditional e-commerce systems and emerging metaverse economies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-payments", "labels": [ "Digital Payments", "Digital Payment" ], "is_subclass_of": [ "Financial Transactions" ], "wikilinks": [ "Cross-Border Transfer", "Payment Processor", "Virtual Purchases", "Blockchain", "Digital Wallet", "E-Commerce", "Financial Transactions", "metaverse", "Network Infrastructure" ] }, { "id": "digital-performance-capture", "title": "Digital Performance Capture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Performance Capture is a type of Creative Media Domain in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-performance-capture", "labels": [ "Digital Performance Capture" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Audio Recording Equipment", "Facial Capture System", "Live Performance", "Motion Capture System", "Optical Sensors", "Performance Animation", "Real-Time Solver", "Reality Capture Workflow", "Skeletal Animation", "SMPTE ST 2119", "Synchronization System", "Voice Recording System", "Character Rigging", "ComputeLayer", "Computer Vision", "CreativeMediaDomain", "Digital Actor Creation", "Marker-Based Tracking", "PhysicalLayer", "Real-Time Character Animation" ] }, { "id": "digital-platform", "title": "Digital Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Digital Platform is a software-mediated environment that enables interaction, transaction, and value exchange between multiple user groups, often exploiting network effects to grow. In spatial computing and Web3 contexts, digital platforms host virtual experiences, marketplaces, and decentralised applications, frequently integrating blockchain infrastructure for ownership and governance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-platform", "labels": [ "Digital Platform" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-pound", "title": "Digital Pound", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Digital Pound is a proposed central bank digital currency for the United Kingdom issued by the Bank of England. It is intended for retail payments and remains under design and consultation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:digital-pound", "labels": [ "Digital Pound" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [ "Bank of England", "Payment System", "CBDC Cross-Border Settlement", "Central Bank Digital Currency", "https://www.bankofengland.co.uk/the-digital-pound", "https://www.bankofengland.co.uk/paper/2023/the-digital-pound-consultation-paper" ] }, { "id": "digital-preservation", "title": "Digital Preservation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Digital Preservation is the managed set of processes, standards, and technologies applied to ensure that digital objects remain accessible, authentic, and usable over extended time periods, spanning decades or centuries. It encompasses format migration, bit-level integrity verification via checksumming, redundant and geographically distributed storage, and comprehensive provenance tracking through metadata standards such as OAIS and PREMIS. Effective digital preservation requires active management of hardware and software obsolescence, format dependency chains, and institutional policy frameworks that together prevent the loss or corruption of cultural, scientific, and organisational digital assets. It is distinguished from simple backup by its explicit concern with long-term interpretability and authenticity rather than mere bit-level recovery.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:digital-preservation", "labels": [ "Digital Preservation" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-product-passport", "title": "digital product passport", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "A Digital Product Passport (DPP) is a standardised, machine-readable digital record that captures and communicates authoritative information about a product's material composition, manufacturing origin, environmental footprint, repair history, and end-of-life options throughout its full lifecycle. Mandated across product categories by the EU Ecodesign for Sustainable Products Regulation (ESPR) and related instruments such as the EU Battery Regulation, DPPs are accessible to consumers, regulators, recyclers, and supply chain actors via data carriers such as QR codes, NFC tags, or RFID. They serve as a foundational infrastructure of circular economy policy by enabling traceability, transparency, and interoperability across complex global supply chains. DPPs are frequently implemented atop distributed ledger or verifiable credential technologies to ensure provenance integrity and tamper-evident audit trails.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:digital-product-passport", "labels": [ "Digital Product Passport", "EU Digital Product Passport", "Product Passport" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "digital-property-rights", "title": "Digital Property Rights", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Property Rights are the legally and technically enforceable claims governing ownership, transfer, and use of digital assets and virtual content. They encompass intellectual property protections, on-chain ownership assertions via non-fungible tokens, licensing frameworks, and governance rules that determine who can access, modify, or commercialise digital objects within virtual and mixed-reality environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-property-rights", "labels": [ "Digital Property Rights", "User-Owned Digital Property" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-public-goods", "title": "Digital Public Goods", "domain": "governance", "domain_name": "Governance", "definition": "Open-source software, open data, open AI models, open standards, and open content that adhere to privacy and other applicable laws and best practices, do no harm by design, and help attain sustainable development goals; formalised by the UN-endorsed Digital Public Goods Alliance standard, they are freely adoptable and adaptable building blocks for digital public infrastructure worldwide.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:digital-public-goods", "labels": [ "Digital Public Goods" ], "is_subclass_of": [ "Public Goods" ], "wikilinks": [ "Public Goods", "Digital Commons", "Open Source Software" ] }, { "id": "digital-real-estate", "title": "Digital Real Estate", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Real Estate is a type of Virtual Economy in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-real-estate", "labels": [ "Digital Real Estate", "Virtual Real Estate" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Advertising Space", "Development Rights", "Event Hosting", "Land Registry", "Metaverse 101", "Ownership Token", "Property Development", "Property Metadata", "ApplicationLayer", "Blockchain", "Blockchain Infrastructure", "Digital Wallet", "Land Parcel", "Metaverse Platform", "MiddlewareLayer", "NFT Standards", "Smart Contracts", "Spatial Computing", "Spatial Coordinates", "Virtual Commerce" ] }, { "id": "digital-regulation", "title": "Digital Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Digital Regulation is the body of law, policy instruments, technical standards, and enforcement mechanisms that govern the development, deployment, operation, and use of digital technologies \u2014 including artificial intelligence systems, online platforms, data-driven services, and digital communications infrastructure. It addresses algorithmic accountability, platform liability, data sovereignty, content moderation obligations, cybersecurity requirements, and cross-border regulatory harmonisation, aiming to balance innovation incentives with fundamental rights protections, market competition, and public safety objectives. Digital Regulation operates across multiple jurisdictional layers \u2014 supranational, national, and sector-specific \u2014 and increasingly relies on regulatory sandboxes, conformity assessments, and co-regulatory models that involve both government bodies and industry stakeholders. Its scope has expanded rapidly with the proliferation of AI, cloud computing, and large-scale data ecosystems, making it a central mechanism for translating societal values into enforceable technical and operational constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:digital-regulation", "labels": [ "Digital Regulation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-repository", "title": "Digital Repository", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A managed storage and access system for digital objects, assets, and metadata designed to ensure long-term preservation, discoverability, and integrity of digital content through standardised ingest, storage, and retrieval processes supporting metaverse archives and institutional collections.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-repository", "labels": [ "Digital Repository", "Trusted Digital Repository" ], "is_subclass_of": [ "Platform and Environment", "Data Management" ], "wikilinks": [ "Access Controls", "Content Access", "Blockchain", "Data Management", "Digital Preservation", "Metadata Management", "Metadata Standards", "metaverse", "Storage Infrastructure" ] }, { "id": "digital-rights-management-extended", "title": "Digital Rights Management (Extended)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Rights Management (Extended) is a type of Virtual Economy Domain in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-rights-management-extended", "labels": [ "Digital Rights Management (Extended)" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Access Control Module", "Content Delivery Network", "Content Encryption Engine", "Content Licensing", "Content Protection Infrastructure", "Identity Verification System", "ISO/IEC 21000 MPEG-21", "License Management System", "Payment Gateway", "Piracy Prevention", "Usage Rights Enforcement", "Usage Tracking System", "W3C Web DRM", "Blockchain", "Blockchain Network", "CreativeMediaDomain", "Cryptographic Key Management", "MiddlewareLayer", "Revenue Distribution", "Smart Contract" ] }, { "id": "digital-rights-management", "title": "Digital Rights Management", "domain": "governance", "domain_name": "Governance", "definition": "Technology systems and protocols that control access to, distribution of, and usage rights for digital content including media, software, and virtual assets, increasingly leveraging blockchain and smart contracts for transparent, decentralised rights enforcement and automated royalty distribution.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:digital-rights-management", "labels": [ "Digital Rights Management", "Rights Management", "Rights Management System", "Usage Rights Enforcement" ], "is_subclass_of": [ "Content Protection" ], "wikilinks": [ "Encryption", "License Management", "Piracy Prevention", "Access Control", "Blockchain", "Content Protection", "metaverse", "Royalty Distribution", "Smart Contracts" ] }, { "id": "digital-rights", "title": "Digital Rights", "domain": "governance", "domain_name": "Governance", "definition": "The human rights and freedoms applicable to digital contexts including privacy, data protection, freedom of expression, access to information, and digital ownership within virtual environments, metaverse platforms, and online spaces, increasingly codified through international frameworks and national legislation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-rights", "labels": [ "Digital Rights" ], "is_subclass_of": [ "Human Rights" ], "wikilinks": [ "Data Sovereignty", "Digital Freedom", "eIDAS", "Enforcement Mechanisms", "GDPR", "Privacy Protection", "Technical Safeguards", "Blockchain", "Digital Services Act", "EU AI Act", "Human Rights", "Legal Framework", "metaverse" ] }, { "id": "digital-ritual", "title": "Digital Ritual", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured virtual ceremonial process that recreates, adapts, or innovates traditional ritual practices in metaverse environments, enabling communities to perform symbolic cultural, religious, or social ceremonies through coordinated digital performances, shared virtual spaces, and meaningful p...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:digital-ritual", "labels": [ "Digital Ritual" ], "is_subclass_of": [ "Platform and Environment", "Virtual Community Practice" ], "wikilinks": [ "Audio-Visual Environment", "Ceremonial Space", "Commemoration Event", "Community Bonding", "Cultural Expression System", "Cultural Festival", "Cultural Protocol", "Digital Religion Studies", "Event Orchestration", "Initiation Rite", "Memorial Service", "Religious Ceremony", "Ritual Design", "Symbolic Enactment", "Synchronization Protocol", "Virtual Worlds Research", "ApplicationLayer", "Avatar System", "Community Governance", "Participant Authentication" ] }, { "id": "digital-safety", "title": "Digital Safety", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Safety encompasses the policies, technical controls, and design practices that protect individuals from harm in digital and spatial computing environments, including cybersecurity threats, privacy violations, online harassment, and misuse of immersive technologies. It applies governance frameworks, access controls, and content moderation mechanisms to ensure safe participation across metaverse, VR, and telecollaboration platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:digital-safety", "labels": [ "Digital Safety" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing", "Telecollaboration" ] }, { "id": "digital-securities", "title": "Digital Securities", "domain": "finance", "domain_name": "Finance", "definition": "Regulated financial instruments \u2014 equities, bonds, fund units, and other investment products \u2014 issued, recorded, and transferred as tokens on distributed ledgers, combining the legal character of traditional securities with programmable compliance, fractional ownership, and near-instant settlement, and implemented through permissioned token standards such as ERC-1400 and ERC-3643.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:digital-securities", "labels": [ "Digital Securities" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [ "Security Token", "ERC-1400", "ERC-3643" ] }, { "id": "digital-security", "title": "Digital Security", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Security encompasses the technical controls, protocols, and governance frameworks protecting digital systems, data, and identities from unauthorised access, tampering, and exploitation. It integrates cryptographic mechanisms, access management policies, threat detection, and incident response to ensure confidentiality, integrity, and availability of digital assets and infrastructure across networked environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-security", "labels": [ "Digital Security" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "digital-services-act", "title": "Digital Services Act", "domain": "governance", "domain_name": "Governance", "definition": "A comprehensive EU regulation establishing a legal framework for digital services accountability, content moderation requirements, platform transparency obligations, and user protection measures across online intermediaries, marketplaces, and social platforms, with implications for metaverse and virtual world governance.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:digital-services-act", "labels": [ "Digital Services Act", "EU Digital Services Act" ], "is_subclass_of": [ "Digital Regulation" ], "wikilinks": [ "Compliance Systems", "Content Transparency", "Moderation Infrastructure", "Platform Accountability", "Reporting Mechanisms", "User Protection", "Digital Regulation", "metaverse", "Telecollaboration" ] }, { "id": "digital-signal-processing", "title": "Digital Signal Processing", "domain": "data", "domain_name": "Data", "definition": "The mathematical and computational techniques for representing, transforming, and manipulating signals in digital form, including filtering, spectral analysis, compression, and enhancement \u2014 enabling real-time audio processing, video analysis, spatial audio for VR/AR, and sensor data interpretation critical for immersive and intelligent systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-signal-processing", "labels": [ "Digital Signal Processing" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [ "Algorithms", "Audio Enhancement", "DSP Hardware", "Sensor Input", "Sensors", "metaverse", "Signal Processing", "Spatial Audio", "Video Processing" ] }, { "id": "digital-signal-processor", "title": "Digital Signal Processor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital signal processor (DSP) is a specialised microprocessor architecture optimised for the high-throughput numerical operations of digital signal processing, such as multiply-accumulate, filtering and fast transforms. DSPs feature hardware support for fixed and floating-point arithmetic, parallel datapaths and efficient memory access to process audio, sensor and communication signals in real time at low power. They are a core building block of audio systems, embedded devices and spatial computing hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:digital-signal-processor", "labels": [ "Digital Signal Processor" ], "is_subclass_of": [ "Audio System" ], "wikilinks": [] }, { "id": "digital-signature-verification", "title": "Digital Signature Verification", "domain": "security", "domain_name": "Security", "definition": "The cryptographic process of validating the authenticity and integrity of digitally signed data by applying the signer's public key to confirm that the signature was produced by the corresponding private key and that the signed content has not been altered.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-signature-verification", "labels": [ "Digital Signature Verification", "ECDSA Signature Verification", "Electronic Signature", "Signature Verification" ], "is_subclass_of": [ "Cryptographic Verification" ], "wikilinks": [ "Document Integrity", "Hash Algorithm", "Signature Algorithm", "Transaction Authentication", "Blockchain", "Cryptographic Verification", "metaverse", "Non-Repudiation", "Public Key" ] }, { "id": "digital-signature", "title": "Digital Signature", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Digital Signature is a cryptographic primitive consisting of three probabilistic polynomial-time algorithms (KeyGen, Sign, Verify) operating over an asymmetric keypair (sk, pk) such that, for any message m drawn from the message space M, Sign(sk, m) produces a signature \u03c3 that Verify(pk, m,...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-signature", "labels": [ "Digital Signature", "Digital-Signature", "DigitalSignature" ], "is_subclass_of": [ "Cryptographic Primitive", "Asymmetric Cryptography", "Authentication Mechanism", "Public-Key Cryptosystem" ], "wikilinks": [ "ACME", "Adobe Sign", "Anderson 2020 Security Engineering 3rd Edition", "ANSI X9", "ANSI X9.62 ECDSA", "Asymmetric Cryptography", "Authentication Mechanism", "Base58 Encoding", "Bernstein 2006 Curve25519", "Bernstein Duif Lange Schwabe Yang 2012 Ed25519", "BIP-340", "BIP-340 Schnorr Signatures secp256k1", "BIP-341", "BIP-342", "Bitcoin Cash", "BLAKE3", "BLS Signature", "Boneh Lynn Shacham 2001 Short Signatures from Weil Pairing", "CAdES", "Certificate Authority" ] }, { "id": "digital-signatures", "title": "Digital Signatures", "domain": "security", "domain_name": "Security", "definition": "A cryptographic mechanism that uses asymmetric key pairs to produce a verifiable seal on digital data, ensuring authenticity, integrity, and non-repudiation of messages, transactions, and documents across distributed and decentralised systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-signatures", "labels": [ "Digital Signatures", "HTTP Signatures", "RSA Signatures" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Authentication", "Data Integrity", "Signature Algorithm", "Blockchain", "Cryptography", "Hash Function", "metaverse", "Non-Repudiation", "Private Key" ] }, { "id": "digital-society-harms", "title": "Digital Society Harms", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital Society Harms is the catalogue of empirically documented and structurally anticipated negative impacts arising from the deployment of digital, algorithmic, and AI systems at population scale across information, psychological, economic, social, human-rights, child-safety, environmental, an...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-society-harms", "labels": [ "Digital Society Harms" ], "is_subclass_of": [ "AI Governance and Ethics", "Sociotechnical Risk", "AI Risks", "Human Rights", "Digital Safety", "Platform Governance" ], "wikilinks": [ "Ada Lovelace Harms Framework", "Ada Lovelace Institute Harms Framework", "Algorithmic Amplification", "Algorithmic Audit", "Attention Economy", "Cambridge Analytica ICO Enforcement 2018", "Center for Countering Digital Hate Deadly by Design 2022", "Child Safety Harms", "Child Safety Online", "Civil Society Oversight", "Council of Europe Framework Convention on AI", "Council of Europe Framework Convention on AI CETS 225", "Council of Europe HUDERIA", "Crawford Atlas of AI 2021", "Democratic Resilience", "Digital Rights Advocacy", "Economic Harms", "Empirical Harm Measurement", "Environmental Harms", "EU AI Act Regulation 2024 1689" ] }, { "id": "digital-society-monetary-theory-node", "title": "Digital Society Monetary Theory Node", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A research and funding node documenting monetary theory, digital currency mechanisms, and value-exchange models relevant to digital society contexts. It examines base money, fiduciary media, cryptocurrency properties, and the interplay between decentralised and state-backed monetary systems in virtual and metaverse economies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-society-monetary-theory-node", "labels": [ "Digital Society Monetary Theory Node", "Money from DigiSoc" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Convergence" ] }, { "id": "digital-society-surveillance", "title": "Digital Society Surveillance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital Society Surveillance is the systematic observation, collection, processing, and analysis of personal data, communications, movements, and behavioural signals generated by individuals participating in digital infrastructure \u2014 encompassing state mass-surveillance programmes, commercial surv...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-society-surveillance", "labels": [ "Digital Society Surveillance" ], "is_subclass_of": [ "AI Governance and Ethics", "Privacy", "Civil Liberties", "Data Ethics", "Political Economy", "Human Rights", "Digital Rights" ], "wikilinks": [ "Advertising Technology", "AI Classification Systems", "AI Surveillance", "Algorithmic Governance", "ALPR Networks", "Anonymous Communication", "Behavioural Analytics", "Behavioural Prediction", "Biometric Classification", "Biometric Databases", "Biometric Tracking", "Bossware", "Civil Liberties", "Clearview AI", "Commercial Profiling", "Commercial Spyware", "Data Brokers", "Data Ethics", "End-to-End Encryption", "EthicsDomain" ] }, { "id": "digital-society", "title": "Digital Society", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The emergent social order in which digital networks, platforms, and data systems become primary mediators of economic activity, civic participation, cultural expression, and interpersonal relationships. Digital society encompasses the institutions, norms, rights, and governance mechanisms needed to ensure equitable, secure, and rights-respecting participation in digital environments including the metaverse.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-society", "labels": [ "Digital Society" ], "is_subclass_of": [ "Platform and Environment", "Metaverse and Telecollaboration" ], "wikilinks": [ "owl:Thing", "Telecollaboration" ] }, { "id": "digital-sovereignty", "title": "Digital Sovereignty", "domain": "governance", "domain_name": "Governance", "definition": "Digital Sovereignty is a policy, regulatory, and technical framework through which states, organisations, or individuals assert meaningful control over their digital infrastructure, data flows, algorithms, and standards, reducing dependency on foreign or monopolistic technology providers and maintaining the capacity to make autonomous decisions about their digital environment. It encompasses data localisation requirements, national cloud infrastructure, domestic AI capability development, preference for open-source technology stacks, and regulatory authority over platforms operating within a jurisdiction. The concept balances competitive participation in the global digital economy against strategic autonomy, security, and cultural self-determination.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:digital-sovereignty", "labels": [ "Digital Sovereignty" ], "is_subclass_of": [ "Data Sovereignty" ], "wikilinks": [] }, { "id": "digital-surface-model", "title": "Digital Surface Model", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:digital-surface-model", "labels": [ "Digital Surface Model" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "digital-tax-compliance-node", "title": "Digital Tax Compliance Node", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital Tax Compliance Node is a type of Virtual Economy Infrastructure in the artificial intelligence domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-tax-compliance-node", "labels": [ "Digital Tax Compliance Node" ], "is_subclass_of": [ "AI Governance and Ethics", "Virtual Economy Infrastructure", "Regulatory Technology" ], "wikilinks": [ "Audit Trail Generation", "Automated Tax Filing", "Cross-border Tax Settlement", "EU DAC7", "Identity Verification System", "Jurisdiction Mapping Service", "OECD Digital Tax Framework", "Real-time Compliance", "Regulatory Database", "Regulatory Reporting Module", "Tax Calculation Engine", "Transaction Ledger", "Transaction Monitor", "Blockchain", "Blockchain Network", "Digital Payment System", "MiddlewareLayer", "Smart Contract", "TrustAndGovernanceDomain", "VirtualEconomyDomain" ] }, { "id": "digital-taxonomy-registry", "title": "Digital Taxonomy Registry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Taxonomy Registry is a type of Metadata Repository in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-taxonomy-registry", "labels": [ "Digital Taxonomy Registry" ], "is_subclass_of": [ "Standards and Interoperability", "Metadata Repository" ], "wikilinks": [ "Analytics & Reporting", "API Interface", "Change Management Process", "Classification Scheme Database", "Cross-Platform Categorization", "Database Management System", "ISO 11179", "OECD Crypto-Asset Registry", "Quality Assurance System", "Semantic Interoperability", "Unique Identifier System", "Versioning System", "Authentication Service", "Blockchain", "ComputationAndIntelligenceDomain", "DataLayer", "Governance Framework", "Metadata Repository", "MiddlewareLayer", "Regulatory Compliance" ] }, { "id": "digital-technology-access-equity", "title": "Digital Technology Access Equity", "domain": "governance", "domain_name": "Governance", "definition": "Digital Technology Access Equity is the principle that all individuals and communities, regardless of geographic location, socioeconomic status, disability, gender, age, or ethnicity, should have fair and meaningful access to digital technologies, connectivity infrastructure, and the skills required to participate fully in digital economies. The concept extends beyond physical access to broadband or devices to encompass affordability, digital literacy, relevant content, and the capacity to derive genuine benefit from AI, cloud computing, and platform ecosystems. It directly shapes policy frameworks, international development agendas, and corporate social responsibility obligations by treating technology access as a prerequisite for social and economic inclusion. Addressing equity gaps requires coordinated interventions across infrastructure deployment, education, regulatory mandates, and localised content production.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-technology-access-equity", "labels": [ "Digital Technology Access Equity", "Digital Equity", "Equity" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Education and AI" ] }, { "id": "digital-technology", "title": "Digital Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Technology refers to electronic tools, systems, and platforms that generate, store, or process information in binary or digital form. It encompasses computing hardware, software, networks, and embedded systems that collectively enable the creation, transformation, and distribution of digital content and services, forming the foundational layer of the modern digital economy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:digital-technology", "labels": [ "Digital Technology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-terrain-model", "title": "Digital Terrain Model", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:digital-terrain-model", "labels": [ "Digital Terrain Model" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "digital-transformation", "title": "Digital Transformation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The comprehensive integration of digital technologies across all areas of business and society, fundamentally changing how organisations operate, deliver value, and engage with customers through technologies such as AI, cloud computing, IoT, data analytics, and immersive platforms; encompassing cultural, process, and structural change alongside technology adoption.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-transformation", "labels": [ "Digital Transformation", "DigitalTransformation" ], "is_subclass_of": [ "Organizational Change" ], "wikilinks": [ "Business Innovation", "Change Management", "Customer Experience", "Operational Efficiency", "Cloud Computing", "Computer Vision", "Data Analytics", "metaverse", "Organizational Change" ] }, { "id": "digital-trust", "title": "Digital Trust", "domain": "governance", "domain_name": "Governance", "definition": "Digital trust is the confidence that users, organisations, and systems place in the security, privacy, reliability, and integrity of digital services, identities, and transactions. It is established through verifiable mechanisms such as cryptography, certificates, identity assurance, and transparent governance rather than personal familiarity. Digital trust is foundational to e-commerce, online identity, and inter-organisational collaboration, where parties must rely on counterparties and infrastructure they cannot directly inspect.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:digital-trust", "labels": [ "Digital Trust" ], "is_subclass_of": [ "Trust" ], "wikilinks": [] }, { "id": "digital-twin-collaboration", "title": "Digital Twin Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The use of shared virtual replicas of physical assets, processes, or environments as collaborative workspaces where geographically distributed teams simultaneously inspect, analyse, simulate, and modify digital representations synchronised with real-world counterparts through sensor data streams...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:digital-twin-collaboration", "labels": [ "Digital Twin Collaboration", "TELE-300-digital-twin-collaboration" ], "is_subclass_of": [ "Telepresence", "Telecollaboration", "TELE-002-telecollaboration" ], "wikilinks": [ "CollaborativeSimulation", "IEC 63278", "IndustryFourPointZero", "IoT", "ISO 19650", "ISO 23247", "OGC CityGML", "TELE-002-telecollaboration", "TELE-020-virtual-reality-telepresence", "TELE-150-webrtc", "TELE-203-haptic-feedback-telepresence", "TELE-301-virtual-office-spaces", "TELE-302-shared-whiteboards", "CloudComputing", "DigitalTwin" ] }, { "id": "digital-twin-construction", "title": "Digital Twin Construction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The application of digital twin technology within the architecture, engineering, and construction (AEC) industry, extending BIM capabilities through real-time sensor integration and IoT connectivity to create dynamic virtual replicas of buildings and infrastructure throughout their lifecycle.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-construction", "labels": [ "Digital Twin Construction" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin" ], "wikilinks": [ "BIM", "Building Lifecycle Management", "Cloud Platform", "Construction Optimization", "Computer Vision", "Digital Twin", "IoT Sensors", "metaverse", "Predictive Maintenance" ] }, { "id": "digital-twin-creation-pipeline", "title": "Digital Twin Creation Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The systematic workflow and technology stack for generating digital twins, combining 3D scanning techniques such as LiDAR, photogrammetry, and structured-light scanning with AI processing to create accurate virtual replicas of physical assets, environments, or systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-creation-pipeline", "labels": [ "Digital Twin Creation Pipeline" ], "is_subclass_of": [ "Platform and Environment", "3D Content Pipeline" ], "wikilinks": [ "3D Content Pipeline", "3D Scanning", "Asset Digitization", "Point Cloud Processing", "Real-Time Monitoring", "Virtual Replica Creation", "Computer Vision", "metaverse", "Photogrammetry" ] }, { "id": "digital-twin-creation", "title": "Digital Twin Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of constructing virtual replicas of physical entities that dynamically reflect real-time conditions through continuous bidirectional data linkage, enabling simulation, monitoring, and optimisation across an asset's entire lifecycle from design through decommissioning.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-creation", "labels": [ "Digital Twin Creation" ], "is_subclass_of": [ "Digital Modeling" ], "wikilinks": [ "IoT Integration", "Lifecycle Management", "Real-Time Monitoring", "Sensor Data", "3D Modeling", "Computer Vision", "Digital Modeling", "metaverse", "Predictive Analytics" ] }, { "id": "digital-twin-data-assimilation", "title": "Digital Twin Data Assimilation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-data-assimilation", "labels": [ "Digital Twin Data Assimilation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "digital-twin-ecosystem", "title": "Digital Twin Ecosystem", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The comprehensive network of interconnected technologies, platforms, and stakeholders that enable digital twin deployment, including IoT sensors, edge computing, cloud platforms, AI analytics, and visualisation systems working toger to create, maintain, and derive value from virtual replicas of p...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-ecosystem", "labels": [ "Digital Twin Ecosystem" ], "is_subclass_of": [ "Platform and Environment", "Technology Ecosystem" ], "wikilinks": [ "Cloud Platform", "Cross-Platform Analytics", "Enterprise Digital Twins", "IoT Infrastructure", "Computer Vision", "Data Standards", "metaverse", "System Integration", "Technology Ecosystem" ] }, { "id": "digital-twin-framework", "title": "Digital Twin Framework", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The architectural blueprint and standards-based modology for implementing digital twins within enterprise and industrial contexts, defining the layered structure from IoT foundation through application layer, and ensuring interoperability across cyber-physical systems and industrial metaverse pla...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-framework", "labels": [ "Digital Twin Framework" ], "is_subclass_of": [ "Platform and Environment", "Enterprise Architecture" ], "wikilinks": [ "Data Models", "IEEE", "IEEE P3144", "Integration Standards", "Scalable Deployment", "Computer Vision", "Enterprise Architecture", "metaverse", "Reference Architecture", "Standards Compliance", "System Interoperability" ] }, { "id": "digital-twin-generation", "title": "Digital Twin Generation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The automated or semi-automated process of creating digital twin models using AI, machine learning, and advanced 3D capture technologies, enabling rapid production of virtual replicas with reduced manual effort and accelerated deployment timelines for industrial and enterprise applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-generation", "labels": [ "Digital Twin Generation" ], "is_subclass_of": [ "Platform and Environment", "Automated Modeling" ], "wikilinks": [ "3D Capture", "AI Training", "Mass Digitization", "Rapid Prototyping", "Automated Modeling", "Computer Vision", "Generative AI", "Machine Learning", "metaverse" ] }, { "id": "digital-twin-infrastructure", "title": "Digital Twin Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The foundational computing, networking, and data management systems required to deploy and operate digital twins at scale, encompassing cloud platforms, edge computing nodes, IoT gateways, and the connectivity fabric that enables real-time data flow between physical assets and their virtual count...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:digital-twin-infrastructure", "labels": [ "Digital Twin Infrastructure", "City Infrastructure Twin" ], "is_subclass_of": [ "Computing and Cloud", "Computing Infrastructure" ], "wikilinks": [ "5G Connectivity", "Cloud Platform", "Edge Analytics", "Scalable Deployment", "Sensor Input", "Computing Infrastructure", "Edge Computing", "metaverse", "Real-Time Processing" ] }, { "id": "digital-twin-interop-protocol", "title": "Digital Twin Interop Protocol", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A standardised API and communication framework enabling the exchange of state, simulation data, and behaviour models between heterogeneous digital twin systems, defining data formats, query interfaces, and synchronisation semantics so twins built on different platforms can interoperate and compose into federated simulations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-twin-interop-protocol", "labels": [ "Digital Twin Interop Protocol" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "API Specification", "Authentication Layer", "Data Exchange Format", "Data Serialization", "Federated Simulation", "GraphQL", "ISO/IEC 23247", "MQTT", "OPC UA", "Query Interface", "Real-Time State Sync", "REST API", "Twin Composition", "WebSocket", "Autonomous Robot", "Cross-Platform Digital Twins", "DataLayer", "Digital Twin Framework", "Identity Management", "InfrastructureDomain" ] }, { "id": "digital-twin-synchronisation-bus", "title": "Digital Twin Synchronisation Bus", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Twin Synchronisation Bus is a type of Digital Twin Infrastructure in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-twin-synchronisation-bus", "labels": [ "Digital Twin Synchronisation Bus" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Bidirectional Data Flow", "Conflict Resolution Module", "Distributed Twin Orchestration", "Event Log", "Event Stream Processor", "ISO 23247 Addendum", "IV) Data Layer", "Message Broker", "Message Queue", "Multi-Instance State Coherence", "Publish-Subscribe Pattern", "State Store", "State Synchronization Engine", "Autonomous Robot", "Digital Twin Infrastructure", "Distributed System", "Event-Driven Architecture", "InfrastructureDomain", "Network Protocol", "Real-Time Digital Twin Synchronization" ] }, { "id": "digital-twin-technology", "title": "Digital Twin Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Twin Technology is the discipline of constructing and operating persistent, synchronised virtual replicas of physical objects, systems, or environments that continuously ingest real-time sensor and telemetry data to mirror the state, behaviour, and lifecycle of their physical counterparts. Rooted in model-based engineering, it combines IoT connectivity, physics-based simulation, data analytics, and 3D visualisation to enable predictive maintenance, design optimisation, remote monitoring, and what-if scenario analysis without physical intervention. The technology spans the full asset lifecycle\u2014from design and commissioning through operation and decommissioning\u2014and underpins industrial metaverse platforms, smart-city infrastructure, and autonomous system validation. Standardisation efforts led by ISO/IEC JTC 1 and the Industrial Internet Consortium (IIC) are consolidating reference architectures and interoperability frameworks across vendor ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:digital-twin-technology", "labels": [ "Digital Twin Technology" ], "is_subclass_of": [ "Cyber Physical Systems" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "digital-twin-of-society-dto-s", "title": "Digital Twin of Society (DToS)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Twin of Society (DToS) is a type of Spatial Computing in the spatial computing domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-twin-of-society-dto-s", "labels": [ "Digital Twin of Society (DToS)" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Agent-Based Simulation", "Census Data", "City Infrastructure Twin", "Crisis Management", "Economic Model", "Environmental Sensor Network", "ETSI GR ARF 010", "Geographic Information System", "Policy Simulation", "Population Simulation", "Real-time City Data", "Siemens Industrial Metaverse", "Smart City Ecosystem", "Social Network Analysis", "Sustainability Optimization", "Traffic Management System", "Urban Data Platform", "Urban Planning", "ApplicationLayer", "Autonomous Robot" ] }, { "id": "digital-twin-of-the-customer", "title": "Digital Twin of the Customer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Digital Twin of the Customer is a dynamic virtual replica that mirrors an individual customer's behaviours, preferences, interactions, and decision-making patterns within a spatially-aware digital environment, enabling real-time simulation and predictive analysis of customer journeys. It integrates IoT telemetry, AI-driven behavioural modelling, and immersive 3D visualisations to create a continuously updated model reflecting how customers engage with products, services, and physical or digital spaces. Organisations use these twins to personalise experiences, predict service needs, and test interventions before live deployment.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:digital-twin-of-the-customer", "labels": [ "Digital Twin of the Customer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "DID Nostr Identity", "MetaverseDomain" ] }, { "id": "digital-twin", "title": "Digital Twin", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital Twin is a spatial computing concept and a type of Digital Twin Technology. that enables Collaborative Design, Process Optimisation. comprising Control Interface, IoT Sensor Data.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:digital-twin", "labels": [ "Digital Twin", "Digital Twin Visualization", "Digital Twins", "Digital-Twin", "DigitalTwin", "Enterprise Digital Twins", "Industrial Digital Twin" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "3D Visualisation", "Collaborative Design", "Control Interface", "Data Aggregation", "IEEE", "Industrial Internet Consortium", "IoT Infrastructure", "IoT Sensor Data", "ISO 23247", "Process Optimisation", "Real-Time Synchronisation", "Remote Monitoring", "3D Model", "Artificial Intelligence", "Blockchain", "Machine Learning", "Predictive Maintenance", "Robotics", "Simulation Engine", "Telecollaboration" ] }, { "id": "digital-wallet", "title": "Digital Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A software application or hardware device that stores private keys and enables users to manage, send, and receive cryptocurrencies and digital assets on blockchain networks, with self-custody wallets providing complete user control over private keys without third-party intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:digital-wallet", "labels": [ "Digital Wallet" ], "is_subclass_of": [ "Cryptocurrency Storage" ], "wikilinks": [ "DeFi Access", "Transaction Signing", "AI Agent System", "Asset Management", "Blockchain Network", "Cryptocurrency Storage", "Cryptographic Security", "Private Key" ] }, { "id": "digital-watermarking", "title": "Digital Watermarking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Digital watermarking is the technique of embedding identifying or authenticating information directly into digital content such as images, audio, video, text, or model outputs, ideally so the mark is imperceptible yet recoverable. A watermark may be robust, surviving compression and editing, or fragile, breaking on tampering to signal alteration. It is increasingly used to mark AI-generated media for provenance and to support copyright protection and content authentication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:digital-watermarking", "labels": [ "Digital Watermarking" ], "is_subclass_of": [ "Content Provenance", "Information Hiding", "Media Security", "Digital Media Authentication" ], "wikilinks": [] }, { "id": "digital-well-being-index", "title": "Digital Well-Being Index", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Composite indicator assessing psychological, social, physical, and temporal impacts of extended virtual engagement, providing quantitative measures of healthy metaverse usage patterns.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:digital-well-being-index", "labels": [ "Digital Well-Being Index" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Activity Logging", "Behavioral Tracking", "Cognitive Load Measurements", "Emotional Wellness Scores", "Health Data Integration", "Healthy Engagement Recommendations", "Physical Activity Indicators", "Platform Governance Framework", "Platform Health Reports", "Screen Time Metrics", "Self-Report Surveys", "Sleep Impact Assessment", "Social Engagement Scores", "Temporal Analysis Tools", "Usage Alerts", "Usage Analytics", "User Health Monitoring System", "WHO Digital Well-Being Metrics", "Application Layer", "Metaverse Psychology Profile" ] }, { "id": "digital-workplace-platform", "title": "Digital Workplace Platform", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Digital Workplace Platform (DWP) is an integrated software environment \u2014 combining communication, collaboration, content management, workflow automation, and people analytics into a unified, identity-governed ecosystem \u2014 that serves as the primary operational substrate for distributed knowledge-w...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:digital-workplace-platform", "labels": [ "Digital Workplace Platform" ], "is_subclass_of": [ "Workspace Tools", "Communication Technology", "Enterprise Software Infrastructure", "Distributed Collaboration Technology", "Cloud Platform", "Employee Experience System", "Intranet Evolution" ], "wikilinks": [ "AI Copilot Layer", "Asynchronous Workflows", "Cloud Platform", "Cloud Storage", "Collaborative Document Editing", "Confluence", "Consumer Messaging Apps", "Cross-Functional Collaboration", "Data Encryption Standards", "Data Residency Controls", "DEI Analytics", "Digital Employee Experience", "Digital Onboarding", "DistributedCollaborationDomain", "Distributed Collaboration Technology", "Employee Engagement", "EmployeeExperienceDomain", "Employee Experience System", "Employee Self-Service", "Employee Wellbeing" ] }, { "id": "dijkstra-algorithm", "title": "Dijkstra Algorithm", "domain": "robotics", "domain_name": "Robotics", "definition": "A graph search algorithm that finds the shortest path from a source node to all other nodes in a weighted graph with non-negative edge weights. It systematically explores nodes in order of increasing distance from the source, guaranteeing optimal solutions.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:dijkstra-algorithm", "labels": [ "Dijkstra Algorithm", "RB-1018-dijkstra-algorithm" ], "is_subclass_of": [ "Navigation and Planning", "Path Planning", "RB-1016-path-planning", "Graph Search", "Graph Search Algorithm", "A-Star Algorithm", "Bellman-Ford Algorithm" ], "wikilinks": [ "Bellman-Ford Algorithm", "Completeness", "Edsger Dijkstra 1956", "GPS Systems", "Graph Representation", "Graph Search Algorithm", "Graph Theory", "Network Routing", "Non-negative Weights", "Optimality", "Priority Queue", "RB-1016-path-planning", "RB-1017-rrt-algorithm", "Shortest Path", "Weighted Graph", "A-Star Algorithm", "Graph Search", "Navigation", "Path Planning", "Robotics" ] }, { "id": "dike", "title": "Dike", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dike", "labels": [ "Dike", "Dyke" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "dilution-of-precision", "title": "Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dilution-of-precision", "labels": [ "Dilution of Precision" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "dimensionality-reduction", "title": "Dimensionality Reduction", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Dimensionality Reduction is the process of transforming high-dimensional data into a lower-dimensional representation while preserving important structural properties and relationships. It addresses the curse of dimensionality, reduces computational costs, enables visualization, removes noise, and improves model performance by eliminating redundant or irrelevant features.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:dimensionality-reduction", "labels": [ "Dimensionality Reduction", "Nonlinear Dimensionality Reduction" ], "is_subclass_of": [ "AI Technique", "Unsupervised Learning" ], "wikilinks": [ "Computational Efficiency", "Curse of Dimensionality", "Data Visualization", "Feature Selection", "Blockchain", "Digital Twin", "Feature Engineering", "Unsupervised Learning" ] }, { "id": "direct-air-capture", "title": "Direct Air Capture", "domain": "governance", "domain_name": "Governance", "definition": "Direct Air Capture (DAC) is a class of carbon-dioxide-removal technologies that chemically extract CO2 directly from ambient air, rather than from a concentrated flue-gas stream. Captured CO2 is then either permanently stored underground or used as a feedstock, producing measurable negative emissions. DAC is energy-intensive because atmospheric CO2 is highly dilute, so its viability depends on low-carbon energy and durable storage or carbon-market incentives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:direct-air-capture", "labels": [ "Direct Air Capture" ], "is_subclass_of": [ "Climate Policy" ], "wikilinks": [] }, { "id": "direct-answer-prompting", "title": "Direct Answer Prompting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Direct answer prompting is a prompting strategy in which a large language model is instructed to produce a final answer immediately, without showing intermediate reasoning steps. It contrasts with chain-of-thought approaches by optimising for brevity, latency and cost on tasks where extended reasoning offers little benefit. The technique is used when an answer is expected to be retrievable or shallow, trading interpretability and complex-reasoning accuracy for efficiency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:direct-answer-prompting", "labels": [ "Direct Answer Prompting" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "direct-preference-optimisation", "title": "Direct Preference Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An alignment method that directly uses preference data to fine-tune language models without training a separate reward model or using reinforcement learning, offering a simpler and more stable alternative to RLHF. DPO reparameterises the reward model objective to optimise the policy directly on preference comparison pairs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:direct-preference-optimisation", "labels": [ "Direct Preference Optimisation", "Direct Preference Optimization" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "Direct Preference Optimization", "MetaverseDomain" ] }, { "id": "direct3d", "title": "Direct3D", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Direct3D is the 3D graphics application programming interface within Microsoft's DirectX collection, providing low-level access to the graphics processing unit for rendering geometry, shading, and compute workloads on Windows and Xbox platforms. It exposes the rendering pipeline through programmable shaders, command buffers, and resource state management, with modern revisions such as Direct3D 12 offering explicit, low-overhead control over GPU memory and parallelism. It is the dominant native graphics interface on Microsoft platforms and a primary backend target for game engines and real-time rendering systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:direct3d", "labels": [ "Direct3D" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "directx", "title": "DirectX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "DirectX is a collection of application programming interfaces developed by Microsoft for handling multimedia tasks, especially graphics and gaming, on Windows and Xbox platforms. Its Direct3D component provides low-level access to graphics hardware for real-time rendering, while related APIs cover input, audio and compute. By abstracting diverse hardware behind a common interface, it lets developers target a wide range of GPUs through a single programming model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:directx", "labels": [ "DirectX" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "directed-acyclic-graph-execution", "title": "Directed Acyclic Graph Execution", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Directed acyclic graph (DAG) execution is a computation model in which tasks are nodes connected by directed dependency edges that contain no cycles, so the graph defines a partial order of operations. An execution engine performs a topological sort and runs nodes as soon as their inputs are ready, enabling parallelism, caching of unchanged subgraphs, and deterministic recomputation. It is the scheduling backbone of workflow engines and node-based authoring tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:directed-acyclic-graph-execution", "labels": [ "Directed Acyclic Graph Execution" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "directed-acyclic-graph", "title": "Directed Acyclic Graph", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A directed acyclic graph (DAG) is a graph whose edges have direction and which contains no directed cycles, so no path returns to its starting vertex. This structure naturally encodes ordered dependencies, enabling a topological ordering of vertices and making DAGs foundational for scheduling, dependency resolution, version histories, and certain distributed-ledger designs. The absence of cycles guarantees that dependency chains terminate, which underpins many algorithms built on top of the structure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:directed-acyclic-graph", "labels": [ "Directed Acyclic Graph" ], "is_subclass_of": [ "Graph Theory" ], "wikilinks": [] }, { "id": "directly-responsible-individual", "title": "Directly Responsible Individual", "domain": "governance", "domain_name": "Governance", "definition": "A management principle assigning a single, specific person ultimate accountability for a particular outcome or project.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:directly-responsible-individual", "labels": [ "Directly Responsible Individual" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "directory-service", "title": "Directory Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A directory service is a specialised database that stores, organises, and provides access to information about the entities of a network \u2014 users, groups, devices, and services \u2014 optimised for high-volume reads and hierarchical lookup. Accessed through protocols such as LDAP, it underpins authentication, authorisation, and resource discovery in enterprise environments, with implementations including Active Directory and OpenLDAP. Directory services centralise identity data so that credentials and access policies can be managed once and enforced consistently across many systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:directory-service", "labels": [ "Directory Service", "Directory Services" ], "is_subclass_of": [ "Identity Provider" ], "wikilinks": [] }, { "id": "disaster-mapping", "title": "Disaster Mapping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:disaster-mapping", "labels": [ "Disaster Mapping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "disaster-recovery", "title": "Disaster Recovery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Disaster recovery (DR) is the set of policies, tools, and procedures enabling an organisation to restore its IT systems, data, and operations following a disruptive event such as hardware failure, cyberattack, natural disaster, or human error. It is quantified by Recovery Time Objective (RTO) and Recovery Point Objective (RPO), and encompasses backup strategies, replication architectures, and tested failover procedures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:disaster-recovery", "labels": [ "Disaster Recovery" ], "is_subclass_of": [ "Resilience" ], "wikilinks": [] }, { "id": "disaster-response", "title": "Disaster Response", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Disaster response is the coordinated use of sensing, mapping and robotic systems to assess damage, locate survivors and direct relief in the aftermath of natural or human-made catastrophes. Spatial-computing techniques fuse aerial imagery, LiDAR, satellite remote sensing and ground-robot telemetry into situational maps that guide responders. Speed, robustness under degraded conditions, and accurate geospatial localisation are the defining requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:disaster-response", "labels": [ "Disaster Response" ], "is_subclass_of": [ "Spatial Computing", "Remote Sensing" ], "wikilinks": [] }, { "id": "disclosure-requirements", "title": "Disclosure Requirements", "domain": "governance", "domain_name": "Governance", "definition": "Disclosure requirements are legal or regulatory obligations to reveal specified information to consumers, regulators, or the public so they can make informed decisions. In technology governance they cover material facts such as the use of automated decision-making, data practices, risks, and conflicts of interest. They underpin transparency regimes and consumer protection by reducing information asymmetry between providers and users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:disclosure-requirements", "labels": [ "Disclosure Requirements", "Disclosure Notification" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "discovery-layer", "title": "Discovery Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Functional layer responsible for search, navigation, and exposure of metaverse experiences and assets through indexing, search engines, and recommendation systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:discovery-layer", "labels": [ "Discovery Layer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Asset Browsing", "Content Catalog", "Content Discovery", "Content Indexer", "Experience Navigation", "MSF Taxonomy 2025", "Personalized Recommendations", "Query Interface", "Recommendation System", "Data Layer", "Data Storage", "InfrastructureDomain", "Metadata Registry", "Metadata Schema", "Search Engine", "Telecollaboration" ] }, { "id": "discreet-log-contracts", "title": "Discreet Log Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A scheme for executing conditional Bitcoin payments based on signed outcomes from external oracles, where the contract logic stays off-chain and only the settled transaction is broadcast.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:discreet-log-contracts", "labels": [ "Discreet Log Contracts", "Discreet Log Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Bitcoin", "Schnorr Signature", "Cryptographic Proof", "Atomic Swap", "Bitcoin Script", "Smart Contract" ] }, { "id": "discrete-cosine-transform", "title": "Discrete Cosine Transform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The discrete cosine transform, DCT, is an integral transform, closely related to the discrete Fourier transform, that expresses a finite sequence of data points as a sum of cosine functions oscillating at different frequencies, concentrating most signal energy into a small number of low-frequency coefficients. This energy-compaction property makes it the core building block of lossy image and video codecs such as JPEG and MPEG, which quantise and discard high-frequency coefficients to achieve compression. Video codecs apply the DCT, or block-based variants of it, to spatial blocks of pixel data before entropy coding.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:discrete-cosine-transform", "labels": [ "Discrete Cosine Transform" ], "is_subclass_of": [ "Fourier Transform" ], "wikilinks": [] }, { "id": "discrete-event-simulation", "title": "Discrete Event Simulation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Discrete event simulation is a modelling technique that represents a system as a sequence of distinct events occurring at specific points in time, advancing the simulation clock directly from one event to the next rather than in fixed time steps. Each event triggers state changes and may schedule further events, typically processed in timestamp order via a priority queue. It is widely used to model queuing systems, logistics networks, and multi-agent interactions where continuous-time simulation would be computationally wasteful.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:discrete-event-simulation", "labels": [ "Discrete Event Simulation", "Discrete-Event Simulation" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "discrete-logarithm-problem", "title": "Discrete Logarithm Problem", "domain": "security", "domain_name": "Security", "definition": "The discrete logarithm problem (DLP) is the computational task of finding the integer exponent x given a generator g and the value g^x within a finite cyclic group such as a multiplicative group modulo a prime or the point group of an elliptic curve. It is widely believed to be intractable for classical computers when the group is suitably large, and this presumed hardness underpins much of public-key cryptography. The elliptic-curve variant (ECDLP) offers equivalent security with smaller keys than the finite-field variant.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:discrete-logarithm-problem", "labels": [ "Discrete Logarithm Problem" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "discrete-mathematics", "title": "Discrete Mathematics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Discrete mathematics is the branch of mathematics concerned with structures that are fundamentally discrete rather than continuous, including sets, graphs, combinatorics, logic and number theory. It provides the formal foundations for computer science, underpinning algorithm analysis, data structures and cryptographic systems that rely on integer arithmetic and modular structures. Computer science curricula treat it as a prerequisite discipline, and fields such as number theory build directly on its results.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:discrete-mathematics", "labels": [ "Discrete Mathematics" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "discriminative-model", "title": "Discriminative Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A discriminative model is a class of machine learning model that directly learns the conditional probability of a target label given the observed input, rather than modelling how the data itself is generated. It focuses on the decision boundary that separates classes, which often yields strong predictive accuracy on classification and regression tasks. Discriminative models contrast with generative models, which learn the joint distribution of inputs and outputs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:discriminative-model", "labels": [ "Discriminative Model" ], "is_subclass_of": [ "Supervised Learning", "AI Technique" ], "wikilinks": [] }, { "id": "discriminator-network", "title": "Discriminator Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A discriminator network is the adversarial component of a generative adversarial network that learns to distinguish real data samples from those synthesised by the generator. Trained as a binary classifier, it outputs a probability that a given input is genuine, and its gradients provide the learning signal that pushes the generator toward producing more realistic outputs. The discriminator and generator are locked in a minimax game whose equilibrium yields a generator whose samples are indistinguishable from real data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:discriminator-network", "labels": [ "Discriminator Network" ], "is_subclass_of": [ "Generative Adversarial Networks", "Neural Network", "Binary Classifier", "Deep Generative Model" ], "wikilinks": [] }, { "id": "disentangled-representation", "title": "Disentangled Representation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A disentangled representation is a learned representation in which distinct, semantically meaningful factors of variation in the data \u2014 such as an object's shape, colour and pose \u2014 are captured by separate, largely independent dimensions of the latent space, so that changing one factor leaves the others unaffected. It is a goal of representation learning that improves interpretability and enables controlled generation, since manipulating a single latent dimension produces a predictable, isolated change in the output. Variational autoencoders and related generative models are commonly used to encourage disentanglement, for example through additional regularisation terms that penalise correlation between latent dimensions.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:disentangled-representation", "labels": [ "Disentangled Representation" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "disinformation", "title": "Disinformation", "domain": "governance", "domain_name": "Governance", "definition": "Disinformation is false or misleading information created and spread deliberately to deceive, manipulate, or cause harm \u2014 typically to advance a political, financial, or strategic objective. Its defining feature is intent: unlike misinformation, which is false content shared without the awareness that it is wrong, disinformation is engineered and disseminated in bad faith, often through coordinated inauthentic behaviour, fabricated sources, or synthetic media such as deepfakes. It threatens information integrity and media authenticity by corrupting the shared factual basis on which public discourse and democratic decision-making depend, and it is a central concern of trust-and-safety, electoral-security, and content-provenance work.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:disinformation", "labels": [ "Disinformation" ], "is_subclass_of": [ "Trust and Safety" ], "wikilinks": [ "Information Integrity", "Synthetic Media", "Deepfakes" ] }, { "id": "disparate-impact", "title": "Disparate Impact", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Disparate Impact is a legal doctrine and AI fairness concept denoting the condition where a facially neutral policy, practice, or algorithmic decision system produces outcomes that disproportionately disadvantage a legally protected group relative to a comparator group, regardless of discriminatory intent. Measured via the four-fifths rule or statistical significance tests, it is legally actionable in employment (US EEOC guidelines), lending, housing, and insurance, and is the analytical basis for algorithmic bias audits under the EU AI Act and GDPR. Remediation requires either demonstrating business necessity or adopting less discriminatory alternatives.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:disparate-impact", "labels": [ "Disparate Impact" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Fairness" ], "wikilinks": [ "EU Anti-Discrimination Directives", "UK Equality Act 2010", "US EEOC Uniform Guidelines", "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer" ] }, { "id": "disparity-map", "title": "Disparity Map", "domain": "robotics", "domain_name": "Robotics", "definition": "A disparity map is an image in which each pixel encodes the horizontal displacement of corresponding points between the two views of a stereo pair. Because disparity is inversely proportional to scene depth, the map converts directly to a depth map given the camera baseline and focal length. It is the core intermediate product of passive stereo vision, computed by rectifying the images and searching for correspondences along epipolar lines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:disparity-map", "labels": [ "Disparity Map" ], "is_subclass_of": [ "Stereo Vision" ], "wikilinks": [] }, { "id": "display-calibration", "title": "Display Calibration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Display calibration is the process of measuring and adjusting a display device's photometric and colorimetric characteristics\u2014luminance, white point, gamma or EOTF, and colour gamut\u2014to conform to a defined target standard or ICC colour profile. It employs colorimetric measurement instruments (colorimeters, spectrophotometers) to sample the display output and generates correction data (LUTs or ICC profiles) applied by the operating system or display hardware to compensate for manufacturing variation and age-related drift. Calibration is mandatory in colour-critical workflows including digital cinema (DCI-P3), broadcast (Rec. 709, Rec. 2020), medical imaging, and visual effects production.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:display-calibration", "labels": [ "Display Calibration" ], "is_subclass_of": [ "Calibration" ], "wikilinks": [] }, { "id": "display-capture", "title": "Display Capture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Display capture is the acquisition of the live pixel contents of a screen, window, or application surface as a video stream for recording or transmission. On the web it is exposed through the Screen Capture API's getDisplayMedia method, which prompts the user to choose a surface and returns a media stream subject to permission. It is the foundational capability behind screen recording and real-time screen sharing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:display-capture", "labels": [ "Display Capture" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "display-hardware", "title": "Display Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The physical devices and display technologies that render visual content for virtual, augmented, and mixed reality experiences, including VR headsets, AR glasses, and related optical systems that create immersive visual interfaces for metaverse applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:display-hardware", "labels": [ "Display Hardware" ], "is_subclass_of": [ "Display and Rendering", "Hardware" ], "wikilinks": [ "Display Technology", "Graphics Processing", "Immersive Visualization", "Optical Systems", "Sensor Input", "Hardware", "metaverse", "Mixed Reality", "Spatial Computing" ] }, { "id": "display-metrology", "title": "Display Metrology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Standardized measurement equipment and instruments for assessing visual performance parameters of XR displays, including colorimeters, photometers, and specialised testing hardware.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:display-metrology", "labels": [ "Display Metrology" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "Colorimeter", "Compliance Testing", "Contrast Ratio Meter", "Display Calibration", "Environmental Control", "ETSI GR ARF 010", "ISO 9241-303", "Luminance Meter", "Measurement Protocols", "Performance Validation", "Photometer", "Sensor Input", "Visual Quality Assessment", "Calibration Standards", "IEEE P2733 Standards", "InteractionDomain", "NetworkLayer", "Resolution Test Chart", "XR Testing Infrastructure" ] }, { "id": "display-technology", "title": "Display Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Display Technology encompasses the hardware substrates, optical systems, and electronic driving circuits used to present visual information to human observers across form factors ranging from flat panels and projection systems to head-mounted microdisplays and retinal projectors. Core substrate families include LCD with quantum dot backlights, OLED, microLED, and laser scanning systems, each offering distinct trade-offs in brightness, contrast ratio, colour gamut, refresh rate, and power consumption. For spatial computing and extended reality applications, display technology must additionally address field of view, vergence-accommodation conflict, waveguide efficiency, and eye-box uniformity. Advances in microdisplay resolution, diffractive waveguide engineering, and foveated rendering pipelines are critical enablers of lightweight, socially acceptable XR headsets and the broader convergence of physical and digital environments.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:display-technology", "labels": [ "Display Technology", "Traditional Display Technology" ], "is_subclass_of": [ "Display Hardware" ], "wikilinks": [] }, { "id": "dispute-resolution-mechanism", "title": "Dispute Resolution Mechanism", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Agreed process and framework for resolving conflicts between metaverse participants through mediation, arbitration, or other structured resolution mods.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:dispute-resolution-mechanism", "labels": [ "Dispute Resolution Mechanism" ], "is_subclass_of": [ "Legal and Regulatory", "Legal Framework" ], "wikilinks": [ "Automated Enforcement", "Conflict Resolution", "Conflict Resolution Protocol", "Dispute Classification System", "Evidence Management", "Evidence Submission System", "Fair Adjudication", "Mediation Process", "UNCITRAL ODR Rules", "Arbitration Process", "Blockchain", "E-Contract Arbitration", "Governance Framework", "Identity Verification", "Legal Framework", "Middleware Layer", "Participant Protection", "Smart Contract", "TrustAndGovernanceDomain", "Trust Infrastructure" ] }, { "id": "dispute-resolution", "title": "Dispute Resolution", "domain": "governance", "domain_name": "Governance", "definition": "Dispute Resolution encompasses the processes and mechanisms by which conflicting parties reach a binding or agreed settlement without necessarily resorting to formal court litigation, including negotiation, mediation, arbitration, and online dispute resolution (ODR) platforms. In blockchain and smart-contract contexts, dispute resolution refers to the programmatic or semi-programmatic adjudication of disagreements arising from contract ambiguity, code exploits, or off-chain facts that automated execution cannot verify, using decentralised arbitration protocols, escrow mechanisms, and cryptoeconomic incentives to produce impartial outcomes. Effective dispute resolution balances finality, cost, speed, and procedural fairness across both traditional legal frameworks and emerging decentralised governance structures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dispute-resolution", "labels": [ "Dispute Resolution", "Cross-Border Dispute Resolution", "Dispute Resolution Forum", "Dispute Resolution Process" ], "is_subclass_of": [ "Dispute Resolution Mechanism" ], "wikilinks": [] }, { "id": "disruptive-technology", "title": "Disruptive Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Disruptive Technology refers to innovations that initially address simple or underserved applications with more accessible and affordable solutions, then move upmarket to displace established incumbents and redefine entire industries. Following Christensen's framework, disruptive technologies succeed not through direct head-on competition but by creating new value networks and business models that incumbents cannot easily replicate. Contemporary examples include generative AI, blockchain, edge computing, and spatial computing.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:disruptive-technology", "labels": [ "Disruptive Technology" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Affordability", "ChristensenTheory", "DigitalPhotography", "DisruptiveInnovation", "ElectricVehicles", "InnovationDomain", "InnovatorsDislemma", "InnovatorsSolution", "MarketCreation", "ParadigmShift", "PersonalComputer", "Scalability", "Simplicity", "Smartphones", "StreamingServices", "SustainingTechnology", "ValueNetwork", "Accessibility", "Blockchain" ] }, { "id": "distance-metric", "title": "Distance Metric", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A distance metric is a function that quantifies how dissimilar two data points are, satisfying non-negativity, identity, symmetry and the triangle inequality. In machine learning it defines the geometry of a feature space and thereby governs nearest-neighbour search, clustering and similarity-based retrieval. Choosing or learning an appropriate metric is often as important as the model itself for tasks driven by proximity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:distance-metric", "labels": [ "Distance Metric" ], "is_subclass_of": [ "Machine Learning", "AI Technique", "Statistics", "Mathematical Analysis" ], "wikilinks": [] }, { "id": "distributed-ai-training", "title": "Distributed AI Training", "domain": "ai", "domain_name": "Ai", "definition": "Distributed AI training is the practice of training machine-learning models across many compute nodes in parallel to handle datasets and model sizes that exceed a single machine. It uses strategies such as data parallelism, model and tensor parallelism, and pipeline parallelism, coordinated by collective communication and gradient synchronisation. It is essential for training large neural networks within feasible time and memory budgets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-ai-training", "labels": [ "Distributed AI Training", "Cross-Organisational AI Training" ], "is_subclass_of": [ "AI Infrastructure", "Distributed Computing", "Machine Learning" ], "wikilinks": [] }, { "id": "distributed-architecture", "title": "Distributed Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Network design pattern allowing multi-node operation of a shared virtual world with coordinated state management across geographic or logical boundaries.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:distributed-architecture", "labels": [ "Distributed Architecture" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "CAP Theorem", "Distributed Consensus", "Distributed Nodes", "Distributed Systems Theory", "ETSI ARF 010", "Geographic Distribution", "High Availability", "Load Balancing", "Peer-to-Peer Networking", "Replication Strategy", "Scalability", "Synchronization Protocols", "Blockchain", "Consensus Protocol", "Data Layer", "Decentralization", "Fault Tolerance", "InfrastructureDomain", "Network Infrastructure", "Network Layer" ] }, { "id": "distributed-authentication-architecture", "title": "Distributed Authentication Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A decentralised identity authentication and management framework leveraging blockchain technology and self-sovereign identity (SSI) principles to enable secure, privacy-preserving user authentication across multiple metaverse platforms without relying on centralised credential storage.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:distributed-authentication-architecture", "labels": [ "Distributed Authentication Architecture" ], "is_subclass_of": [ "Security and Identity", "Identity Management" ], "wikilinks": [ "Cross-Platform Interoperability", "DID Nostr Identity", "Identity Management", "metaverse" ] }, { "id": "distributed-collaboration", "title": "Distributed Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Distributed Collaboration is the set of technologies, protocols, and organisational practices that enable geographically or temporally dispersed individuals and teams to work together on shared tasks, artefacts, and decisions. It encompasses both synchronous modalities\u2014real-time video conferencing, shared virtual workspaces, co-presence mechanisms\u2014and asynchronous modalities such as version-controlled repositories, threaded discussion, and document co-authoring. The field draws on distributed systems theory, human-computer interaction, and organisational science to address challenges of latency, consistency, access control, and remote team coordination. In emerging spatial and AI-augmented contexts, distributed collaboration extends to avatar-based co-presence, AI-assisted meeting facilitation, and decentralised consensus workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-collaboration", "labels": [ "Distributed Collaboration" ], "is_subclass_of": [ "Thing" ], "wikilinks": [] }, { "id": "distributed-communication", "title": "Distributed Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed communication is the exchange of messages between processes running on separate machines across a network, forming the substrate of distributed systems. It encompasses paradigms such as remote procedure calls, message queues, publish-subscribe, and streaming, each managing serialisation, addressing, ordering, and failure handling. Middleware abstracts these mechanics so application components can interact reliably despite network partitions and latency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-communication", "labels": [ "Distributed Communication" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "distributed-computing", "title": "Distributed Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed Computing is a computational paradigm in which networked autonomous nodes (processes, machines, datacentres or geographic regions) coordinate through message passing over partially synchronous networks to solve problems no single node can solve alone or to scale capacity beyond a sing...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:distributed-computing", "labels": [ "Distributed Computing", "Distributed Computing Infrastructure", "Distributed Processing", "Ray Distributed Computing" ], "is_subclass_of": [ "Computing and Cloud", "Network and Communication", "Parallel Computing", "Networked Systems", "Concurrent Computing", "Coordination Systems" ], "wikilinks": [ "Abadi 2012 PACELC", "Actor Model", "Apache Flink", "Apache Kafka", "Apache Software Foundation", "Apache Software Foundation Documentation", "Apache Spark", "Blockchain Consensus", "Brewer 2000 CAP Conjecture", "Bulk Synchronous Parallel", "Byzantine Generals Problem", "Cambridge Computer Laboratory History", "CAP Theorem", "Carbone et al 2015 Apache Flink", "Castro Liskov 1999 PBFT", "Centralised Computing", "Cloud Native", "CNCF", "CNCF Annual Survey 2024", "Concurrent Computing" ] }, { "id": "distributed-consensus", "title": "Distributed Consensus", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Distributed consensus is the fundamental computer science problem of achieving reliable agreement on a shared value or sequence of values across a set of independent processes in a distributed system, despite the possibility of node crashes, network partitions, message delays, and Byzantine (arbitrarily malicious) behaviour. The problem is formalised through three core properties: agreement (all non-faulty nodes decide the same value), validity (the decided value was proposed by some participant), and termination (every non-faulty node eventually decides). The Fischer-Lynch-Paterson (FLP) impossibility theorem establishes that deterministic consensus in a fully asynchronous system is impossible even with one crash-faulty process, forcing all practical protocols to make synchrony assumptions or adopt probabilistic termination.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:distributed-consensus", "labels": [ "Distributed Consensus", "Distributed Systems Consensus", "DistributedConsensus" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "distributed-data-structure", "title": "Distributed Data Structure", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A Distributed Data Structure is an abstract organisational framework for storing, managing, and accessing data across multiple networked computing nodes without centralised coordination. It partitions or replicates data across independent nodes employing consensus protocols to maintain consistency and availability, providing the foundational storage architecture for blockchain systems and other decentralised platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-data-structure", "labels": [ "Distributed Data Structure", "DistributedDataStructure" ], "is_subclass_of": [ "Network Component", "Blockchain Entity" ], "wikilinks": [ "Block Tree", "Byzantine Fault Tolerance Papers", "CAP Theorem", "Distributed Hash Table", "Distributed Systems: Principles and Paradigms", "State Tree", "Trie", "AI Agent System", "Blockchain", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "Merkle Tree", "Transaction Pool" ] }, { "id": "distributed-databases", "title": "Distributed Databases", "domain": "data", "domain_name": "Data", "definition": "A distributed database stores and manages data across multiple networked nodes, presenting a unified logical database while partitioning and replicating data for scale and resilience. It must reconcile the trade-offs of the CAP theorem, choosing among strong consistency, availability, and partition tolerance through consensus, quorum, or conflict-resolution strategies. It underpins large-scale applications that exceed the capacity or fault-tolerance limits of a single server.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-databases", "labels": [ "Distributed Databases", "Distributed Database", "Partition-tolerant Databases" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "distributed-decision-making", "title": "Distributed Decision Making", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Distributed decision making encompasses the full spectrum of collective and algorithmic processes by which multiple autonomous agents, nodes, organisations, or individuals arrive at binding or coordinating choices without relying on a single central authority.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:distributed-decision-making", "labels": [ "Distributed Decision Making", "Distributed Decision Network" ], "is_subclass_of": [ "Workspace Tools", "Organisational Governance", "Distributed Systems", "Collective Intelligence", "Social Choice Theory", "Coordination Mechanisms" ], "wikilinks": [ "Appeal Mechanism", "Arrow Impossibility Theorem", "Asynchronous Collaboration Patterns", "Bitcoin Improvement Proposals", "Centralised Decision Making", "Collective Action Problems", "Collective Intelligence", "Command-and-Control Organisations", "Commit-Reveal Schemes", "Common-Pool Resource Governance", "ComputationalSocialChoiceDomain", "Conflict Resolution", "Consensus Protocols", "Coordination Mechanisms", "DAO Treasury Management", "Decentralised Autonomous Organisations", "Decision Documentation", "Decision Traceability", "Delegation Mechanisms", "DistributedCollaborationDomain" ] }, { "id": "distributed-file-system", "title": "Distributed File System", "domain": "data", "domain_name": "Data", "definition": "A distributed file system presents a unified file or namespace interface over storage that is physically spread across many networked machines. It transparently handles data placement, replication, fault tolerance, and concurrent access so that clients interact with remote, partitioned storage much as they would with a local file system. Such systems scale capacity and throughput beyond a single node while tolerating individual machine failures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:distributed-file-system", "labels": [ "Distributed File System" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "distributed-governance", "title": "Distributed Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed governance is a model of collective decision-making in which authority over a protocol, network or organisation is spread across many independent participants rather than held by a central body. In blockchain systems it is typically enacted through on-chain mechanisms such as token-weighted or quadratic voting, proposals and treasury control, allowing stakeholders to direct upgrades, parameters and resource allocation. It aims to align incentives, resist capture and make governance transparent and verifiable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:distributed-governance", "labels": [ "Distributed Governance" ], "is_subclass_of": [ "Decentralised Governance" ], "wikilinks": [] }, { "id": "distributed-hash-table", "title": "Distributed Hash Table", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed hash table (DHT) is a decentralised data structure that partitions a key-value store across a set of participating nodes so that each node is responsible for only a fraction of the total keyspace, with lookups routed through a structured overlay network in O(log n) hops without any central coordinator. Nodes join and leave dynamically, and the system rebalances key responsibility through consistent hashing or an XOR-metric routing algorithm, tolerating high churn without degrading availability. DHTs form the foundational lookup and routing primitive of peer-to-peer networks, underpinning decentralised content addressing, peer discovery, and distributed storage at global scale.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:distributed-hash-table", "labels": [ "Distributed Hash Table", "Distributed Hash Table Routing" ], "is_subclass_of": [ "Distributed System" ], "wikilinks": [] }, { "id": "distributed-identity", "title": "Distributed Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed Identity (used interchangeably with decentralised identity and self-sovereign identity / SSI) is the architectural paradigm in which natural persons, legal entities, devices and digital objects own and present cryptographically-verifiable identifiers and attribute claims without depen...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:distributed-identity", "labels": [ "Distributed Identity" ], "is_subclass_of": [ "Network Component", "Digital Identity", "Decentralised Web", "Identity Management", "Privacy-Preserving Technology" ], "wikilinks": [ "Age Verification", "Allen 2016 Path to Self-Sovereign Identity", "AnonCreds", "BBS Plus Signatures", "Bluesky AT Protocol", "Boneh Boyen Shacham 2004 Short Group Signatures", "Camenisch & Lysyanskaya 2002 CL Signatures", "Cameron 2005 Laws of Identity", "CBOR", "Centralized Identity Provider", "Credential Schema", "Cross-Border Recognition", "Cryptographic Accumulator", "Cryptographic Wallet", "CryptographyDomain", "Decentralized Identity Foundation", "DID Document", "DIDComm", "Digital Government Services", "DNS" ] }, { "id": "distributed-inference", "title": "Distributed Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Distributed inference is the execution of a machine learning model's forward pass across multiple devices or machines so that models too large or too demanding for a single accelerator can serve predictions. It partitions the model and its computation using strategies such as tensor, pipeline and data parallelism, and coordinates the resulting workers with high-bandwidth interconnects. Distributed inference is essential for serving very large language and vision models at acceptable latency and throughput.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-inference", "labels": [ "Distributed Inference" ], "is_subclass_of": [ "Model Serving", "AI Infrastructure" ], "wikilinks": [] }, { "id": "distributed-key-generation", "title": "Distributed Key Generation", "domain": "security", "domain_name": "Security", "definition": "Distributed Key Generation (DKG) is a cryptographic protocol in which a group of mutually distrusting participants collaboratively compute a shared public key together with secret key shares, without any single party ever learning or holding the complete private key. Each participant contributes randomness so that the final key material is the joint product of all honest parties, providing resilience against compromise of individual nodes. DKG underpins threshold signatures and secure multi-party computation, removing the single point of failure inherent in a centrally generated key.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-key-generation", "labels": [ "Distributed Key Generation" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "distributed-ledger-technology-dlt", "title": "Distributed Ledger Technology (DLT)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed database infrastructure using cryptographic consensus mechanisms to maintain immutable, tamper-resistant records across decentralised peer-to-peer networks without centralised authority.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:distributed-ledger-technology-dlt", "labels": [ "Distributed Ledger Technology (DLT)", "DLT Ledger Design" ], "is_subclass_of": [ "Computing and Cloud", "Network Infrastructure" ], "wikilinks": [ "Block Structure", "Cryptographic Algorithm", "Cryptographic Hash Function", "Decentralized Application", "Distributed Network", "Immutable Record", "ISO 22739", "NIST Blockchain Technology Overview", "Trustless Transaction", "Blockchain", "Byzantine Fault Tolerance", "Consensus Protocol", "Cryptocurrency", "Cryptography", "Data Replication", "Digital Asset", "Digital Signature", "Distributed System", "InfrastructureDomain", "InfrastructureLayer" ] }, { "id": "distributed-ledger-technology", "title": "Distributed Ledger Technology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed Ledger Technology (DLT) is a class of decentralised database protocols in which transaction records are replicated, shared, and synchronised across multiple networked nodes without a central administrator. Consensus mechanisms \u2014 including proof-of-work, proof-of-stake, and Byzantine fault-tolerant protocols \u2014 ensure agreement on the canonical ledger state across participants who may not mutually trust one another. DLT encompasses permissionless public networks such as Bitcoin and Ethereum as well as permissioned enterprise ledgers such as Hyperledger Fabric and R3 Corda, spanning use cases in finance, supply chain, identity management, and tokenised assets. The technology achieves tamper-evidence through cryptographic chaining of records, making retrospective alteration computationally infeasible.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-ledger-technology", "labels": [ "Distributed Ledger Technology" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "distributed-ledger", "title": "Distributed Ledger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Distributed Ledger (or Distributed Ledger Technology, DLT) is a cryptographically-secured, append-only data structure replicated across a set of independently-operated nodes that collectively reach agreement on the canonical sequence and validity of state transitions through a deterministic", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:distributed-ledger", "labels": [ "Distributed Ledger", "DistributedLedger", "Permissioned Distributed Ledger" ], "is_subclass_of": [ "Network Component", "Data Structure", "Replicated State Machine", "Append-Only Log", "Distributed System", "Multi-Party Database" ], "wikilinks": [ "Androulaki 2018 Hyperledger Fabric EuroSys", "Append-Only Log", "Atomic Settlement", "Avalanche", "Baird 2016 Hashgraph Consensus", "Bank for International Settlements", "Bank of England", "BIS 2023 Annual Economic Report Unified Ledger", "BIS Innovation Hub", "BlackRock BUIDL", "BoE BIS 2023 Project Rosalind Phase 1", "BoE HMT 2023 Digital Pound Consultation", "Brewer 2000 CAP Theorem", "Brown Carlyle Grigg Hearn 2016 Corda Introduction", "Cambridge Centre for Alternative Finance", "Castro Liskov 1999 PBFT", "CBDC", "CCAF 2024 Global Cryptoasset Benchmarking Study", "Centralised Ledger", "Channel Isolation Pattern" ] }, { "id": "distributed-logseq-knowledge-network", "title": "Distributed Logseq Knowledge Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "LogNet is a networked knowledge-management infrastructure layer that interconnects Logseq graph nodes, enabling cross-graph querying, link resolution, and distributed publishing of structured ontology pages. It provides the network substrate over which linked-data references between pages are resolved and semantic annotations are aggregated, acting as the connective tissue between individual Logseq knowledge graphs and wider ontology repositories.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:distributed-logseq-knowledge-network", "labels": [ "Distributed Logseq Knowledge Network", "lognet" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "distributed-protocol", "title": "Distributed Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Distributed Protocol is a formally specified set of rules, message formats, and procedures governing communication, coordination, and state synchronization among independent nodes in a distributed network without centralized control.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:distributed-protocol", "labels": [ "Distributed Protocol" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity" ], "wikilinks": [ "Block Propagation Protocol", "Byzantine Agreement Protocols", "Distributed Systems: Concepts and Design", "Peer Discovery Protocol", "Peer-to-Peer Networks", "State Synchronization Protocol", "Transaction Broadcast Protocol", "Agreement Protocol", "AI Agent System", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "Consensus Mechanism", "Gossip Protocol" ] }, { "id": "distributed-sensing", "title": "Distributed Sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Distributed sensing is the collection and fusion of environmental information from multiple spatially separated sensing agents, such as the robots in a multi-robot system or swarm, rather than from a single centralised sensor. It exploits the combined spatial coverage and redundancy of many sensors to build a more complete, robust picture of the environment than any single agent could obtain alone. Distributed sensing is central to swarm robotics applications such as environmental monitoring, search and rescue, and cooperative mapping.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:distributed-sensing", "labels": [ "Distributed Sensing" ], "is_subclass_of": [ "Multi-Robot Systems" ], "wikilinks": [] }, { "id": "distributed-storage", "title": "Distributed Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed storage is a class of storage system architecture in which data is partitioned, replicated, and managed across multiple physically separate nodes or clusters to achieve scalability, fault tolerance, and high availability beyond the capacity of any single machine. Such systems employ replication protocols, erasure coding, consistent hashing, and distributed consensus algorithms to maintain data integrity and consistency under node failure, network partition, and concurrent access. Examples include object stores such as Amazon S3, distributed file systems such as HDFS and Ceph, and NewSQL databases with sharded storage layers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:distributed-storage", "labels": [ "Distributed Storage", "Distributed Storage System", "DistributedStorage" ], "is_subclass_of": [ "Distributed System" ], "wikilinks": [] }, { "id": "distributed-system-architecture", "title": "Distributed System Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An architectural paradigm for metaverse systems that distributes computing resources, data storage, and processing across multiple interconnected nodes to achieve scalability, fault tolerance, and low-latency experiences whilst supporting s of concurrent users.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:distributed-system-architecture", "labels": [ "Distributed System Architecture" ], "is_subclass_of": [ "Platform and Environment", "System Architecture" ], "wikilinks": [ "Autonomous Robot", "metaverse", "Metaverse Infrastructure", "System Architecture" ] }, { "id": "distributed-system-protocol", "title": "Distributed System Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A Distributed System Protocol is a formal specification of rules, message formats, and coordination procedures that govern how autonomous nodes in a networked system communicate, synchronise state, and jointly accomplish tasks without centralised control. Such protocols address the fundamental challenges of partial failure, network partitioning, and asynchronous message delivery described by the CAP theorem, providing mechanisms for consensus, leader election, gossip dissemination, and fault recovery. They underpin peer-to-peer networks, blockchain infrastructures, distributed databases, and large-scale cloud orchestration systems. Correctness properties \u2014 safety, liveness, and eventual consistency \u2014 are formally analysed and proven against adversarial models such as Byzantine fault tolerance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:distributed-system-protocol", "labels": [ "Distributed System Protocol" ], "is_subclass_of": [ "Protocol Layer" ], "wikilinks": [ "Blockchain" ] }, { "id": "distributed-system", "title": "Distributed System", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed system is a collection of autonomous computing nodes interconnected by a network, coordinating their actions through message passing to appear as a single coherent system to end users or applications. The architecture is governed by fundamental trade-offs formalised in the CAP theorem \u2014 a system can guarantee at most two of consistency, availability, and partition tolerance simultaneously. Correctness in the presence of node failures and malicious actors is addressed by Byzantine Fault Tolerance protocols and consensus mechanisms such as Paxos, Raft, and Practical Byzantine Fault Tolerance. Canonical instantiations span peer-to-peer networks, blockchain ledgers, microservices architectures, distributed databases, and large-scale cloud infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:distributed-system", "labels": [ "Distributed System" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Blockchain" ] }, { "id": "distributed-systems-security", "title": "Distributed Systems Security", "domain": "security", "domain_name": "Security", "definition": "Distributed systems security is the discipline of protecting confidentiality, integrity and availability across systems whose components run on separate machines and communicate over untrusted networks. It addresses threats unique to distribution, including partial failure, Byzantine participants, replay and partition attacks, and the absence of a single trusted authority. Techniques span authenticated and encrypted channels, fault-tolerant consensus, access control and threat modelling tailored to the multi-node setting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-systems-security", "labels": [ "Distributed Systems Security" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "distributed-systems-theory", "title": "Distributed Systems Theory", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The theoretical study of computational systems whose components run on separate networked machines and coordinate by passing messages, encompassing fundamental impossibility results, consistency models, and formal proofs of consensus and replication protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-systems-theory", "labels": [ "Distributed Systems Theory", "Distributed Consistency" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Algorithm", "Consensus Algorithm", "Fault Tolerance", "Blockchain", "Distributed Systems" ] }, { "id": "distributed-systems", "title": "Distributed Systems", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Collections of independent computing nodes that coordinate through message passing to present a unified service, providing fault tolerance, horizontal scalability, and geographic distribution. Distributed systems are foundational to metaverse platforms, blockchain networks, and large-scale AI inference pipelines, where no single node holds all state and consistency guarantees (CAP theorem trade-offs) govern design choices.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:distributed-systems", "labels": [ "Distributed Systems", "Distributed Systems Infrastructure", "DistributedSystems" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "distributed-team-collaboration", "title": "Distributed Team Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Distributed team collaboration encompasses the practices, tools, and organisational patterns that enable geographically dispersed individuals working across multiple time zones to coordinate effectively on shared goals without co-location. It integrates asynchronous communication protocols, synchronous video conferencing, shared digital workspaces, version-controlled artefacts, and social norms around documentation and decision-making transparency to replicate \u2014 and often surpass \u2014 the coordination bandwidth of co-located teams. The discipline has grown from a niche practice of open-source communities and multinational corporations into a mainstream organisational capability following the global shift to remote work accelerated by the 2020 pandemic.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-team-collaboration", "labels": [ "Distributed Team Collaboration" ], "is_subclass_of": [ "Distributed Collaboration", "Workspace Tools" ], "wikilinks": [] }, { "id": "distributed-teams", "title": "Distributed Teams", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed teams are groups of collaborators who work from different geographic locations and often across time zones, coordinating primarily through digital communication and collaboration tools. They rely on asynchronous workflows, shared documentation, and recorded or AI-assisted meetings to maintain alignment without co-location. They have become a dominant model of knowledge work, trading the spontaneity of the office for flexibility and access to a global talent pool.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:distributed-teams", "labels": [ "Distributed Teams", "Distributed Team", "Multilingual Teams" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "distributed-teamwork", "title": "Distributed Teamwork", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Distributed teamwork is the coordinated effort of team members who are geographically dispersed and collaborate primarily through digital and immersive tools rather than co-located presence. It addresses the challenges of time-zone separation, reduced informal contact, and asynchronous decision-making by combining shared artefacts, communication norms, and presence technologies. In spatial computing, distributed teamwork increasingly leverages virtual environments that recreate co-presence and shared spatial context for remote collaborators.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-teamwork", "labels": [ "Distributed Teamwork" ], "is_subclass_of": [ "Distributed Collaboration", "Workspace Tools" ], "wikilinks": [] }, { "id": "distributed-tracing", "title": "Distributed Tracing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Distributed tracing is an observability technique that follows a single request as it propagates across the many services of a distributed system, recording the timing and causal relationships of each operation. Each unit of work is captured as a span, and spans linked by a shared trace identifier form a trace that reconstructs the request's end-to-end path. It is essential for diagnosing latency, dependencies and failures in microservice architectures where no single component holds the full picture.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-tracing", "labels": [ "Distributed Tracing" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "distributed-training", "title": "distributed training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Distributed training is a machine learning paradigm that partitions computation, data, and model parameters across multiple processors, accelerators, or networked nodes so that training jobs too large or too slow for a single device can complete at scale. The four primary parallelism axes are data parallelism (each worker processes a distinct data shard and aggregates gradients), model parallelism (layers assigned to different devices), tensor parallelism (individual weight matrices sharded across devices), and pipeline parallelism (the forward pass staged across devices as a micro-batch pipeline). These strategies are composed into multi-dimensional schemes \u2014 commonly called 3D parallelism \u2014 that balance compute, memory, and communication trade-offs for workloads ranging from fine-tuning to pre-training frontier models with hundreds of billions of parameters.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:distributed-training", "labels": [ "Distributed Training", "Distributed Training Recovery", "Large-Scale Distributed Training" ], "is_subclass_of": [ "AI Infrastructure", "Machine Learning", "High-Performance Computing" ], "wikilinks": [] }, { "id": "distributed-transaction", "title": "Distributed Transaction", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed transaction is a unit of work whose operations span two or more independent data stores, services or network nodes, yet must complete with all-or-nothing atomicity across every participant. Coordinating such a transaction requires protocols that agree on a single outcome despite partial failures, network partitions and concurrent activity at each site. Classical coordination uses atomic commit protocols, while modern systems often relax strict atomicity for availability using compensating workflows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-transaction", "labels": [ "Distributed Transaction" ], "is_subclass_of": [ "Transaction Processing" ], "wikilinks": [] }, { "id": "distributed-trust", "title": "Distributed Trust", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed trust is a model in which confidence in a system arises from the collective behaviour of many independent participants rather than from a single trusted authority. Through cryptography, consensus and economic incentives, no single party needs to be trusted for the system as a whole to behave correctly. It is the foundational principle behind blockchains and other decentralised infrastructures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:distributed-trust", "labels": [ "Distributed Trust" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "distributed-validator-technology", "title": "Distributed Validator Technology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Distributed Validator Technology (DVT) splits the signing key and duties of a single blockchain validator across multiple independent nodes, so that consensus on each validator action requires threshold agreement among the participating operators. It uses threshold cryptography to reconstruct or aggregate signatures without any single node holding the complete private key, removing single points of failure. DVT is used in proof-of-stake networks such as Ethereum to improve validator fault tolerance, decentralisation, and resilience against slashing from individual node outages.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:distributed-validator-technology", "labels": [ "Distributed Validator Technology" ], "is_subclass_of": [ "Validator" ], "wikilinks": [] }, { "id": "distributed-work", "title": "Distributed Work", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Organisational models, practices, and enabling technologies that allow geographically dispersed teams to collaborate effectively whilst maintaining productivity, cohesion, and wellbeing. Distributed work encompasses remote, hybrid, and globally distributed team configurations, supported by asynchronous communication norms, shared documentation practices, and virtual presence technologies ranging from video conferencing to XR meeting spaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:distributed-work", "labels": [ "Distributed Work", "DistributedWork" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Blockchain" ] }, { "id": "distributor", "title": "Distributor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Distributor is a natural or legal person in the AI supply chain, other than the provider or the importer, that makes an AI system available on the Union market without modifying it. Distributors bear verification and cooperation duties under the EU AI Act Article 24, including confirming CE marking, ensuring documentation completeness, and informing authorities of suspected non-compliance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:distributor", "labels": [ "Distributor" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Blockchain", "MetaverseDomain" ] }, { "id": "disturbance-rejection", "title": "Disturbance Rejection", "domain": "robotics", "domain_name": "Robotics", "definition": "Disturbance rejection is the capacity of a control system to maintain desired output behaviour despite unmeasured external disturbances or model uncertainties acting on the plant. It is typically achieved through feedback control, integral action, or disturbance observers that estimate and cancel the disturbance's effect. In robotics it is essential for maintaining stable setpoints and trajectories when a robot is subject to external forces, friction, or sensor noise.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:disturbance-rejection", "labels": [ "Disturbance Rejection" ], "is_subclass_of": [ "Feedback Control" ], "wikilinks": [] }, { "id": "diversity-non-discrimination-and-fairness", "title": "Diversity, Non-Discrimination, and Fairness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Diversity, Non-Discrimination, and Fairness is a foundational trustworthiness dimension of responsible AI that requires systems to avoid unfair bias against protected characteristics (sex, race, religion, disability, age, sexual orientation), ensure equitable treatment and outcomes across demographic groups, implement accessibility and universal design for people with diverse abilities, and enable inclusive stakeholder participation throughout the AI development lifecycle. It encompasses three interdependent pillars: unfair-bias avoidance through pre-processing data corrections, in-processing fairness constraints, and post-processing adjustments; accessibility and universal design aligned with WCAG and the European Accessibility Act; and participatory design methodologies that co-create systems with affected communities. Legal mandates including the EU AI Act and GDPR enforce these requirements with substantial penalties.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:diversity-non-discrimination-and-fairness", "labels": [ "Diversity, Non-Discrimination, and Fairness" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU Charter Article 21", "ISO/IEC TR 24027", "WCAG", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin", "EU AI Act" ] }, { "id": "do-ra", "title": "DoRA", "domain": "ai", "domain_name": "Ai", "definition": "DoRA (Weight-Decomposed Low-Rank Adaptation) is a parameter-efficient fine-tuning method that decomposes pretrained weights into separate magnitude and direction components, applying low-rank updates only to the directional component while learning the magnitude independently. By separating these two degrees of freedom, DoRA more closely mirrors the learning dynamics of full fine-tuning than standard LoRA, improving accuracy on many tasks at comparable parameter cost and without added inference latency once merged. It is used to adapt large language and vision models efficiently.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:do-ra", "labels": [ "DoRA", "DORA" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning", "Low-Rank Adaptation", "Transfer Learning" ], "wikilinks": [] }, { "id": "docker-containerisation-platform", "title": "Docker Containerisation Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Docker is an open-source platform that automates the deployment, scaling, and management of applications by packaging them together with their runtime dependencies into lightweight, portable containers built on Linux kernel primitives (namespaces and cgroups). Unlike virtual machines, Docker containers share the host operating system kernel, providing process and filesystem isolation with far lower overhead while guaranteeing consistent execution across heterogeneous computing environments. Launched in 2013 by Docker Inc., the platform introduced an intuitive developer-facing toolchain, a layered image format, and the Docker Hub public registry, collectively mainstreaming container technology and catalysing the cloud-native ecosystem. Docker standardised container packaging through the Open Container Initiative (OCI) specification and remains the dominant interface for building, distributing, and running container images in both development and production contexts.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:docker-containerisation-platform", "labels": [ "Docker Containerisation Platform", "Docker" ], "is_subclass_of": [ "Software Platform" ], "wikilinks": [] }, { "id": "document-comments", "title": "Document Comments", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Document Comments are annotations anchored to specific sections of a shared document that allow collaborators to ask questions, suggest changes, or provide feedback without altering the primary content. They support threaded replies, @mentions, and resolution states so discussions are traceable and actionable. This asynchronous feedback mechanism is essential for distributed review workflows across time zones.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:document-comments", "labels": [ "Document Comments", "Comments" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "document-processing", "title": "Document Processing", "domain": "ai", "domain_name": "Ai", "definition": "The automated ingestion, interpretation, and transformation of documents \u2014 scans, PDFs, forms, invoices, contracts, and email \u2014 into structured, machine-actionable data, combining optical character recognition, layout analysis, classification, and entity extraction; in its AI-driven form (intelligent document processing) it feeds downstream automation, search, and analytics with high-accuracy extracted content.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:document-processing", "labels": [ "Document Processing" ], "is_subclass_of": [ "Business Process Automation" ], "wikilinks": [ "Optical Character Recognition", "Enterprise Search", "Robotic Process Automation" ] }, { "id": "document-retrieval", "title": "Document Retrieval", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Document Retrieval is the process of identifying and returning relevant documents from a corpus in response to an information need expressed as a query. It forms the foundational layer of search engines, question-answering systems, and retrieval-augmented generation pipelines. Retrieval methods range from sparse keyword matching to dense neural embedding approaches that encode semantic similarity. Effectiveness is typically measured using metrics such as precision, recall, mean reciprocal rank, and normalised discounted cumulative gain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:document-retrieval", "labels": [ "Document Retrieval" ], "is_subclass_of": [ "Information Retrieval", "Natural Language Processing", "Machine Learning" ], "wikilinks": [] }, { "id": "document-store", "title": "Document Store", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A document store is a category of NoSQL database that persists, retrieves, and manages data as self-describing documents, typically encoded as JSON, BSON, or XML. Each document is a flexible, schema-optional aggregate that groups related data together, allowing nested structures and varying fields across records. Document stores favour horizontal scalability and developer-friendly data modelling over the rigid tabular schema of relational databases.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:document-store", "labels": [ "Document Store" ], "is_subclass_of": [ "Database", "NoSQL Database" ], "wikilinks": [] }, { "id": "document-summarisation", "title": "Document Summarisation", "domain": "ai", "domain_name": "Ai", "definition": "Document summarisation is the natural language processing task of producing a concise, faithful representation of the salient information in one or more source documents. It encompasses extractive approaches, which select and concatenate important spans, and abstractive approaches, which generate new text that paraphrases the content. Modern systems are built predominantly on transformer-based large language models and are evaluated for informativeness, coherence, and factual consistency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:document-summarisation", "labels": [ "Document Summarisation" ], "is_subclass_of": [ "Natural Language Processing", "Text Generation", "Content Generation" ], "wikilinks": [] }, { "id": "document-verification", "title": "Document Verification", "domain": "finance", "domain_name": "Finance", "definition": "Document verification is the process of confirming that an identity or supporting document presented by a customer is authentic, unaltered, and belongs to the presenting individual. In financial onboarding it combines optical capture, security-feature and template checks, data extraction, and cross-referencing against issuing-authority records or biometric liveness, forming a core step within know-your-customer and anti-money-laundering controls.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:document-verification", "labels": [ "Document Verification" ], "is_subclass_of": [ "Know Your Customer" ], "wikilinks": [] }, { "id": "documentation-as-code", "title": "Documentation As Code", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Documentation as Code is a software-engineering practice in which technical documentation is authored in plain-text markup, stored in version control alongside source code, and built and published through the same automated pipelines used for software. It applies developer workflows such as pull requests, code review, linting, and continuous integration to documentation. The approach keeps docs synchronised with code and improves consistency, traceability, and collaboration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:documentation-as-code", "labels": [ "Documentation As Code" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "documentation-generation", "title": "Documentation Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Documentation Generation is the automated production of human-readable technical documentation\u2014API references, code comments, user guides, and release notes\u2014using large language models and natural language generation pipelines. By coupling static analysis, code execution traces, and prompt engineering, these systems reduce the documentation burden on developers while improving consistency and coverage.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:documentation-generation", "labels": [ "Documentation Generation" ], "is_subclass_of": [ "AI Application", "Generative AI", "Natural Language Generation", "Natural Language Processing" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Digital Twin", "Generative Ai" ] }, { "id": "documentation-standards", "title": "Documentation Standards", "domain": "standards", "domain_name": "Standards", "definition": "Formal specifications and technical guidelines established by standards bodies to ensure interoperability, consistency, and quality across metaverse platforms, encompassing terminology, data formats, interfaces, and ethical considerations for immersive technology development.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:documentation-standards", "labels": [ "Documentation Standards" ], "is_subclass_of": [ "Technical Standards" ], "wikilinks": [ "IEEE 2048.101-2023", "IEEE 3079-2020", "IEEE 7014-2024", "IEEE P2048", "IEEE P7016", "ISO/IEC 18039:2019", "ITU", "Metaverse Standards Forum", "Autonomous Robot", "metaverse", "Technical Standards" ] }, { "id": "documentation", "title": "Documentation", "domain": "data", "domain_name": "Data", "definition": "Documentation is the structured body of written, diagrammatic, or interactive material that describes the purpose, design, behaviour, and use of a system, dataset, process, or software artefact. It serves as the primary medium through which knowledge about an artefact is transferred between its creators and its users, maintainers, and auditors. Effective documentation spans reference material, conceptual explanations, tutorials, and procedural guides, and is increasingly treated as a versioned, testable component of the artefact itself rather than an afterthought.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:documentation", "labels": [ "Documentation" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "domain-access-control", "title": "Domain Access Control", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Domain Access Control is the set of mechanisms that govern which users, avatars, or services may enter, view, or modify resources within a bounded virtual-world domain or server cluster. In metaverse platforms it enforces ownership and permission boundaries at the level of a hosted domain, mediating connection requests and content rights. It is foundational to multi-tenant virtual environments where independent operators host interoperable spaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:domain-access-control", "labels": [ "Domain Access Control" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [] }, { "id": "domain-adaptation", "title": "domain adaptation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Domain adaptation is a sub-field of transfer learning that addresses the domain shift problem: a model trained on a labelled source distribution degrades when applied to a target distribution whose marginal or conditional statistics differ. Methods fall into feature-alignment approaches (learning domain-invariant representations via adversarial training, maximum mean discrepancy minimisation, or optimal transport), instance re-weighting schemes that correct for covariate shift, and self-training or pseudo-labelling strategies that exploit unlabelled target data. It spans unsupervised, semi-supervised, and multi-source settings and underpins practical deployment of models in NLP, computer vision, speech recognition, and scientific computing wherever labelled target data is scarce or costly.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:domain-adaptation", "labels": [ "Domain Adaptation", "Domain Adaptation Without Fine-Tuning" ], "is_subclass_of": [ "Transfer Learning", "AI Technique", "Machine Learning", "Representation Learning" ], "wikilinks": [] }, { "id": "domain-expert-contact-index", "title": "Domain Expert Contact Index", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A curated personal contact and collaboration index tracking domain experts, industry practitioners, and potential collaborators relevant to AI, immersive technology, and spatial computing initiatives. Entries record relationship context, project overlap, and follow-up actions to support relationship management and opportunity development.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:domain-expert-contact-index", "labels": [ "Domain Expert Contact Index", "PEOPLE", "People Analytics Dashboard" ], "is_subclass_of": [ "Knowledge Management" ], "wikilinks": [ "Could", "Education and AI", "Large Language Models" ] }, { "id": "domain-model", "title": "Domain Model", "domain": "data", "domain_name": "Data", "definition": "A Domain Model is a conceptual representation of the entities, attributes, relationships, and rules of a particular problem domain, independent of any specific software implementation. It captures the shared vocabulary and structural constraints that stakeholders agree describe the domain, serving as a blueprint for data schemas, APIs, and system behaviour. Standardised domain models enable interoperability by giving distinct systems a common semantic reference.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:domain-model", "labels": [ "Domain Model" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "domain-name-system", "title": "Domain Name System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Domain Name System (DNS) is a hierarchical, distributed naming system that translates human-readable domain names into the numerical addresses used to locate computers and services on a network. It is a foundational component of the Internet, resolving names through a delegated tree of authoritative servers, recursive resolvers and caching layers. DNS also carries service, security and routing metadata that many higher-level protocols depend upon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:domain-name-system", "labels": [ "Domain Name System" ], "is_subclass_of": [ "Networking Infrastructure" ], "wikilinks": [] }, { "id": "domain-ontology", "title": "Domain Ontology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Domain Ontology is a formal, explicit, machine-readable specification of a shared conceptualisation restricted to a delimited subject domain (clinical medicine, financial instruments, cultural heritage, manufacturing robotics, gene products, scholarly publications, e-commerce products), followi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:domain-ontology", "labels": [ "Domain Ontology", "DomainOntology" ], "is_subclass_of": [ "Software Engineering", "Data Management", "Formal Specification", "Ontology", "Knowledge Representation Artifact", "Controlled Vocabulary", "Conceptual Model" ], "wikilinks": [ "Annotation Property", "Application Ontology", "Ashburner et al 2000 Gene Ontology", "Automated Reasoning", "Axiom Set", "Baader Calvanese McGuinness Nardi Patel-Schneider 2010 Description Logic Handbook", "Bennett Allemang et al 2018 FIBO", "Berners-Lee 2006 Linked Data Design Issues", "Bizer Heath Berners-Lee 2009 Linked Data Story So Far", "Caufield et al 2024 OntoGPT SPIRES", "Class Hierarchy", "Closed-World Negation as Failure", "Conceptual Model", "Conceptualisation", "Data Property", "Description Logic", "Document Schema", "Doerr 2003 CIDOC CRM", "Domain Expert", "Edge et al 2024 Graph RAG Microsoft Research" ] }, { "id": "domain-randomisation", "title": "Domain Randomisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Domain randomisation is a technique for training robot and agent policies in simulation by randomising the parameters of the simulated environment, such as textures, lighting, dynamics, masses and sensor noise, so that the real world appears as just another variation. By forcing a policy to be robust across a wide distribution of simulated conditions, the approach narrows the reality gap and improves zero-shot or few-shot transfer of policies trained purely in simulation to physical hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:domain-randomisation", "labels": [ "Domain Randomisation", "Domain Randomization" ], "is_subclass_of": [ "Robo Actuation and Control", "Robo Actuation and Control" ], "wikilinks": [] }, { "id": "domain-driven-design", "title": "Domain-Driven Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Domain-driven design (DDD) is an approach to software development that centres the design on a deep, shared model of the business domain, expressed in a ubiquitous language common to engineers and domain experts. It provides strategic patterns for partitioning large systems into bounded contexts and tactical patterns such as aggregates, entities and value objects for structuring the model within each context. DDD aims to keep complex software aligned with the evolving realities of the business it serves.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:domain-driven-design", "labels": [ "Domain-Driven Design", "Domain Driven Design" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "domain-specific-llms", "title": "Domain-Specific LLMs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Large language models that are pre-trained or fine-tuned on specialized corpora to achieve superior performance in a particular field, such as finance or law.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:domain-specific-llms", "labels": [ "Domain-Specific LLMs" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "domain", "title": "Domain", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Domain is a coherent and bounded sphere of knowledge or subject area within formal ontology engineering, establishing the scope of conceptualisation\u2014the set of entities, relationships, and axioms characterising a particular area of discourse. Domains serve as fundamental organising principles enabling modular knowledge organisation and interoperability between specialised knowledge systems, realised through namespace declarations, import mechanisms, and modular ontology structures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:domain", "labels": [ "Domain" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "DNSSEC", "W3C OWL 2", "Blockchain" ] }, { "id": "doppler-radar", "title": "Doppler Radar", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:doppler-radar", "labels": [ "Doppler Radar" ], "is_subclass_of": [ "Radar Instrument" ], "wikilinks": [] }, { "id": "dot-product", "title": "Dot Product", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The dot product is a linear algebra operation that combines two vectors of equal dimension into a single scalar by summing the products of their corresponding components. Geometrically it measures how much two vectors align, and it is proportional to the cosine of the angle between them scaled by their magnitudes. In machine learning the dot product underlies similarity measures such as cosine similarity, attention scoring, and nearest-neighbour vector search.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:dot-product", "labels": [ "Dot Product" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "double-materiality", "title": "Double Materiality", "domain": "governance", "domain_name": "Governance", "definition": "A principle of sustainability reporting under which a topic must be disclosed if it is material from either of two perspectives: financial materiality, where sustainability matters create risks or opportunities that affect the company's cash flows and enterprise value, and impact materiality, where the company's own activities and value chain materially affect people or the environment. Adopted as the mandatory lens of the EU Corporate Sustainability Reporting Directive and its ESRS standards, it determines the scope of every disclosure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:double-materiality", "labels": [ "Double Materiality" ], "is_subclass_of": [ "Materiality Assessment" ], "wikilinks": [ "Materiality Assessment", "Corporate Sustainability Reporting", "Corporate Sustainability Reporting Directive" ] }, { "id": "double-spending", "title": "Double Spending", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Double spending is the fraudulent attempt to spend the same digital asset more than once by broadcasting conflicting transactions to different parts of a blockchain network before they are confirmed. It represents the fundamental security problem that consensus mechanisms are designed to prevent in distributed ledger systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:double-spending", "labels": [ "Double Spending", "Double Spend", "Double-Spend Attack", "Double-Spend Problem", "Double-Spending", "Double-Spending Problem", "Double-spending", "double-spending" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "double-blind-review", "title": "Double-Blind Review", "domain": "ai", "domain_name": "Ai", "definition": "Double-blind review is a peer review process in which both the identity of the authors and the identity of the reviewers are withheld from each other, intended to reduce bias arising from reputation, affiliation or personal relationships. It is widely used by academic conferences and journals, including major machine learning venues, to assess submissions on the basis of their technical merit alone. Implementing double-blind review typically requires anonymised manuscripts and a submission system that separates author and reviewer identities throughout the process.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:double-blind-review", "labels": [ "Double-Blind Review" ], "is_subclass_of": [ "Peer Review" ], "wikilinks": [] }, { "id": "double-spend-prevention", "title": "Double-Spend Prevention", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Double-spend prevention refers to the set of cryptographic, consensus-based, and protocol-level mechanisms that ensure a given unit of digital value cannot be spent more than once within a payment or transaction system. This problem is fundamental to digital money because, unlike physical currency, digital data can be trivially copied; preventing double-spending without a trusted central authority was the key unsolved challenge that Satoshi Nakamoto's Bitcoin whitepaper addressed through the combination of a public transaction ledger and proof-of-work consensus.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:double-spend-prevention", "labels": [ "Double-Spend Prevention", "Double Spend Prevention", "Double Spending Prevention", "Double-Spending Prevention" ], "is_subclass_of": [ "Blockchain Security" ], "wikilinks": [] }, { "id": "dr-o-hare-writing-for-log-seq", "title": "Dr O'Hare Writing for LogSeq", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A prescriptive style guide authored by Dr John O'Hare specifying the structural, syntactic, and tonal conventions for writing public-facing knowledge graph pages in Logseq. It mandates nested bullet-point outlines, UK English prose, Logseq wiki-link syntax, dense academic-conversational authorial voice, and heavy inline citation with external links.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:dr-o-hare-writing-for-log-seq", "labels": [ "Dr O'Hare Writing for LogSeq" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Author YEAR", "hulsmann2008ethics", "Nakamoto 2008", "WikiLink", "WikiLinks", "cypherpunk", "Digital Asset Risks" ] }, { "id": "drake", "title": "Drake", "domain": "robotics", "domain_name": "Robotics", "definition": "Drake is an open-source C++/Python toolbox for model-based design and verification of robotics systems, originally developed at MIT and maintained by the Toyota Research Institute. It provides rigorous multibody dynamics, kinematics, and optimisation tooling, including analytical gradients, contact modelling, and trajectory optimisation. Drake is widely used for simulating manipulators and mobile robots and for solving collision-detection and motion-planning problems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:drake", "labels": [ "Drake", "Drake Toolbox" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "dreamlab-creative-technology-collective", "title": "DreamLab Creative Technology Collective", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Dreamlab is a UK-based creative technologist collective offering multidisciplinary services spanning AI and machine learning, virtual production, immersive XR experiences, generative media, spatial computing, and strategic R&D partnerships. It operates at the intersection of advanced AI infrastructure, real-time rendering, and immersive storytelling for creative and enterprise clients.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:dreamlab-creative-technology-collective", "labels": [ "DreamLab Creative Technology Collective", "Dreamlab" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "drift-ice", "title": "Drift Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:drift-ice", "labels": [ "Drift Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "driver-software", "title": "Driver Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "System-level software components that enable communication between operating systems and extended reality (XR) hardware devices, translating high-level application commands into hardware-specific instructions for VR headsets, AR glasses, haptic controllers, and spatial computing peripherals.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:driver-software", "labels": [ "Driver Software", "Driver Implementation", "Vendor Driver" ], "is_subclass_of": [ "Platform and Environment", "System Software" ], "wikilinks": [ "Sensor Input", "Extended Reality Xr", "metaverse", "System Software" ] }, { "id": "drone-navigation", "title": "Drone Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Drone navigation is the set of methods by which unmanned aerial vehicles determine position, plan routes and control flight to reach objectives.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:drone-navigation", "labels": [ "Drone Navigation" ], "is_subclass_of": [ "Mobile Robotics", "Navigation and Planning" ], "wikilinks": [ "Inertial Measurement Unit", "Path Planning", "Mapping", "SLAM", "Mobile Robotics", "https://en.wikipedia.org/wiki/Unmanned_aerial_vehicle", "https://ardupilot.org/" ] }, { "id": "dropout", "title": "Dropout", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Dropout is a regularisation technique for neural network training in which a randomly selected fraction of neuron activations is set to zero during each forward pass, preventing neurons from co-adapting and forcing the network to learn redundant representations. By randomly deactivating 20\u201350% of units per training step, dropout acts as an ensemble method \u2014 each mini-batch trains a slightly different network architecture \u2014 significantly reducing overfitting on limited training datasets. At inference time, all neurons are active but their outputs are scaled by the retention probability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:dropout", "labels": [ "Dropout", "Dropout Regularisation", "Dropout Training" ], "is_subclass_of": [ "Regularisation" ], "wikilinks": [ "ISO/IEC 22989:2022", "NIST AI RMF", "ArtificialIntelligenceDomain", "Autonomous Robot" ] }, { "id": "drought-monitoring", "title": "Drought Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:drought-monitoring", "labels": [ "Drought Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "drug-discovery-ai", "title": "Drug Discovery AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Drug Discovery AI encompasses artificial intelligence systems that accelerate pharmaceutical research and development through automated molecular design, virtual screening, target identification, toxicity prediction, and clinical trial optimisation. These systems integrate cheminformatics, molecular modelling, and machine learning \u2014 including graph neural networks and generative models \u2014 to reduce drug development timelines and costs whilst improving candidate success rates.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:drug-discovery-ai", "labels": [ "Drug Discovery AI" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Generative Models", "Molecular Design", "Computer Vision", "Graph Neural Network", "Infrastructure", "Medical AI", "MetaverseDomain" ] }, { "id": "drug-discovery", "title": "drug discovery", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Drug discovery is the multidisciplinary scientific process of identifying and validating novel therapeutic compounds that modulate disease-relevant biological targets, spanning target identification, hit discovery, lead optimisation, and preclinical candidate nomination. The modern discipline applies machine learning, deep learning, graph neural networks, and generative modelling to predict molecular properties such as binding affinity, selectivity, and ADMET profiles \u2014 dramatically accelerating virtual screening and de novo molecular design. Structural biology tools including protein structure prediction (e.g. AlphaFold) underpin structure-based drug design by supplying accurate 3D target models at proteome scale. The field sits at the intersection of cheminformatics, structural biology, computational chemistry, and clinical informatics, with AI methods increasingly embedded throughout the entire development pipeline from target discovery through biomarker identification and patient stratification.", "entityType": "Class", "qualityScore": 0.92, "maturity": "emerging", "iri": "urn:ngm:class:drug-discovery", "labels": [ "Drug Discovery" ], "is_subclass_of": [ "AI Application", "Computational Biology", "Biomedical Informatics", "Computer-Aided Drug Design" ], "wikilinks": [] }, { "id": "dublin-core", "title": "Dublin Core", "domain": "data", "domain_name": "Data", "definition": "Dublin Core is a standardised set of fifteen core metadata elements defined by the Dublin Core Metadata Initiative (DCMI) to describe digital and physical resources in a simple, interoperable manner. It provides a lowest-common-denominator vocabulary applicable across libraries, repositories, the web, and cross-domain data exchange.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:dublin-core", "labels": [ "Dublin Core", "Dublin Core Metadata Initiative" ], "is_subclass_of": [ "Metadata Standard" ], "wikilinks": [] }, { "id": "dune-analytics", "title": "Dune Analytics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Dune Analytics is a platform for querying and visualising blockchain data through user-created dashboards and shared queries.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dune-analytics", "labels": [ "Dune Analytics" ], "is_subclass_of": [ "Blockchain Analytics" ], "wikilinks": [ "Blockchain Analytics", "Data Visualisation", "Ethereum" ] }, { "id": "dutch-auction", "title": "Dutch Auction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Dutch auction is a descending-price auction mechanism in which the offered price starts high and falls over time until a bidder accepts, with the first acceptance determining the clearing price. In blockchain and token markets it is implemented in smart contracts to distribute tokens or NFTs, set initial offering prices, and liquidate collateral, with the falling-price schedule encoded on-chain. Variants include single-item and multi-unit uniform-price formats used for fairer price discovery.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:dutch-auction", "labels": [ "Dutch Auction" ], "is_subclass_of": [ "Auction Mechanism" ], "wikilinks": [] }, { "id": "dwarf-planet", "title": "Dwarf Planet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dwarf-planet", "labels": [ "Dwarf Planet" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "dynamic-batching", "title": "Dynamic Batching", "domain": "ai", "domain_name": "Ai", "definition": "Dynamic batching is a serving-system technique that aggregates multiple independently arriving inference requests into a single batch for joint GPU execution, without requiring a fixed batch size determined at service startup. Requests are collected over a short time window or until a target batch size is reached, then processed together in one forward pass, amortising the fixed overhead of GPU kernel launches and memory transfers across requests. This substantially increases GPU utilisation and throughput compared with processing each request independently, at the cost of a small, controllable increase in per-request latency.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:dynamic-batching", "labels": [ "Dynamic Batching" ], "is_subclass_of": [ "Artificial Intelligence", "Inference Optimisation", "Model Serving", "Batch Processing", "Request Scheduling" ], "wikilinks": [] }, { "id": "dynamic-character-animation", "title": "Dynamic Character Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time procedural animation techniques that enable 3D avatars and characters to move, express, and respond dynamically to user input and environmental stimuli in metaverse environments, utilising motion capture, rigging systems, and AI-driven motion synthesis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:dynamic-character-animation", "labels": [ "Dynamic Character Animation" ], "is_subclass_of": [ "Content and Assets", "3D Animation" ], "wikilinks": [ "3D Animation", "Computer Vision", "Immersive Experiences", "metaverse" ] }, { "id": "dynamic-coordinate-reference-system", "title": "Dynamic Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dynamic-coordinate-reference-system", "labels": [ "Dynamic Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "dynamic-lighting", "title": "Dynamic Lighting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Dynamic lighting is a real-time rendering technique in which light sources, shadows, and indirect illumination are computed per-frame based on the current state of a scene, allowing lights to move, change intensity or colour, and interact with animated geometry without relying on pre-baked static lighting data. It is foundational to believable 3D environments in games, virtual production, and spatial computing applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dynamic-lighting", "labels": [ "Dynamic Lighting", "Lighting System" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "dynamic-model", "title": "Dynamic Model", "domain": "ai", "domain_name": "Ai", "definition": "A Dynamic Model is a mathematical or computational representation of a system that explicitly captures how the system's state evolves over time in response to inputs, internal dynamics, and disturbances. Distinguished from static models by their time-varying state equations\u2014typically differential equations for continuous systems or recurrence relations for discrete systems\u2014dynamic models are fundamental to control engineering, physics simulation, robotics, and economic forecasting. They may be physics-derived from first principles, identified from data using system identification techniques, or learned end-to-end from observations using neural networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dynamic-model", "labels": [ "Dynamic Model" ], "is_subclass_of": [ "Simulation", "Mathematical Model", "Predictive Model", "Probabilistic Model" ], "wikilinks": [] }, { "id": "dynamic-pricing", "title": "Dynamic Pricing", "domain": "ai", "domain_name": "Ai", "definition": "Dynamic Pricing is a strategy in which the price of a good or service is adjusted in real time based on demand, supply, competitor prices, inventory, and customer signals. It is typically driven by machine-learning models that forecast willingness to pay and optimise revenue or other objectives. Common in e-commerce, ride-hailing, travel, and logistics, it relies on continuous data feeds and automated decisioning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dynamic-pricing", "labels": [ "Dynamic Pricing", "Dynamic Pricing Engine" ], "is_subclass_of": [ "AI Application", "Optimisation", "Decision Support System", "Revenue Management" ], "wikilinks": [] }, { "id": "dynamic-programming", "title": "Dynamic Programming", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Dynamic programming is an algorithmic technique for solving problems by breaking them into overlapping subproblems whose solutions are stored and reused rather than recomputed. It applies to problems exhibiting optimal substructure, combining subproblem solutions to construct an optimal whole. By memoising or tabulating intermediate results it converts exponential brute-force searches into polynomial-time algorithms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:dynamic-programming", "labels": [ "Dynamic Programming" ], "is_subclass_of": [ "Programming Paradigm", "Optimisation Algorithm" ], "wikilinks": [] }, { "id": "dynamic-range", "title": "Dynamic Range", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:dynamic-range", "labels": [ "Dynamic Range" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "dynamic-scalable-bft", "title": "Dynamic Scalable BFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Dynamic Scalable BFT (DSBFT) is an optimised Byzantine fault-tolerant consensus protocol that combines Distributed Key Generation (DKG) with BLS aggregate signatures to allow a validator committee to reach consensus with O(n) rather than O(n\u00b2) message complexity, while supporting dynamic membership\u2014nodes may join or leave the committee without requiring a full protocol restart or trusted dealer for key setup.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:dynamic-scalable-bft", "labels": [ "Dynamic Scalable BFT" ], "is_subclass_of": [ "Protocol and Consensus", "Practical Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Practical Byzantine Fault Tolerance" ] }, { "id": "dynamical-systems-theory", "title": "Dynamical Systems Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The mathematical study of systems that evolve over time according to fixed rules, focusing on long-term behaviour, stability, attractors, bifurcations, and qualitative structure of trajectories in state space.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dynamical-systems-theory", "labels": [ "Dynamical Systems Theory" ], "is_subclass_of": [ "AI Research Area", "owl:Thing", "Applied Mathematics", "Mathematical Physics" ], "wikilinks": [ "Differential Equations", "Linear Algebra", "Complex Systems", "Feedback Loop", "owl:Thing" ] }, { "id": "dynamical-systems", "title": "Dynamical Systems", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Dynamical systems is the mathematical study of how the state of a system evolves over time according to fixed rules. It applies to physics, biology, engineering, and control.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:dynamical-systems", "labels": [ "Dynamical Systems" ], "is_subclass_of": [ "AI Research Area", "owl:Thing", "Applied Mathematics Domain", "Mathematical Modelling" ], "wikilinks": [ "Control Theory", "Physics Simulation Engine", "owl:Thing", "https://en.wikipedia.org/wiki/Dynamical_system", "https://mathworld.wolfram.com/DynamicalSystem.html" ] }, { "id": "dynamics", "title": "Dynamics", "domain": "robotics", "domain_name": "Robotics", "definition": "Dynamics is the branch of mechanics that studies the forces and torques that cause motion and the resulting accelerations of bodies. In robotics it provides the equations of motion that relate joint forces to accelerations, essential for force control, simulation, and model-based control. It contrasts with kinematics, which describes motion without reference to the forces that produce it.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:dynamics", "labels": [ "Dynamics", "Forward Dynamics" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "e-commerce", "title": "E-Commerce", "domain": "finance", "domain_name": "Finance", "definition": "E-Commerce (electronic commerce) is the commercial exchange of goods, services, and information conducted over digital networks, encompassing business-to-consumer (B2C), business-to-business (B2B), consumer-to-consumer (C2C), and direct-to-consumer (D2C) transaction models. It relies on integrated layers of digital payment infrastructure, identity and authentication, logistics coordination, and data-driven personalisation to operate at scale. Modern e-commerce platforms leverage cloud infrastructure, recommendation engines, and APIs to connect buyers, sellers, and fulfilment networks globally. It is a mature commercial domain that continues to evolve through mobile-first experiences, conversational commerce, and blockchain-based payment rails.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:e-commerce", "labels": [ "E-Commerce", "E-Commerce Fulfilment", "Electronic Commerce" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "e-contract-arbitration", "title": "E-Contract Arbitration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Online dispute resolution process specifically designed for resolving conflicts arising from smart contract execution, code interpretation, or automated transaction failures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:e-contract-arbitration", "labels": [ "E-Contract Arbitration" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Automated Dispute Resolution", "Blockchain Transaction Log", "Code Interpretation Service", "Contract Analysis Process", "Contract Enforcement", "Fair Adjudication", "On-Chain Evidence Verification", "Smart Contract Code", "Smart Contract Governance", "Smart Contract Standards", "Transaction Reversal", "UNCITRAL ODR Model", "Arbitration Decision Engine", "Arbitrator Expertise", "Blockchain", "Dispute Resolution Mechanism", "Identity Verification", "Legal Framework", "Middleware Layer", "TrustAndGovernanceDomain" ] }, { "id": "e-waste", "title": "E-Waste", "domain": "governance", "domain_name": "Governance", "definition": "E-waste is discarded electrical and electronic equipment, including devices that have reached the end of their useful life. Managing it involves recycling, recovery, and safe disposal.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:e-waste", "labels": [ "E-Waste", "Electronic Waste" ], "is_subclass_of": [ "Circular Economy" ], "wikilinks": [ "Sustainability", "Hardware", "Circular Economy", "https://www.unep.org/topics/chemicals-and-pollution-action/pollution-and-health/e-waste", "https://globalewaste.org" ] }, { "id": "eba-travel-rule-guidelines", "title": "EBA Travel Rule Guidelines", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The EBA Travel Rule Guidelines are guidance issued by the European Banking Authority on how payment service providers and crypto-asset service providers must comply with the EU Transfer of Funds Regulation. They specify the originator and beneficiary information that must accompany transfers of funds and crypto-assets, implementing the FATF Travel Rule within EU law. They define data fields, thresholds, and responsibilities for intermediaries to support anti-money-laundering controls.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eba-travel-rule-guidelines", "labels": [ "EBA Travel Rule Guidelines" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "eba", "title": "EBA", "domain": "finance", "domain_name": "Finance", "definition": "The European Banking Authority (EBA) is an independent EU regulatory agency established in 2011 to ensure consistent prudential regulation and supervision across the European banking sector. It develops binding technical standards and guidelines that national supervisors must implement, conducts EU-wide stress tests to assess banking system resilience, and plays a central role in developing anti-money laundering supervisory frameworks. The EBA's remit has expanded into digital finance, including technical standards for crypto-asset service providers under the MiCA Regulation and guidelines on ICT risk management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eba", "labels": [ "EBA" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "ebsi", "title": "EBSI", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The European Blockchain Services Infrastructure (EBSI) is a pan-European permissioned blockchain network operated by EU member states and the European Commission to deliver cross-border public services. It anchors verifiable credentials and decentralised identifiers, enabling trusted exchange of diplomas, official documents, and identity attestations between governments. EBSI is a leading example of a public-sector consortium blockchain supporting self-sovereign identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ebsi", "labels": [ "EBSI" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "ecb-digital-euro-regulation-proposal", "title": "ECB Digital Euro Regulation Proposal", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ECB Digital Euro Regulation Proposal is the European Commission's draft legislative package, published in 2023, that would establish the legal basis for a digital euro. It sets out proposed rules for legal-tender status, privacy protections, distribution by supervised intermediaries, and offline functionality before adoption by EU co-legislators. As a proposal it frames the policy options and obligations that any future digital-euro framework would codify.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ecb-digital-euro-regulation-proposal", "labels": [ "ECB Digital Euro Regulation Proposal" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "ecb-digital-euro-regulation", "title": "ECB Digital Euro Regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ECB Digital Euro Regulation is the legal and regulatory framework governing a potential central bank digital currency issued by the European Central Bank for the euro area. It defines the digital euro's legal-tender status, privacy safeguards, holding limits, and the roles of the Eurosystem and intermediaries in distribution. It establishes the rules under which a retail CBDC would operate alongside cash and commercial bank money.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ecb-digital-euro-regulation", "labels": [ "ECB Digital Euro Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "ecb", "title": "ECB", "domain": "governance", "domain_name": "Governance", "definition": "The European Central Bank, the central bank responsible for monetary policy across the member states that use the euro and for supervisory functions within the banking union.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ecb", "labels": [ "ECB", "ECB Banking Supervision" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [ "Central Bank", "Monetary Policy", "Financial Stability", "Financial System" ] }, { "id": "eccv", "title": "ECCV", "domain": "ai", "domain_name": "Ai", "definition": "ECCV, the European Conference on Computer Vision, is a biennial premier venue for research in computer vision and pattern recognition, held in alternating years with ICCV to provide the community with annual high-quality publication opportunities. Founded in 1990, it attracts thousands of submissions covering topics from low-level image processing and 3D reconstruction to high-level scene understanding and video analysis. ECCV proceedings published through Springer LNCS constitute one of the most-cited bodies of literature in artificial intelligence. Acceptance at ECCV carries high prestige, signalling rigorous peer review and methodological significance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:eccv", "labels": [ "ECCV" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "ecdsa-cryptography", "title": "ECDSA Cryptography", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ECDSA (Elliptic Curve Digital Signature Algorithm) is a public-key signature scheme that uses the algebra of elliptic curves over finite fields to produce compact, efficient digital signatures. It provides strong security with much smaller keys than RSA, making it the dominant signing algorithm in blockchains such as Bitcoin and Ethereum. ECDSA enables verifiable authorship and integrity of messages and transactions without revealing the private key.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ecdsa-cryptography", "labels": [ "ECDSA Cryptography" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "ecdsa", "title": "ECDSA", "domain": "security", "domain_name": "Security", "definition": "Elliptic Curve Digital Signature Algorithm (ECDSA) is a cryptographic primitive that uses elliptic curve mathematics to generate and verify digital signatures, underpinning transaction authentication in Bitcoin, Ethereum, and most public blockchain networks. ECDSA provides non-repudiation, integrity verification, and ownership proof with compact key sizes relative to RSA equivalents.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ecdsa", "labels": [ "ECDSA", "ECDSA Signature" ], "is_subclass_of": [ "Security and Identity", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "eda-software", "title": "EDA Software", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Electronic Design Automation software used to design and test integrated circuits and semiconductor devices, serving as a critical bottleneck in chip manufacturing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:eda-software", "labels": [ "EDA Software" ], "is_subclass_of": [ "Semiconductor" ], "wikilinks": [] }, { "id": "eip-process", "title": "EIP Process", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The EIP Process is the governance workflow by which Ethereum Improvement Proposals are submitted, reviewed, and standardised. It defines proposal types (Core, Networking, Interface, ERC, Meta), status stages from Draft through Final, and the role of editors and community review in reaching rough consensus. It is the mechanism through which protocol changes and token standards such as ERC-20 are coordinated across the Ethereum ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eip-process", "labels": [ "EIP Process" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "eip-1271", "title": "EIP-1271", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum standard defining a method for smart contracts to validate signatures on their behalf via a standard interface. It enables contract accounts to verify that a signature is valid.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:eip-1271", "labels": [ "EIP-1271" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "eip-1559", "title": "eip-1559", "domain": "blockchain", "domain_name": "Blockchain", "definition": "EIP-1559 is an Ethereum Improvement Proposal, activated in the London hard fork (August 2021), that replaced Ethereum's original first-price gas auction with a protocol-determined base fee which adjusts algorithmically each block based on whether the prior block consumed more or less than the target gas. The base fee is burned \u2014 permanently removed from circulating ether supply \u2014 rather than paid to block producers, while an optional priority fee (tip) compensates validators for timely inclusion. This two-tier mechanism improves fee predictability, bounds user overpayment, introduces a deflationary supply dynamic, and partially constrains miner/validator extractable value arising from fee competition.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:eip-1559", "labels": [ "EIP-1559" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "eip-2981", "title": "EIP-2981", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum standard defining a royalty interface that allows non-fungible and multi-token contracts to signal a royalty amount and recipient for a sale. It standardises retrieval of royalty information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:eip-2981", "labels": [ "EIP-2981", "EIP-2981 Royalty Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "eip-4337", "title": "EIP-4337", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum Improvement Proposal defining account abstraction through a higher-layer UserOperation mechanism, without requiring changes to the core protocol. It enables smart contract accounts with custom validation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:eip-4337", "labels": [ "EIP-4337" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "eip-4844", "title": "EIP-4844", "domain": "blockchain", "domain_name": "Blockchain", "definition": "EIP-4844, known as Proto-Danksharding or 'Shard Blob Transactions', is an Ethereum Improvement Proposal activated on the Ethereum mainnet in March 2024 that introduces a new transaction type carrying temporary binary large objects (blobs) of data. Each blob is approximately 128 KB, persists for a short window (roughly two weeks) rather than being stored permanently in Ethereum state, and is made accessible to smart contracts only via a cryptographic commitment. The proposal dramatically reduces the data posting costs for Layer-2 rollups by creating a dedicated, cheap data-availability lane, providing a foundation for full Danksharding without requiring immediate shard implementation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:eip-4844", "labels": [ "EIP-4844" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "eip-712", "title": "EIP-712", "domain": "blockchain", "domain_name": "Blockchain", "definition": "EIP-712 (Ethereum Improvement Proposal 712) is a standard for hashing and signing typed structured data in Ethereum, enabling wallets to display human-readable transaction information instead of opaque hexadecimal byte strings before a user signs a message. It defines a domain separator (chainId, contract address, version) that binds a signature to a specific deployment context to prevent cross-chain or cross-contract replay attacks, and a type-encoding scheme that recursively hashes structured fields to produce a deterministic 32-byte digest that the ECDSA private key signs. EIP-712 is the foundation of off-chain signing workflows \u2014 gasless approvals, meta-transactions, permit() functions (EIP-2612), delegated votes, and order book signatures in decentralised exchanges \u2014 making it one of the most widely deployed Ethereum standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eip-712", "labels": [ "EIP-712", "EIP-712 Signature", "EIP-712 Signatures", "EIP-712 Typed Signatures", "EIP-712 Typed Signing", "EIP-712 Typed Structured Data" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "eip-7702", "title": "EIP-7702", "domain": "standards", "domain_name": "Standards", "definition": "EIP-7702 is an Ethereum Improvement Proposal that lets externally owned accounts temporarily act with smart contract code to support account abstraction features.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:eip-7702", "labels": [ "EIP-7702" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "eip", "title": "EIP", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum Improvement Proposal (EIP) is the formal, versioned design document through which changes to the Ethereum protocol, its standards and its processes are proposed, debated and ratified. Categories include core protocol changes, networking and interface proposals, and application-level standards known as ERCs such as the token interfaces. Each proposal moves through defined statuses from draft to final, providing a transparent, community-driven mechanism for evolving the network.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:eip", "labels": [ "EIP", "Ethereum Improvement Proposal" ], "is_subclass_of": [ "Ethereum", "Governance Proposal" ], "wikilinks": [] }, { "id": "electra", "title": "ELECTRA", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ELECTRA (Efficiently Learning an Encoder that Classifies Token Replacements Accurately) is a transformer pre-training method that trains a discriminator to detect replaced tokens rather than reconstructing masked inputs, using a generator-discriminator architecture. This replaced token detection task utilises all positions in a sequence, yielding substantially greater sample efficiency than masked language modelling with less than one quarter of the compute required by comparable models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:electra", "labels": [ "ELECTRA" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "emc-standard", "title": "EMC Standard", "domain": "robotics", "domain_name": "Robotics", "definition": "EMC standard ensures robots neither emit electromagnetic interference that disrupts other equipment nor experience susceptibility to external electromagnetic noise that degrades performance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:emc-standard", "labels": [ "EMC Standard", "EMC Directive", "IEC 61000 EMC Standards for Control Valves" ], "is_subclass_of": [ "Safety and Standards", "Robot Standard", "Electromagnetic Compatibility" ], "wikilinks": [ "Compliance Criteria", "Control Systems", "Electrical Safety", "Electromagnetic Compatibility", "Electromagnetic Theory", "Emissions Limit", "Fibre Optic Isolation", "Filter Implementation", "Grounding Practice", "IEC 61000", "Immunity Requirement", "Industrial Deployment", "Measurement Equipment", "PCB Layout", "Reliable Operation", "Shielding Design", "Test Facility", "Test Procedure", "AI Agent System", "Communication Protocol" ] }, { "id": "en-301-549", "title": "EN 301 549", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "EN 301 549 is the European standard specifying accessibility requirements for information and communication technology products and services, including web, software, hardware, and documents. It harmonises with the Web Content Accessibility Guidelines and is referenced by EU legislation such as the Web Accessibility Directive and the European Accessibility Act. Conformance is a legal procurement requirement for public-sector ICT in the EU.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:en-301-549", "labels": [ "EN 301 549" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ens-dao", "title": "ENS DAO", "domain": "governance", "domain_name": "Governance", "definition": "ENS DAO is the decentralised autonomous organisation that governs the Ethereum Name Service (ENS), the primary naming infrastructure for the Ethereum ecosystem, using the ENS governance token ($ENS) to enable token-weighted voting on protocol upgrades, treasury allocation, and price oracle parameter changes through a Governor Bravo-compatible on-chain governance contract. Established in November 2021 via a retroactive airdrop of 25 million ENS tokens to historical registrants and contributors, the DAO holds a treasury of tens of millions of dollars in ETH and ENS, oversees the root multi-sig controlling the ENS root keys, and appoints stewards for working groups covering Meta-Governance, Ecosystem, Public Goods, and Community operations. ENS DAO represents one of the largest and most active protocol DAOs by voter participation and treasury size.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ens-dao", "labels": [ "ENS DAO" ], "is_subclass_of": [ "DAO" ], "wikilinks": [] }, { "id": "ens", "title": "ens", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Name Service (ENS) is a decentralised, on-chain naming system built on Ethereum that maps human-readable names ending in .eth to Ethereum addresses, content hashes, multi-coin addresses, and arbitrary text records via ERC-137 compliant smart contracts and pluggable resolver contracts. It functions as the Web3 analogue of DNS, replacing opaque hexadecimal addresses with memorable, self-sovereign labels backed by ERC-721 NFTs and governed by the ENS DAO. Off-chain resolution via CCIP-Read (EIP-3668) extends the system to conventional databases with on-chain verification, reducing gas overhead for high-frequency updates whilst preserving trustless resolution semantics.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:ens", "labels": [ "ENS", "ENS (Ethereum Name Service)", "ENS Name" ], "is_subclass_of": [ "Decentralised Identity" ], "wikilinks": [] }, { "id": "erc-1155", "title": "erc-1155", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-1155 is an Ethereum multi-token standard (formalised as EIP-1155) that enables a single smart contract to manage an arbitrary number of token types \u2014 fungible, non-fungible, and semi-fungible \u2014 each identified by a uint256 token ID with per-address per-ID balance tracking. Pioneered by Witek Radomski of Enjin in 2018, it introduced safeBatchTransferFrom for atomic multi-token transfers in a single transaction, significantly reducing gas costs relative to deploying separate ERC-20 or ERC-721 contracts for each asset type. The standard's onERC1155Received and onERC1155BatchReceived hooks enforce safe-transfer guarantees on receiver contracts, preventing tokens from being permanently locked. It is widely adopted across blockchain gaming, NFT marketplaces, and on-chain economies requiring a mixture of consumable items and unique collectibles.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:erc-1155", "labels": [ "ERC-1155" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [] }, { "id": "erc-1400", "title": "ERC-1400", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-1400 is a composite Ethereum token standard that defines a framework for security tokens representing regulated financial instruments such as equities, bonds, and real estate on-chain. It extends the ERC-20 fungible token model with partition-based token tranches, on-chain transfer restrictions enforced via controller logic, document attachment for prospectuses and legal agreements, and forced transfer capabilities required for regulatory compliance. The standard is designed to satisfy securities law requirements across multiple jurisdictions by embedding compliance logic directly into the smart contract layer.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:erc-1400", "labels": [ "ERC-1400", "ERC-1400 Security Token" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-20-token-standard", "title": "ERC-20 Token Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ERC-20 Token Standard is the foundational Ethereum Request for Comments specification that defines a common interface for fungible tokens on the Ethereum blockchain, enabling seamless interoperability between token contracts, wallets, decentralised exchanges, and other smart contract systems. Proposed by Fabian Vogelsteller in 2015 and formalised as an Ethereum Improvement Proposal, it specifies six mandatory functions\u2014totalSupply, balanceOf, transfer, transferFrom, approve, and allowance\u2014and two events. ERC-20 standardisation catalysed the 2017 ICO boom and remains the dominant token interface in decentralised finance, with thousands of tokens deployed to this specification. Its simplicity has made it the basis for numerous extended standards including ERC-777, ERC-1400, and ERC-3643.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:erc-20-token-standard", "labels": [ "ERC-20 Token Standard", "ERC-20", "ERC-20 Standard" ], "is_subclass_of": [ "Enterprise Token Standards" ], "wikilinks": [] }, { "id": "erc-20-tokens", "title": "ERC-20 Tokens", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum standard defining a common interface for fungible tokens, including functions for transfers, balances and allowances. It is the widely adopted standard for interchangeable tokens on Ethereum.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-20-tokens", "labels": [ "ERC-20 Tokens", "ERC-20 Token" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-20-votes", "title": "ERC-20 Votes", "domain": "standards", "domain_name": "Standards", "definition": "An extension pattern building on the ERC-20 fungible token standard to track voting power and historical balances for on-chain governance. It associates token holdings with delegated voting rights.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-20-votes", "labels": [ "ERC-20 Votes" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-20", "title": "erc-20", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-20 is the Ethereum Request for Comments token standard that defines a mandatory six-function interface \u2014 totalSupply, balanceOf, transfer, transferFrom, approve, and allowance \u2014 plus optional name, symbol, and decimals metadata, enabling any conformant fungible token to interoperate seamlessly with wallets, decentralised exchanges, and smart-contract protocols without bespoke integration. Proposed by Fabian Vogelsteller and Vitalik Buterin in November 2015 and formalised as Ethereum Improvement Proposal 20 (EIP-20), it became the foundational primitive of the DeFi ecosystem and the predominant format for utility tokens, governance tokens, stablecoins, and wrapped assets. The standard defines a delegated-transfer pattern via approve and transferFrom that allows smart contracts to spend tokens on a holder's behalf, and it underpins the liquidity infrastructure of all major decentralised exchanges, lending protocols, and yield-aggregators on EVM-compatible networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:erc-20", "labels": [ "ERC-20" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [] }, { "id": "erc-2612-permit", "title": "ERC-2612 Permit", "domain": "standards", "domain_name": "Standards", "definition": "ERC-2612 adds a permit function to ERC-20 tokens, allowing approvals to be granted through signed messages rather than separate on-chain transactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-2612-permit", "labels": [ "ERC-2612 Permit", "ERC-2612" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-3475", "title": "ERC-3475", "domain": "standards", "domain_name": "Standards", "definition": "ERC-3475 is an Ethereum standard defining an interface for abstract bonds, supporting multiple redemption conditions within a single token contract.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-3475", "labels": [ "ERC-3475", "ERC-3475 Abstract Bond" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-3525", "title": "ERC-3525", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum standard defining a semi-fungible token interface in which each token has an identifier and a value within a shared slot. It combines features of fungible and non-fungible tokens.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-3525", "labels": [ "ERC-3525" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-3643", "title": "ERC-3643", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-3643 is an Ethereum token standard, also known as T-REX (Token for Regulated EXchanges), that enables the issuance and management of permissioned security tokens on public blockchain infrastructure. It embeds on-chain identity verification and compliance rule enforcement into the token transfer logic, ensuring that only verified, eligible investors can hold and transfer regulated digital assets. The standard extends ERC-20 with an identity registry, modular compliance modules, and claim-based permissioning, making it suitable for tokenised securities, real-world assets, and other regulated financial instruments. ERC-3643 was standardised through the Ethereum Improvement Proposal process and has been adopted by multiple institutional asset tokenisation platforms.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:erc-3643", "labels": [ "ERC-3643", "ERC-3643 (T-REX)", "ERC-3643 T-REX" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-4337-account-abstraction", "title": "ERC-4337 Account Abstraction", "domain": "standards", "domain_name": "Standards", "definition": "An Ethereum standard defining account abstraction using a higher-layer pseudo-transaction object called a UserOperation, without changes to the core protocol. It enables smart contract wallets with custom validation logic.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-4337-account-abstraction", "labels": [ "ERC-4337 Account Abstraction" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-4337", "title": "ERC-4337", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-4337 is an Ethereum Request for Comment specifying an account abstraction standard that enables smart contract accounts to function as first-class transaction originators without requiring changes to the Ethereum consensus protocol. It introduces a separate transaction processing pipeline \u2014 the user operation mempool \u2014 in which bundlers aggregate user operations (UserOps), simulate them for validity, and submit them as bundled transactions to the EntryPoint smart contract, which validates and executes them on behalf of the respective smart contract wallets. Paymasters, a complementary component, allow third parties to sponsor gas fees, enabling gasless transactions and alternative fee payment tokens.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:erc-4337", "labels": [ "ERC-4337" ], "is_subclass_of": [ "Account Abstraction" ], "wikilinks": [] }, { "id": "erc-4626", "title": "ERC-4626", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-4626 is a finalised Ethereum Improvement Proposal that specifies a standardised application-programming interface for tokenised yield-bearing vaults, extending the ERC-20 fungible token standard with deposit, withdrawal, and share-accounting functions. It allows any vault \u2014 whether backed by lending protocols, automated market makers, or staking contracts \u2014 to expose a uniform interface so that aggregators, routers, and analytics tools can interact with all vaults without bespoke adapters. The standard mandates methods such as deposit, mint, withdraw, redeem, and convertToAssets alongside preview variants that return expected amounts without state changes, enabling safe, read-only price discovery. Adoption of ERC-4626 reduces integration friction across decentralised finance, enabling composable yield strategies and dramatically lowering the cost of building multi-vault products.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:erc-4626", "labels": [ "ERC-4626" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-721-standard", "title": "ERC-721 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-721 is an Ethereum token standard defining a minimal interface for non-fungible tokens (NFTs), where each token carries a unique identifier that distinguishes it from every other token in the same contract. It specifies ownership tracking, safe transfer functions, and an optional metadata extension for associating URIs with token properties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:erc-721-standard", "labels": [ "ERC-721 Standard", "ERC-721" ], "is_subclass_of": [ "NFT Standard" ], "wikilinks": [] }, { "id": "erc-721", "title": "erc-721", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC-721 is the Ethereum Request for Comments standard that defines the interface for non-fungible tokens (NFTs), where each token is uniquely identified by a uint256 token ID paired with an owner address stored immutably on-chain. Unlike ERC-20 fungible tokens, every ERC-721 token is distinct and non-interchangeable, making the standard suitable for representing provably unique digital or tokenised physical assets. The mandatory interface specifies ownership queries, safe and unsafe transfer methods, and operator approval delegation, while an optional metadata extension enables each token to reference off-chain JSON describing name, description, and media. Proposed by William Entriken, Dieter Shirley, Jacob Evans, and Nastassia Sachs and finalised as an Ethereum standard in January 2018, ERC-721 underpins digital art, collectibles, gaming items, domain names, and real-world asset tokenisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:erc-721", "labels": [ "ERC-721", "ERC-721 NFTs" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "erc-7540", "title": "ERC-7540", "domain": "standards", "domain_name": "Standards", "definition": "ERC-7540 extends the ERC-4626 tokenised vault interface to support asynchronous deposit and redemption requests.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-7540", "labels": [ "ERC-7540" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc-7683", "title": "ERC-7683", "domain": "standards", "domain_name": "Standards", "definition": "ERC-7683 is an Ethereum standard proposing a common interface for cross-chain intents and order execution between networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:erc-7683", "labels": [ "ERC-7683" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Ethereum", "Technical Standard" ] }, { "id": "erc1155-standard", "title": "ERC1155 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multi-token standard supporting both fungible and non-fungible tokens in a single contract.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:erc1155-standard", "labels": [ "ERC1155 Standard", "ERC-1155 Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "erc1155-token", "title": "ERC1155 Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC1155 is an Ethereum multi-token standard that enables a single smart contract to manage an arbitrary number of fungible, non-fungible, and semi-fungible token classes simultaneously. By batching multiple token operations into a single transaction, ERC1155 dramatically reduces gas consumption compared to deploying separate ERC-20 or ERC-721 contracts for each asset class, making it the dominant standard for blockchain gaming, NFT collections with tiered rarity, and on-chain item inventories.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:erc1155-token", "labels": [ "ERC1155 Token" ], "is_subclass_of": [ "Token and Asset", "Semi-Fungible Token" ], "wikilinks": [ "Blockchain", "Semi-Fungible Token" ] }, { "id": "erc1400-standard", "title": "ERC1400 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum security token standard (ERC-1400) combining fungible token behaviour with regulatory compliance controls, including partitioned balances, transfer restrictions, forced transfers for legal recovery, and on-chain document attachment, enabling the issuance and lifecycle management of regulated securities on public blockchains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:erc1400-standard", "labels": [ "ERC1400 Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "erc1400-token", "title": "ERC1400 Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An ERC1400 Token is an Ethereum-based security token standard that extends the ERC20 interface with partitioned token tranches, forced transfer capabilities, controller-initiated operations, and on-chain document management, enabling issuers to represent regulated financial securities on a public or permissioned blockchain. The standard enforces transfer restrictions through a canTransfer validation hook that integrates with off-chain compliance logic \u2014 such as AML/KYC whitelists and jurisdictional eligibility rules \u2014 returning standardised EIP-1066 status codes rather than simple booleans. ERC1400 is positioned as an umbrella framework composing interoperability sub-standards (ERC1410 for partitions, ERC1594 for transfers, ERC1643 for documents, ERC1644 for controller operations), making it suitable for tokenising equities, bonds, real estate, and fund units subject to securities law.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:erc1400-token", "labels": [ "ERC1400 Token" ], "is_subclass_of": [ "Token and Asset", "Security Token" ], "wikilinks": [ "Blockchain", "Security Token" ] }, { "id": "erc20-standard", "title": "ERC20 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Technical standard for fungible tokens on ereum, defining required mods and events for token contracts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:erc20-standard", "labels": [ "ERC20 Standard", "ERC-20 Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "erc20-token", "title": "ERC20 Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An ERC20 Token is a fungible token implementing the ERC-20 interface standard on the Ethereum blockchain, which defines a uniform set of six mandatory functions (totalSupply, balanceOf, transfer, transferFrom, approve, allowance) enabling interoperability between token contracts, decentralised exchanges, wallets, and DeFi protocols without bespoke integration. ERC-20 became the dominant token standard following its formalisation in 2015 and is used for utility tokens, governance tokens, stablecoins, and wrapped assets. The standard's fungibility means all token units are identical and mutually interchangeable, contrasting with ERC-721 non-fungible tokens that represent distinct assets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:erc20-token", "labels": [ "ERC20 Token", "ERC-20 Token" ], "is_subclass_of": [ "Token and Asset", "Fungible Token" ], "wikilinks": [ "Blockchain", "Fungible Token" ] }, { "id": "erc20", "title": "ERC20", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC20 is the Ethereum fungible token standard (Ethereum Improvement Proposal 20) defining a mandatory common interface for token transfers, approvals, and balance queries. It specifies six required functions and two events enabling interoperable smart contract interaction, forming the technical foundation of decentralised finance, governance tokens, and digital asset ecosystems on EVM-compatible chains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:erc20", "labels": [ "ERC20" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Balance Tracking", "EVM", "Token Transfer", "BlockchainDomain", "Ethereum", "Smart Contract" ] }, { "id": "erc20-votes-standard", "title": "ERC20Votes Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC20Votes is an extension of the ERC-20 fungible-token standard that adds on-chain governance capabilities by tracking historical voting power through checkpointed balances and supporting delegation of votes. It records a snapshot of each account's balance at every block where it changes, so a governance contract can query voting weight at a specific past block, preventing double-voting and flash-loan manipulation. Implemented in widely audited libraries, it is the dominant pattern for token-based voting in decentralised autonomous organisations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:erc20-votes-standard", "labels": [ "ERC20Votes Standard", "ERC-20Votes Delegation Standard", "ERC20Votes Extension" ], "is_subclass_of": [ "ERC20 Standard" ], "wikilinks": [] }, { "id": "erc3643-standard", "title": "ERC3643 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissioned token standard with on-chain identity verification for regulatory compliance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:erc3643-standard", "labels": [ "ERC3643 Standard", "ERC-3643 Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "erc3643-token", "title": "ERC3643 Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ERC-3643 (T-REX) standard is an Ethereum permissioned token framework designed for compliant security token issuance, embedding on-chain identity verification and transfer-restriction logic that enforces investor eligibility, jurisdiction rules, and AML/KYC requirements at the smart contract level. Transfer authorisation is governed by an on-chain identity registry and a modular compliance module, enabling automated regulatory enforcement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:erc3643-token", "labels": [ "ERC3643 Token" ], "is_subclass_of": [ "Token and Asset", "Security Token" ], "wikilinks": [ "Blockchain", "Security Token" ] }, { "id": "erc721-standard", "title": "ERC721 Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Technical standard for non-fungible tokens on ereum, defining unique token identification and ownership tracking.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:erc721-standard", "labels": [ "ERC721 Standard" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "erc721-token", "title": "ERC721 Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An ERC721 Token is a non-fungible token (NFT) on the Ethereum blockchain that implements the ERC-721 open standard, which assigns each token a unique integer identifier and tracks ownership via the ownerOf(tokenId) function. Unlike fungible ERC-20 tokens, every ERC-721 token is distinct and non-interchangeable, enabling verifiable digital ownership of unique assets such as digital art, collectibles, in-game items, and real-world asset representations on decentralised marketplaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:erc721-token", "labels": [ "ERC721 Token" ], "is_subclass_of": [ "Token and Asset", "Non-Fungible Token (NFT)" ], "wikilinks": [ "Non Fungible Token", "Blockchain" ] }, { "id": "erc721", "title": "ERC721", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ERC721 is the Ethereum non-fungible token standard enabling unique, individually identified digital assets each with a distinct tokenId and metadata URI. It specifies eight mandatory functions and three events supporting ownership transfer, provenance tracking, and approval management, underpinning NFT markets, digital art, gaming items, and tokenised real-world assets on EVM-compatible blockchains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:erc721", "labels": [ "ERC721" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Intellectual Property", "Ownership Transfer", "BlockchainDomain", "Digital Asset Management", "Ethereum", "Provenance Tracking", "Smart Contract" ] }, { "id": "esg-investing", "title": "esg investing", "domain": "finance", "domain_name": "Finance", "definition": "ESG Investing is an investment methodology that systematically integrates Environmental, Social, and Governance criteria into the analysis, selection, and management of portfolios, enabling investors to assess material sustainability risks and ethical impact alongside conventional financial returns. Environmental factors encompass carbon emissions, resource consumption, and climate transition exposure; social factors cover labour practices, supply-chain conditions, human rights, and community impact; governance factors evaluate board composition, executive remuneration, audit independence, and shareholder rights. ESG signals are incorporated through negative screening, positive best-in-class selection, engagement and proxy voting, or the construction of fully ESG-integrated thematic funds, each reflecting different theories of how non-financial data predicts long-run financial performance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:esg-investing", "labels": [ "ESG Investing", "ESG Investment" ], "is_subclass_of": [ "Sustainable Finance" ], "wikilinks": [] }, { "id": "esg-reporting", "title": "ESG Reporting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ESG Reporting is the composite corporate disclosure discipline by which obligated and voluntary entities \u2014 listed companies, large private undertakings, banks, insurers, asset managers, investment funds and a rapidly expanding perimeter of public-interest organisations \u2014 measure, assure and publi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:esg-reporting", "labels": [ "ESG Reporting", "ESG Disclosure", "ESG Reporting System" ], "is_subclass_of": [ "Governance and Regulation", "Corporate Disclosure", "Regulatory Reporting", "Non-Financial Reporting", "Sustainability Disclosure", "Stakeholder Reporting" ], "wikilinks": [ "AI Sustainability Copilot", "Article 8 Fund Classification", "Article 9 Fund Classification", "AssuranceLayer", "Battiston Mandel Monasterolo Schutze Visentin 2017 Climate Stress Test Nature", "Bingler Kraus Leippold Webersinke 2022 ClimateBert Finance Research Letters", "Board-Level Oversight", "Bolton Despres Pereira da Silva Samama Svartzman 2020 Green Swan BIS", "Capital Allocation Efficiency", "Carbon Accounting Software", "Carbon Border Adjustment Compliance", "Carbon Border Adjustment Mechanism", "Carney 2015 Tragedy of the Horizon Lloyd's", "CDP", "CDP Disclosure Data 2024", "CDP Questionnaires", "Climate Risk Pricing", "Climate Scenario Analysis", "Climate Scenario Database", "Climate Scenario Modelling" ] }, { "id": "esg-compliant-blockchain", "title": "ESG-Compliant Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An ESG-compliant blockchain is a distributed-ledger network designed and operated to meet environmental, social, and governance criteria, most notably by minimising energy use through low-power consensus and by offsetting or sourcing renewable energy. It typically pairs proof-of-stake or other efficient consensus with transparent emissions reporting and inclusive governance. Such networks aim to make blockchain adoption defensible under corporate sustainability and disclosure requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:esg-compliant-blockchain", "labels": [ "ESG-Compliant Blockchain" ], "is_subclass_of": [ "Blockchain", "Governance and Regulation" ], "wikilinks": [] }, { "id": "esg", "title": "ESG", "domain": "governance", "domain_name": "Governance", "definition": "Environmental, social and governance criteria used to evaluate the sustainability and ethical impact of an organisation's operations and investments.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:esg", "labels": [ "ESG", "ESG Data", "ESG Data Disclosure", "ESG Framework", "ESG Rating" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Sustainability", "Corporate Governance", "Transparency", "Accountability", "Audit" ] }, { "id": "esma", "title": "ESMA", "domain": "governance", "domain_name": "Governance", "definition": "The European Securities and Markets Authority (ESMA) is the European Union's independent financial regulator responsible for safeguarding the stability and integrity of EU securities markets. It develops technical standards, supervises certain market participants such as credit-rating agencies, and coordinates national regulators to ensure consistent application of EU financial law. ESMA plays a central role in regimes such as MiCA for crypto-assets and in securities and regulatory-reporting rules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:esma", "labels": [ "ESMA" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "etl-pipeline", "title": "ETL Pipeline", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Extract-Transform-Load pipeline that automates the movement of data from heterogeneous source systems, applies normalisation and enrichment transformations, and loads the results into target data stores such as data warehouses or feature stores. ETL pipelines are foundational to data engineering and AI/ML workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:etl-pipeline", "labels": [ "ETL Pipeline", "ETL", "ETL Process" ], "is_subclass_of": [ "AI Infrastructure", "Artificial Intelligence", "Data Pipeline", "Data Engineering" ], "wikilinks": [ "Data Engineering", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Blockchain", "Digital Twin" ] }, { "id": "etsi-arf-010", "title": "ETSI ARF 010", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ETSI ARF 010 is a technical deliverable produced by the ETSI Industry Specification Group on Augmented Reality Framework (ISG ARF), defining architectural principles, terminology, and interoperability requirements for augmented reality systems and services. It establishes a common reference architecture that allows AR components\u2014such as tracking engines, scene anchors, and rendering pipelines\u2014to interoperate across heterogeneous platforms and devices. The specification addresses how spatial anchors, coordinate systems, and world models are managed and shared, enabling consistent AR experiences in multi-vendor and multi-device deployments. ETSI ARF 010 serves as a foundational normative document within the ARF series, complemented by companion deliverables covering use cases, protocols, and conformance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:etsi-arf-010", "labels": [ "ETSI ARF 010" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-domain-ai-creative-media", "title": "ETSI Domain AI + Creative Media", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-domain marker for metaverse components that combine artificial intelligence capabilities with creative media applications such as generative content, procedural generation, and AI-assisted authoring.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ai-creative-media", "labels": [ "ETSI Domain AI + Creative Media", "ETSI Domain AI Creative Media" ], "is_subclass_of": [ "AI Research Area", "ETSI Domain AI", "CreativeMediaDomain", "ComputationAndIntelligenceDomain" ], "wikilinks": [ "ETSI Domain Taxonomy", "ETSI GS MEC", "Generative Content Classification", "AI Art Categorization", "ApplicationLayer", "ComputationAndIntelligenceDomain", "Computer Vision", "CreativeMediaDomain", "ETSI Domain AI" ] }, { "id": "etsi-domain-ai-data-mgmt", "title": "ETSI Domain AI + Data Mgmt", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-domain marker for metaverse components combining artificial intelligence with data management capabilities including ML pipelines, intelligent data processing, analytics, and AI-driven data governance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ai-data-mgmt", "labels": [ "ETSI Domain AI + Data Mgmt", "ETSI Domain AI Data Mgmt" ], "is_subclass_of": [ "ETSI Domain AI", "InfrastructureDomain", "ComputationAndIntelligenceDomain" ], "wikilinks": [ "ETSI Domain Taxonomy", "ETSI GS MEC", "Intelligent Analytics Categorization", "ML Pipeline Classification", "ApplicationLayer", "ComputationAndIntelligenceDomain", "Computer Vision", "ETSI Domain AI", "InfrastructureDomain" ] }, { "id": "etsi-domain-ai-governance", "title": "ETSI Domain AI + Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-domain marker for metaverse components combining artificial intelligence with governance frameworks including AI ics, explainability, bias detection, regulatory compliance, and responsible AI systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ai-governance", "labels": [ "ETSI Domain AI + Governance", "ETSI Domain AI Governance" ], "is_subclass_of": [ "AI Governance and Ethics", "ETSI Domain AI", "AI Governance", "TrustAndGovernanceDomain" ], "wikilinks": [ "AI Ethics Classification", "ETSI Domain Taxonomy", "ETSI GS MEC", "Explainability Categorization", "ApplicationLayer", "Blockchain", "ComputationAndIntelligenceDomain", "ETSI Domain AI", "TrustAndGovernanceDomain" ] }, { "id": "etsi-domain-ai-human-interface", "title": "ETSI Domain AI + Human Interface", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Cross-domain marker for metaverse components combining artificial intelligence with human interaction systems including conversational AI, gesture recognition, emotion detection, and intelligent user experience adaptation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ai-human-interface", "labels": [ "ETSI Domain AI + Human Interface", "ETSI Domain AI Human Interface" ], "is_subclass_of": [ "AI Research Area", "ETSI Domain AI", "Human Computer Interaction", "InteractionDomain" ], "wikilinks": [ "Conversational AI Classification", "ETSI Domain Taxonomy", "ETSI GS MEC", "Intelligent UX Categorization", "ApplicationLayer", "ComputationAndIntelligenceDomain", "ETSI Domain AI", "InteractionDomain", "Telecollaboration" ] }, { "id": "etsi-domain-ai", "title": "ETSI Domain AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Domain marker concept for categorising metaverse components related to artificial intelligence, machine learning, and computational intelligence capabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ai", "labels": [ "ETSI Domain AI" ], "is_subclass_of": [ "AI Research Area", "ComputationAndIntelligenceDomain", "Standardisation Domain Concept" ], "wikilinks": [ "AI Service Classification", "ETSI Domain Taxonomy", "ETSI GS MEC", "Intelligence Layer Categorization", "ApplicationLayer", "ComputationAndIntelligenceDomain", "Computer Vision", "ETSI Domain AI Creative Media", "ETSI Domain AI Data Mgmt", "ETSI Domain AI Governance", "ETSI Domain AI Human Interface" ] }, { "id": "etsi-domain-application-creative", "title": "ETSI Domain Application + Creative", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-domain marker for metaverse application components focused on creative industries including digital art, music production, animation, film, design tools, and creative collaboration platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-application-creative", "labels": [ "ETSI Domain Application + Creative" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Collaboration Tool Categorization", "Creative Application Classification", "ETSI Domain Taxonomy", "ETSI GS MEC", "ApplicationLayer", "Computer Vision", "CreativeMediaDomain", "InfrastructureDomain" ] }, { "id": "etsi-domain-application-education", "title": "ETSI Domain Application + Education", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-domain marker for metaverse application components focused on education and training including virtual classrooms, immersive learning environments, educational simulations, and collaborative learning platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-application-education", "labels": [ "ETSI Domain Application + Education" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Education Application Classification", "ETSI Domain Taxonomy", "ETSI GS MEC", "ApplicationLayer", "InfrastructureDomain", "Learning Platform Categorization", "Telecollaboration", "VirtualSocietyDomain" ] }, { "id": "etsi-domain-application-health", "title": "ETSI Domain Application + Health", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-domain marker for metaverse application components focused on healthcare and wellness including telemedicine platforms, medical training simulations, therapeutic VR applications, and health monitoring systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-application-health", "labels": [ "ETSI Domain Application + Health" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ETSI Domain Taxonomy", "ETSI GS MEC", "Medical Platform Categorization", "ApplicationLayer", "Healthcare Application Classification", "InfrastructureDomain", "Telecollaboration", "VirtualSocietyDomain" ] }, { "id": "etsi-domain-application-industry", "title": "ETSI Domain Application + Industry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-domain marker for metaverse application components focused on industrial applications including manufacturing simulations, industrial digital twins, predictive maintenance, remote operations, and industrial training systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-application-industry", "labels": [ "ETSI Domain Application + Industry" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ETSI Domain Taxonomy", "ETSI GS MEC", "Industrial Application Classification", "Manufacturing Platform Categorization", "ApplicationLayer", "Autonomous Robot", "InfrastructureDomain", "VirtualEconomyDomain" ] }, { "id": "etsi-domain-application-tourism", "title": "ETSI Domain Application + Tourism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cross-domain marker for metaverse application components focused on tourism and hospitality including virtual tours, destination previews, cultural heritage experiences, and travel planning platforms in immersive environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-application-tourism", "labels": [ "ETSI Domain Application + Tourism" ], "is_subclass_of": [ "Platform and Environment", "Etsi Metaverse Domain Taxonomy" ], "wikilinks": [ "Cultural Experience Categorization", "ETSI Domain Taxonomy", "ETSI GS MEC", "Tourism Application Classification", "ApplicationLayer", "InfrastructureDomain", "Telecollaboration", "VirtualSocietyDomain" ] }, { "id": "etsi-domain-data", "title": "ETSI Domain Data", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Domain categorisation for data management, storage, analytics, AI/ML systems, and intelligence capabilities processing information in metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data", "labels": [ "ETSI Domain Data" ], "is_subclass_of": [ "Platform and Environment", "Etsi Metaverse Domain Model" ], "wikilinks": [ "AI Systems", "ETSI GR MEC 032", "Intelligence", "ISO 23257", "ApplicationLayer", "ComputationAndIntelligenceDomain", "Computer Vision", "Data Analytics", "Data Processing", "Data Storage", "ETSI Metaverse Domain Model", "Machine Learning", "Predictive Analytics" ] }, { "id": "etsi-domain-identity-and-trust", "title": "ETSI Domain Identity and Trust", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An ETSI governance domain establishing identity management, authentication, and trust infrastructure for virtual environments. It integrates decentralised identity models, verifiable credentials, biometric authentication, zero-trust architecture, and multi-party trust platforms aligned with eIDAS 2.0 and ISO/IEC 24760, enabling secure, privacy-respecting identity assurance across distributed metaverse platforms and consortia.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-identity-and-trust", "labels": [ "ETSI Domain Identity and Trust" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "BiometricAuthentication", "DecentralisedIdentity", "eIDAS", "ETSI", "ETSI GR MEC 032", "ISO/IEC 24760", "W3C DID Core", "ZeroTrustArchitecture", "DID Nostr Identity", "DigitalIdentity", "MetaverseDomain", "VerifiableCredentials", "VirtualEnvironment" ] }, { "id": "etsi-domain-immersive-reality-capture-crossover", "title": "ETSI Domain Immersive + Reality Capture Crossover", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Domain categorization marker indicating metaverse systems operating at the intersection of immersive interaction capabilities and reality capture technologies for photorealistic virtual environment creation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-immersive-reality-capture-crossover", "labels": [ "ETSI Domain Immersive + Reality Capture Crossover" ], "is_subclass_of": [ "Platform and Environment", "Etsi Metaverse Domain Model" ], "wikilinks": [ "ETSI GR MEC 032", "ISO 23257", "Photorealistic Immersion", "ApplicationLayer", "CreativeMediaDomain", "ETSI Domain Immersive", "ETSI Domain Reality Capture", "ETSI Metaverse Domain Model", "InteractionDomain", "Spatial Interaction", "Volumetric Capture" ] }, { "id": "etsi-domain-security-and-privacy", "title": "ETSI Domain Security and Privacy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The ETSI Domain Security and Privacy is a governance framework protecting metaverse ecosystems through comprehensive controls spanning cryptography, access management, data governance, and compliance measures. It addresses threat surface mapping, post-quantum cryptographic resilience, privacy-enhancing computation, digital evidence chain of custody, and psychological profiling safeguards for users of immersive environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-security-and-privacy", "labels": [ "ETSI Domain Security and Privacy", "Security & Privacy" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "GDPR", "Privacy-Enhancing Computation", "Cross-Border Data Transfer Rule", "Digital Evidence Chain of Custody", "MetaverseDomain", "Metaverse Psychology Profile", "Post-Quantum Cryptography", "Privacy-Enhancing Computation (PEC)", "Privacy Impact Assessment", "Privacy Impact Assessment (PIA)", "Security Layer", "Threat Surface Map", "Token Custody Service", "Zero-Trust Architecture (ZTA)" ] }, { "id": "etsi-domain-taxonomy", "title": "etsi domain taxonomy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The ETSI Domain Taxonomy is a hierarchical classification framework produced by the European Telecommunications Standards Institute for systematically organising AI-related technical specifications, standards, and work items across application domains such as transportation, healthcare, telecommunications, and public safety. It provides a controlled vocabulary and structured ontology enabling consistent referencing of AI standards across ETSI technical committees and supports cross-domain applicability analysis. The taxonomy underpins ETSI's AI standardisation roadmap and aligns with CEN-CENELEC and ISO/IEC JTC 1/SC 42 standardisation activities. As a living classification instrument, it maps broad application sectors down to specific sub-domains, allowing AI systems and standards to receive multiple domain affiliations and enabling gap analysis across the European standardisation ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-taxonomy", "labels": [ "ETSI Domain Taxonomy" ], "is_subclass_of": [ "AI Governance and Ethics", "Knowledge Organisation System", "Standardisation Framework" ], "wikilinks": [] }, { "id": "etsi-domain-creative-media", "title": "ETSI Domain: Creative Media", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Domain marker for ETSI metaverse categorisation covering creative content production, 3D modelling, rendering, and multimedia authoring for virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-creative-media", "labels": [ "ETSI Domain: Creative Media" ], "is_subclass_of": [ "Content and Assets", "Etsi Metaverse Domain Taxonomy" ], "wikilinks": [ "3D Content Creation", "Content Pipeline", "Creative Tools", "ETSI GR MEC 032", "Graphics Processing", "Multimedia Authoring", "Scene Design", "ApplicationLayer", "Asset Format Standards", "Asset Management", "Avatar Customization", "Computer Vision", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "Rendering Pipeline", "Virtual World Building" ] }, { "id": "etsi-domain-data-management-ai", "title": "ETSI Domain: Data Management + AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Crossover domain for ETSI metaverse categorisation addressing data infrastructure supporting AI/ML workflows, training data management, model versioning, and inference serving.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management-ai", "labels": [ "ETSI Domain: Data Management + AI" ], "is_subclass_of": [ "ETSI Domain AI + Data Mgmt", "ETSI Domain: Data Management", "ETSI Metaverse Domain Taxonomy" ], "wikilinks": [ "AI & Machine Learning", "Data Pipelines", "ETSI GR MEC 032", "Experiment Tracking", "ML Operations", "MLOps Infrastructure", "Model Deployment", "Model Registry", "ApplicationLayer", "Computer Vision", "Data Management", "Data Versioning", "ETSI Metaverse Domain Taxonomy", "Feature Store", "InfrastructureDomain", "Training Data Repository" ] }, { "id": "etsi-domain-data-management-creative-media", "title": "ETSI Domain: Data Management + Creative Media", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Crossover domain for ETSI metaverse categorisation addressing data infrastructure supporting creative content workflows, asset management, and version control systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management-creative-media", "labels": [ "ETSI Domain: Data Management + Creative Media" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Asset Database", "Asset Version Control", "Collaborative Authoring", "Content Distribution", "Content Pipeline", "Creative Media", "Distributed Storage", "ETSI GR MEC 032", "Version Control", "ApplicationLayer", "Computer Vision", "Data Management", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "Media Library", "Metadata Management" ] }, { "id": "etsi-domain-data-management-cultural-heritage", "title": "ETSI Domain: Data Management + Cultural Heritage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Crossover domain for ETSI metaverse categorisation addressing data preservation and management systems for cultural heritage digitisation, archival, and accessibility.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management-cultural-heritage", "labels": [ "ETSI Domain: Data Management + Cultural Heritage", "Cultural Heritage Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Cultural Heritage Digitization", "Educational Outreach", "ETSI GR MEC 032", "Heritage Database", "Long-term Preservation", "Metadata Schemas", "Preservation System", "Public Access", "Access Control", "ApplicationLayer", "Archival Standards", "Data Management", "Digital Archive", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "Telecollaboration" ] }, { "id": "etsi-domain-data-management-ethics", "title": "ETSI Domain: Data Management + Ethics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Crossover domain for ETSI metaverse categorisation addressing ical data handling, privacy-preserving storage, consent management, and responsible data governance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management-ethics", "labels": [ "ETSI Domain: Data Management + Ethics" ], "is_subclass_of": [ "AI Governance and Ethics", "ETSI Domain: Data Management", "ETSI Metaverse Domain Taxonomy" ], "wikilinks": [ "Anonymization", "Audit Logging", "Ethics & Law", "ETSI GR MEC 032", "GDPR", "Privacy Controls", "Privacy Regulations", "User Control", "ApplicationLayer", "Blockchain", "Compliance Verification", "Consent Management", "Data Management", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "Privacy-Preserving Analytics" ] }, { "id": "etsi-domain-data-management-security", "title": "ETSI Domain: Data Management + Security", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Crossover domain for ETSI metaverse categorisation addressing secure data storage, encrypted databases, access control systems, and data protection mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management-security", "labels": [ "ETSI Domain: Data Management + Security" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Access Control Enforcement", "Authentication Systems", "Data-at-Rest Protection", "Encrypted Storage", "Encryption Algorithms", "ETSI GR MEC 032", "Key Management", "Security & Privacy", "Security Audit", "Threat Detection", "Access Control", "ApplicationLayer", "Blockchain", "Data Management", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain" ] }, { "id": "etsi-domain-data-management", "title": "ETSI Domain: Data Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Domain marker for ETSI metaverse categorisation covering data storage, processing, synchronisation, and lifecycle management for distributed virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-data-management", "labels": [ "ETSI Domain: Data Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Caching Infrastructure", "Consistency Protocols", "Cross-Platform Synchronization", "Data Lifecycle", "Data Synchronization", "Database Systems", "ETSI GR MEC 032", "Replication Mechanisms", "State Persistence", "ApplicationLayer", "Autonomous Robot", "Data Analytics", "Data Processing", "Data Storage", "Distributed Systems", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain" ] }, { "id": "etsi-domain-ethics-and-law", "title": "ETSI Domain: Ethics & Law", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Domain marker for ETSI metaverse categorisation covering ical frameworks, legal compliance, regulatory requirements, and responsible governance structures for virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-ethics-and-law", "labels": [ "ETSI Domain: Ethics & Law" ], "is_subclass_of": [ "AI Governance and Ethics", "ETSI Metaverse Domain Taxonomy" ], "wikilinks": [ "Ethical Frameworks", "ETSI GR MEC 032", "GDPR", "Legal Accountability", "Legal Compliance", "Regulatory Systems", "Rights Management", "User Protection", "ApplicationLayer", "Blockchain", "Compliance Monitoring", "Content Moderation Standards", "Digital Services Act", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "Policy Enforcement", "Responsible AI" ] }, { "id": "etsi-domain-governance-and-compliance", "title": "ETSI Domain: Governance & Compliance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Crossover domain for ETSI metaverse categorisation addressing organisational governance structures, compliance verification systems, and regulatory adherence mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-governance-and-compliance", "labels": [ "ETSI Domain: Governance & Compliance", "ETSI_Domain_Governance_Compliance" ], "is_subclass_of": [ "AI Governance and Ethics", "ETSI Metaverse Domain Taxonomy" ], "wikilinks": [ "Audit Systems", "Audit Trails", "Automated Compliance", "ETSI GR MEC 032", "Governance Frameworks", "Industry Regulations", "Reporting Tools", "ApplicationLayer", "Blockchain", "Compliance Monitoring", "ETSI Metaverse Domain Taxonomy", "InfrastructureDomain", "ISO Standards", "Policy Enforcement", "Regulatory Standards", "Risk Management" ] }, { "id": "etsi-domain-governance-and-ethics", "title": "ETSI Domain: Governance & Ethics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Crossover domain for ETSI metaverse categorisation addressing ical governance frameworks, responsible decision-making processes, and value-aligned organisational structures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-governance-and-ethics", "labels": [ "ETSI Domain: Governance & Ethics", "ETSI_Domain_Governance___Ethics" ], "is_subclass_of": [ "Legal and Regulatory", "AI Governance and Ethics" ], "wikilinks": [ "Ethical Decision-Making", "Ethical Principles", "Ethics & Law", "Ethics Committee", "ETSI GR MEC 032", "Governance Models", "Stakeholder Accountability", "Stakeholder Engagement", "Value Framework", "ApplicationLayer", "Blockchain", "ETSI Metaverse Domain Taxonomy", "Governance", "Governance Board", "InfrastructureDomain", "Value Alignment" ] }, { "id": "etsi-eni-008", "title": "ETSI ENI 008", "domain": "standards", "domain_name": "Standards", "definition": "ETSI ENI 008 is a deliverable of the ETSI Experiential Networked Intelligence group, which works on AI-based network management and optimisation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:etsi-eni-008", "labels": [ "ETSI ENI 008", "ETSI ENI" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-gr-arf-007", "title": "ETSI GR ARF 007", "domain": "standards", "domain_name": "Standards", "definition": "ETSI GR ARF 007 is a group report from the ETSI Augmented Reality Framework activity addressing aspects of AR interoperability and architecture.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:etsi-gr-arf-007", "labels": [ "ETSI GR ARF 007" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-gr-arf-010", "title": "ETSI GR ARF 010", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ETSI GR ARF 010 is a Group Report produced by the ETSI Industry Specification Group on Augmented Reality Framework (ISG ARF), defining architectural concepts, terminology, and interoperability requirements for augmented reality systems. It specifies a reference architecture that decouples AR content pipelines, tracking subsystems, and rendering layers to enable multi-vendor interoperability. The report addresses how AR devices, services, and platforms should communicate using standardised interfaces covering world anchoring, coordinate systems, and content delivery. It serves as a foundational normative reference for the broader ARF specification suite, informing subsequent standards such as ETSI GS ARF 003.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:etsi-gr-arf-010", "labels": [ "ETSI GR ARF 010" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-gr-mec-032", "title": "ETSI GR MEC 032", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ETSI GR MEC 032 is a Group Report published by the ETSI Industry Specification Group for Multi-access Edge Computing (ISG MEC) that addresses network slicing support and integration with MEC systems, defining how MEC platform capabilities can be exposed and leveraged across sliced 5G network infrastructure. The report analyses the relationship between Multi-access Edge Computing and Network Slicing, identifying reference architectures, use cases, and the interplay between MEC management entities and network slice management functions. It provides normative guidance for operators and vendors integrating edge computing workloads with 5G core network slicing mechanisms, covering aspects such as slice selection for MEC application placement, resource isolation guarantees, and inter-slice MEC service continuity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:etsi-gr-mec-032", "labels": [ "ETSI GR MEC 032" ], "is_subclass_of": [ "MultiAccessEdgeComputing" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-gs-mec-003", "title": "ETSI GS MEC 003", "domain": "standards", "domain_name": "Standards", "definition": "An ETSI Group Specification within the Multi-access Edge Computing (MEC) series defining the framework and reference architecture for MEC. It describes functional elements and reference points of an edge computing platform.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:etsi-gs-mec-003", "labels": [ "ETSI GS MEC 003" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ETSI", "Technical Standard" ] }, { "id": "etsi-mec", "title": "ETSI MEC", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ETSI MEC (Multi-access Edge Computing) is a set of standards from the European Telecommunications Standards Institute defining an open framework for running applications at the edge of mobile and fixed networks, close to end users. It specifies APIs and a reference architecture so applications can access low-latency compute, radio-network information, and location services hosted at base stations or aggregation points. MEC is foundational to latency-sensitive use cases such as AR, autonomous systems, and IoT.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:etsi-mec", "labels": [ "ETSI MEC", "ETSI ISG MEC", "ETSI MEC Specification Series", "MEC Platform" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "etsi", "title": "etsi", "domain": "standards", "domain_name": "Standards", "definition": "ETSI (European Telecommunications Standards Institute) is a not-for-profit, independent standardisation organisation headquartered in Sophia Antipolis, France, that produces globally applicable ICT standards covering telecommunications, broadcasting, electronic signatures, cybersecurity, and emerging technologies such as 5G, network functions virtualisation (NFV), and multi-access edge computing (MEC). Founded in 1988 under the auspices of the European Commission, ETSI operates a membership model open to industry, governments, and research bodies worldwide, producing European Norms (ENs), ETSI Standards (ES), and Technical Specifications (TS) that feed into EU regulatory frameworks including eIDAS, the EU AI Act, and the European Cybersecurity Act. Its Industry Specification Groups (ISGs) provide an accelerated track for industry-led work on rapidly evolving topics such as quantum-safe cryptography, AI-assisted network management, and zero-trust architecture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:etsi", "labels": [ "ETSI", "ETSI EN 303 645", "ETSI EN 319 series", "ETSI GR MEC 002", "ETSI Standards" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "etsidomain-classification", "title": "ETSIDomainClassification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ETSIDomainClassification is a formal taxonomy from the European Telecommunications Standards Institute categorising technology domains, standards, and governance areas to enable systematic organisation and cross-domain coordination. It partitions the metaverse and digital ecosystem into Infrastructure, Interaction, Trust and Governance, and Computation and Intelligence domains, enabling standards alignment, regulatory mapping, and interoperability discovery across heterogeneous technical contexts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsidomain-classification", "labels": [ "ETSIDomainClassification" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "European Telecommunications Standards Institute", "Autonomous Robot", "MetaverseDomain" ] }, { "id": "etsi-domain-governance-economy", "title": "ETSI_Domain_Governance_Economy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ETSI domain intersection addressing economic governance, financial regulation, and market oversight mechanisms within metaverse ecosystems. These Governance Frameworks integrate Monetary Policy, Market Supervision, and Economic Oversight to ensure stable financial operations.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:etsi-domain-governance-economy", "labels": [ "ETSI_Domain_Governance_Economy" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Economic Oversight", "Economic Policy", "ETSI", "ETSI GR MEC 032", "ETSI GR PDL 030", "ETSI GR PDL 034", "Financial Governance", "Governance Frameworks", "ISO 23257", "Market Supervision", "Metaverse Standards Forum", "Monetary Policy", "Asset Management", "Blockchain", "MetaverseDomain" ] }, { "id": "etsi-domain-governance-security", "title": "ETSI_Domain_Governance_Security", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ETSI domain crossover representing security governance policies, risk management frameworks, and compliance enforcement across metaverse infrastructure and services. These Security Governance Systems coordinate Access Controls, Threat Management, and Compliance Verification.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:etsi-domain-governance-security", "labels": [ "ETSI_Domain_Governance_Security" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Access Controls", "ETSI", "ETSI GR MEC 032", "ETSI TR 104 168", "ETSI TS 133 501", "ISO 23257", "ISO 27001", "Security Governance Systems", "Security Verification", "Threat Management", "Threat Prevention", "Vulnerability Management", "Blockchain", "Compliance Verification", "MetaverseDomain" ] }, { "id": "etsi-domain-governance-society", "title": "ETSI_Domain_Governance_Society", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An ETSI governance domain addressing societal impacts, inclusion, community wellbeing, and cultural considerations for metaverse and digital infrastructure.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:etsi-domain-governance-society", "labels": [ "ETSI_Domain_Governance_Society" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Community Participation", "Cultural Respect", "ETSI", "ETSI GR MEC 032", "ISO 23257", "Metaverse Standards Forum", "Social Equity", "Accessibility Standards", "MetaverseDomain", "Telecollaboration" ] }, { "id": "etsi-domain-human-interface", "title": "ETSI_Domain_Human_Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An ETSI governance domain addressing user interaction design, accessibility, and human-computer interfaces within virtual environments. It defines standards for gesture, voice, gaze, and biometric input modalities, ensuring metaverse platforms prioritise usable and accessible experiences through alignment with ISO 9241 ergonomics standards, ETSI GR ARF 007/010, and inclusive design principles.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-human-interface", "labels": [ "ETSI_Domain_Human_Interface" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "ETSI", "ETSI GR ARF 007", "ETSI GR ARF 010", "ETSI GR MEC 032", "InclusiveDesign", "InteractionDesign", "ISO 9241", "ISO/IEC JTC 1/SC 24", "UserExperience", "AccessibilityStandards", "MetaverseDomain", "Telecollaboration", "VirtualEnvironment" ] }, { "id": "etsi-domain-human-interface-governance", "title": "ETSI_Domain_Human_Interface_Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An ETSI subdomain addressing governance mechanisms, policy frameworks, and institutional structures governing human-centric systems in VirtualEnvironment|virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:etsi-domain-human-interface-governance", "labels": [ "ETSI_Domain_Human_Interface_Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "DecentralisedGovernance", "ETSI", "ETSI GR MEC 032", "ISO 23257", "RoleBasedAccess", "WCAG 2.1", "Blockchain", "ConsentManagement", "MetaverseDomain", "PolicyEnforcement", "VirtualEnvironment" ] }, { "id": "etsi-domain-human-interface-ux", "title": "ETSI_Domain_Human_Interface_UX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An ETSI subdomain focusing on user experience design, usability engineering, and interaction paradigms optimising satisfaction, efficiency, and accessibility in immersive digital environments. It covers design principles, usability testing, accessibility compliance, inclusive UX design, and AI-driven adaptive interfaces aligned with ISO 9241-110 dialogue principles and ETSI GR ARF 010 specifications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-human-interface-ux", "labels": [ "ETSI_Domain_Human_Interface_UX" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "AccessibilityCompliance", "DesignPrinciples", "ETSI", "ETSI GR ARF 007", "ETSI GR ARF 010", "ETSI GR MEC 032", "InclusiveUXDesign", "ISO 9241-110", "UsabilityTesting", "MetaverseDomain", "Telecollaboration", "VirtualEnvironment" ] }, { "id": "etsi-domain-immersive-experiences", "title": "ETSI_Domain_Immersive_Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An ETSI standardisation domain encompassing technologies, frameworks, and practices for creating immersive digital experiences including Virtual Reality, Augmented Reality, Extended Reality, and Spatial Computing. Governed by ETSI GR CIM 052 and GR ARF 020, this domain addresses content rendering, real-time interaction, sensory feedback, and environmental simulation to deliver seamless, engaging immersive experiences across heterogeneous hardware from HMDs to mobile AR.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-immersive-experiences", "labels": [ "ETSI_Domain_Immersive_Experiences" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ContentRendering", "EnvironmentalSimulation", "ETSI", "ExtendedReality", "RealTimeInteraction", "SensoryFeedback", "AugmentedReality", "Computer Vision", "MetaverseDomain", "SpatialComputing", "VirtualReality" ] }, { "id": "etsi-domain-infrastructure", "title": "ETSI_Domain_Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An ETSI governance domain governing the foundational systems, networks, and computational resources enabling Metaverse platforms at scale. It encompasses network infrastructure, cloud and edge computing, data centres, content delivery, latency management, and hardware abstraction, providing reliable, performant, and secure technical foundations for distributed virtual environment operation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-infrastructure", "labels": [ "ETSI_Domain_Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "DataCenter", "ETSI", "NetworkSecurity", "SystemResilience", "6G Network Slice", "CloudComputing", "Cloud Rendering Service", "Compute Layer", "Content Delivery Network (CDN)", "Context Awareness", "Distributed Ledger Technology (DLT)", "EdgeComputing", "Edge Computing Node", "Edge Mesh Network", "Edge Network", "Edge Orchestration", "Hardware Abstraction Layer (HAL)", "Infrastructure Layer", "Latency", "Latency Management Protocol" ] }, { "id": "etsi-domain-infrastructure-data", "title": "ETSI_Domain_Infrastructure_Data", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The ETSI Infrastructure Data domain is the foundational technical and architectural framework enabling trustworthy, sovereign data exchange within metaverse and digital ecosystems through decentralised data spaces, spatial intelligence integration, and policy-driven governance standards. It operationalises Trustworthy Data Spaces via NGSI-LD APIs, Federated Catalogs, Policy Enforcement Points, and F5G backbone connectivity to address data sovereignty, security, and real-to-virtual integration across distributed networks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-infrastructure-data", "labels": [ "ETSI_Domain_Infrastructure_Data" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "NGSI-LD", "MetaverseDomain", "Storage Layer" ] }, { "id": "etsi-domain-infrastructure-governance", "title": "ETSI_Domain_Infrastructure_Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ETSI governance framework and mechanisms overseeing infrastructure domains through trust domain separation, policy enforcement, and security management implementing local policy control across distributed network components.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:etsi-domain-infrastructure-governance", "labels": [ "ETSI_Domain_Infrastructure_Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain", "Resilience Metric" ] }, { "id": "etsi-domain-infrastructure-immersive", "title": "ETSI_Domain_Infrastructure_Immersive", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A functional domain classification within the European Telecommunications Standards Institute (ETSI) metaverse technical architecture framework (GS MEC 003, GS ARF 003) that encompasses the foundational technological infrastructure required to deliver immersive experiences including extended real...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-infrastructure-immersive", "labels": [ "ETSI_Domain_Infrastructure_Immersive" ], "is_subclass_of": [ "Computing and Cloud", "ETSIDomainClassification", "InfrastructureLayer", "TechnicalArchitectureFramework", "StandardsBasedTaxonomy" ], "wikilinks": [ "3GPP TS 23.501 5G System Architecture", "5GStandAloneArchitecture", "BandwidthCapacity", "ComputationalThroughput", "ContentDeliveryNetwork", "DistributedCaching", "DistributedStorage", "ETSI GS ARF 003 Augmented Reality Framework", "ETSI GS MEC 003 Multi-Access Edge Computing Framework", "ETSI GS NFV Network Functions Virtualization", "ETSI ISG F5G Fifth Generation Fixed Network", "ExtendedReality", "FiberOpticsInfrastructure", "GPUComputeInfrastructure", "ISO/IEC 23090-14 Scene Description for MPEG Media", "Khronos OpenXR 1.0 Specification", "MetaverseEnvironment", "MultiAccessEdgeComputing", "NetworkSlicing", "RealtimeSynchronization" ] }, { "id": "etsi-domain-infrastructure-interop", "title": "ETSI_Domain_Infrastructure_Interop", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ETSI infrastructure interoperability domain providing standardised frameworks enabling disparate systems, services, and domains to exchange data and operate cohesively across organisational boundaries. It underpins cross-domain data integration through NGSI-LD interfaces, intent-based network management, spectrum sharing frameworks, and EU Data Act Article 35 mandates for centralised interoperability standards repositories.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-infrastructure-interop", "labels": [ "ETSI_Domain_Infrastructure_Interop" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "NGSI-LD", "Hardware-/Platform-Agnostic", "MetaverseDomain" ] }, { "id": "etsi-domain-infrastructure-security", "title": "ETSI_Domain_Infrastructure_Security", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ETSI infrastructure security domain safeguarding metaverse and edge computing environments through NFV architectural frameworks, cross-domain authentication standards, and decentralised identity systems. It addresses multi-party trust models, privacy-preserving mechanisms, and security lifecycle management coordinated with 3GPP and ITU-T for ultra-edge and terminal-edge deployments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-infrastructure-security", "labels": [ "ETSI_Domain_Infrastructure_Security" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "MetaverseDomain", "Quantum Network Node" ] }, { "id": "etsi-domain-interoperability", "title": "ETSI_Domain_Interoperability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ETSI domain consolidating standards enabling different systems and organisations to work together effectively through technical and semantic data exchange mechanisms, shared vocabularies such as NGSI-LD and SAREF, and standardised protocols. It covers avatar interoperability, digital twin synchronisation, state synchronisation, and portability of digital assets across metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-interoperability", "labels": [ "ETSI_Domain_Interoperability" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "NGSI-LD", "SAREF", "3D Scene Exchange Protocol (SXP)", "API Standard", "Avatar Interoperability", "Compatibility Process", "Digital Twin Interop Protocol", "Digital Twin Synchronisation Bus", "Discovery Layer", "glTF (3D File Format)", "Interoperability", "Interoperability Framework", "MetaverseDomain", "Multiverse", "Persistence", "Platform Layer", "Portability", "Service Layer", "State Synchronization", "Universal Manifest" ] }, { "id": "etsi-domain-interoperability-creative", "title": "ETSI_Domain_Interoperability_Creative", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The ETSI Domain Interoperability/Creative is a functional domain classification defined by the European Telecommunications Standards Institute that addresses the intersection of technical interoperability standards and creative content workflows within metaverse and extended reality ecosystems. It encompasses the standards, protocols, and toolchains\u2014including glTF 2.0, USD, WebXR, OpenXR, and MPEG-I Scene Description\u2014that enable digital creative assets such as three-dimensional models, animations, and immersive environments to be authored once and deployed across heterogeneous platforms without proprietary lock-in or fidelity degradation. The domain integrates digital rights management frameworks and semantic metadata schemas to preserve creator attribution, licensing terms, and provenance across cross-platform distribution pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-interoperability-creative", "labels": [ "ETSI_Domain_Interoperability_Creative" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "CreativeDomain", "ETSI ISG ARF Augmented Reality Framework", "ETSI ISG MEC Multi-access Edge Computing", "European Commission Digital Decade 2030", "InteroperabilityDomain", "ISO/IEC 21122-3 Watermarking", "ISO/IEC 23090-14 Scene Description", "Khronos Group glTF 2.0 Specification", "Khronos OpenXR 1.1", "Metaverse Standards Forum", "Pixar USD Universal Scene Description", "W3C WebXR Device API", "Metaverse Content Pipeline", "MetaverseDomain" ] }, { "id": "etsi-domain-reality-capture", "title": "ETSI_Domain_Reality_Capture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An ETSI standardisation domain enabling digital reconstruction of physical environments and objects through 3D scanning, photogrammetry, sensor fusion, depth sensing, and motion capture to create semantically labelled digital twins for metaverse and immersive applications. Governed by ETSI GS ARF 004-6 and GR ARF 010, this domain defines interoperability requirements for real-time mesh generation, volumetric video coding (ISO V3C/V-PCC), and integration with scene management and digital twin platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-reality-capture", "labels": [ "ETSI_Domain_Reality_Capture", "ETSI Domain Reality Capture" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "DigitalTwin", "Digital Twin", "Human Capture & Recognition", "Human Capture & Recognition", "MetaverseDomain", "Motion Capture Rig", "Photogrammetry", "Reality Capture System" ] }, { "id": "etsi-domain-reality-capture-creative", "title": "ETSI_Domain_Reality_Capture_Creative", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ETSI domain addressing real-time digital performance capture and creative content generation for immersive experiences. It integrates motion capture, facial expression tracking, gesture recognition, and AI-enhanced animation synthesis to transform physical performances into high-fidelity digital representations for metaverse avatars, virtual events, and interactive entertainment applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-reality-capture-creative", "labels": [ "ETSI_Domain_Reality_Capture_Creative" ], "is_subclass_of": [ "Content and Assets", "Digital Performance Capture" ], "wikilinks": [ "Digital Performance Capture", "MetaverseDomain" ] }, { "id": "etsi-domain-virtual-economy", "title": "ETSI_Domain_Virtual_Economy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ETSI domain encompassing economic systems, digital assets, and financial mechanisms enabling value exchange within metaverse and virtual environments through crypto tokens, cryptocurrencies, CBDCs, NFTs, and virtual assets. It coordinates decentralised exchanges, creator economy mechanisms, creator royalty tokens, carbon credit tokenisation, and standards development integrating virtual economies with traditional finance and regulatory compliance frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-virtual-economy", "labels": [ "ETSI_Domain_Virtual_Economy" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "NFT", "CarbonCreditToken", "Carbon Credit Token", "CentralBankDigitalCurrency", "Central Bank Digital Currency (CBDC)", "CreatorEconomy", "Creator Economy", "CreatorRoyaltyToken", "Creator Royalty Token", "CryptoToken", "Crypto Token", "Cryptocurrency", "Cryptocurrency", "DecentralizedExchange", "Decentralized Exchange (DEX)", "DigitalAsset", "Digital Asset", "DigitalGoods", "Digital Goods", "Digital Goods Registry" ] }, { "id": "etsi-domain-virtual-society", "title": "ETSI_Domain_Virtual_Society", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ETSI_Domain_Virtual_Society is a standardisation domain within the European Telecommunications Standards Institute (ETSI) that addresses the technical foundations and interoperability requirements of virtual society infrastructures, including metaverse ecosystems, Extended Reality (XR) environments, decentralised digital identity, virtual economies, and interoperable virtual worlds. It produces technical reports and specifications that coordinate standardisation work across ETSI technical bodies, liaising with ISO/IEC, ITU-T, 3GPP, W3C, and the Metaverse Standards Forum to ensure coherent standards coverage. The domain spans immersive media, avatar portability, blockchain-based identity and ownership, privacy-preserving authentication, and the governance frameworks required for responsible virtual society deployment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:etsi-domain-virtual-society", "labels": [ "ETSI_Domain_Virtual_Society" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "3D assets", "3D Audio", "3D Web Interoperability", "3GPP", "5G/6G", "advanced audio coding (AAC)", "AI", "AI-enhanced AR/VR", "AR/VR interfaces", "avatar standards", "avatars", "Blockchain-based solutions", "Blockchain integration", "blockchain technology", "cloud streaming", "Cyber Resilience Act", "DAOs", "DeFi", "Decentraland", "decentralized identity" ] }, { "id": "eu-ai-act-article-50", "title": "EU AI Act Article 50", "domain": "governance", "domain_name": "Governance", "definition": "EU AI Act Article 50 establishes transparency obligations that apply to providers and deployers of specific categories of AI system within the European Union's Artificial Intelligence Act regulatory framework. The article requires that AI systems designed to interact with natural persons, or that generate synthetic audio, image, video, or text content, must disclose their AI nature to affected individuals. Providers of deep fake content generation tools are required to label outputs in machine-readable form, and broadcast media and online platforms carrying synthetic media must carry appropriate disclosures. The article represents the EU's approach to mandating content authentication and countering AI-generated deception at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:eu-ai-act-regulatory-instrument-article-50", "labels": [ "EU AI Act Article 50", "EU AI Act Article 50 (2024)" ], "is_subclass_of": [ "EU AI Act Regulatory Instrument" ], "wikilinks": [] }, { "id": "eu-ai-act-regulatory-instrument-article-53", "title": "EU AI Act Article 53", "domain": "governance", "domain_name": "Governance", "definition": "The article of the EU Artificial Intelligence Act (Regulation (EU) 2024/1689) that imposes obligations on providers of general-purpose AI models: maintaining technical documentation, supplying information to downstream integrators, putting in place a policy to comply with EU copyright law including text-and-data-mining opt-outs, and publishing a sufficiently detailed public summary of the content used for training. It is the principal legal hook connecting web scraping and training-data practices to EU regulatory enforcement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:eu-ai-act-regulatory-instrument-article-53", "labels": [ "EU AI Act Article 53" ], "is_subclass_of": [ "EU AI Act" ], "wikilinks": [ "EU AI Act", "Transparency", "Training Data", "AI Scrapers" ] }, { "id": "eu-ai-act-regulatory-instrument", "title": "EU AI Act Regulatory Instrument", "domain": "governance", "domain_name": "Governance", "definition": "The EU AI Act (Regulation (EU) 2024/1689) is the European Union's comprehensive, legally binding horizontal regulation for artificial intelligence, establishing a four-tier risk-based classification system that prohibits certain AI practices outright, imposes strict pre-market conformity obligations on high-risk AI systems, mandates transparency duties for limited-risk AI, and introduces a dedicated governance regime for general-purpose AI models. Published in the Official Journal in July 2024 and progressively entering into force through 2025\u20132027, it represents the world's first binding horizontal AI law and creates a de facto global compliance benchmark through its extraterritorial scope. Enforcement is shared between national market surveillance authorities, the European AI Office, and the European AI Board, with penalties reaching up to \u20ac35 million or 7% of global annual turnover.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:eu-ai-act-regulatory-instrument", "labels": [ "EU AI Act Regulatory Instrument", "AI Act", "AI Act Compliance", "EU AI Act", "EU AI Act Article 10", "EU AI Act Article 27" ], "is_subclass_of": [ "AI Regulation" ], "wikilinks": [] }, { "id": "eu-ai-act", "title": "EU AI Act", "domain": "governance", "domain_name": "Governance", "definition": "The European Union's Artificial Intelligence Act (Regulation (EU) 2024/1689), the world's first comprehensive horizontal legal framework for artificial intelligence, which classifies AI systems into risk tiers \u2014 unacceptable, high, limited and minimal \u2014 and imposes proportionate obligations on providers and deployers, including conformity assessment, data governance, transparency, human oversight, robustness and post-market monitoring, with extraterritorial reach over any system placed on the EU market and penalties of up to 7% of global annual turnover.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:eu-ai-act", "labels": [ "EU AI Act" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Regulation", "Risk Based Regulation", "AI Governance", "Algorithmic Bias" ] }, { "id": "eu-digital-finance-strategy", "title": "EU Digital Finance Strategy", "domain": "finance", "domain_name": "Finance", "definition": "The EU Digital Finance Strategy is the European Commission's policy framework, adopted in 2020, for enabling digital transformation of the EU financial sector while managing the associated risks. It sets out priorities including removing fragmentation in the digital single market, adapting the EU regulatory framework to facilitate digital innovation, promoting data-driven finance, and addressing digital operational resilience. The strategy provided the policy basis for subsequent EU regulations governing crypto-assets, such as MiCA.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:eu-digital-finance-strategy", "labels": [ "EU Digital Finance Strategy" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "eu-green-deal", "title": "EU Green Deal", "domain": "governance", "domain_name": "Governance", "definition": "The EU Green Deal is the European Union's overarching policy package aimed at making Europe climate-neutral by 2050. It sets binding emissions-reduction targets and coordinates legislation across energy, industry, transport, agriculture, and finance, including circular-economy and sustainable-finance measures. As a governance framework it shapes regulation, funding, and reporting obligations that ripple through technology and supply-chain sectors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eu-green-deal", "labels": [ "EU Green Deal" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "eu-hleg-ai", "title": "EU HLEG AI", "domain": "governance", "domain_name": "Governance", "definition": "The European Commission High-Level Expert Group on Artificial Intelligence (EU HLEG AI) is a multidisciplinary advisory body convened by the European Commission in 2018, comprising 52 experts from academia, industry, and civil society. It produced the foundational Ethics Guidelines for Trustworthy AI (April 2019), establishing seven key requirements \u2014 human agency and oversight, technical robustness and safety, privacy and data governance, transparency, diversity and fairness, societal and environmental wellbeing, and accountability \u2014 that shaped subsequent EU AI policy. The group also delivered a self-assessment tool (ALTAI) and sector-specific policy recommendations that directly informed the European AI Strategy and the EU AI Act legislative process.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:eu-hleg-ai", "labels": [ "EU HLEG AI", "EU High-Level Expert Group on AI" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [ "AI Ethics", "AI Regulation", "Trustworthy AI", "AI Governance" ] }, { "id": "eu-mi-ca-regulation", "title": "EU MiCA Regulation", "domain": "finance", "domain_name": "Finance", "definition": "The EU Markets in Crypto-Assets Regulation is a European Union framework establishing harmonised rules for crypto-asset issuance, trading and service provision, covering asset-referenced tokens, e-money tokens, and crypto-asset service providers across EU member states.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eu-mi-ca-regulation", "labels": [ "EU MiCA Regulation" ], "is_subclass_of": [ "Crypto Regulation" ], "wikilinks": [ "Regulatory Frameworks", "Investor Protection", "Stablecoin", "Crypto Regulation", "https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32023R1114", "https://www.esma.europa.eu/esmas-activities/digital-finance-and-innovation/markets-crypto-assets-regulation-mica" ] }, { "id": "eu-taxonomy", "title": "EU Taxonomy", "domain": "governance", "domain_name": "Governance", "definition": "The EU Taxonomy is a classification system that defines which economic activities count as environmentally sustainable under European Union law. It sets technical screening criteria across climate and environmental objectives so that investors, companies, and regulators apply a common definition of green activity. The taxonomy underpins sustainable-finance disclosure and ESG reporting by preventing greenwashing through standardised criteria.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eu-taxonomy", "labels": [ "EU Taxonomy", "EU Taxonomy Regulation 2020/852", "Green Taxonomy", "Taxonomy Regulation" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "eu-union-customs-code", "title": "EU Union Customs Code", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The EU Union Customs Code (UCC) is the legal framework governing customs rules and procedures for goods entering, leaving, or moving within the European Union. It standardises declarations, tariff classification, valuation, origin rules, and the move toward fully electronic, paperless customs processing. The UCC is the reference regime for customs and trade-facilitation systems operating across the EU single market.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eu-union-customs-code", "labels": [ "EU Union Customs Code" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "eudi-wallet", "title": "EUDI Wallet", "domain": "security", "domain_name": "Security", "definition": "The EUDI Wallet (European Digital Identity Wallet) is a mobile application framework mandated under the revised eIDAS regulation, enabling European citizens and residents to store, present, and selectively disclose identity attributes and credentials across member states. It allows holders to prove identity, sign documents, and share verifiable attestations with public and private services while retaining control over what data is revealed. The wallet is the user-facing instrument of the EU's interoperable cross-border identity scheme.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:eudi-wallet", "labels": [ "EUDI Wallet" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "evm-compatibility", "title": "EVM Compatibility", "domain": "blockchain", "domain_name": "Blockchain", "definition": "EVM compatibility is the property of a blockchain or execution environment that allows it to run smart contracts compiled for the Ethereum Virtual Machine without modification, supporting the same bytecode, opcodes and account model. Compatible chains can reuse Ethereum tooling, wallets, contracts and developer skills, which lowers the cost of porting applications and bootstrapping liquidity. It is the foundation of the wider EVM ecosystem spanning Layer-2 rollups, sidechains and alternative Layer-1 networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:evm-compatibility", "labels": [ "EVM Compatibility" ], "is_subclass_of": [ "Smart Contract Platform", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "evm-compatible-blockchain", "title": "EVM-Compatible Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An EVM-compatible blockchain is a network whose execution environment implements the Ethereum Virtual Machine semantics, allowing it to run unmodified Ethereum smart contract bytecode and reuse Ethereum tooling. Compatibility lets developers deploy Solidity contracts, wallets and infrastructure across many chains with minimal changes, fostering an interoperable multi-chain ecosystem. Such chains include layer-2 rollups, sidechains and alternative layer-1 networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:evm-compatible-blockchain", "labels": [ "EVM-Compatible Blockchain", "EVM Compatible Blockchain", "EVM-Compatible Chain" ], "is_subclass_of": [ "Blockchain", "Blockchain Network" ], "wikilinks": [] }, { "id": "evm", "title": "EVM", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Virtual Machine (EVM) is a stack-based, sandboxed, quasi-Turing-complete virtual machine that executes smart contract bytecode on the Ethereum network and EVM-compatible blockchains. It defines a deterministic computation environment in which all nodes independently execute the same transactions to reach identical state transitions, using a gas metering system to bound computation and prevent denial-of-service attacks. The EVM specification encompasses opcodes, memory model, call semantics, and the gas cost schedule governing the economic cost of each computational step. As an open industry standard maintained by the Enterprise Ethereum Alliance, the EVM has been adopted by hundreds of alternative blockchain networks, establishing it as the dominant cross-chain execution environment.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:evm", "labels": [ "EVM", "EVM Bytecode", "EVM Chain", "EVM Equivalence" ], "is_subclass_of": [ "Virtual Machine" ], "wikilinks": [] }, { "id": "early-stopping", "title": "Early Stopping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A regularisation technique that terminates model training when validation performance ceases to improve for a configurable number of epochs (the patience parameter), preventing overfitting by restoring the best checkpoint before performance degraded. Early stopping balances training progress against generalisation to unseen data and is most effective when combined with other regularisation techniques such as dropout and weight decay.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:early-stopping", "labels": [ "Early Stopping", "Stopping Criterion" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Gaussian Splatting", "MetaverseDomain" ] }, { "id": "earth-observation-analysis-ready-data", "title": "Earth Observation Analysis Ready Data", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-analysis-ready-data", "labels": [ "Earth Observation Analysis Ready Data" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-applications", "title": "Earth Observation Applications", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-applications", "labels": [ "Earth Observation Applications" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-collection", "title": "Earth Observation Collection", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-collection", "labels": [ "Earth Observation Collection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-archive", "title": "Earth Observation Data Archive", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-archive", "labels": [ "Earth Observation Data Archive" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-assimilation", "title": "Earth Observation Data Assimilation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-assimilation", "labels": [ "Earth Observation Data Assimilation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-cube", "title": "Earth Observation Data Cube", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-cube", "labels": [ "Earth Observation Data Cube" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-discovery", "title": "Earth Observation Data Discovery", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-discovery", "labels": [ "Earth Observation Data Discovery" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-processing-level", "title": "Earth Observation Data Processing Level", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-processing-level", "labels": [ "Earth Observation Data Processing Level" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-product", "title": "Earth Observation Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-product", "labels": [ "Earth Observation Data Product" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-data-quality", "title": "Earth Observation Data Quality", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-data-quality", "labels": [ "Earth Observation Data Quality" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-granule", "title": "Earth Observation Granule", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-granule", "labels": [ "Earth Observation Granule" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-instrument", "title": "Earth Observation Instrument", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-instrument", "labels": [ "Earth Observation Instrument" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "earth-observation-level-0-data-product", "title": "Earth Observation Level 0 Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-0-data-product", "labels": [ "Earth Observation Level 0 Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-1-processing-state", "title": "Earth Observation Level 1 Processing State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-1-processing-state", "labels": [ "Earth Observation Level 1 Processing State", "L1" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-level-1a-data-product", "title": "Earth Observation Level 1A Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-1a-data-product", "labels": [ "Earth Observation Level 1A Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-1b-data-product", "title": "Earth Observation Level 1B Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-1b-data-product", "labels": [ "Earth Observation Level 1B Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-2-data-product", "title": "Earth Observation Level 2 Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-2-data-product", "labels": [ "Earth Observation Level 2 Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-2-processing-state", "title": "Earth Observation Level 2 Processing State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-2-processing-state", "labels": [ "Earth Observation Level 2 Processing State", "L2" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-level-3-data-product", "title": "Earth Observation Level 3 Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-3-data-product", "labels": [ "Earth Observation Level 3 Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-3-processing-state", "title": "Earth Observation Level 3 Processing State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-3-processing-state", "labels": [ "Earth Observation Level 3 Processing State", "L3" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-level-4-data-product", "title": "Earth Observation Level 4 Data Product", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-4-data-product", "labels": [ "Earth Observation Level 4 Data Product" ], "is_subclass_of": [ "Earth Observation Data Product" ], "wikilinks": [] }, { "id": "earth-observation-level-4-processing-state", "title": "Earth Observation Level 4 Processing State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-level-4-processing-state", "labels": [ "Earth Observation Level 4 Processing State", "L4" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-processing", "title": "Earth Observation Processing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-processing", "labels": [ "Earth Observation Processing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation-satellite", "title": "Earth Observation Satellite", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-satellite", "labels": [ "Earth Observation Satellite" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "earth-observation-sensor-fusion", "title": "Earth Observation Sensor Fusion", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-sensor-fusion", "labels": [ "Earth Observation Sensor Fusion" ], "is_subclass_of": [ "Sensor Fusion" ], "wikilinks": [] }, { "id": "earth-observation-sensor", "title": "Earth Observation Sensor", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-sensor", "labels": [ "Earth Observation Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "earth-observation-spectrometer", "title": "Earth Observation Spectrometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-spectrometer", "labels": [ "Earth Observation Spectrometer" ], "is_subclass_of": [ "Earth Observation Instrument" ], "wikilinks": [] }, { "id": "earth-observation-and-geospatial-sensing", "title": "Earth Observation and Geospatial Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation-and-geospatial-sensing", "labels": [ "Earth Observation and Geospatial Sensing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-observation", "title": "Earth Observation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-observation", "labels": [ "Earth Observation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-orientation-parameter", "title": "Earth Orientation Parameter", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-orientation-parameter", "labels": [ "Earth Orientation Parameter" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-science-data-record", "title": "Earth Science Data Record", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-science-data-record", "labels": [ "Earth Science Data Record", "ESDR" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-system-phenomena", "title": "Earth System Phenomena", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-system-phenomena", "labels": [ "Earth System Phenomena" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-centred-earth-fixed-frame", "title": "Earth-centred Earth-fixed Frame", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-centred-earth-fixed-frame", "labels": [ "Earth-centred Earth-fixed Frame" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "earth-centred-inertial-frame", "title": "Earth-centred Inertial Frame", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:earth-centred-inertial-frame", "labels": [ "Earth-centred Inertial Frame" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "eccentric-anomaly", "title": "Eccentric Anomaly", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:eccentric-anomaly", "labels": [ "Eccentric Anomaly" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "eclair", "title": "Eclair", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Eclair is an open-source, production-grade implementation of the Bitcoin Lightning Network protocol, written in Scala and developed by ACINQ. It implements the full BOLT specification suite, enabling trustless off-chain payment channels, multi-hop routing, and peer-to-peer gossip between Lightning nodes. Eclair serves as the backend engine for ACINQ's Phoenix mobile wallet and is widely deployed as a routing node by businesses and individuals seeking high-throughput, low-latency Bitcoin micropayments. It is interoperable with other conformant Lightning implementations such as LND and Core Lightning through shared adherence to the BOLT standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:eclair", "labels": [ "Eclair" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning Network", "Payment Channel", "Bitcoin" ] }, { "id": "eclipse-attack", "title": "Eclipse Attack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Eclipse Attack is a peer-to-peer network attack in which an adversary monopolises all of a target node's inbound and outbound connections, isolating it from the honest network. The eclipsed node is fed a fabricated view of the blockchain, enabling the attacker to double-spend against that node, delay its transaction confirmations, or waste its mining resources on a private fork. Countermeasures include connection diversity, random peer selection, and detecting network-level routing anomalies.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:eclipse-attack", "labels": [ "Eclipse Attack" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Network Component", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "eclipse-power-budget", "title": "Eclipse Power Budget", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:eclipse-power-budget", "labels": [ "Eclipse Power Budget" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ecliptic-coordinate-system", "title": "Ecliptic Coordinate System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ecliptic-coordinate-system", "labels": [ "Ecliptic Coordinate System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ecliptic-latitude", "title": "Ecliptic Latitude", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ecliptic-latitude", "labels": [ "Ecliptic Latitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ecliptic-longitude", "title": "Ecliptic Longitude", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ecliptic-longitude", "labels": [ "Ecliptic Longitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ecma-international", "title": "Ecma International", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ecma International is an industry standards organisation that develops and maintains specifications for information and communication technology, most notably the ECMAScript language standard (ECMA-262). It operates through technical committees that produce royalty-free, openly published standards adopted across the software industry. Ecma collaborates with bodies such as ISO and IEC to give its specifications wider international standing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ecma-international", "labels": [ "Ecma International", "ECMA International" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "ecmascript", "title": "Ecmascript", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ECMAScript is the standardised specification of the scripting language commonly implemented as JavaScript, defining its syntax, semantics, and core libraries. Maintained by Ecma International as the ECMA-262 standard and developed through the TC39 process, it provides a stable, versioned contract that browser and runtime vendors implement. ECMAScript underpins client-side web programming and a large ecosystem of server-side and tooling runtimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ecmascript", "labels": [ "Ecmascript", "ECMAScript" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "ecodesign-for-sustainable-products-regulation", "title": "Ecodesign For Sustainable Products Regulation", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The Ecodesign for Sustainable Products Regulation (ESPR) is an EU framework that sets ecodesign requirements to improve the durability, reusability, repairability, and recyclability of products placed on the EU market. It introduces the Digital Product Passport to carry product sustainability and supply-chain data, and bans the destruction of unsold goods in some categories. ESPR is a cornerstone of the EU circular-economy agenda.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ecodesign-for-sustainable-products-regulation", "labels": [ "Ecodesign For Sustainable Products Regulation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "econometrics", "title": "Econometrics", "domain": "data", "domain_name": "Data", "definition": "Econometrics is the application of statistical and mathematical methods to economic data in order to test hypotheses, estimate relationships, and forecast economic phenomena. It combines economic theory with regression analysis, time-series modelling, and causal-inference techniques to quantify effects such as elasticities and policy impacts. Econometrics is foundational to empirical economics and increasingly overlaps with machine learning for prediction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:econometrics", "labels": [ "Econometrics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "economic-competitiveness", "title": "Economic Competitiveness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Economic competitiveness is the capacity of a firm, sector or economy to sustain productivity, innovation and market share relative to peers, typically assessed through measures such as productivity growth, export performance and capacity to attract investment and talent. In the context of artificial intelligence, it refers to how AI adoption and capability affect a firm's or nation's relative economic standing. Investment in AI capability and competitive dynamics within AI markets are both treated as drivers of broader economic competitiveness.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:economic-competitiveness", "labels": [ "Economic Competitiveness" ], "is_subclass_of": [ "National Competitiveness" ], "wikilinks": [] }, { "id": "economic-development", "title": "Economic Development", "domain": "governance", "domain_name": "Governance", "definition": "Economic development is the sustained process of improving a region's productive capacity, living standards, and institutional capabilities, typically measured through indicators such as employment, infrastructure investment, and income growth. It is pursued through policy instruments including industrial strategy, financial inclusion programmes, and public investment in skills and infrastructure. National and regional strategies increasingly treat technology adoption as a lever for economic development.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:economic-development", "labels": [ "Economic Development" ], "is_subclass_of": [ "Economic Growth" ], "wikilinks": [] }, { "id": "economic-exchange", "title": "Economic Exchange", "domain": "finance", "domain_name": "Finance", "definition": "Economic exchange is the voluntary transfer of goods, services, information, or rights between parties according to mutually agreed terms of value equivalence, constituting the fundamental transaction unit through which market economies allocate resources, generate price signals, and produce social surplus. It encompasses the full spectrum from primitive barter to complex multi-party financial instrument transactions, requiring coordination mechanisms \u2014 markets, prices, contracts, platforms \u2014 to match counterparties, communicate value, and enforce agreements. Modern digital economic exchange increasingly occurs through automated market makers, decentralised exchange protocols, and algorithmic trading systems that reduce friction and enable global, continuous, programmatic commerce.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:economic-exchange", "labels": [ "Economic Exchange" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "economic-finality", "title": "Economic Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Economic finality is a guarantee in proof-of-stake blockchains that reverting a finalised block would require an attacker to forfeit an economically prohibitive amount of staked capital through slashing. Rather than relying on probabilistic confirmation depth, it makes reversal irrational by binding the cost of an attack to a large, destroyable bond. A block is economically final once enough validators have attested to it that any conflicting chain would entail penalising at least a quantified fraction of the total stake. It is the cryptoeconomic foundation of settlement assurance in modern staking-based consensus.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:economic-finality", "labels": [ "Economic Finality" ], "is_subclass_of": [ "Finality" ], "wikilinks": [] }, { "id": "economic-governance", "title": "Economic Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Economic Governance is the set of institutions, rules and decision-making processes through which economic activity is steered, monitored and held accountable. In digital and metaverse contexts it spans token issuance, treasury management, incentive design and market oversight, extending traditional monetary and fiscal governance into decentralised digital economies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:economic-governance", "labels": [ "Economic Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "economic-growth", "title": "Economic Growth", "domain": "finance", "domain_name": "Finance", "definition": "Economic Growth is the increase over time in the value of goods and services produced by an economy, typically measured as the rise in real gross domestic product. It is driven by accumulation of capital, growth in the labour force, and improvements in productivity from technology and innovation. Sustained growth is a primary objective of economic policy and a frequently cited benefit of technology adoption.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:economic-growth", "labels": [ "Economic Growth" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "economic-layer", "title": "Economic Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Economic Layer is the abstraction level at which concrete economic mechanisms are implemented in blockchain and distributed systems, encompassing fee markets, reward distribution formulae, automated market makers, staking yield calculations, and game-theoretic incentive structures. It is distinct from conceptual economic theory in that it specifies exact parameters, formulas, and on-chain behaviours.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:economic-layer", "labels": [ "Economic Layer", "EconomicLayer" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "economic-mechanism", "title": "Economic Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Economic Mechanism is a structured system of incentives, penalties, and resource allocation rules designed to align participant behavior with desired network outcomes, drawing from mechanism design theory to create game-theoretic environments where rational self-interested actors produce collectively beneficial outcomes without central coordination.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:economic-mechanism", "labels": [ "Economic Mechanism", "EconomicMechanism" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity" ], "wikilinks": [ "Fee Burning", "Gas Mechanism", "Liquidity Mining", "Mechanism Design Theory", "Slashing", "Staking Reward", "Tokenomics Research", "Transaction Fee Economics", "Block Reward", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "Token", "Transaction Fee" ] }, { "id": "economic-model", "title": "Economic Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An economic model is a simplified representation of an economy or market used to analyse behaviour and predict outcomes. It expresses relationships between variables such as supply, demand, prices and output.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:economic-model", "labels": [ "Economic Model", "Economic Incentive Design", "Regional Economic Policy" ], "is_subclass_of": [ "Economics", "Computational Modelling", "Mathematical Modelling" ], "wikilinks": [ "Economics", "Agent-Based Modelling", "Simulation", "Game Theory" ] }, { "id": "economic-multiplier-effect", "title": "Economic Multiplier Effect", "domain": "economics", "domain_name": "Economics", "definition": "The macroeconomic phenomenon where an initial increase in spending or income generates a larger total increase in economic activity through subsequent rounds of consumption.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:economic-multiplier-effect", "labels": [ "Economic Multiplier Effect" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "economic-parameters", "title": "Economic Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The configurable variables and constraints that govern virtual economy behaviour in metaverse environments, including token supply mechanisms, transaction fees, inflation rates, reward structures, staking parameters, and liquidity controls that shape economic interactions between users and digital assets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:economic-parameters", "labels": [ "Economic Parameters" ], "is_subclass_of": [ "Platform and Environment", "Virtual Economy" ], "wikilinks": [ "metaverse", "Virtual Economy" ] }, { "id": "economic-participation", "title": "Economic Participation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The mechanisms and opportunities enabling users to engage in economic activities within metaverse environments, including earning, spending, trading, and governance participation through cryptocurrency, NFTs, and decentralised autonomous organisation (DAO) structures.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:economic-participation", "labels": [ "Economic Participation" ], "is_subclass_of": [ "Platform and Environment", "Virtual Economy" ], "wikilinks": [ "Play-to-Earn", "metaverse", "Virtual Economy" ] }, { "id": "economic-security", "title": "Economic Security", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cost of attack vs reward within blockchain systems, providing essential functionality for distributed ledger technology operations and properties.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:economic-security", "labels": [ "Economic Security" ], "is_subclass_of": [ "DeFi and Economics", "Economic Mechanism", "Blockchain Entity", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "economic-substance-test", "title": "Economic Substance Test", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The economic substance test is a regulatory analysis used to determine whether a transaction or arrangement has genuine business purpose and economic effect beyond obtaining a favourable legal or tax outcome, distinguishing legitimate commercial activity from arrangements structured purely to evade classification. In securities regulation it is applied alongside frameworks such as the Howey test when assessing whether an arrangement constitutes an investment contract. Regulators use the test to look past the formal structure of a transaction to its underlying economic reality.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:economic-substance-test", "labels": [ "Economic Substance Test" ], "is_subclass_of": [ "Securities Regulation" ], "wikilinks": [] }, { "id": "economic-systems", "title": "Economic Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Economic Systems are the structured sets of rules, institutions, and mechanisms that govern the production, distribution, and consumption of goods, value, and digital assets within a platform or environment. In spatial computing and Web3 contexts, economic systems encompass tokenomics, smart-contract-enforced incentive structures, and decentralised exchange mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:economic-systems", "labels": [ "Economic Systems", "Economic System" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "economics", "title": "Economics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Economics, as applied to the digital and AI-driven era, is the systematic study of how scarce resources are allocated through markets, institutions, and mechanisms when the primary inputs and outputs are information goods, algorithmic capabilities, autonomous agents, and cryptographic assets.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:economics", "labels": [ "Economics", "Information Economics", "Post-Scarcity Economics" ], "is_subclass_of": [ "AI Research Area", "Complexity Science", "Social Science", "Political Economy", "Information Theory", "Institutional Economics", "Behavioural Economics" ], "wikilinks": [ "Agent-Based Models", "Antitrust Policy", "Auction Theory", "Bank of England", "Behavioural Economics", "Behavioural Science", "BIS", "Carbon Market Design", "Causal Inference", "Complexity Modelling", "Complexity Science", "Computational Complexity", "Contract Theory", "Econometrics", "EconomicsDomain", "Empirical IO", "Financial Conduct Authority", "Game Theory", "General Equilibrium Models", "IMF" ] }, { "id": "ecosystem-connectivity", "title": "Ecosystem Connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ecosystem Connectivity is the degree to which distinct platforms, services, and participants in a technology ecosystem can discover, interoperate with, and exchange value across one another. It depends on shared standards, compatible interfaces, and gateways that link otherwise siloed systems. High connectivity amplifies network effects and lowers switching and integration costs across the ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ecosystem-connectivity", "labels": [ "Ecosystem Connectivity" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "ed25519", "title": "ed25519", "domain": "security", "domain_name": "Security", "definition": "Ed25519 is a high-performance elliptic-curve digital signature scheme instantiated on the Twisted Edwards curve over the prime field GF(2^255-19), offering approximately 128 bits of security with deterministic signing and compact 64-byte signatures. Designed by Bernstein, Duif, Lange, Schwabe, and Yang, it is engineered to resist side-channel attacks through constant-time implementation properties and eliminates the weak-nonce vulnerabilities that affect ECDSA. Ed25519 is standardised in RFC 8032 and FIPS 186-5, and is widely deployed across blockchain networks, SSH authentication, TLS handshakes, and decentralised identity systems.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:ed25519", "labels": [ "Ed25519", "Ed25519 Signature", "Ed25519 Signatures" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "ed-dsa", "title": "EdDSA", "domain": "security", "domain_name": "Security", "definition": "EdDSA (Edwards-curve Digital Signature Algorithm) is a high-performance digital signature scheme based on twisted Edwards elliptic curves, standardised in RFC 8032. It provides deterministic signing\u2014eliminating the random number generation vulnerabilities that afflicted earlier schemes like ECDSA\u2014whilst offering strong security with compact key and signature sizes. The most widely deployed instantiation is Ed25519, which operates over Curve25519 and produces 64-byte signatures with 128-bit security. EdDSA is extensively used in secure communications protocols, blockchain systems, verifiable credentials, and decentralised identity frameworks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ed-dsa", "labels": [ "EdDSA" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "edge-ai-accelerator", "title": "Edge AI Accelerator", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Specialised silicon or programmable hardware designed to execute AI inference workloads directly on edge or IoT devices, without routing data to the cloud. Designs prioritise energy efficiency, low latency, and a small physical footprint while sustaining the throughput demanded by real-time computer vision, natural language processing, and sensor fusion tasks. Representative architectures include neural processing units (NPUs), purpose-built ASICs, FPGAs configured for neural-network dataflows, and neuromorphic chips that mimic sparse spiking computation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:edge-ai-accelerator", "labels": [ "Edge AI Accelerator", "Hailo Accelerator" ], "is_subclass_of": [ "AI Infrastructure", "AI Hardware", "Artificial Intelligence" ], "wikilinks": [ "AI Hardware", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Autonomous Robot", "Digital Twin", "Edge Computing" ] }, { "id": "edge-ai-accelerators", "title": "Edge AI Accelerators", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Edge AI Accelerators are specialised hardware processors \u2014 including Neural Processing Units (NPUs), Tensor Processing Units (TPUs), FPGAs, and custom ASICs \u2014 designed to execute machine learning inference workloads on resource-constrained edge devices with dramatically higher throughput and energy efficiency than general-purpose CPUs. By exploiting the inherent parallelism of neural network matrix operations and using low-precision arithmetic (INT8, FP16), edge accelerators achieve 5\u2013100x performance gains over CPUs at 10\u201350x lower power per inference, enabling real-time AI on mobile, embedded, automotive, and IoT platforms without cloud connectivity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:edge-ai-accelerators", "labels": [ "Edge AI Accelerators" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "ConceptualLayer", "Digital Twin" ] }, { "id": "edge-ai-security", "title": "Edge AI Security", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The discipline of protecting machine learning systems deployed on distributed edge devices against adversarial attacks, model theft, data poisoning, physical tampering, and unauthorised access, while respecting the severe resource constraints of embedded and IoT environments. Edge AI Security integrates hardware-level protections (Trusted Execution Environments, secure boot), software hardening (model encryption, differential privacy), and algorithmic defences (certified robustness, Byzantine-robust aggregation) into defence-in-depth architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:edge-ai-security", "labels": [ "Edge AI Security" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer" ] }, { "id": "edge-ai-system", "title": "Edge AI System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Edge AI System is a distributed computing architecture that deploys trained machine learning models directly onto edge devices, sensors, and gateways at the network periphery, enabling local inference without requiring continuous cloud connectivity. Processing occurs close to the data source, achieving sub-millisecond latency, reduced bandwidth consumption, and enhanced data privacy. Edge AI Systems must fit within the memory, power, and computational constraints of embedded hardware through model compression techniques such as quantisation, pruning, and knowledge distillation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:edge-ai-system", "labels": [ "Edge AI System" ], "is_subclass_of": [ "AI Infrastructure", "Distributed Systems", "Embedded Systems" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer" ] }, { "id": "edge-ai-for-smart-cities", "title": "Edge AI for Smart Cities", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Edge AI for Smart Cities is the deployment of machine learning inference directly on distributed urban infrastructure\u2014smart cameras, IoT sensors, edge gateways, and computing nodes embedded in roads, buildings, and public spaces\u2014enabling real-time autonomous decision-making for traffic management, public safety monitoring, environmental sensing, and energy optimisation without requiring centrally routed cloud processing. This architecture achieves sub-second response latencies, preserves citizen privacy by processing video locally, and overcomes bandwidth constraints that make continuous cloud uplink of high-resolution urban sensor streams impractical.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:edge-ai-for-smart-cities", "labels": [ "Edge AI for Smart Cities" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "ConceptualLayer", "Digital Twin" ] }, { "id": "edge-ai", "title": "Edge AI", "domain": "ai", "domain_name": "Ai", "definition": "Edge AI is the deployment and execution of artificial intelligence inference workloads directly on edge devices\u2014such as smartphones, IoT sensors, surveillance cameras, and embedded systems\u2014close to the data source, rather than in centralised cloud data centres. It reduces latency, preserves data privacy, enables offline operation, and cuts bandwidth costs by processing data locally using optimised neural network models and dedicated hardware accelerators such as neural processing units (NPUs). Edge AI encompasses model compression techniques (quantisation, pruning, knowledge distillation), specialised deployment runtimes, and on-device training paradigms such as federated learning. It bridges the domains of machine learning, embedded systems, and distributed computing to enable intelligent applications at the network periphery.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:edge-ai", "labels": [ "Edge AI", "AI at the Edge" ], "is_subclass_of": [ "Edge Computing", "On-Device Machine Learning" ], "wikilinks": [] }, { "id": "edge-computing-architecture", "title": "Edge Computing Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A distributed computing paradigm that positions computational resources closer to end-user devices such as VR headsets and AR glasses, reducing latency, improving responsiveness, and enabling scalable metaverse experiences by offloading processing from centralised cloud servers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:edge-computing-architecture", "labels": [ "Edge Computing Architecture" ], "is_subclass_of": [ "Platform and Environment", "Distributed System Architecture" ], "wikilinks": [ "Low-Latency Experiences", "Distributed System Architecture", "metaverse" ] }, { "id": "edge-computing-layer", "title": "Edge Computing Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Edge Computing Layer is the stratum that places computation and storage near the source of data, away from centralised facilities. It sits above the Hardware and Network strata at the periphery and below the application and inference workloads it hosts locally. It contains edge nodes, local schedulers, and the synchronisation logic that links edge to core.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:edge-computing-layer", "labels": [ "Edge Computing Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Hardware Layer", "Network Layer", "Inference Layer", "Application Layer", "Latency", "Distributed Computing", "owl:Thing" ] }, { "id": "edge-computing-node", "title": "Edge Computing Node", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical computing resource deployed near data sources to reduce latency for immersive applications through localized processing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:edge-computing-node", "labels": [ "Edge Computing Node", "EdgeComputingNode" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Bandwidth Optimization", "Cooling System", "ETSI ARF 010", "GPU", "IEEE P2048-3", "Local Data Processing", "Low Latency Processing", "Memory Module", "Network Connectivity", "Network Interface", "Physical Housing", "Processor", "Real-time Analytics", "Storage Unit", "ComputationAndIntelligenceDomain", "ComputeLayer", "Edge Mesh Network", "Edge Network", "InfrastructureDomain", "PhysicalLayer" ] }, { "id": "edge-computing", "title": "Edge Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Edge Computing is a distributed computing paradigm that relocates computation, storage, and intelligence from centralised hyperscale data centres towards the topological extremities of the network \u2014 into base stations, on-premises micro data centres, customer-premises gateways, vehicles, industri...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:edge-computing", "labels": [ "Edge Computing", "Edge Computing Infrastructure", "Edge Computing Stack", "EdgeComputing" ], "is_subclass_of": [ "Computing and Cloud", "Network and Communication", "Distributed Computing", "Cloud Computing", "Network Function Virtualisation", "Latency-Sensitive Infrastructure", "Decentralised Architecture" ], "wikilinks": [ "3GPP", "3GPP TS 23.558 Architecture for Enabling Edge Applications", "5G", "5G Core", "5G Network", "6G", "Autonomous Vehicles", "Banbury et al. 2021 MLPerf Tiny Benchmark", "Bandwidth Efficiency", "Beutel et al. 2022 Flower Federated Learning Framework", "Bonomi et al. 2012 Fog Computing and Its Role in IoT", "Cao et al. 2023 Overview on Edge Computing Research", "Centralised Cloud Computing", "Cloudflare Workers V8 Isolates Architecture", "Cloudlet Architecture", "CNCF", "ComputeAndNetworkingDomain", "Container Image", "Container Runtime", "Containerisation" ] }, { "id": "edge-deployment", "title": "Edge Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Edge deployment is the practice of running machine learning models on or near the devices where data is generated, rather than in a centralised cloud. It reduces inference latency, preserves bandwidth, improves privacy by keeping data local and enables operation under intermittent connectivity. Edge deployment usually depends on model compression and hardware-aware optimisation to fit models within the compute, memory and power budgets of edge hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:edge-deployment", "labels": [ "Edge Deployment" ], "is_subclass_of": [ "Model Deployment", "Edge Computing", "Edge AI" ], "wikilinks": [] }, { "id": "edge-detection", "title": "Edge Detection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Edge detection is a computer vision technique that identifies points in a digital image where brightness changes sharply, marking the boundaries of objects, surfaces and textures. It typically computes image gradients and applies thresholding to produce a binary or magnitude map of edges. As a low-level feature operator it underpins higher-level tasks such as segmentation, object detection and shape analysis.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:edge-detection", "labels": [ "Edge Detection" ], "is_subclass_of": [ "Feature Detection" ], "wikilinks": [] }, { "id": "edge-gateway", "title": "Edge Gateway", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An edge gateway is a network device or software component deployed at the boundary between a local edge environment and wider network infrastructure\u2014such as the internet or a cloud backend\u2014that performs protocol translation, security enforcement, traffic routing, and local data preprocessing before forwarding selected data to upstream systems. Edge gateways aggregate data from IoT devices, sensors, and local compute nodes, filter and normalise it, enforce access control policies, and reduce bandwidth consumption by processing and compressing data at the point of collection rather than transmitting raw streams to the cloud. They are essential components of edge computing architectures in industrial IoT, smart cities, autonomous systems, and distributed XR infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:edge-gateway", "labels": [ "Edge Gateway" ], "is_subclass_of": [ "Edge Computing Node" ], "wikilinks": [] }, { "id": "edge-inference", "title": "Edge Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The execution of machine learning model inference on local edge devices rather than in centralised cloud infrastructure, close to where data is generated, enabling low-latency, privacy-preserving and bandwidth-efficient AI applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "growing", "iri": "urn:ngm:class:edge-inference", "labels": [ "Edge Inference", "Edge AI Inference", "Edge Inference Runtime" ], "is_subclass_of": [ "Edge AI", "Machine Learning Inference", "Distributed Computing" ], "wikilinks": [ "Inference Engine", "Knowledge Distillation", "Edge AI", "Edge Computing" ] }, { "id": "edge-layer", "title": "Edge Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Edge Layer is the cross-cutting stratum at the periphery of a system, where it meets external devices, users, and data sources. It sits above local hardware and network resources and below the application workloads it serves close to origin. It contains edge gateways, local caches, and the boundary logic that mediates between core and periphery.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:edge-layer", "labels": [ "Edge Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Network Layer", "Hardware Layer", "Edge Computing Layer", "Application Layer", "Content Delivery Network", "Internet of Things", "owl:Thing" ] }, { "id": "edge-mesh-network", "title": "Edge Mesh Network", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Decentralized interconnection of edge computing nodes providing dynamic load balancing, redundancy, and peer-to-peer communication for distributed workloads.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:edge-mesh-network", "labels": [ "Edge Mesh Network" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Decentralized Computation", "Distributed Coordination Protocol", "Dynamic Load Balancing", "ETSI ARF 010", "Failover Mechanism", "IEEE P2048-3", "Load Balancer", "Mesh Router", "Network Connectivity", "Redundant Processing", "Routing Protocol", "Service Discovery", "ComputationAndIntelligenceDomain", "Edge Computing Node", "Fault Tolerance", "InfrastructureDomain", "NetworkLayer", "PhysicalLayer" ] }, { "id": "edge-network", "title": "Edge Network", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Distributed set of computing nodes providing local processing close to users to improve performance, reduce latency, and optimize bandwidth for immersive applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:edge-network", "labels": [ "Edge Network" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Bandwidth Optimization", "Connectivity Fabric", "Coordination Protocol", "Distributed Processing", "ETSI ARF 010", "IEEE P2048-3", "Latency Reduction", "Load Distribution Service", "Network Management System", "Orchestration Layer", "Regional Compute", "ComputationAndIntelligenceDomain", "ComputeLayer", "Edge Computing Node", "InfrastructureDomain", "Network Infrastructure", "NetworkLayer" ] }, { "id": "edge-orchestration", "title": "Edge Orchestration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of dynamically coordinating, allocating, and balancing computational tasks between edge nodes and cloud infrastructure to optimize latency, resource utilization, and quality of experience for immersive metaverse applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:edge-orchestration", "labels": [ "Edge Orchestration" ], "is_subclass_of": [ "Platform and Environment", "Edge Computing Architecture" ], "wikilinks": [ "Adaptive Workload Distribution", "Decision Framework", "Distributed Computing Infrastructure", "Edge Computing Nodes", "ETSI ENI 008", "IEEE P2048-3", "Load Balancing System", "Low-Latency Computing", "Monitoring System", "Network Performance Metrics", "Optimized Resource Utilization", "Orchestration Policy", "Resource Availability Data", "Resource Discovery", "Resource Monitor", "Scalable Processing", "Service Level Agreements", "Task Allocation Engine", "Workload Scheduler", "Edge Computing Architecture" ] }, { "id": "edge-server", "title": "Edge Server", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A server deployed at a network point of presence geographically and topologically close to end users, which caches and serves content, terminates TLS connections, and increasingly executes application logic on behalf of a distant origin server; edge servers are the building blocks of content delivery networks and edge computing platforms, cutting round-trip latency, absorbing traffic spikes and DDoS load, and reducing bandwidth demand on origin infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:edge-server", "labels": [ "Edge Server" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [ "Network Infrastructure", "Content Delivery Network", "Origin Server", "Edge Computing" ] }, { "id": "edge-cloud-collaboration", "title": "Edge-Cloud Collaboration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Edge-Cloud Collaboration is a hybrid architecture that dynamically partitions AI workloads between resource-constrained edge devices and powerful cloud infrastructure, optimising end-to-end latency, bandwidth utilisation, energy consumption, and inference accuracy through adaptive offloading, model splitting, early exit, and cascaded inference strategies. The architecture enables edge devices to handle time-sensitive inference locally while delegating computationally intensive or contextually uncertain tasks to the cloud, achieving the complementary strengths of both deployment tiers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:edge-cloud-collaboration", "labels": [ "Edge-Cloud Collaboration" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AWS IoT Greengrass", "Azure IoT Edge", "ETSI MEC", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin" ] }, { "id": "education-metaverse", "title": "Education Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A virtual platform that provides immersive educational experiences through interconnected digital learning environments, enabling collaborative instruction, skills development, and knowledge transfer across distributed participants.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:education-metaverse", "labels": [ "Education Metaverse" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Assessment System", "Collaboration Tool", "Content Authoring System", "Educational Content", "Gesture Recognition", "IEEE 2888.1-2023", "IMS Global Learning Consortium", "Remote Education", "Skills Training", "3D Rendering Engine", "ApplicationLayer", "Avatar System", "Collaborative Learning", "CreativeMediaDomain", "Identity Management", "Immersive Learning", "Learning Analytics", "Learning Module", "Metaverse Application Platform", "Network Infrastructure" ] }, { "id": "education-technology", "title": "education technology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Education Technology (EdTech) is the systematic application of digital tools, AI-driven systems, data analytics, and immersive platforms to enhance teaching efficacy, personalise learning pathways, and streamline educational administration across formal, informal, and professional learning contexts. Modern EdTech leverages adaptive learning algorithms that adjust content difficulty and pacing to individual learner performance, natural language processing for automated tutoring and essay scoring, and extended-reality environments that create immersive simulations for fields such as medicine, engineering, and the sciences. It bridges pedagogical theory with computational infrastructure, drawing on learning science research to design interventions at scale while grappling with data-privacy obligations and algorithmic-bias risks inherent in learner-profiling systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:education-technology", "labels": [ "Education Technology" ], "is_subclass_of": [ "AI Application", "Digital Service", "Learning System" ], "wikilinks": [] }, { "id": "education-and-ai", "title": "Education and AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Education and AI (Artificial Intelligence in Education, AIED) denotes the deployment of machine-learning systems\u2014intelligent tutoring systems (ITS) such as Carnegie Learning Mathia at 600K+ US students across + districts, ALEKS at 25M+ cumulative users Knewton (acquired Wiley 2019 for ~after rais...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:education-and-ai", "labels": [ "Education and AI", "AI in Education" ], "is_subclass_of": [ "AI Application", "Applied AI", "Educational Technology", "Human-Centred AI", "Conversational AI Applications" ], "wikilinks": [ "Accessibility Provisioning", "Adaptive Learning", "AI Detection", "AI Literacy", "AI Literacy Curriculum", "AI Tutor", "ALEKS", "Anderson Corbett Koedinger 1995 Cognitive Tutors", "Applied AI", "AppliedAIDomain", "Assessment Framework", "Assessment Integrity", "Audrey Watters", "Automated Grading", "Bastani et al 2024 GenAI Harm Learning Wharton", "Bayesian Knowledge Tracing", "BBC Bitesize", "Behaviourist Drill And Practice", "Ben Williamson", "Bloom 1984 2-Sigma Problem" ] }, { "id": "education", "title": "Education", "domain": "governance", "domain_name": "Governance", "definition": "Education is the structured process of facilitating learning, knowledge and skills through teaching, study and training. It occurs in formal institutions and through informal and online means.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:education", "labels": [ "Education" ], "is_subclass_of": [ "Workspace Tools", "owl:Thing" ], "wikilinks": [ "Assistive Technology", "Accessibility", "owl:Thing" ] }, { "id": "educational-credentials", "title": "Educational Credentials", "domain": "security", "domain_name": "Security", "definition": "Educational Credentials are formal attestations of learning achievement, such as diplomas, degrees, certificates, and micro-credentials, issued by an educational authority. When expressed as verifiable credentials they can be cryptographically signed and held by the learner, allowing tamper-evident, privacy-preserving verification without contacting the issuer. They are a primary use case for decentralised, self-sovereign identity systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:educational-credentials", "labels": [ "Educational Credentials" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "educational-methodology", "title": "Educational Methodology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Educational Methodology is a structured approach to the design, delivery, and evaluation of learning experiences, encompassing pedagogical theories, instructional design models, and assessment frameworks. In spatial computing contexts it includes immersive, experiential, and simulation-based learning paradigms enabled by XR technologies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:educational-methodology", "labels": [ "Educational Methodology" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "educational-narrative", "title": "Educational Narrative", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive storytelling techniques employed in metaverse learning environments that place learners within engaging storylines, utilising narrative transportation to make complex concepts more approachable and significantly improve learning outcomes through active participation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:educational-narrative", "labels": [ "Educational Narrative" ], "is_subclass_of": [ "Content and Assets", "Immersive Learning" ], "wikilinks": [ "Personalised Learning", "Immersive Learning", "metaverse" ] }, { "id": "educational-technology", "title": "Educational Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Educational Technology (EdTech) is the systematic application of digital tools, platforms, pedagogical frameworks, and immersive media to the design, delivery, and assessment of learning experiences. In spatial-computing and metaverse contexts, it encompasses virtual classrooms, simulation-based training, adaptive learning systems, and collaborative virtual environments that extend education beyond physical and temporal constraints. The field draws on learning science, human-computer interaction, artificial intelligence, and networked infrastructure to personalise instruction and reduce barriers to access. Mature EdTech deployments span K-12, higher education, corporate training, and professional development, increasingly leveraging extended reality modalities and AI-driven tutoring systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:educational-technology", "labels": [ "Educational Technology" ], "is_subclass_of": [ "Metaverse Technology" ], "wikilinks": [ "Metaverse Technology" ] }, { "id": "effective-isotropic-radiated-power", "title": "Effective Isotropic Radiated Power", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:effective-isotropic-radiated-power", "labels": [ "Effective Isotropic Radiated Power" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "efficiency-ai", "title": "Efficiency AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence primarily to reduce costs, streamline existing workflows, and perform current tasks with fewer resources or human intervention.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:efficiency-ai", "labels": [ "Efficiency AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "eigen-layer", "title": "EigenLayer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "EigenLayer is an Ethereum-based restaking protocol that allows ETH stakers to extend their cryptoeconomic security \u2014 the stake already deposited to validate the Ethereum consensus layer \u2014 to additional decentralised services called Actively Validated Services (AVS). By opting into EigenLayer smart contracts, stakers grant the protocol the right to apply slashing conditions from multiple AVS operators simultaneously, enabling new protocols such as data availability layers, bridges, oracles, and sequencers to bootstrap economic security without deploying their own native token staking systems. EigenLayer fundamentally reuses and resells Ethereum's security budget.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:eigen-layer", "labels": [ "EigenLayer", "Eigen" ], "is_subclass_of": [ "Blockchain Protocol" ], "wikilinks": [] }, { "id": "el-salvador", "title": "El Salvador", "domain": "blockchain", "domain_name": "Blockchain", "definition": "El Salvador is a Central American nation-state that in September 2021 became the first country in the world to adopt Bitcoin as legal tender alongside the United States dollar, enacting the Bitcoin Law (Ley Bitcoin). The policy required all businesses capable of providing the technology to accept Bitcoin for goods and services, deployed the government-issued Chivo Wallet to facilitate adoption, and established the Lightning Network as the primary payment rail for low-value retail transactions. El Salvador serves in this knowledge graph as a primary real-world case study for sovereign-level cryptocurrency adoption, digital-currency monetary policy, and the intersection of blockchain infrastructure with national governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:el-salvador", "labels": [ "El Salvador" ], "is_subclass_of": [ "Governance and Regulation", "Sovereign Entity" ], "wikilinks": [ "Cryptocurrency", "Bitcoin", "Lightning Network", "Entity" ] }, { "id": "electric-actuator", "title": "Electric Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "Electric actuator converts electrical energy into controlled mechanical motion through electromagnetic forces, providing the primary means of actuation in modern robots.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:electric-actuator", "labels": [ "Electric Actuator" ], "is_subclass_of": [ "Actuation and Control", "Robot Actuator", "Electromechanical Systems" ], "wikilinks": [ "AC Induction Motors", "Brush DC Motors", "Brushless DC Motors", "Control Signal", "Dynamic Responsiveness", "Electrical Power", "Electroactive Polymers", "Electromechanical Systems", "Energy-Constrained Robots", "Force Exertion", "Gearboxes", "Linear Actuators", "Linear Servo", "Load Support", "Mechanical Coupling", "Mechanical Interface", "Motion Generation", "Motor Driver", "Motor Winding", "Output Transmission" ] }, { "id": "electric-linear-actuator", "title": "Electric Linear Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "An actuator that converts rotary electrical motor output into controlled linear displacement using a mechanical transmission such as a lead screw, rack-and-pinion, or belt drive. Electric linear actuators offer precise position control, programmable stroke lengths, and clean operation, making them preferable to pneumatic or hydraulic counterparts in many robotics and automation applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:electric-linear-actuator", "labels": [ "Electric Linear Actuator", "Linear Actuator" ], "is_subclass_of": [ "Actuation and Control", "Electric Actuator" ], "wikilinks": [ "Electric Actuator", "Robotics" ] }, { "id": "electric-motor", "title": "Electric Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "Electric motor converts electrical energy into rotational mechanical power through electromagnetic forces, forming the most widespread actuation technology in modern robotics.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:electric-motor", "labels": [ "Electric Motor" ], "is_subclass_of": [ "Actuation and Control", "Electric Actuator", "Rotating Machinery" ], "wikilinks": [ "Bearings", "Brush DC Motors", "Brushless DC Motors", "CAN Bus", "Electrical Power Supply", "Electromagnetic Theory", "Heat Dissipation", "Joint Actuation", "Linear Motors", "Mechanical Load", "Mobile Robots", "Motor Driver Electronics", "Power Electronics", "Power Terminals", "Rotating Machinery", "Rotational Motion", "Rotor", "Servo Motors", "Shaft", "Speed Control" ] }, { "id": "electric-propulsion", "title": "Electric Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "Electric propulsion uses electrical power to accelerate a propellant. It produces much higher exhaust velocity than most chemical systems, so a spacecraft can obtain a given total impulse with less propellant. The trade is low thrust: manoeuvres that take minutes with chemical propulsion may require days or months with an electric thruster.[^1]", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:electric-propulsion", "labels": [ "Electric Propulsion" ], "is_subclass_of": [ "Spacecraft Propulsion" ], "wikilinks": [] }, { "id": "electric-vehicle", "title": "Electric Vehicle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An electric vehicle (EV) is a road vehicle propelled wholly or partly by one or more electric motors, drawing energy from an onboard rechargeable battery pack rather than, or in addition to, an internal combustion engine. Its core subsystems include the battery pack, a battery management system that monitors cell health and state of charge, a power electronics inverter, and one or more electric traction motors. Electric vehicles are a central pillar of transport decarbonisation and increasingly participate in the wider electricity grid as flexible, distributed energy storage.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:electric-vehicle", "labels": [ "Electric Vehicle" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "electrical-power", "title": "Electrical Power", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Electrical Power is the rate at which electrical energy is transferred or consumed by a system, measured in watts. It is the fundamental resource that drives compute infrastructure, robotic actuators, sensors, and communications equipment. The availability, capacity, and efficiency of electrical power are primary constraints on data-centre scale and on the autonomy of mobile and robotic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:electrical-power", "labels": [ "Electrical Power" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "electricity-consumption", "title": "Electricity Consumption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Electricity Consumption is the total quantity of electrical energy used by a system over a period, typically measured in kilowatt- or terawatt-hours. For blockchains it quantifies the energy drawn by mining, validation, and node operation, and is a central metric in debates over their environmental impact. Indices such as the Cambridge Bitcoin Electricity Consumption Index estimate network-wide consumption from hardware and hashrate data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:electricity-consumption", "labels": [ "Electricity Consumption" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "electromagnetic-compatibility-testing", "title": "Electromagnetic Compatibility Testing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:electromagnetic-compatibility-testing", "labels": [ "Electromagnetic Compatibility Testing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "electromyography", "title": "Electromyography", "domain": "robotics", "domain_name": "Robotics", "definition": "Electromyography is the measurement of the electrical activity produced by skeletal muscles, captured by surface or intramuscular electrodes. The resulting signals reflect motor-unit recruitment and can be processed to infer intended movement and muscular effort. In robotics it provides a biosignal interface for prosthetic control, exoskeleton actuation and gesture recognition.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:electromyography", "labels": [ "Electromyography" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "electronic-design-automation", "title": "Electronic Design Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Electronic design automation (EDA) is the category of software tools used to design, simulate, verify, and physically lay out electronic systems such as integrated circuits and printed circuit boards. Modern chips containing tens of billions of transistors cannot be designed by hand, so EDA tools perform logic synthesis, placement and routing, timing and power analysis, and formal verification, drawing heavily on combinatorial optimisation and SAT solving. The market is dominated by Synopsys, Cadence, and Siemens EDA, making the sector a strategic chokepoint in the semiconductor supply chain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:electronic-design-automation", "labels": [ "Electronic Design Automation" ], "is_subclass_of": [ "CAD Software" ], "wikilinks": [ "CAD Software", "ASIC", "Sat Solving", "Automated Design", "Semiconductor" ] }, { "id": "electronic-health-record", "title": "Electronic Health Record", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An electronic health record (EHR) is a longitudinal digital record of a patient's health information maintained across care episodes and, increasingly, across providers. It consolidates demographics, diagnoses, medications, results and clinical notes into a structured, queryable store that supports care delivery, decision support and analytics. EHRs depend on interoperability standards and strong privacy controls to be shared safely between systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:electronic-health-record", "labels": [ "Electronic Health Record", "Electronic Health Records" ], "is_subclass_of": [ "Healthcare Records" ], "wikilinks": [] }, { "id": "electronic-signature", "title": "Electronic Signature", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Electronic Signature is any electronic data that is logically associated with a document and used to indicate the signatory's acceptance or authentication of its content, spanning simple typed names through to qualified digital signatures backed by cryptographic certificates. Legal frameworks such as eIDAS in the EU and ESIGN/UETA in the US establish the legal equivalence of electronic signatures to handwritten ones under defined conditions. Advanced forms rely on public-key infrastructure to provide non-repudiation and integrity assurances.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:electronic-signature", "labels": [ "Electronic Signature" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "electrospray-thruster", "title": "Electrospray Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:electrospray-thruster", "labels": [ "Electrospray Thruster" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "elements-project", "title": "Elements Project", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Elements Project is an open-source blockchain platform, derived from Bitcoin Core, that serves as a testbed and reference implementation for advanced features such as confidential transactions, asset issuance, and federated sidechains. Maintained primarily by Blockstream, it provides the codebase underpinning the Liquid Network and allows developers to experiment with extensions that may later be proposed for Bitcoin itself. It packages cryptographic enhancements like confidential assets and amounts into a deployable, Bitcoin-compatible client.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:elements-project", "labels": [ "Elements Project" ], "is_subclass_of": [ "Open Source Software" ], "wikilinks": [] }, { "id": "eleven-labs", "title": "ElevenLabs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ElevenLabs is a company that develops artificial intelligence software for speech synthesis and voice generation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:eleven-labs", "labels": [ "ElevenLabs" ], "is_subclass_of": [ "Generative AI", "Speech Processing", "AI Company" ], "wikilinks": [ "Deep Learning", "Text-to-Speech", "Voice Cloning", "Generative AI" ] }, { "id": "ellipsoidal-height", "title": "Ellipsoidal Height", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ellipsoidal-height", "labels": [ "Ellipsoidal Height" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "elliptic-curve-cryptography", "title": "Elliptic Curve Cryptography", "domain": "security", "domain_name": "Security", "definition": "A public-key cryptographic system based on the algebraic structure of elliptic curves over finite fields, providing strong security guarantees with shorter key lengths than RSA. ECC underpins digital signatures (ECDSA), key agreement (ECDH), and identity operations throughout blockchain infrastructure, TLS, and secure communications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:elliptic-curve-cryptography", "labels": [ "Elliptic Curve Cryptography", "Elliptic Curve Point Multiplication", "Elliptic-Curve Cryptography", "EllipticCurveCryptography" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "elliptic-curve-diffie-hellman", "title": "Elliptic Curve Diffie-Hellman", "domain": "security", "domain_name": "Security", "definition": "Elliptic Curve Diffie-Hellman (ECDH) is a key-agreement protocol in which two parties each combine their own private elliptic-curve key with the other party's public key to derive an identical shared secret, without transmitting that secret over the network. It provides the same forward-secrecy properties as classical Diffie-Hellman key exchange but with much smaller key sizes for equivalent security, because it relies on the elliptic curve discrete logarithm problem. It is the key-exchange mechanism used in TLS 1.3 and other modern TLS-based encryption to establish ephemeral session keys.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:elliptic-curve-diffie-hellman", "labels": [ "Elliptic Curve Diffie-Hellman" ], "is_subclass_of": [ "Elliptic Curve Cryptography" ], "wikilinks": [] }, { "id": "elliptic-curve-group", "title": "Elliptic Curve Group", "domain": "security", "domain_name": "Security", "definition": "An elliptic curve group is the algebraic structure formed by the points on an elliptic curve defined over a finite field, together with a geometrically defined point-addition operation that satisfies the group axioms. The apparent difficulty of the discrete logarithm problem within this group, recovering a scalar multiplier from a known point and its scalar multiple, is the hard mathematical problem that elliptic curve cryptography relies on for security. Elliptic curve groups underpin cryptographic primitives such as Schnorr signatures and Pedersen commitments used in blockchain systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:elliptic-curve-group", "labels": [ "Elliptic Curve Group" ], "is_subclass_of": [ "Elliptic Curve Cryptography" ], "wikilinks": [] }, { "id": "elliptic", "title": "Elliptic", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Elliptic is a blockchain analytics and financial crime compliance company that provides software tools to trace, analyse, and risk-score cryptocurrency transactions across multiple blockchain networks. Its technology maps on-chain flows to real-world entities \u2014 exchanges, darknet markets, ransomware wallets, and sanctioned addresses \u2014 enabling financial institutions, crypto businesses, and law enforcement to detect and prevent money laundering, terrorist financing, and sanctions evasion. Elliptic's transaction monitoring and wallet screening solutions are used by regulated virtual-asset service providers globally to satisfy AML/CFT obligations under frameworks such as the FATF Travel Rule and MiCA.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:elliptic", "labels": [ "Elliptic" ], "is_subclass_of": [ "Blockchain Analytics" ], "wikilinks": [ "Blockchain Analytics", "Anti-Money Laundering", "Transaction Monitoring" ] }, { "id": "elliptical-galaxy", "title": "Elliptical Galaxy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:elliptical-galaxy", "labels": [ "Elliptical Galaxy" ], "is_subclass_of": [ "Galaxy" ], "wikilinks": [] }, { "id": "elliptical-orbit", "title": "Elliptical Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:elliptical-orbit", "labels": [ "Elliptical Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "email-corpus-retrieval-architecture", "title": "Email Corpus Retrieval Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Email search refers to the set of techniques, architectures, and tooling used to index, query, and retrieve relevant messages from large email corpora. Modern self-hosted email search stacks combine traditional full-text indexing (BM25-based engines such as Tantivy or Xapian) with dense vector embeddings and hybrid retrieval strategies to support semantic and keyword queries at scale. LLM-based reranking and optional graph-database layers for entity-relationship queries extend precision and recall beyond keyword matching, enabling systems to handle corpora of hundreds of thousands of messages with sub-second latency.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:email-corpus-retrieval-architecture", "labels": [ "Email Corpus Retrieval Architecture", "email search" ], "is_subclass_of": [ "Search Engine" ], "wikilinks": [] }, { "id": "embankment", "title": "Embankment", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:embankment", "labels": [ "Embankment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "embedded-ai-frameworks", "title": "Embedded AI Frameworks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software infrastructure and tooling optimised for deploying and running machine learning models on resource-constrained embedded systems and edge devices, with footprints of 100 KB to 10 MB, supporting INT8/FP16 quantisation, NPU/FPGA/DSP hardware acceleration abstraction, and streamlined memory allocation to avoid heap fragmentation on microcontrollers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:embedded-ai-frameworks", "labels": [ "Embedded AI Frameworks" ], "is_subclass_of": [ "AI Infrastructure", "Edge AI", "AI Inference", "AI Model Inference Engine" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "ConceptualLayer", "Digital Twin" ] }, { "id": "embedded-system", "title": "Embedded System", "domain": "robotics", "domain_name": "Robotics", "definition": "An embedded system is a dedicated computing system designed to perform specific functions within a larger mechanical or electrical system, typically under real-time constraints. It combines a microcontroller or microprocessor with firmware and tightly coupled hardware such as sensors and actuators, often operating with limited memory, power, and processing resources. Embedded systems are ubiquitous in consumer devices, vehicles, industrial equipment, and robotics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:embedded-system", "labels": [ "Embedded System" ], "is_subclass_of": [ "Robotics", "Real-Time Control" ], "wikilinks": [] }, { "id": "embedded-systems", "title": "Embedded Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Embedded Systems are specialised computing systems designed and deployed to perform dedicated functions within larger mechanical, electronic, or cyber-physical host devices, operating under strict resource constraints, real-time deadlines, and tight hardware-software coupling. They comprise microcontrollers, microprocessors, FPGAs, ASICs, sensors, actuators, and the firmware or RTOS environments that orchestrate them. Unlike general-purpose computers, embedded systems are optimised for a specific task domain \u2014 motor control, signal processing, network communication, or safety-critical actuation \u2014 and must satisfy reliability, power-budget, thermal, and often functional-safety requirements simultaneously. They form the foundational computational layer beneath robotics, industrial automation, consumer electronics, automotive ECUs, medical devices, and the Internet of Things.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:embedded-systems", "labels": [ "Embedded Systems", "Automotive Embedded System", "Embedded Software Stack", "Embedded System", "Embedded System-on-Chip" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "embedding-layer", "title": "Embedding Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An embedding layer is a trainable component of a neural network that maps discrete tokens or categorical indices to dense, continuous vector representations. It is implemented as a lookup table whose rows are learned vectors, transforming sparse one-hot inputs into low-dimensional embeddings that capture semantic and relational structure. Embedding layers are the standard entry point for language, recommendation, and sequence models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:embedding-layer", "labels": [ "Embedding Layer" ], "is_subclass_of": [ "Neural Network", "Trainable Layer" ], "wikilinks": [] }, { "id": "embedding-model", "title": "embedding model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An embedding model is a neural network trained to project discrete or high-dimensional inputs\u2014such as tokens, sentences, images, audio segments, or knowledge-graph nodes\u2014into dense, fixed-dimensional vector representations in a continuous latent space that preserves semantic and structural relationships. The geometry of the resulting embedding space encodes similarity: semantically related inputs map to vectors with high cosine similarity, enabling downstream tasks such as semantic search, clustering, classification, and retrieval-augmented generation. Embedding models are trained using objectives such as contrastive learning, masked language modelling, or cross-modal alignment, and are evaluated on standardised benchmarks including MTEB and BEIR.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:embedding-model", "labels": [ "Embedding Model" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network", "Representation Learning" ], "wikilinks": [] }, { "id": "embedding-search", "title": "Embedding Search", "domain": "ai", "domain_name": "Ai", "definition": "Embedding search is a retrieval paradigm in which queries and documents are encoded into dense vector representations in a shared semantic space, and similarity\u2014typically measured by cosine distance or dot product\u2014is used to rank and retrieve the most relevant items. Unlike keyword-based search, embedding search captures semantic relatedness, enabling matches on meaning rather than exact lexical overlap.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:embedding-search", "labels": [ "Embedding Search" ], "is_subclass_of": [ "Semantic Search", "Information Retrieval", "Neural Information Retrieval", "Dense Retrieval" ], "wikilinks": [] }, { "id": "embedding-space", "title": "Embedding Space", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An embedding space is a continuous vector space into which discrete or high-dimensional objects \u2014 such as words, images, or graph nodes \u2014 are mapped so that geometric relationships encode semantic similarity. Learned by models during representation learning, the space arranges related items close together and supports operations like nearest-neighbour search and analogy via vector arithmetic. It is the substrate underlying semantic search, retrieval, and many downstream machine-learning tasks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:embedding-space", "labels": [ "Embedding Space" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "embedding", "title": "Embedding", "domain": "ai", "domain_name": "Ai", "definition": "An embedding is a learned mapping from discrete or high-dimensional objects\u2014such as words, sentences, images, graphs, or code\u2014into a continuous, low-dimensional vector space, such that semantically or functionally similar inputs are mapped to geometrically proximate vectors. Embeddings are fundamental to modern machine learning, enabling downstream tasks including similarity search, classification, clustering, and retrieval through compact, transferable representations. The quality of an embedding space is typically measured by how well geometric proximity in the vector space reflects semantic or functional similarity in the original domain.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:embedding", "labels": [ "Embedding", "Embedding Representations", "Hyperbolic Embedding", "Semantic Embedding" ], "is_subclass_of": [ "Representation Learning", "Feature Extraction", "Machine Learning Technique" ], "wikilinks": [] }, { "id": "embeddings", "title": "Embeddings", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Embeddings are dense, low-dimensional vector representations of discrete objects\u2014words, sentences, images, code, graphs, or arbitrary entities\u2014learned by neural networks such that geometric relationships in the vector space correspond to semantic or functional relationships between the original objects. The core property is that semantically similar inputs map to nearby vectors, enabling tasks like similarity search, clustering, and retrieval to be performed as efficient geometric operations. Embeddings are the foundational representation layer of modern deep learning, underpinning language models, recommendation systems, search engines, and retrieval-augmented generation pipelines. They transform high-dimensional sparse inputs into compact continuous representations that downstream neural architectures\u2014particularly attention-based transformers\u2014can process and reason over.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:embeddings", "labels": [ "Embeddings", "Input Embeddings", "Sentence Embeddings" ], "is_subclass_of": [ "Representation Learning", "Latent Space Model", "Deep Learning" ], "wikilinks": [] }, { "id": "embodied-ai-simulation", "title": "embodied ai simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Embodied AI Simulation is the practice of training, evaluating, and validating autonomous agents and robotic systems within physically accurate, interactive virtual environments before deployment on real hardware. These simulation platforms render high-fidelity physics (rigid and soft-body dynamics, fluid simulation, contact forces), sensor models (cameras, LiDAR, IMU, depth sensors), and procedurally generated scenes to expose agents to diverse scenarios at scale. Sim-to-real transfer methods \u2014 including domain randomisation, domain adaptation, and curriculum learning \u2014 are applied to close the fidelity gap between simulated and real-world conditions. The discipline spans household robot manipulation tasks, autonomous vehicle testing, humanoid locomotion, and multi-agent coordination, bridging artificial intelligence research with physical robotic deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:embodied-ai-simulation", "labels": [ "Embodied AI Simulation" ], "is_subclass_of": [ "Virtual Environment" ], "wikilinks": [] }, { "id": "embodied-ai", "title": "embodied ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Embodied AI is a research paradigm holding that intelligence emerges from the continuous sensorimotor interaction of an agent with its physical or simulated environment, rather than from purely symbolic or disembodied language-based reasoning. Embodied agents perceive the world through sensors \u2014 cameras, proprioceptive IMUs, force-torque sensors, tactile arrays \u2014 and act upon it through actuators, learning to navigate, manipulate objects, and cooperate via reinforcement learning or imitation learning in physics simulators. The field unifies robotics, cognitive science, and deep learning, with applications spanning household manipulation, autonomous navigation, humanoid motor control, and grounded natural language understanding. A central hypothesis is that richer, more transferable representations arise from interactive physical engagement with an environment rather than from passive statistical learning over corpora.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:embodied-ai", "labels": [ "Embodied AI", "Embodied Agent" ], "is_subclass_of": [ "AI Research Area", "Machine Learning Discipline", "Robotics" ], "wikilinks": [] }, { "id": "embodied-cognition", "title": "Embodied Cognition", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Embodied cognition is the theory that cognitive processes are deeply shaped by the body's sensorimotor systems and its interaction with the physical environment, rather than residing solely in abstract symbol manipulation in the brain. It holds that perception, action, and the body's morphology constrain and constitute thought, so understanding concepts is grounded in bodily experience. The view informs human-computer interaction and immersive design, where natural movement and spatial presence improve learning and comprehension. It contrasts with classical computationalism that treats the mind as disembodied information processing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:embodied-cognition", "labels": [ "Embodied Cognition" ], "is_subclass_of": [ "Human-Computer Interaction" ], "wikilinks": [] }, { "id": "embodied-interaction", "title": "Embodied Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Embodied interaction is an approach to human-computer interaction in which the user's body, physical movement, and spatial context are the primary means of engaging with digital systems. It draws on theories of embodied cognition, treating perception and action as inseparable, and is central to virtual, augmented, and mixed reality where users act through avatars and natural gesture. The paradigm prioritises tangible, gestural, and full-body input over abstract symbolic command.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:embodied-interaction", "labels": [ "Embodied Interaction" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "embodied-minds", "title": "Embodied Minds", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Embodied Minds denotes the theoretical and engineering position that genuine cognition, intelligence and meaning-making cannot be reduced to disembodied symbol manipulation or text-token prediction but instead arises from the dynamic, sensorimotor coupling of a physical body with a structured env...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:embodied-minds", "labels": [ "Embodied Minds" ], "is_subclass_of": [ "AI Research Area", "Cognitive Science", "Artificial Intelligence", "Philosophy of Mind", "Robotics", "Situated Cognition" ], "wikilinks": [ "4E Cognition", "ActionLayer", "Active Inference", "Actuators", "Affordance", "AlgorithmLayer", "Assistive Robotics", "Autonomous Vehicles", "Bayesian Brain Hypothesis", "Behaviour-Based Robotics", "Bender Koller 2020 Climbing Towards NLU", "Black et al 2024 Pi0 VLA Flow Model", "Body Schema", "Brohan et al 2023 RT-2", "Brooks 1986 Subsumption Architecture", "Brooks 1991 Intelligence Without Representation", "Cartesian Dualism", "Chi et al 2023 Diffusion Policy", "Clark 1997 Being There", "Clark 2008 Supersizing the Mind" ] }, { "id": "embodied-presence", "title": "Embodied Presence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Embodied presence is the sense of being physically located within a virtual or remote environment, often through an avatar or robotic proxy. It is studied in virtual reality and telepresence.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:embodied-presence", "labels": [ "Embodied Presence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Avatar", "Virtual World", "Embodied AI", "Telepresence", "https://en.wikipedia.org/wiki/Immersion_(virtual_reality)", "https://en.wikipedia.org/wiki/Telepresence" ] }, { "id": "embodiment", "title": "Embodiment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The sense of inhabiting a body \u2014 in immersive systems, the subjective experience that a virtual body or avatar is one's own, arising from synchronised visuomotor and multisensory feedback and decomposable into self-location, agency, and body ownership; a foundational construct for presence, social VR, telepresence, and embodied theories of cognition.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:embodiment", "labels": [ "Embodiment" ], "is_subclass_of": [ "Embodied Cognition" ], "wikilinks": [ "Embodied Cognition", "Presence", "Avatar", "Proprioception" ] }, { "id": "emergence", "title": "Emergence", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Emergence is the cross-disciplinary phenomenon in which qualitatively novel structures, behaviours, capabilities or properties arise at a higher level of organisation in a system as a non-trivial collective consequence of the interactions among its lower-level constituents, where the macro-level ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:emergence", "labels": [ "Emergence" ], "is_subclass_of": [ "Computing and Cloud", "Complex Systems Concept", "Cross-Scale Phenomenon", "Collective Behaviour", "Non-Reductive Property", "Organisational Pattern" ], "wikilinks": [ "Adaptive Behaviour", "Agent-Based Modelling", "Agent-Based Models", "Aggregative Property", "AI Capability Forecasting", "Albantakis et al 2023 IIT 4.0", "Alexander 1920 Space Time and Deity", "Anderson 1972 More is Different", "Axelrod 1984 Evolution of Cooperation", "Bak 1996 How Nature Works", "Bak Tang Wiesenfeld 1987 Self-Organized Criticality", "Barabasi Albert 1999 Emergence of Scaling", "Behavioural Economics", "Bettencourt et al 2007 Urban Scaling", "Broad 1925 The Mind and Its Place in Nature", "Brown et al 2020 GPT-3 In-Context Learning", "Cellular Automata", "Chalmers 2006 Strong and Weak Emergence", "Chaos Theory", "CognitiveScienceDomain" ] }, { "id": "emergent-behavior", "title": "Emergent Behavior", "domain": "ai", "domain_name": "Ai", "definition": "Emergent Behavior is complex, system-level behaviour that arises from the interactions of many simpler components or agents and is not explicitly programmed into any individual part. In AI-driven game agents and open-world simulations it produces lifelike, unscripted dynamics from local rules and agent decisions. Emergence is valued for richness and replayability but can be hard to predict, test, and control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:emergent-behavior", "labels": [ "Emergent Behavior", "Emergent Behaviour" ], "is_subclass_of": [ "AI Agent System", "Complex Adaptive Systems", "Complex Systems Science" ], "wikilinks": [] }, { "id": "emergent-capabilities", "title": "Emergent Capabilities", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Emergent Capabilities are abilities that appear in large language models at scale but are absent or near-random in smaller models, seemingly arising abruptly as parameters, data, or compute increase. Examples include multi-step reasoning, in-context learning, and instruction following. Their unpredictability complicates capability forecasting and is central to debates about scaling and AI safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:emergent-capabilities", "labels": [ "Emergent Capabilities", "Emergent Abilities", "Emergent Capability" ], "is_subclass_of": [ "Large Language Models", "Emergent Behavior", "Scaling Laws" ], "wikilinks": [] }, { "id": "emergent-gameplay", "title": "Emergent Gameplay", "domain": "ai", "domain_name": "Ai", "definition": "Emergent Gameplay refers to complex, often unanticipated play behaviours, strategies, and narratives that arise spontaneously from the interaction of relatively simple, well-defined game rules, systems, and player agency, rather than being explicitly scripted or designed by the game's creators. It occurs when the combinatorial space of game mechanics, physics simulations, AI agent behaviours, and player decisions produces outcomes that exceed the designers' explicit intentions, creating novel experiences from the bottom up. Emergent gameplay is a hallmark of open-world, sandbox, and simulation genres, and is increasingly studied in the context of AI game agents and procedural content generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:emergent-gameplay", "labels": [ "Emergent Gameplay" ], "is_subclass_of": [ "Game Mechanics", "Complex Adaptive Systems", "Systems Theory", "Simulation" ], "wikilinks": [] }, { "id": "emission-factors", "title": "Emission Factors", "domain": "governance", "domain_name": "Governance", "definition": "Emission Factors are coefficients that quantify the average amount of a pollutant\u2014typically greenhouse gases expressed in CO2-equivalent\u2014released per unit of an activity, such as per kilowatt-hour of electricity consumed or per kilometre driven by a particular vehicle class. They are derived from empirical measurement campaigns and modelling studies, and published by bodies such as the IPCC, the IEA, and national environmental agencies. Emission factors are the fundamental input to carbon accounting frameworks such as the GHG Protocol, enabling organisations to calculate their Scope 1, 2, and 3 inventories from activity data. Regular revision of these factors reflects technological change and regional grid decarbonisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:emission-factors", "labels": [ "Emission Factors", "Emission Factor Databases" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "emission-schedule", "title": "Emission Schedule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Emission Schedule is the predetermined, protocol-encoded timeline that specifies the rate at which new tokens are minted and distributed to participants over the lifetime of a blockchain network. It governs how the total supply of a token expands from genesis toward any eventual supply cap or steady-state inflation rate, directly shaping the economic incentives for validators, miners, and stakers. Well-designed emission schedules balance early bootstrapping of network security with long-term sustainability once adoption is established.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:emission-schedule", "labels": [ "Emission Schedule", "Predictable Issuance Schedule", "Token Emission Schedule" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "emissions-reporting", "title": "Emissions Reporting", "domain": "governance", "domain_name": "Governance", "definition": "Emissions reporting is the structured disclosure of an organisation's greenhouse gas output across defined scopes and reporting periods, typically following standards such as the GHG Protocol. It converts underlying carbon accounting data into public or regulatory filings that stakeholders and auditors can verify. Frameworks such as the Crypto Climate Accord require signatories to publish emissions reporting to demonstrate progress toward stated climate commitments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:emissions-reporting", "labels": [ "Emissions Reporting" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "emissions-trading-scheme", "title": "Emissions Trading Scheme", "domain": "finance", "domain_name": "Finance", "definition": "An emissions trading scheme is a market-based policy instrument that caps the total quantity of greenhouse gases that regulated entities may emit and allows them to trade emission allowances. By placing a price on carbon through tradable permits it incentivises reductions where they are cheapest to achieve. Schemes such as the EU ETS and the UK ETS operate on a cap-and-trade basis and form a central pillar of climate policy and carbon markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:emissions-trading-scheme", "labels": [ "Emissions Trading Scheme" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [] }, { "id": "emissivity", "title": "Emissivity", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:emissivity", "labels": [ "Emissivity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "emotion-aware-interaction", "title": "Emotion Aware Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Affective computing technologies integrated into metaverse systems that identify users emotional cues through facial expressions, body language, and voice tones, enabling context-aware, meaningful interactions that enhance genuine human-like experiences in virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:emotion-aware-interaction", "labels": [ "Emotion Aware Interaction", "Emotion-Aware Interaction" ], "is_subclass_of": [ "Interaction Technology", "Human-Computer Interaction" ], "wikilinks": [ "Emotionally Intelligent Metaverse", "Human-Computer Interaction", "metaverse" ] }, { "id": "emotion-recognition", "title": "Emotion Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Emotion recognition is the computational task of inferring a person's affective state from signals such as facial expressions, voice prosody, language and physiological measurements. It draws on affective computing and machine learning to classify or estimate emotions along discrete categories or continuous dimensions. The technology raises significant accuracy, bias and privacy concerns that constrain responsible deployment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:emotion-recognition", "labels": [ "Emotion Recognition" ], "is_subclass_of": [ "Affective Computing" ], "wikilinks": [] }, { "id": "emotional-analytics-engine", "title": "Emotional Analytics Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AI module analyzing affective states from facial, voice, or physiological data to enable adaptive agent responses and affective computing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:emotional-analytics-engine", "labels": [ "Emotional Analytics Engine" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Adaptive User Interface", "Affective Computing System", "Affective State Predictor", "Biometric Sensors", "Facial Expression Analyzer", "IEEE Affective Computing 2023", "Mental Health Monitoring", "Neural Networks", "Physiological Sensor Processor", "Privacy Protection", "Sensor Data Stream", "Sentiment Classification Model", "Speech Processing", "User Experience Analytics Platform", "Voice Emotion Detector", "ComputationAndIntelligenceDomain", "Computer Vision", "Emotion-Aware Interaction", "Machine Learning Model", "MiddlewareLayer" ] }, { "id": "emotional-immersion", "title": "Emotional Immersion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Psychological absorption and empathetic engagement experienced during virtual interaction, characterized by affective resonance with virtual content and reduced awareness of physical surroundings.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:emotional-immersion", "labels": [ "Emotional Immersion" ], "is_subclass_of": [ "Platform and Environment", "Immersion" ], "wikilinks": [ "ACM", "Affective Design", "Audio Design", "Emotional Resonance", "Empathetic Connection", "Interaction Design", "PresentationLayer", "Sensory Feedback", "Story Engagement", "User Engagement", "Visual Fidelity", "ApplicationLayer", "Immersion", "InteractionDomain", "Narrative Content" ] }, { "id": "emotional-intelligence", "title": "Emotional Intelligence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Emotional Intelligence in AI refers to the capacity of artificial systems to recognise, interpret, and respond appropriately to human emotional states, expressed through text, voice, facial expression, or physiological signals. It extends classical AI with affective computing capabilities, enabling machines to calibrate their outputs based on a user's emotional context. Applications span conversational agents, digital humans, therapeutic tools, and immersive experience design where user engagement depends on emotionally-resonant interaction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:emotional-intelligence", "labels": [ "Emotional Intelligence" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Blockchain" ] }, { "id": "empathetic-ai", "title": "Empathetic AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Empathetic AI is a subfield of Affective Computing and Human-Computer Interaction concerned with computational systems that perceive, model, and respond to human affective states \u2014 including emotions, sentiment, stress, mood, and social cues \u2014 to produce contextually appropriate, emotiona...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:empathetic-ai", "labels": [ "Empathetic AI" ], "is_subclass_of": [ "AI Application", "Affective Computing", "Cognitive AI", "Human-Computer Interaction", "Conversational AI", "Multimodal AI" ], "wikilinks": [ "Adaptive Learning Systems", "Affective Computing", "AffectiveComputingDomain", "Affective Dialogue Management", "Affective Language Models", "AIApplicationDomain", "Attention Mechanisms", "Baltrusaitis et al. 2018 OpenFace 2.0 IEEE FG", "Breazeal 2003 Sociable Robots Robotics Autonomous Systems", "Cassell 2000 Embodied Conversational Agents MIT Press", "Categorical Emotion Classification", "Cold AI Interaction", "Companion AI", "Cowen & Keltner 2017 27 Distinct Emotion Categories PNAS", "Crisis Intervention", "D'Mello & Graesser 2012 Affective States Complex Learning", "DialogueLayer", "Dimensional Emotion Models", "Ekman & Friesen 1978 Facial Action Coding System", "Elder Care Robots" ] }, { "id": "empirical-experimental-design", "title": "Empirical Experimental Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An experiment is a structured empirical procedure designed to test a hypothesis or evaluate a system under controlled conditions, producing observable and repeatable results. In the context of AI and software engineering it encompasses both scientific investigations (measuring model behaviour, benchmarking performance) and engineering trials (A/B tests, canary deployments, feature flags) aimed at generating evidence to guide design decisions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:empirical-experimental-design", "labels": [ "Empirical Experimental Design", "Experiment Reproducibility", "experiment" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "back_hashcash-denial_2002" ] }, { "id": "employee-incentives", "title": "Employee Incentives", "domain": "governance", "domain_name": "Governance", "definition": "Structures and mechanisms, such as leaderboards, rewards, or recognition titles, designed to motivate and guide the behavior of workforce members toward organizational goals.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:employee-incentives", "labels": [ "Employee Incentives" ], "is_subclass_of": [ "Venture Capital" ], "wikilinks": [] }, { "id": "employment-contract-restructuring-analysis", "title": "Employment Contract Restructuring Analysis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Role reorg denotes an organisational restructuring process in which employment or contractual arrangements are renegotiated, typically comparing permanent employment contracts with temporary or agency-based assignments across dimensions such as salary, holiday entitlement, pension eligibility, notice periods, and additional benefits. Such analysis underpins workforce planning decisions in academic and research institutions, balancing income stability against contractual flexibility and total compensation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:employment-contract-restructuring-analysis", "labels": [ "Employment Contract Restructuring Analysis", "Role reorg" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "employment-social-contract-under-automation", "title": "Employment Social Contract Under Automation", "domain": "governance", "domain_name": "Governance", "definition": "The evolving set of expectations, obligations and entitlements linking workers, employers and the state around employment, income security and social protection, examined specifically under conditions of technological displacement driven by automation, robotics and artificial intelligence. The concept encompasses both descriptive analysis of how existing labour-market settlements are stressed by AI-driven task substitution and normative debates about how governments, firms and civil society should renegotiate the terms of work, welfare and the distribution of productivity gains. It spans labour economics, political philosophy and public policy, treating technological change as a structural challenge to long-standing institutions such as collective bargaining, unemployment insurance and occupational licensing.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:employment-social-contract-under-automation", "labels": [ "Employment Social Contract Under Automation", "Social contract and jobs" ], "is_subclass_of": [ "AI Governance and Ethics", "Labour Market Policy" ], "wikilinks": [ "Governance", "AI Adoption", "Artificial Intelligence", "Governance Domain" ] }, { "id": "encoder-decoder-architecture", "title": "Encoder Decoder Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network design pattern comprising an encoder that compresses an input sequence into a latent representation and a decoder that generates an output sequence from that representation, enabling sequence-to-sequence mappings for tasks such as machine translation, summarisation, and image captioning. The architecture underpins the original Transformer (Vaswani et al., 2017) and models such as T5 and BART.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:encoder-decoder-architecture", "labels": [ "Encoder Decoder Architecture", "Encoder-Decoder", "Encoder-Decoder Architecture" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "encoder-decoder", "title": "Encoder-Decoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The encoder-decoder is a neural network architecture pattern in which an encoder maps a variable-length input into an intermediate representation and a decoder generates a variable-length output conditioned on that representation. It underpins sequence-to-sequence learning for machine translation, speech recognition, and speech synthesis, and appears in both recurrent and transformer instantiations, usually augmented with an attention mechanism so the decoder can consult the full encoded input at every generation step rather than a single fixed-size vector.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:encoder-decoder", "labels": [ "Encoder-Decoder", "Encoder Decoder" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Neural Network", "Attention Mechanism", "Machine Translation", "Transformer", "Sequence To Sequence Learning" ] }, { "id": "encoder", "title": "Encoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The component in an encoder-decoder architecture that processes an input sequence and produces contextualised representations via stacked self-attention and position-wise feed-forward layers; used as the representation-learning backbone in transformer models such as BERT and T5.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:encoder", "labels": [ "Encoder", "Encoder Network", "Observation Encoder", "Vision Encoder" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "artificial intelligence", "MetaverseDomain" ] }, { "id": "encrypted-storage", "title": "Encrypted Storage", "domain": "security", "domain_name": "Security", "definition": "Encrypted Storage is the protection of data at rest by transforming it with cryptographic algorithms so that it is unreadable without the corresponding decryption keys. It can operate at the disk, filesystem, database, or object level, and is essential for confidentiality of sensitive data such as biometric templates. Proper key management, including secure key generation, rotation, and access control, is critical to its effectiveness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:encrypted-storage", "labels": [ "Encrypted Storage" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "encryption-protocol", "title": "Encryption Protocol", "domain": "security", "domain_name": "Security", "definition": "An Encryption Protocol is a defined set of rules and message exchanges that uses cryptographic algorithms to establish keys and protect the confidentiality and integrity of data in transit or at rest. It specifies handshakes, cipher negotiation, key exchange, and authentication so that interoperating parties can communicate securely. TLS 1.3 is a widely deployed example securing internet traffic.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:encryption-protocol", "labels": [ "Encryption Protocol" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "encryption-scheme", "title": "Encryption Scheme", "domain": "security", "domain_name": "Security", "definition": "An Encryption Scheme is a cryptographic construction comprising key-generation, encryption, and decryption algorithms that together transform plaintext into ciphertext and back under a key. Schemes are categorised as symmetric (shared key, e.g. AES) or asymmetric (public/private key, e.g. RSA, ECC), and are evaluated by their security definitions such as semantic security. They are the building blocks of confidentiality in cryptographic protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:encryption-scheme", "labels": [ "Encryption Scheme", "BFV Scheme", "BGV Scheme" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "encryption-service", "title": "Encryption Service", "domain": "security", "domain_name": "Security", "definition": "An encryption service is a system or component that provides cryptographic protection of data, typically offering key management and encryption operations through an interface. It supports confidentiality of stored and transmitted data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:encryption-service", "labels": [ "Encryption Service" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Symmetric Encryption", "Privacy", "Cloud Computing", "Cryptography", "https://csrc.nist.gov/glossary/term/encryption", "https://www.nist.gov/cryptography" ] }, { "id": "encryption", "title": "encryption", "domain": "security", "domain_name": "Security", "definition": "Encryption is the cryptographic process of transforming plaintext data into ciphertext using a defined algorithm and secret key, rendering the data unintelligible to any party that does not possess the corresponding decryption key, thereby ensuring confidentiality. Symmetric schemes such as AES-GCM use a single shared secret for both encryption and decryption, providing authenticated encryption with associated data (AEAD) in a single pass; asymmetric schemes such as RSA-OAEP and ECDH use mathematically linked key pairs where the public key encrypts and the private key decrypts. Hybrid constructions combine both paradigms \u2014 using asymmetric key exchange to establish a shared session key and then symmetric ciphers for bulk data \u2014 as exemplified by TLS 1.3. Encryption is foundational to data-at-rest protection, data-in-transit security, end-to-end encrypted messaging, confidential computing, and post-quantum cryptography.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:encryption", "labels": [ "Encryption", "Hybrid Encryption" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "end-effector", "title": "End Effector", "domain": "robotics", "domain_name": "Robotics", "definition": "End Effector - A task-specific tool or manipulator mounted at the Robot Wrist that physically interacts with the environment (gripper, welder, drill, camera), translating robotic control commands into productive work through mechanical, electrical, or pneumatic actuation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:end-effector", "labels": [ "End Effector", "End Effectors", "End-Effector", "End-Effector Control", "End-Effector Platform", "End-Effector Pose", "Robot End-Effector" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Mechanical Component" ], "wikilinks": [ "Assembly Operations", "Control Interface", "Force Feedback", "Manipulation System", "Material Handling", "Mechanical Interface", "Precision Manufacturing", "Robot Arm", "Robot Wrist", "Mechanical Component", "Robotics", "RoboticsDomain", "Smart Contract" ] }, { "id": "end-of-life-disposal", "title": "End-of-life Disposal", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:end-of-life-disposal", "labels": [ "End-of-life Disposal" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "end-to-end-encrypted-collaboration", "title": "End-to-End Encrypted Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "End-to-end encrypted collaboration refers to real-time co-editing and communication systems in which content is encrypted on the sender's device and can only be decrypted by intended recipients, with no plaintext accessible to intermediary servers. Implementing E2EE in collaborative contexts requires careful key management, since features like conflict resolution, server-side search, and access control must operate on ciphertext or be handled entirely client-side. Protocols such as Matrix's Megolm and systems built on the Signal Protocol provide practical frameworks for achieving this in group collaboration scenarios.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:end-to-end-encrypted-collaboration", "labels": [ "End-to-End Encrypted Collaboration" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "end-to-end-encryption", "title": "End-to-End Encryption", "domain": "security", "domain_name": "Security", "definition": "End-to-end encryption (E2EE) is a communication security model in which data is encrypted on the sender's device and can only be decrypted by the intended recipient's device, ensuring that no intermediate party\u2014including service providers, network operators, or infrastructure owners\u2014can access the plaintext content. It combines asymmetric key exchange with symmetric session encryption to provide confidentiality, integrity, and authenticity without trusting intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:end-to-end-encryption", "labels": [ "End-to-End Encryption" ], "is_subclass_of": [ "Encryption" ], "wikilinks": [] }, { "id": "end-to-end-learning", "title": "End-to-End Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A training paradigm in which a single differentiable model is optimised to map raw inputs directly to final task outputs \u2014 pixels to steering angles, waveforms to transcripts, text to text \u2014 with all intermediate representations learned jointly by gradient descent rather than specified as hand-engineered features or separately built pipeline stages; it trades the modularity, interpretability, and testability of engineered pipelines for the ability to discover representations that human designers would not, and dominates wherever data and compute are abundant.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:end-to-end-learning", "labels": [ "End-to-End Learning" ], "is_subclass_of": [ "Deep Learning" ], "wikilinks": [ "Deep Learning", "Feature Engineering", "Representation Learning", "Perception System" ] }, { "id": "endorsement-policy", "title": "Endorsement Policy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Endorsement Policy is a rule, used in permissioned blockchains such as Hyperledger Fabric, that specifies which organisations' peers must execute and cryptographically sign a transaction proposal before it is considered valid. It encodes the trust and approval requirements for a smart contract, for example requiring signatures from a majority or a named set of consortium members. Endorsement policies are central to multi-party governance of enterprise distributed ledgers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:endorsement-policy", "labels": [ "Endorsement Policy" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "endpoint-detection-and-response", "title": "Endpoint Detection and Response", "domain": "security", "domain_name": "Security", "definition": "Endpoint Detection and Response (EDR) is a cybersecurity technology that continuously records endpoint activity, detects suspicious or malicious behaviour, and provides investigation and response capabilities on hosts such as laptops, servers and workstations. Lightweight agents stream rich telemetry to an analytics backend that applies behavioural detection, threat intelligence and anomaly models to surface and contain threats that bypass preventive controls. EDR enables analysts to investigate incidents, isolate compromised hosts and remediate threats, and increasingly feeds extended detection and response and managed services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:endpoint-detection-and-response", "labels": [ "Endpoint Detection and Response" ], "is_subclass_of": [ "Threat Detection" ], "wikilinks": [] }, { "id": "endpoint-security", "title": "Endpoint Security", "domain": "security", "domain_name": "Security", "definition": "The practice of protecting end-user and edge devices \u2014 laptops, desktops, servers, mobiles, and increasingly IoT hardware \u2014 from compromise, combining preventive controls such as anti-malware, disk encryption, patching, and application allow-listing with detective and responsive capabilities delivered by endpoint detection and response (EDR) agents. As perimeter defences have weakened under remote work and cloud adoption, the endpoint has become the primary battleground and telemetry source for enterprise defence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:endpoint-security", "labels": [ "Endpoint Security" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "Cybersecurity", "Defense In Depth", "Network Security" ] }, { "id": "energy-attribute-certificates", "title": "Energy Attribute Certificates", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Energy Attribute Certificates (EACs) are tradable instruments that each represent proof that one megawatt-hour of electricity was generated from a specific source, typically renewable. They decouple the environmental attributes of energy from the underlying physical electricity, allowing those attributes to be sold, retired, or tracked independently. In blockchain contexts they are increasingly tokenised to provide auditable, fraud-resistant accounting of clean-energy consumption and to offset network energy use.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:energy-attribute-certificates", "labels": [ "Energy Attribute Certificates", "EnergyTag Granular Certificate", "EnergyTag Granular Certificate Standard" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "energy-consumption", "title": "Energy Consumption", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Energy consumption is the total quantity of energy drawn by a system, process, or device over a defined time period, typically expressed in kilowatt-hours (kWh) or joules. In computing and digital infrastructure contexts it encompasses the electrical power used by processors, memory, networking equipment, and cooling systems. It is a foundational metric in evaluating the environmental footprint, operational cost, and sustainability compliance of data centres, blockchain networks, AI training pipelines, and distributed systems. Minimising energy consumption without sacrificing throughput or reliability is a core design constraint across hardware architecture, consensus mechanisms, and large-scale deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:energy-consumption", "labels": [ "Energy Consumption", "AI Energy Consumption", "Energy Consumption Monitoring", "EnergyConsumptionMeasurement" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Proof of Stake", "Sustainability" ] }, { "id": "energy-efficiency", "title": "Energy Efficiency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Energy efficiency is the property of a system that achieves a given output or function while consuming the least possible energy. In computing, robotics, and control systems it is measured as useful work per unit of energy, and is improved through better algorithms, hardware, scheduling, and motion optimisation. It is a core sustainability objective because reductions in energy draw lower operating cost and carbon footprint without sacrificing performance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:energy-efficiency", "labels": [ "Energy Efficiency", "Energy Efficiency Calculation" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "energy-infrastructure", "title": "Energy Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The physical systems and facilities, including power plants, transmission lines, and grids, required to generate, transmit, and distribute electricity.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:energy-infrastructure", "labels": [ "Energy Infrastructure" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "energy-management", "title": "Energy Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Energy management is the systematic process of monitoring, controlling, and optimising the generation, distribution, and consumption of energy in facilities, industrial processes, or grid infrastructure to minimise cost, reduce waste, and meet sustainability and regulatory targets. It encompasses metering infrastructure, energy auditing, demand-side management, load forecasting, and integration of renewable generation, governed by standards such as ISO 50001. Modern energy management systems (EMS) use real-time telemetry from IoT sensors, predictive analytics, and automated control loops to balance supply and demand dynamically.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:energy-management", "labels": [ "Energy Management", "Energy Management System", "EnergyMix" ], "is_subclass_of": [ "Power Management" ], "wikilinks": [] }, { "id": "energy-optimization", "title": "Energy Optimisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The systematic adjustment of how a building, facility, or physical system consumes, stores, and sources energy in order to minimise cost, waste, and carbon emissions while maintaining required service levels such as comfort, output, and safety. Energy optimisation combines metered consumption data, simulation of thermal and electrical behaviour, and control strategies\u2014from schedule tuning to model-predictive control\u2014often driven by a digital twin that lets operators test interventions virtually before applying them to the real asset.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:energy-optimization", "labels": [ "Energy Optimisation", "Energy Optimization" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [ "Energy Management", "Construction Digital Twin", "Energy Efficiency" ] }, { "id": "energy-policy", "title": "Energy Policy", "domain": "governance", "domain_name": "Governance", "definition": "The set of government decisions and regulations that govern the production, distribution, and consumption of energy resources.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:energy-policy", "labels": [ "Energy Policy" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "energy-storage", "title": "Energy Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Energy storage is the capture of energy produced at one time for use at a later time, balancing supply and demand across electrical, mechanical, thermal, and chemical media. In power systems it smooths the variability of renewable generation, provides grid services such as frequency regulation, and improves resilience. Technologies range from electrochemical batteries and pumped hydro to thermal and hydrogen storage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:energy-storage", "labels": [ "Energy Storage" ], "is_subclass_of": [ "Power Systems" ], "wikilinks": [] }, { "id": "energy-and-power", "title": "Energy and Power", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Energy and Power, within the Infrastructure domain, denotes the integrated sociotechnical system governing the generation, transmission, distribution, storage, and consumption of electrical energy in contexts directly relevant to AI Data Centres, Bitcoin Mining operations, and the bro...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:energy-and-power", "labels": [ "Energy and Power" ], "is_subclass_of": [ "Computing and Cloud", "Software Engineering", "Grid Infrastructure", "Infrastructure", "Sustainability", "Critical Infrastructure", "Carbon Footprint Measurement" ], "wikilinks": [ "AI Data Centres", "Balancing Services", "Battery Energy Storage", "Battery Storage", "Behind-the-Meter Generation", "Capacity Market", "Capacity Markets", "Carbon Intensity", "Carbon Offsetting", "Climate Policy", "ContractualLayer", "Critical Infrastructure", "Data Centre", "Demand Forecasting", "Demand Response", "DigitalEconomyDomain", "Electrification", "Energy Storage", "EPRI Standards", "FERC Regulations" ] }, { "id": "enforcement-action", "title": "Enforcement Action", "domain": "governance", "domain_name": "Governance", "definition": "An enforcement action is a formal measure taken by a regulatory or supervisory authority against an entity that has violated laws, rules, or licence conditions. It can include fines, cease-and-desist orders, licence revocation, restitution to affected parties, or referral for prosecution. Enforcement actions are the operative mechanism by which consumer-protection and market-conduct regulations are made binding.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enforcement-action", "labels": [ "Enforcement Action", "SEC Enforcement Action" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "enforcement-mechanism", "title": "Enforcement Mechanism", "domain": "governance", "domain_name": "Governance", "definition": "An enforcement mechanism is any technical, legal, or procedural instrument through which rules, policies, or contractual obligations are made to take effect and violations are detected and sanctioned. In digital systems, enforcement mechanisms range from smart contracts that execute conditions automatically to access control systems that block unauthorised actions. In regulatory contexts they include fines, licence revocations, and injunctions applied by supervisory authorities. Effective enforcement mechanisms are characterised by their ability to detect non-compliance, impose credible consequences, and do so at a cost proportionate to the harm prevented.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:enforcement-mechanism", "labels": [ "Enforcement Mechanism", "Disclosure Enforcement Mechanism", "Enforcement Mechanisms", "Enforcement Powers" ], "is_subclass_of": [ "Compliance Control" ], "wikilinks": [] }, { "id": "engagement-optimisation", "title": "Engagement Optimisation", "domain": "ai", "domain_name": "Ai", "definition": "Engagement optimisation is the use of machine-learning models and feedback loops to maximise a user's interaction with a system, measured by signals such as session length, return frequency, and response rate. It typically learns from behavioural data to personalise content, timing, and prompts. While it can improve usefulness, it raises ethical concerns when optimisation targets attention or dependency rather than user benefit.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:engagement-optimisation", "labels": [ "Engagement Optimisation" ], "is_subclass_of": [ "AI Technique", "Machine Learning", "Reinforcement Learning", "Personalised Interaction", "Recommender System", "Bayesian Optimisation" ], "wikilinks": [] }, { "id": "enhanced-due-diligence", "title": "Enhanced Due Diligence", "domain": "finance", "domain_name": "Finance", "definition": "Enhanced due diligence (EDD) is a heightened level of customer scrutiny applied by regulated institutions to relationships and transactions that present elevated money-laundering, terrorist-financing, or sanctions risk. Going beyond standard customer due diligence, it requires deeper verification of identity and beneficial ownership, investigation of source of funds and wealth, and ongoing intensified monitoring. EDD is mandated for higher-risk categories such as politically exposed persons, customers in high-risk jurisdictions, and complex or unusually large transactions. It is a cornerstone of the risk-based approach embedded in anti-money-laundering and know-your-customer regulatory regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:enhanced-due-diligence", "labels": [ "Enhanced Due Diligence" ], "is_subclass_of": [ "Customer Due Diligence" ], "wikilinks": [] }, { "id": "enhanced-vegetation-index", "title": "Enhanced Vegetation Index", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:enhanced-vegetation-index", "labels": [ "Enhanced Vegetation Index" ], "is_subclass_of": [ "Vegetation Index" ], "wikilinks": [] }, { "id": "ensemble-collaborative-intelligence-principle", "title": "Ensemble Collaborative Intelligence Principle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Two Heads Are Better Than One is the principle that collaborative or ensemble approaches to problem-solving outperform individual effort, applied in AI contexts to multi-agent systems, ensemble methods, and human-in-the-loop architectures. It underpins debate-based reasoning, peer-review agent patterns, and consensus mechanisms where diverse model outputs are aggregated to improve accuracy and reduce error.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ensemble-collaborative-intelligence-principle", "labels": [ "Ensemble Collaborative Intelligence Principle", "Two Heads Are Better Than One" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "ensemble-methods", "title": "Ensemble Methods", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Machine learning techniques that combine multiple base models (weak learners) to produce a stronger, more accurate predictor by aggregating their predictions, thereby reducing variance, bias, or both, and achieving better generalisation than any single model alone.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ensemble-methods", "labels": [ "Ensemble Methods", "Ensemble Learning", "Ensemble Method" ], "is_subclass_of": [ "Machine Learning", "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Generalization", "Gradient Boosting", "Model Combination", "Random Forest", "Blockchain", "Digital Twin", "Machine Learning" ] }, { "id": "enterprise-ai-adoption", "title": "Enterprise AI Adoption", "domain": "ai", "domain_name": "Ai", "definition": "Enterprise AI adoption is the organisational process of integrating artificial-intelligence capabilities into business operations, products, and decision-making at scale. It spans strategy, data readiness, platform selection, governance, change management, and measurement of return on investment. Successful adoption depends as much on workflow redesign, skills, and executive sponsorship as on the underlying models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:enterprise-ai-adoption", "labels": [ "Enterprise AI Adoption" ], "is_subclass_of": [ "AI Application", "Technology Adoption", "Digital Transformation", "Organisational Change Management" ], "wikilinks": [] }, { "id": "enterprise-ai-cost-management", "title": "Enterprise AI Cost Management", "domain": "economics", "domain_name": "Economics", "definition": "The strategic and operational practice of monitoring, budgeting, and optimizing the financial expenditure associated with deploying and scaling artificial intelligence workloads within an organization.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:enterprise-ai-cost-management", "labels": [ "Enterprise AI Cost Management" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "enterprise-ai-deployment", "title": "Enterprise AI Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The strategic process of integrating and operationalizing artificial intelligence systems within the internal workflows and IT infrastructure of large organizations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:enterprise-ai-deployment", "labels": [ "Enterprise AI Deployment" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "enterprise-ai-spend", "title": "Enterprise AI Spend", "domain": "economics", "domain_name": "Economics", "definition": "The aggregate financial outlay by organizations for AI tools, services, and infrastructure, often analyzed per employee or by adoption tier.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:enterprise-ai-spend", "labels": [ "Enterprise AI Spend" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "enterprise-ai-strategy", "title": "Enterprise AI Strategy", "domain": "governance", "domain_name": "Governance", "definition": "The organizational planning and decision-making process for integrating AI technologies, balancing factors like cost, control, security, and performance.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:enterprise-ai-strategy", "labels": [ "Enterprise AI Strategy" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "enterprise-ai", "title": "Enterprise Ai", "domain": "ai", "domain_name": "Ai", "definition": "Enterprise AI is the application of artificial-intelligence systems within large organisations to automate workflows, augment knowledge work and inform decisions, subject to governance, security and integration constraints. It spans foundation-model assistants, retrieval-augmented systems over corporate data, and agentic automation embedded in existing software estates. Distinct from consumer AI, it prioritises auditability, data residency, access control and measurable return on investment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:enterprise-ai", "labels": [ "Enterprise Ai", "Enterprise AI" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application", "Digital Transformation", "Knowledge Management" ], "wikilinks": [] }, { "id": "enterprise-architecture", "title": "Enterprise Architecture", "domain": "governance", "domain_name": "Governance", "definition": "A strategic framework for aligning business processes, information systems, and technology infrastructure with organisational goals, increasingly incorporating metaverse technologies such as XR, digital twins, and AI to enable digital transformation and persistent virtual work environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:enterprise-architecture", "labels": [ "Enterprise Architecture", "EnterpriseArchitecture" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [ "Enterprise Metaverse", "metaverse", "System Architecture" ] }, { "id": "enterprise-automation", "title": "Enterprise Automation", "domain": "ai", "domain_name": "Ai", "definition": "Enterprise automation is the application of software and AI to execute repeatable business processes with minimal human intervention across an organisation. It combines techniques such as robotic process automation, workflow orchestration, business rules engines, and increasingly AI agents that handle unstructured tasks. The goal is to reduce cost, error rates, and cycle time while freeing human workers for higher-value activity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enterprise-automation", "labels": [ "Enterprise Automation" ], "is_subclass_of": [ "AI Application", "Intelligent Automation" ], "wikilinks": [] }, { "id": "enterprise-blockchain-architecture", "title": "Enterprise Blockchain Architecture", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissioned, enterprise-grade blockchain infrastructure implementing Consensus Mechanism|consensus mechanisms, Smart Contract|smart contracts, and governance frameworks for organisations requiring controlled participation, privacy, and regulatory compliance\u2014exemplified by", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:enterprise-blockchain-architecture", "labels": [ "Enterprise Blockchain Architecture", "BC-0428-enterprise-blockchain-architecture" ], "is_subclass_of": [ "Network Component", "Smart Contract Platform" ], "wikilinks": [ "BC-0001-blockchain", "BC-0120-consensus-mechanism", "BC-0142-smart-contract", "BC-0426-hyperledger-fabric", "BC-0427-hyperledger-besu", "BC-0429-permissioned-blockchain", "BC-0430-private-channels", "Corda", "BlockchainDomain", "Consensus Mechanism", "Hyperledger Besu", "Hyperledger Fabric", "Smart Contract" ] }, { "id": "enterprise-blockchain", "title": "Enterprise Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Enterprise blockchain refers to permissioned distributed ledger platforms designed specifically for business use cases, providing organisations with controlled access, enhanced privacy, and regulatory compliance capabilities. Unlike public blockchains, enterprise solutions restrict network participation to authenticated entities, enabling secure data sharing, automated business processes through smart contracts, and maintenance of a single source of truth without reliance on central authorities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:enterprise-blockchain", "labels": [ "Enterprise Blockchain", "Enterprise Blockchain Client" ], "is_subclass_of": [ "Distributed Ledger Technology" ], "wikilinks": [ "Business Process Automation", "Blockchain" ] }, { "id": "enterprise-ethereum-alliance-specification", "title": "Enterprise Ethereum Alliance Specification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Enterprise Ethereum Alliance (EEA) Specification is a set of open standards defining how Ethereum technology can be implemented for enterprise use, covering permissioning, privacy, performance, and interoperability requirements. It provides a common reference so that permissioned and private Ethereum-based platforms remain compatible with the public Ethereum ecosystem. Conforming implementations gain a portable, vendor-neutral baseline for business deployments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enterprise-ethereum-alliance-specification", "labels": [ "Enterprise Ethereum Alliance Specification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "enterprise-ethereum-alliance", "title": "Enterprise Ethereum Alliance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Enterprise Ethereum Alliance (EEA) is an industry consortium founded in February 2017 that develops open, blockchain-based specifications for enterprise deployments of Ethereum technology. It standardises the interfaces between private, consortium, and public Ethereum networks, enabling organisations to build interoperable enterprise applications that can bridge permissioned and permissionless environments. The EEA's Client Specification defines conformance requirements for enterprise Ethereum clients, covering private transaction management, permissioning, token standards, and off-chain compute. Membership spans over two hundred organisations including JP Morgan, Microsoft, Accenture, Intel, and ConsenSys, making it one of the largest blockchain standards bodies in the world.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enterprise-ethereum-alliance", "labels": [ "Enterprise Ethereum Alliance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "enterprise-integration", "title": "Enterprise Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Enterprise integration is the discipline of connecting an organisation's disparate applications, data stores and processes so they operate as a coherent whole, exchanging information and coordinating workflows across system and organisational boundaries. It employs patterns such as messaging, APIs, service buses and event-driven architectures to overcome heterogeneity in technology, format and ownership. It underpins enterprise architecture, automation and end-to-end business processes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:enterprise-integration", "labels": [ "Enterprise Integration" ], "is_subclass_of": [ "Infrastructure", "Software Engineering" ], "wikilinks": [] }, { "id": "enterprise-model-post-training", "title": "Enterprise Model Post-Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of fine-tuning or adapting pre-trained foundation models using proprietary enterprise data to align them with specific organizational requirements, domain knowledge, or compliance standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:enterprise-model-post-training", "labels": [ "Enterprise Model Post-Training" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "enterprise-resource-planning", "title": "Enterprise Resource Planning", "domain": "data", "domain_name": "Data", "definition": "Enterprise resource planning (ERP) is an integrated category of business software that unifies core organisational processes - finance, procurement, manufacturing, inventory, human resources and sales - around a shared data model and central database. By recording transactions once and propagating them across modules, ERP systems eliminate data silos and give a consistent, real-time view of operations. Modern ERP is frequently delivered as a cloud service and extended with analytics and supply-chain capabilities.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:enterprise-resource-planning", "labels": [ "Enterprise Resource Planning" ], "is_subclass_of": [ "Business Process Management" ], "wikilinks": [] }, { "id": "enterprise-risk-management", "title": "Enterprise Risk Management", "domain": "governance", "domain_name": "Governance", "definition": "Enterprise Risk Management (ERM) is a structured, organisation-wide approach to identifying, assessing, prioritising, and mitigating risks that could affect an entity's objectives. It integrates financial, operational, strategic, compliance, and reputational risk into a single governance framework with defined ownership and reporting. Established frameworks such as COSO ERM and ISO 31000 provide the reference models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enterprise-risk-management", "labels": [ "Enterprise Risk Management" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "enterprise-search", "title": "enterprise search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Enterprise Search is the organisational capability to retrieve relevant information from heterogeneous internal data sources\u2014documents, databases, emails, intranets, wikis, and code repositories\u2014through a single unified query interface with access-control enforcement. Modern systems combine full-text inverted-index retrieval with dense vector embeddings and semantic search to surface contextually relevant results beyond simple keyword matching. The discipline integrates information retrieval, knowledge management, and applied AI, requiring content connectors, identity-aware access control, relevance tuning, and low-latency indexing pipelines at organisational scale. Retrieval-Augmented Generation (RAG) architectures have become a dominant deployment pattern, passing retrieved document chunks to large language models to synthesise grounded, cited answers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:enterprise-search", "labels": [ "Enterprise Search" ], "is_subclass_of": [ "Information Retrieval", "Knowledge Management System" ], "wikilinks": [] }, { "id": "enterprise-service-bus", "title": "Enterprise Service Bus", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An enterprise service bus is a middleware backbone that connects disparate applications by mediating, routing and transforming messages between them through a common integration layer. It centralises concerns such as protocol bridging, data transformation, message routing and orchestration so that services need not know about each other directly. The pattern is associated with service-oriented architecture and contrasts with lighter, decentralised integration styles.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:enterprise-service-bus", "labels": [ "Enterprise Service Bus" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [] }, { "id": "enterprise-smart-contracts", "title": "Enterprise Smart Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Enterprise smart contracts are self-executing code artefacts deployed on permissioned Distributed Ledger Technology platforms \u2014 Hyperledger Fabric chaincode (Go/Node.js/Java), R3 Corda CorDapps (Kotlin/JVM), Quorum Blockchain and Hyperledger Besu (EVM permissioned with Tessera...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:enterprise-smart-contracts", "labels": [ "Enterprise Smart Contracts" ], "is_subclass_of": [ "Network Component", "Blockchain", "Smart Contracts", "Distributed Application", "Permissioned Compute Artefact", "Business Logic Code", "Multi-Party Agreement" ], "wikilinks": [ "Accord Project", "Accord Project Cicero", "Accord Project Cicero Ergo Specification", "AMD SEV-SNP", "Androulaki 2018 Hyperledger Fabric EuroSys", "Atomic Settlement", "Audit Function", "AWS Nitro Enclaves", "Azure Confidential Ledger", "B3i", "BIS Project Mariana Rosalind Cedar Agor\u00e1 Reports", "BlackRock 2024 BUIDL Tokenised Treasury Fund", "Brown Carlyle Grigg Hearn 2016 Corda Introduction", "Business Logic Code", "Buterin 2014 Ethereum White Paper", "Castro Liskov 1999 PBFT", "CCAF Global Cryptoasset Benchmarking Study", "Certificate Authority", "Chaincode", "Clack Bakshi Braine 2016 Smart Contract Templates UCL CBT" ] }, { "id": "enterprise-software-platform", "title": "Enterprise Software Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An enterprise software platform is an integrated software environment that provides shared services, data, and tooling on which an organisation's business applications are built and run. Examples include ERP, CRM, and blockchain-as-a-service platforms that offer multi-tenant infrastructure, security, integration, and extensibility. The platform model lets enterprises consolidate capabilities and accelerate delivery of line-of-business applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:enterprise-software-platform", "labels": [ "Enterprise Software Platform", "Enterprise Software" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "enterprise-token-standards", "title": "Enterprise Token Standards", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Enterprise Token Standards are the family of blockchain token specifications engineered to represent regulated financial instruments \u2014 equities, bonds, money-market fund units, real estate, private-equity interests, structured products, regulated stablecoins and central-bank-backed deposit tokens...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:enterprise-token-standards", "labels": [ "Enterprise Token Standards" ], "is_subclass_of": [ "Token and Asset", "Token Standard", "Smart Contract Specification", "Digital Asset Standard", "Securities Representation Format", "Regulated Financial Instrument" ], "wikilinks": [ "24-7 Securities Settlement", "Aave Arc", "ABN AMRO", "ACX", "Agor\u00e1", "Aldgate Developments", "Algorand", "Algorand Standard Asset", "Algoreg", "Allen & Overy", "AmFi", "Andrea Maria Cosentino", "Antoinette Schoar", "Apollo", "Aquis Exchange", "Arbitrum", "Archax", "Arweave", "ASA", "Asset Tokenisation" ] }, { "id": "enterprise-training", "title": "Enterprise Training", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Enterprise Training in spatial computing is the use of immersive virtual and augmented reality experiences to upskill, certify and onboard a workforce at organisational scale. It places learners in realistic, interactive simulations of equipment, procedures and hazardous scenarios that would be costly or dangerous to reproduce physically. By combining experiential learning with analytics and learning-management integration, it improves retention, standardises competency and reduces training risk.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:enterprise-training", "labels": [ "Enterprise Training" ], "is_subclass_of": [ "Immersive Experiences" ], "wikilinks": [] }, { "id": "enterprise-workflow", "title": "Enterprise Workflow", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "An enterprise workflow is a defined sequence of tasks and approvals that an organisation uses to carry out a business process. Software systems coordinate and automate these workflows.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:enterprise-workflow", "labels": [ "Enterprise Workflow" ], "is_subclass_of": [ "Workflow Automation" ], "wikilinks": [ "Workflow Automation", "AI Agent", "https://en.wikipedia.org/wiki/Workflow", "https://www.omg.org/spec/BPMN/" ] }, { "id": "entity-linking", "title": "Entity Linking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Entity linking is the natural language processing task of mapping mentions of entities in unstructured text to their corresponding unique identifiers in a target knowledge base such as Wikidata or a domain ontology. It combines candidate generation, which retrieves plausible referents for a surface form, with entity disambiguation, which selects the correct referent using contextual, semantic, and popularity signals. The task resolves ambiguity where one surface form may denote many entities and one entity may be expressed by many surface forms. Entity linking is a foundational step in transforming free text into structured, machine-readable knowledge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:entity-linking", "labels": [ "Entity Linking" ], "is_subclass_of": [ "Information Extraction", "Natural Language Processing" ], "wikilinks": [] }, { "id": "entity-resolution", "title": "Entity Resolution", "domain": "data", "domain_name": "Data", "definition": "Entity resolution is the computational process of determining whether two or more records \u2014 potentially from disparate sources, formats, or schemas \u2014 refer to the same real-world entity, and then linking, merging, or deduplicating them into a single canonical representation. It encompasses blocking strategies that reduce the candidate comparison space, similarity scoring across attributes, and decision logic (deterministic, probabilistic, or ML-based) that determines match, non-match, or possible-match outcomes. Widely applied in master data management, knowledge graph construction, fraud detection, and census processing, it forms a foundational layer of data integration pipelines that require a consistent, unified view of entities such as persons, organisations, products, or locations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:entity-resolution", "labels": [ "Entity Resolution" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [ "Data Quality", "Master Data Management", "Named Entity Recognition", "Data Integration" ] }, { "id": "entity", "title": "Entity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Entity is a discrete, identifiable object or agent within an AI system, simulation, or knowledge model\u2014possessing attributes, state, and potentially the capacity for autonomous action. In multi-agent and knowledge-graph contexts, entities are the primary nodes that bear properties, participate in relations, and serve as the subjects and objects of reasoning; in simulation and digital-twin contexts, they represent physical or virtual objects whose behaviour is modelled and tracked over time.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:entity", "labels": [ "Entity", "Entity Model" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot", "Digital Twin" ] }, { "id": "entropy-coding", "title": "Entropy Coding", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Entropy coding is a class of lossless data-compression techniques that assign shorter codewords to more frequent symbols and longer codewords to rarer ones, approaching the information-theoretic entropy limit of a source. Methods such as Huffman coding and arithmetic coding form the final, lossless stage of most image, audio, and video codecs, packing quantised data into a compact bitstream. Because it discards no information, entropy coding can be perfectly reversed during decoding.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:entropy-coding", "labels": [ "Entropy Coding" ], "is_subclass_of": [ "Lossless Compression" ], "wikilinks": [] }, { "id": "entropy-source", "title": "Entropy Source", "domain": "security", "domain_name": "Security", "definition": "An entropy source is a physical or computational process that produces unpredictable raw data used to seed cryptographic random number generation. Good entropy sources draw on inherently uncertain phenomena, such as electronic noise, timing jitter or radioactive decay, so that their output cannot be predicted or reproduced by an adversary. The quality of an entropy source directly determines the strength of keys, nonces and other security-critical random values derived from it.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:entropy-source", "labels": [ "Entropy Source" ], "is_subclass_of": [ "Random Number Generation" ], "wikilinks": [] }, { "id": "entropy", "title": "Entropy", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A measure of disorder or uncertainty. In thermodynamics it quantifies the unavailable energy in a system, and in information theory it quantifies the average uncertainty or information content of a source.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:entropy", "labels": [ "Entropy" ], "is_subclass_of": [ "Information Theory", "Probability Theory", "Statistical Mechanics" ], "wikilinks": [ "Random Number Generation", "Cryptography", "Information Theory" ] }, { "id": "environment-mapping", "title": "Environment Mapping", "domain": "robotics", "domain_name": "Robotics", "definition": "Environment mapping is the robotics and computer-vision process of building a spatial representation of a robot's surroundings from sensor data such as lidar, cameras, or depth sensors. Representations include occupancy grids, point clouds, and topological or semantic maps that record obstacles, free space, and landmarks. Accurate mapping is a prerequisite for localisation, path planning, and safe autonomous navigation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:environment-mapping", "labels": [ "Environment Mapping", "3D Environment Mapping", "Environment Map" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "environment-model", "title": "Environment Model", "domain": "robotics", "domain_name": "Robotics", "definition": "An environment model is an internal representation maintained by a robot, autonomous agent, or AI system that encodes the geometry, semantics, dynamics, and state of its surrounding physical or virtual world, enabling planning, navigation, and interaction without direct real-time sensor observation of every aspect of the scene. It may range from metric maps and occupancy grids to rich semantic scene graphs and learned neural radiance fields.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:environment-model", "labels": [ "Environment Model" ], "is_subclass_of": [ "Scene Graph" ], "wikilinks": [] }, { "id": "environment", "title": "Environment", "domain": "robotics", "domain_name": "Robotics", "definition": "Environment is the external context in which an agent or system operates, providing inputs and responding to actions, a central concept in robotics and reinforcement learning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:environment", "labels": [ "Environment" ], "is_subclass_of": [ "Navigation and Planning", "owl:Thing" ], "wikilinks": [ "Reinforcement Learning", "Scene Understanding", "owl:Thing" ] }, { "id": "environmental-accounting", "title": "Environmental Accounting", "domain": "governance", "domain_name": "Governance", "definition": "A systematic framework for measuring, tracking, and reporting the environmental costs and impacts associated with economic activities, extending traditional financial accounting to incorporate natural capital, carbon emissions, resource consumption, and waste generation. Environmental accounting supports sustainability reporting, regulatory compliance, and circular economy transitions.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:environmental-accounting", "labels": [ "Environmental Accounting", "Environmental Accounting Standard" ], "is_subclass_of": [ "SustainabilityReporting" ], "wikilinks": [ "owl:Thing" ] }, { "id": "environmental-assessment", "title": "Environmental Assessment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The systematic evaluation of the environmental impact of technological systems, infrastructure, or projects, measuring metrics such as energy consumption, carbon footprint, and e-waste generation. In spatial computing and metaverse contexts, environmental assessment applies to data centre operations, XR hardware lifecycles, and the cumulative sustainability profile of immersive platform deployments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:environmental-assessment", "labels": [ "Environmental Assessment", "Environmental Impact Assessment", "EnvironmentalAssessment" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Computer Vision", "owl:Thing" ] }, { "id": "environmental-asset-market", "title": "Environmental Asset Market", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Blockchain-enabled trading platforms for environmental assets including carbon credits, renewable energy certificates, and biodiversity offsets, utilising tokenisation and smart contracts to enhance transparency, prevent double-counting, and enable fractional ownership of sustainability instruments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:environmental-asset-market", "labels": [ "Environmental Asset Market" ], "is_subclass_of": [ "Governance and Safety", "Digital Asset Market" ], "wikilinks": [ "Sustainable Finance", "Digital Asset Market", "metaverse" ] }, { "id": "environmental-certificate", "title": "Environmental Certificate", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A verifiable credential or token attesting that a digital platform, virtual environment, or computational process meets defined environmental sustainability thresholds. Environmental certificates underpin accountability frameworks for metaverse and spatial computing infrastructure by providing tamper-evident proof of carbon offset, renewable energy use, or compliance with environmental KPIs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:environmental-certificate", "labels": [ "Environmental Certificate", "Environmental Attribute Certificate" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "environmental-control-and-life-support-system", "title": "Environmental Control and Life Support System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:environmental-control-and-life-support-system", "labels": [ "Environmental Control and Life Support System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "environmental-impact-metric", "title": "Environmental Impact Metric", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A quantitative measurement framework for assessing the environmental sustainability of metaverse and digital systems, encompassing energy consumption, carbon emissions, resource efficiency, and ecological footprint across computational infrastructure and user interactions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:environmental-impact-metric", "labels": [ "Environmental Impact Metric" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Data Collection System", "E-Waste Measurement", "Emissions Database", "Energy Consumption Metric", "Energy Metering", "Environmental Optimization", "ESG Reporting System", "EU Ecodesign Directive", "GHG Protocol", "Green Computing Initiative", "Green IT Compliance", "ISO 14040", "Lifecycle Assessment Tool", "Monitoring Infrastructure", "Resource Efficiency Score", "Resource Tracking", "Sustainability Framework", "Sustainability Report", "Sustainability Reporting", "Benchmark Standard" ] }, { "id": "environmental-k-p-i", "title": "Environmental K P I", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Quantitative metrics used to measure, monitor, and report the environmental impact of digital platforms, metaverse infrastructure, and spatial computing systems. Environmental KPIs encompass energy consumption per user session, carbon emissions per compute hour, water usage effectiveness, and percentage of renewable energy sourced, providing governance bodies with evidence for sustainability compliance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:environmental-k-p-i", "labels": [ "Environmental K P I" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "environmental-mapping", "title": "Environmental Mapping", "domain": "robotics", "domain_name": "Robotics", "definition": "Environmental mapping is the computational process of constructing spatial representations of a physical environment from sensor data, enabling autonomous agents or robots to understand, navigate, and interact with their surroundings. It encompasses techniques that transform raw sensory inputs\u2014such as lidar point clouds, camera imagery, and depth data\u2014into structured geometric or semantic maps used for path planning and obstacle avoidance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:environmental-mapping", "labels": [ "Environmental Mapping" ], "is_subclass_of": [ "Mapping" ], "wikilinks": [] }, { "id": "environmental-monitoring", "title": "Environmental Monitoring", "domain": "data", "domain_name": "Data", "definition": "Environmental monitoring is the systematic collection, measurement, and analysis of physical, chemical, and biological parameters of natural and built environments over time, enabling the detection of change, assessment of regulatory compliance, and support of scientific understanding of ecological and climate systems. It encompasses sensor networks, satellite remote sensing, in-situ measurement stations, and the data pipelines that transform raw measurements into actionable environmental intelligence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:environmental-monitoring", "labels": [ "Environmental Monitoring", "IoT Environmental Monitoring" ], "is_subclass_of": [ "Monitoring" ], "wikilinks": [] }, { "id": "environmental-registry", "title": "Environmental Registry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A catalogued, authoritative store of descriptors for physical or virtual environments, recording their spatial bounds, safety parameters, hazard zones, and compliance status. XR systems query the environmental registry during session initialisation to validate that the current space meets operational and safety requirements before permitting immersive interactions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:environmental-registry", "labels": [ "Environmental Registry" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "environmental-sensing", "title": "Environmental Sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Environmental sensing is the acquisition of data about physical surroundings, such as temperature, light, humidity, air quality, sound, or proximity, using sensors. It provides the raw perceptual input that systems use to understand and respond to their context. In robotics, IoT, and context-aware applications it is the foundational layer enabling situational awareness and adaptive behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:environmental-sensing", "labels": [ "Environmental Sensing", "Weather Sensing" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "environmental-sensor", "title": "Environmental Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An environmental sensor is a device that measures ambient physical conditions of its surroundings, such as temperature, humidity, pressure, light level, air quality, or sound. As an exteroceptive sensor it perceives the external world rather than a system's internal state. These sensors supply the contextual data used by robots, augmented-reality systems, and smart-environment applications to adapt their behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:environmental-sensor", "labels": [ "Environmental Sensor", "Environmental Sensors" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "environmental-standards", "title": "Environmental Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Regulatory frameworks and technical specifications governing the environmental sustainability of metaverse infrastructure, including energy efficiency requirements for data centres, carbon footprint disclosure mandates, and e-waste management guidelines for XR hardware.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:environmental-standards", "labels": [ "Environmental Standards" ], "is_subclass_of": [ "Governance and Safety", "Technical Standards" ], "wikilinks": [ "metaverse", "Technical Standards" ] }, { "id": "environmental-sustainability-label", "title": "Environmental Sustainability Label", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Certification process and label indicating compliance with environmental sustainability standards for digital infrastructure, energy consumption, and carbon footprint in metaverse operations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:environmental-sustainability-label", "labels": [ "Environmental Sustainability Label" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Certification Issuance", "Energy Consumption Audit", "Energy Monitoring System", "Green Infrastructure Certification", "Infrastructure Metrics", "ISO 14021", "Measurement Protocols", "Sustainability Compliance System", "Sustainability Reporting", "User Trust Building", "Carbon Accounting", "Carbon Footprint Assessment", "Carbon Neutrality Verification", "Compliance Verification", "Environmental Standards", "Governance Framework", "Middleware Layer", "Third-Party Auditor", "TrustAndGovernanceDomain" ] }, { "id": "environmental-sustainability", "title": "Environmental Sustainability", "domain": "governance", "domain_name": "Governance", "definition": "The responsible stewardship of natural resources and environmental systems in AI development and deployment, minimising ecological harm whilst potentially leveraging AI to address environmental challenges including climate change, biodiversity loss and resource depletion.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:environmental-sustainability", "labels": [ "Environmental Sustainability", "BC-0214-environmental-sustainability" ], "is_subclass_of": [ "Sustainable Development" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "environmental-trading", "title": "Environmental Trading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Environmental Trading encompasses market-based mechanisms for buying and selling environmental assets such as carbon credits, renewable energy certificates, and biodiversity offsets. These mechanisms use blockchain-backed registries and smart contracts to provide transparent, auditable records of environmental value transfers, enabling organisations to meet sustainability obligations through market participation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:environmental-trading", "labels": [ "Environmental Trading" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "environmental-understanding", "title": "Environmental Understanding", "domain": "robotics", "domain_name": "Robotics", "definition": "Environmental understanding is the capability of a robot or autonomous system to perceive, interpret, and build a usable model of its surroundings from sensor data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:environmental-understanding", "labels": [ "Environmental Understanding" ], "is_subclass_of": [ "Scene Understanding" ], "wikilinks": [ "Perception", "Mobile Manipulation", "Environment", "Scene Understanding" ] }, { "id": "environmental-verification", "title": "Environmental Verification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of validating that a physical or virtual environment meets defined safety, compliance, or operational standards before allowing XR interactions to proceed. Encompasses spatial boundary checks, hazard detection, and cross-referencing environment state against authoritative registries to ensure user safety and regulatory compliance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:environmental-verification", "labels": [ "Environmental Verification", "Environmental Impact Verification" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "envoy-proxy", "title": "Envoy Proxy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Envoy is a high-performance, open-source edge and service proxy designed for cloud-native applications, originally built at Lyft and graduated under the CNCF. It provides L3/L4 and L7 traffic management, dynamic configuration via the xDS APIs, observability, and resilience features such as retries, circuit breaking, and rate limiting. Envoy is the default data plane for many service meshes, where it runs as a sidecar alongside each workload.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:envoy-proxy", "labels": [ "Envoy Proxy" ], "is_subclass_of": [ "Reverse Proxy" ], "wikilinks": [] }, { "id": "ephemeris", "title": "Ephemeris", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ephemeris", "labels": [ "Ephemeris" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "epidemiological-modelling", "title": "Epidemiological Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Epidemiological modelling is the use of mathematical and computational models to describe how infectious diseases spread through populations over time. Compartmental models partition a population into states such as susceptible, infected, and recovered, and use differential equations to govern transitions between them, while network and agent-based models capture heterogeneous contact structures. These models inform forecasting, intervention design, and public-health policy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:epidemiological-modelling", "labels": [ "Epidemiological Modelling" ], "is_subclass_of": [ "Computational Modelling", "Mathematical Biology", "Simulation" ], "wikilinks": [] }, { "id": "epipolar-geometry", "title": "Epipolar Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Epipolar Geometry is the projective geometry of stereo vision, describing the geometric relationship between two camera views of the same 3D scene. It is encapsulated in the Fundamental Matrix (uncalibrated cameras) and Essential Matrix (calibrated cameras), which constrain the search for correspondences between images to one-dimensional epipolar lines rather than the full 2D image plane.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:epipolar-geometry", "labels": [ "Epipolar Geometry" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "epistemic-modality-marker", "title": "Epistemic Modality Marker", "domain": "ai", "domain_name": "Ai", "definition": "An Epistemic Modality Marker is a lexical or grammatical element \u2014 such as the modal verbs may, might, could and must, or hedges like perhaps and possibly \u2014 that encodes a speaker's degree of certainty or commitment toward a proposition. Detecting and generating such markers is central to hedge detection, uncertainty-aware natural language processing and calibrated dialogue systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:epistemic-modality-marker", "labels": [ "Epistemic Modality Marker", "Could" ], "is_subclass_of": [ "Natural Language Processing", "Computational Linguistics", "Pragmatics" ], "wikilinks": [ "Dialogue Systems", "Natural Language Processing" ] }, { "id": "epoch", "title": "Epoch", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In machine learning, an Epoch is one complete pass through the entire training dataset, during which model parameters are updated after each constituent batch. The number of epochs is a primary training hyperparameter: too few yield underfitting, whilst too many risk overfitting\u2014a trade-off managed by techniques such as early stopping and learning-rate scheduling. More broadly, an epoch denotes a fixed reference point or interval in time, as used in astronomical coordinate systems (e.g., J2000.0) and geological stratigraphy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:epoch", "labels": [ "Epoch" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Autonomous Robot", "Blockchain" ] }, { "id": "equalized-odds", "title": "Equalized Odds", "domain": "ai", "domain_name": "Ai", "definition": "A group fairness criterion, introduced by Hardt, Price, and Srebro (2016), requiring that a classifier's true positive rate and false positive rate be equal across protected groups \u2014 the prediction must be conditionally independent of group membership given the true outcome; unlike demographic parity it permits base rates to differ between groups, but it conflicts with calibration when base rates differ, and is typically approached by threshold adjustment or constrained training rather than achieved exactly.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:equalized-odds", "labels": [ "Equalized Odds" ], "is_subclass_of": [ "Fairness Metrics" ], "wikilinks": [ "Fairness Metrics", "Algorithmic Fairness", "Fairness", "Bias Mitigation Techniques" ] }, { "id": "equator", "title": "Equator", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:equator", "labels": [ "Equator" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "equatorial-coordinate-system", "title": "Equatorial Coordinate System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:equatorial-coordinate-system", "labels": [ "Equatorial Coordinate System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "equitable-access", "title": "Equitable Access", "domain": "governance", "domain_name": "Governance", "definition": "Equitable access is the principle and practice of ensuring that all people, regardless of ability, income, geography, or background, can reach and use a resource, service, or technology on fair terms. It goes beyond equal access by actively removing barriers and providing accommodations so that disadvantaged groups achieve comparable outcomes. In technology it is closely tied to accessibility standards and inclusive design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:equitable-access", "labels": [ "Equitable Access", "Equitable Metaverse Access" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "erasure-coding", "title": "erasure coding", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Erasure coding is a forward error-correction (FEC) technique that encodes a data object into n encoded fragments (shards or chunks), distributed across nodes or storage devices, such that any k of those n fragments are sufficient to reconstruct the original data without any centralised copy. The redundancy overhead ratio (n \u2212 k) / k is typically far lower than full replication, making erasure coding the preferred durability mechanism in large-scale distributed storage, distributed ledger systems, and content-addressed networks where storage efficiency and fault tolerance are simultaneously required. Foundational schemes include Reed-Solomon codes (based on Galois Field arithmetic), as well as computationally efficient variants such as LDPC, Fountain codes (LT and Raptor), and Cauchy Reed-Solomon; newer constructions couple erasure codes with polynomial commitments (e.g. KZG) to provide data availability proofs in blockchain systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:erasure-coding", "labels": [ "Erasure Coding" ], "is_subclass_of": [ "Forward Error Correction" ], "wikilinks": [] }, { "id": "ergodic-theory", "title": "Ergodic Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Ergodic theory is the branch of mathematics that studies the long-term statistical behaviour of dynamical systems that preserve a measure. Its central result, the ergodic theorem, gives conditions under which the time average of a quantity along a single trajectory equals its average over the whole state space. The theory provides the foundations for understanding when sampling a process over time yields the same information as sampling its underlying distribution, which is essential for Monte Carlo methods and statistical learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ergodic-theory", "labels": [ "Ergodic Theory" ], "is_subclass_of": [ "Dynamical Systems", "Mathematics", "Probability Theory" ], "wikilinks": [] }, { "id": "ergodicity", "title": "Ergodicity", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Ergodicity is the property of a stochastic process whereby its long-run time average, computed along a single sufficiently long trajectory, converges to its ensemble average across all possible states. It is a required condition for Markov chain Monte Carlo methods to converge to the target distribution, since it guarantees that a chain will eventually visit all reachable states in proportion to their stationary probability. Non-ergodic chains can become trapped in subsets of the state space and yield biased samples.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ergodicity", "labels": [ "Ergodicity" ], "is_subclass_of": [ "Stochastic Process" ], "wikilinks": [] }, { "id": "ergonomics", "title": "Ergonomics", "domain": "robotics", "domain_name": "Robotics", "definition": "Ergonomics, also called human factors engineering, is the discipline of designing systems, tools and environments to fit the physical and cognitive characteristics of the people who use them. It studies posture, reach, force, repetition, perception and workload in order to optimise human performance, comfort and safety while reducing fatigue and injury. In robotics and automation it governs the design of workstations, controls and collaborative robots so that machines and humans can operate together efficiently and without harm.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ergonomics", "labels": [ "Ergonomics" ], "is_subclass_of": [ "Human Factors" ], "wikilinks": [] }, { "id": "error-analysis", "title": "Error Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Error analysis is the systematic examination of a model's mistakes to identify patterns, root causes and subgroups where performance is weakest. It typically involves inspecting misclassified examples, confusion matrices and numerical error metrics to distinguish data issues from modelling issues. The findings guide targeted fixes such as additional training data, feature changes or architectural adjustments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:error-analysis", "labels": [ "Error Analysis", "Error analysis" ], "is_subclass_of": [ "Model Evaluation" ], "wikilinks": [] }, { "id": "error-calculation", "title": "Error Calculation", "domain": "robotics", "domain_name": "Robotics", "definition": "Error calculation is the step in a control system that computes the difference between a desired setpoint and the measured process variable. This error term is the input that drives corrective action in feedback controllers such as PID loops. Accurate, low-latency error calculation is essential for stable and responsive position, velocity, and other regulated control behaviours.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:error-calculation", "labels": [ "Error Calculation" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "error-correction", "title": "Error Correction", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Error correction is the set of techniques and mathematical frameworks that detect and rectify errors introduced into data during transmission, storage, or computation, by adding structured redundancy that allows a decoder to infer and restore the original information even when some fraction of the data has been corrupted or lost. Unlike error detection alone, error correction codes (ECCs) carry sufficient redundancy to reconstruct the original codeword without retransmission, at the cost of additional bandwidth or storage overhead. Applications range from deep-space communication and data storage to quantum computing, where error correction is essential to suppress decoherence and enable fault-tolerant operation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:error-correction", "labels": [ "Error Correction", "Error Control Coding", "Error Correcting Code" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "error-handling", "title": "Error Handling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Error Handling is the set of mechanisms, patterns, and strategies in software and systems design for detecting, reporting, and recovering from anomalous conditions that deviate from expected operation. It encompasses exception mechanisms, error codes, retry logic, circuit breakers, fallback strategies, and graceful degradation. Robust error handling is essential to system reliability, security, and maintainability, and is a prerequisite for fault-tolerant distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:error-handling", "labels": [ "Error Handling" ], "is_subclass_of": [ "Infrastructure", "Software Engineering (Infrastructure)" ], "wikilinks": [] }, { "id": "error-recovery", "title": "Error Recovery", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Error recovery is the set of mechanisms by which a system detects that an operation has failed or produced an invalid state and restores correct operation, through techniques such as retries with exponential backoff, checkpointing and rollback, compensating actions, graceful degradation, and escalation to human oversight. In autonomous agent systems it distinguishes robust task execution from brittle scripted automation, and in interaction design it covers helping users notice, diagnose, and undo their own mistakes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:error-recovery", "labels": [ "Error Recovery" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [ "Fault Tolerance", "Autonomous Task Execution", "Agent Orchestrator", "Interaction Design", "Resilience" ] }, { "id": "error-signal", "title": "Error Signal", "domain": "robotics", "domain_name": "Robotics", "definition": "An error signal is the quantity in a feedback control system representing the instantaneous difference between the reference setpoint and the actual measured output. It is the driving input to the controller, which acts to reduce it toward zero. The error signal's magnitude, rate of change, and accumulated value are processed by proportional, derivative, and integral control terms respectively.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:error-signal", "labels": [ "Error Signal" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "escape-velocity", "title": "Escape Velocity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:escape-velocity", "labels": [ "Escape Velocity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "escrow-system", "title": "Escrow System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An escrow system is a mechanism that holds assets or funds with a trusted third party or, on blockchains, in a smart contract until predefined release conditions are met. It reduces counterparty risk in transactions by guaranteeing that neither party can unilaterally seize value before obligations are fulfilled. On-chain escrows use conditional logic and timelocks to release or refund funds automatically.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:escrow-system", "labels": [ "Escrow System", "Escrow", "Trusted Escrow" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "essential-climate-variable", "title": "Essential Climate Variable", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:essential-climate-variable", "labels": [ "Essential Climate Variable" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "etcd", "title": "Etcd", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "etcd is a strongly consistent, distributed key-value store used to hold the critical configuration and coordination data of distributed systems. It uses the Raft consensus algorithm to replicate data across a cluster, providing linearisable reads and writes with reliable failover. etcd is best known as the primary datastore for Kubernetes cluster state, and is also used for service discovery, distributed locking, leader election, and configuration management in many cloud-native platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:etcd", "labels": [ "Etcd", "etcd" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "ethan-mollick", "title": "Ethan Mollick", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Ethan Mollick is an American academic, professor at the Wharton School of the University of Pennsylvania, and widely cited public intellectual on the practical deployment of generative artificial intelligence in work, education, and entrepreneurship. His research bridges management science and AI-adoption theory, investigating how large language models augment human productivity, reshape organisational workflows, and alter pedagogical practice. He is the author of the book 'Co-Intelligence: Living and Working with AI' and is known for his hands-on experimental approach to understanding AI capabilities and their societal implications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ethan-mollick", "labels": [ "Ethan Mollick" ], "is_subclass_of": [ "Artificial Intelligence", "Artificial Intelligence Domain", "Management Science", "Entrepreneurship Research" ], "wikilinks": [ "AI Adoption", "Generative AI", "Prompt Engineering", "Artificial Intelligence Domain" ] }, { "id": "ether-cat", "title": "ethercat", "domain": "robotics", "domain_name": "Robotics", "definition": "EtherCAT (Ethernet for Control Automation Technology) is an open, IEC 61158-standardised real-time Ethernet fieldbus protocol developed by Beckhoff Automation, enabling deterministic, high-bandwidth communication between industrial master controllers and distributed slave devices such as servo drives, I/O modules, and encoders. Its on-the-fly processing architecture allows each slave node to extract addressed data from and insert its response into a propagating Ethernet frame, achieving sub-microsecond synchronisation across large distributed networks. Governed by the EtherCAT Technology Group (ETG), the standard is widely deployed in industrial robotics, CNC machining, semiconductor manufacturing, and laboratory automation, with the Safety over EtherCAT (FSoE) extension providing IEC 61784-3-compliant functional safety over the same physical layer.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:ether-cat", "labels": [ "EtherCAT", "EtherCAT Protocol", "EtherCAT Technology Group" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "ethereum-account", "title": "Ethereum Account", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum account is a state entity on the Ethereum blockchain identified by a 20-byte address and holding a balance, nonce, and optionally code and storage. There are two kinds: externally owned accounts controlled by a private key, and contract accounts controlled by their deployed smart-contract code. Accounts are the unit against which transactions are debited, gas is charged, and state transitions are applied.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ethereum-account", "labels": [ "Ethereum Account" ], "is_subclass_of": [ "Ethereum", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "ethereum-attestation-service", "title": "Ethereum Attestation Service", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Attestation Service (EAS) is an open, public infrastructure for making on-chain and off-chain attestations about any subject using registered, reusable schemas. Each attestation is a signed, timestamped claim that can be verified, revoked, and composed by other applications. EAS provides a neutral primitive for building reputation systems, identity credentials, and trust layers without a centralised issuer.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ethereum-smart-contract-platform-attestation-service", "labels": [ "Ethereum Attestation Service", "Attestation Service" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "ethereum-classic", "title": "Ethereum Classic", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A proof-of-work blockchain that continued the original Ethereum chain after the 2016 DAO hard fork, preserving the unaltered transaction history. It maintains the principle that the ledger should not be reversed by social intervention.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ethereum-smart-contract-platform-classic", "labels": [ "Ethereum Classic" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [ "Mining", "Smart Contract", "DeFi", "Ethereum", "Ethereum Virtual Machine", "Cryptocurrency", "https://ethereumclassic.org/" ] }, { "id": "ethereum-foundation", "title": "Ethereum Foundation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Foundation is a non-profit organisation, registered in Switzerland, that supports the development of the Ethereum protocol and its surrounding research and developer community. It funds core protocol research, client development, security audits and educational initiatives, but does not control the network, which is maintained by a decentralised set of clients, validators and contributors. The Foundation was established around the 2014 to 2015 launch of Ethereum.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ethereum-smart-contract-platform-foundation", "labels": [ "Ethereum Foundation" ], "is_subclass_of": [ "Governance and Regulation", "owl:Thing" ], "wikilinks": [ "Ethereum", "Proof of Stake", "Governance Domain", "Blockchain Domain", "owl:Thing" ] }, { "id": "ethereum-improvement-proposal", "title": "Ethereum Improvement Proposal", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum Improvement Proposal (EIP) is a formal design document that describes a proposed change to the Ethereum protocol, its standards, or its processes, providing the technical specification and rationale for community review. EIPs are categorised into Core proposals affecting consensus, Networking proposals, Interface proposals, and Application-level standards known as ERCs. They follow a defined lifecycle from Draft through Review, Last Call, and Final, serving as the canonical mechanism for coordinating decentralised protocol evolution.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ethereum-improvement-proposal", "labels": [ "Ethereum Improvement Proposal", "Ethereum Improvement Proposals" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "ethereum-name-service", "title": "Ethereum Name Service", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Name Service (ENS) is a decentralised, open-source naming protocol deployed on Ethereum that maps human-readable names (e.g. 'alice.eth') to machine-readable identifiers including wallet addresses, content hashes, and arbitrary metadata stored on-chain. It operates through a hierarchy of smart contracts \u2014 a central registry recording ownership and resolver assignments, and resolver contracts that translate names to resources according to standardised ERC specifications. ENS names are minted as ERC-721 non-fungible tokens, granting cryptographic ownership without reliance on a centralised registrar, and extending DNS-compatible reverse resolution so on-chain addresses can be presented as readable identities. The protocol has become the dominant blockchain naming infrastructure on Ethereum, with millions of registered names and integrations across wallets, browsers, and decentralised applications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:ethereum-smart-contract-platform-name-service", "labels": [ "Ethereum Name Service" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Ethereum Smart Contracts", "Identity", "Smart Contract", "Ethereum" ] }, { "id": "ethereum-smart-contract-platform", "title": "Ethereum Smart Contract Platform", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Ethereum is a decentralised, open-source blockchain platform that introduced Turing-complete smart contracts, enabling programmable, self-executing agreements without trusted intermediaries. Launched in 2015 by Vitalik Buterin and co-founders, it underpins the largest ecosystem of decentralised applications (dApps), decentralised finance (DeFi), and NFT markets. Its transition to proof-of-stake consensus via The Merge in September 2022 dramatically reduced its energy footprint while preserving security.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ethereum-smart-contract-platform", "labels": [ "Ethereum Smart Contract Platform", "BC-0066-ethereum", "Ethereum EIPs", "Ethereum Mainnet", "Ethereum Request for Comments", "Ethereum Specification" ], "is_subclass_of": [ "Blockchain Network" ], "wikilinks": [] }, { "id": "ethereum-smart-contracts", "title": "Ethereum Smart Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Programs deployed to the Ethereum blockchain that execute deterministically on the Ethereum Virtual Machine and maintain state enforced by network consensus, enabling trustless automation of agreements and protocols through immutable on-chain code.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ethereum-smart-contract-platform-smart-contracts", "labels": [ "Ethereum Smart Contracts", "Ethereum Smart Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Ethereum Virtual Machine", "Solidity", "ERC-20", "Ethereum", "Smart Contract" ] }, { "id": "ethereum-virtual-machine", "title": "Ethereum Virtual Machine", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Ethereum Virtual Machine (EVM) is a deterministic, quasi-Turing-complete, stack-based virtual machine that serves as the sandboxed runtime environment for executing compiled smart contract bytecode on the Ethereum blockchain. Every full node in the network independently executes the same sequence of EVM opcodes and must converge to an identical post-execution world state, enforcing global consensus over account balances, contract storage, and transaction receipts. Execution is metered by a resource-accounting unit called Gas, which prevents denial-of-service attacks and prices computational work proportionally to its cost. The EVM's instruction set and state transition function have become a de-facto industry standard, with dozens of EVM-compatible chains and Layer 2 networks implementing the same specification to enable cross-chain portability of smart contracts.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:ethereum-smart-contract-platform-virtual-machine", "labels": [ "Ethereum Virtual Machine" ], "is_subclass_of": [ "Virtual Machine" ], "wikilinks": [ "Smart Contract Platform", "Ethereum Smart Contracts", "Gas", "Virtual Machine" ] }, { "id": "ethereum", "title": "Ethereum", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Ethereum is an open-source, decentralised layer-1 blockchain platform conceived by Vitalik Buterin and launched in 2015, distinguished from Bitcoin by its general-purpose programmability via the Ethereum Virtual Machine (EVM), which executes Turing-complete smart contracts deployed on a shared global state. It transitioned from Proof of Work to Proof of Stake in September 2022 (the Merge), dramatically reducing energy consumption while preserving consensus security through a validator set staking Ether (ETH). Ethereum serves as the foundational settlement layer for the majority of decentralised finance (DeFi) protocols, NFT standards, DAOs, and Layer-2 rollup networks, with its fee market governed by EIP-1559's base-fee-burn mechanism.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:ethereum", "labels": [ "Ethereum", "Ethereum 2.0", "Ethereum Developer Community", "Ethereum Ecosystem", "Ethereum Mainnet", "Ethereum Network", "Ethereum Protocol", "Ethereum State Model" ], "is_subclass_of": [ "Smart Contract Platform", "Blockchain Network" ], "wikilinks": [] }, { "id": "ethernet", "title": "Ethernet", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ethernet is a family of wired networking technologies, standardised as IEEE 802.3, used to connect devices in local area networks (LANs) and data centres via twisted-pair, fibre, and coaxial cabling at speeds from 10 Mbit/s to 400 Gbit/s.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ethernet", "labels": [ "Ethernet", "Automotive Ethernet" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Distributed Systems", "Network Protocol" ] }, { "id": "ethical-ai", "title": "Ethical AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The development, deployment, and use of artificial intelligence systems in accordance with moral principles and values that respect human dignity, rights, and wellbeing, incorporating considerations of fairness, transparency, accountability, privacy, safety, and beneficence throughout the AI lifecycle, whilst promoting human flourishing, social justice, and the common good through deliberate design choices, governance mechanisms, and operational practices that embed ethical reasoning into AI system functioning and organisational decision-making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:ethical-ai", "labels": [ "Ethical AI", "Ethical AI Deployment", "Ethical AI Guidelines", "Ethical AI Operation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Ethics by Design", "NIP-05", "stakeholder engagement", "Trustworthy AI", "Accountability", "AI Governance", "AI Impact Assessment", "Education and AI", "Fairness", "Human Rights", "MetaverseDomain", "Privacy", "Responsible AI", "Transparency" ] }, { "id": "ethical-design-standard", "title": "Ethical Design Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Ethical Design Standard is a codified set of principles and requirements that guides the design of digital products, immersive environments and AI systems so that they respect user autonomy, privacy, accessibility and wellbeing. Such standards translate ethical values into verifiable design criteria, supporting audit, certification and regulatory compliance in spatial computing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ethical-design-standard", "labels": [ "Ethical Design Standard" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "ethical-framework", "title": "Ethical Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured set of moral principles, values, and reasoning methods that guide the development, deployment, and use of AI systems to ensure they respect human dignity, promote well-being, and avoid harm. Ethical frameworks draw from consequentialism, deontology, virtue ethics, and care ethics to address AI-specific dilemmas around fairness, transparency, accountability, and privacy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ethical-framework", "labels": [ "Ethical Framework", "Ethical Guidelines", "EthicalFramework" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Governance Principle" ], "wikilinks": [ "ACM Code of Ethics", "AI Governance Principle", "Asilomar AI Principles", "Autonomy", "Beneficence", "Human Dignity", "IEEE Ethically Aligned Design", "Justice", "Montreal Declaration for Responsible AI", "Non-maleficence", "Accountability", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin", "Fairness", "Privacy", "Transparency" ] }, { "id": "ethical-review-process", "title": "Ethical Review Process", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Ethical Review Process is a structured methodology for evaluating AI systems against ethical frameworks, organisational values, and societal norms, involving expert deliberation, stakeholder consultation, and documented decision-making to ensure responsible AI development and deployment. It applies consequentialist, deontological, and virtue-ethics frameworks to assess fairness, privacy, autonomy, safety, and accountability dimensions, producing approval, conditional approval, deferral, or rejection outcomes with documented rationale.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ethical-review-process", "labels": [ "Ethical Review Process" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "IEEE P7000", "ISO/IEC 42001:2023", "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer", "EU AI Act" ] }, { "id": "ethical-sourcing", "title": "Ethical Sourcing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ical Sourcing is the corporate and institutional procurement discipline that selects, qualifies and monitors suppliers across multi-tier global supply chains against verifiable human rights, labour, environmental and anti-corruption standards \u2014 operationalised through codified frameworks (", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ethical-sourcing", "labels": [ "Ethical Sourcing", "BC-0453-ethical-sourcing" ], "is_subclass_of": [ "Governance and Regulation", "Sustainable Procurement", "Supply Chain Governance", "Corporate Social Responsibility", "Human Rights Due Diligence", "Responsible Business Conduct" ], "wikilinks": [ "AI Risk Scoring", "Amnesty International 2016 This Is What We Die For Cobalt DRC", "Apple 2024 Supplier Responsibility Annual Report 18th Edition", "Apple Supplier Responsibility", "Audit Capacity", "Australian Modern Slavery Act 2018", "Bangladesh Accord", "Barrientos 2019 Gender and Work in Global Value Chains", "BHRRC 2024 Modern Slavery Statements Review", "Biodiversity Conservation", "Blockchain Provenance", "California SB 657", "CCAF 2024 Global Cryptoasset and Sustainable Finance Benchmarking Study", "Certification Infrastructure", "CertificationLayer", "Chain of Custody", "Child Labour Prevention", "Circulor", "Civil Society Oversight", "ComplianceLayer" ] }, { "id": "ethics-and-law-layer", "title": "Ethics & Law Layer", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Framework layer defining norms, rights, and regulations for responsible conduct in metaverse environments through compliance mechanisms, ical AI governance, and legal frameworks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:ethics-and-law-layer", "labels": [ "Ethics & Law Layer" ], "is_subclass_of": [ "AI Governance and Ethics", "Governance Architecture" ], "wikilinks": [ "Audit Logging", "Ethical AI Guidelines", "Ethical Governance", "Ethics Principles", "Legal Compliance", "Legal Regulation Schema", "MSF Taxonomy 2025", "Rights Management System", "Rights Protection", "Compliance Framework", "Governance Architecture", "Identity Management", "Middleware Layer", "Policy Engine", "Regulatory Standards", "Responsible AI", "TrustAndGovernanceDomain", "Trust Framework" ] }, { "id": "ethics-and-law", "title": "Ethics and Law", "domain": "governance", "domain_name": "Governance", "definition": "Ethics and law is the combined domain concerned with the moral principles and binding legal rules that govern the development and deployment of technology. Ethics addresses what ought to be done, while law codifies enforceable obligations such as liability, privacy, and rights. In AI and data governance the two are treated together because responsible systems must satisfy both normative expectations and statutory requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ethics-and-law", "labels": [ "Ethics and Law" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "ethics", "title": "Ethics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Ethics is the systematic study of what is right, good and obligatory, and of the principles that should guide conduct. In the context of artificial intelligence it concerns how systems should be designed, deployed and governed so that their effects on people and society are beneficial, fair and accountable. AI ethics draws on long-standing moral philosophy while addressing new questions raised by autonomous and data-driven systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ethics", "labels": [ "Ethics" ], "is_subclass_of": [ "AI Safety", "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "etsi-domain-immersive", "title": "Etsi Domain Immersive", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The ETSI technical domain addressing standardisation of immersive technologies including virtual reality, augmented reality, and mixed reality systems, encompassing network requirements, quality of experience metrics, and interoperability specifications for XR applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:etsi-domain-immersive", "labels": [ "Etsi Domain Immersive" ], "is_subclass_of": [ "Standards and Interoperability", "ETSI" ], "wikilinks": [ "ETSI Standards" ] }, { "id": "etsi-metaverse-domain-model", "title": "Etsi Metaverse Domain Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A conceptual framework developed by ETSI defining the structural organisation, functional components, and interaction patterns of metaverse systems, providing a standardised reference architecture for telecommunications infrastructure supporting persistent, interconnected virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:etsi-metaverse-domain-model", "labels": [ "Etsi Metaverse Domain Model" ], "is_subclass_of": [ "Platform and Environment", "Domain Model" ], "wikilinks": [ "Domain Model" ] }, { "id": "etsi-metaverse-domain-taxonomy", "title": "Etsi Metaverse Domain Taxonomy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A hierarchical classification system developed by ETSI organising metaverse concepts, technologies, and services into structured categories, enabling consistent terminology, clear domain boundaries, and interoperable standards development across the telecommunications and immersive technology ind...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:etsi-metaverse-domain-taxonomy", "labels": [ "Etsi Metaverse Domain Taxonomy" ], "is_subclass_of": [ "Platform and Environment", "Taxonomy" ], "wikilinks": [ "Standards Interoperability", "Taxonomy" ] }, { "id": "eu-digital-single-market", "title": "Eu Digital Single Market", "domain": "governance", "domain_name": "Governance", "definition": "The EU Digital Single Market is a European Union policy strategy aimed at removing regulatory and technical barriers so that digital goods, services, and data can move freely across member states under a harmonised set of rules. It seeks to extend the principles of the single market to the online economy, covering areas such as e-commerce, data portability, telecommunications, copyright, and cross-border digital services. The strategy underpins later legislative instruments including data-protection, platform-governance, and data-sharing frameworks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:eu-digital-single-market", "labels": [ "Eu Digital Single Market", "Digital Single Market", "EU Digital Single Market" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "euclidean-distance", "title": "Euclidean Distance", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Euclidean distance is the straight-line distance between two points in Euclidean space, computed as the square root of the sum of squared differences across all coordinate dimensions. It is the most widely used distance metric in geometry, statistics, and machine learning, serving as the default measure of dissimilarity in clustering algorithms, nearest-neighbour search, and dimensionality reduction methods. As a special case of the Minkowski distance (p=2), it satisfies the metric axioms of non-negativity, symmetry, and the triangle inequality.", "entityType": "Class", "qualityScore": 0.88, "maturity": "mature", "iri": "urn:ngm:class:euclidean-distance", "labels": [ "Euclidean Distance" ], "is_subclass_of": [ "Distance Metric", "Minkowski Distance", "Linear Algebra" ], "wikilinks": [] }, { "id": "euler-angles", "title": "Euler Angles", "domain": "robotics", "domain_name": "Robotics", "definition": "Euler angles are a representation of a rigid body's three-dimensional orientation as a sequence of three successive rotations about specified coordinate axes, such as roll, pitch, and yaw. They are compact and intuitive for human interpretation but suffer from gimbal lock, a loss of one rotational degree of freedom when two rotation axes align. Euler angles are widely used in inertial measurement unit output and robot pose description, often converted internally to rotation matrices or quaternions to avoid singularities.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:euler-angles", "labels": [ "Euler Angles" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "european-central-bank", "title": "European Central Bank", "domain": "finance", "domain_name": "Finance", "definition": "The European Central Bank (ECB) is the central bank for the euro and the monetary authority of the euro area, responsible for setting monetary policy with a primary mandate of price stability. It administers the single currency, supervises significant banks under the Single Supervisory Mechanism, and manages the Eurosystem alongside national central banks. The ECB is also developing a potential digital euro as a central bank digital currency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:european-central-bank", "labels": [ "European Central Bank" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [] }, { "id": "european-commission", "title": "european commission", "domain": "governance", "domain_name": "Governance", "definition": "The European Commission is the politically independent executive body of the European Union, holding the exclusive right of legislative initiative and responsible for proposing legislation, implementing decisions of the Council of the European Union, managing the EU budget, and upholding EU treaties and law across all 27 member states. In the technology domain it has authored landmark regulatory frameworks including the EU AI Act, the Data Governance Act, the Digital Markets Act, and the Digital Services Act, constituting the most comprehensive digital governance agenda of any jurisdiction. It also directs research funding through Horizon Europe, coordinates AI standardisation mandates to CEN-CENELEC and ETSI, and operates cross-border digital infrastructure such as the European Blockchain Services Infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:european-commission", "labels": [ "European Commission" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "european-parliament", "title": "European Parliament", "domain": "governance", "domain_name": "Governance", "definition": "The European Parliament is the directly elected legislative body of the European Union, sharing law-making and budgetary authority with the Council of the EU. It debates, amends, and adopts EU legislation across domains including financial regulation, data protection, AI, and digital-currency policy. As a co-legislator it is a primary source of the binding rules that shape technology governance across member states.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:european-parliament", "labels": [ "European Parliament" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "evaluation-harness", "title": "Evaluation Harness", "domain": "ai", "domain_name": "Ai", "definition": "An evaluation harness is a software framework that automates the systematic assessment of AI model capabilities across standardised benchmark tasks, providing reproducible prompt formatting, answer extraction, scoring, and aggregated reporting. It enables consistent, comparable measurement of model performance across tasks, modalities, and versions, forming the backbone of LLM leaderboards and model selection workflows.", "entityType": "Class", "qualityScore": 0.88, "maturity": "emerging", "iri": "urn:ngm:class:evaluation-harness", "labels": [ "Evaluation Harness", "EleutherAI lm-evaluation-harness", "Evaluation Framework", "Evaluation Protocol" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation", "Benchmarks", "Reproducibility" ], "wikilinks": [] }, { "id": "evaluation-layer", "title": "Evaluation Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Evaluation Layer is the cross-cutting stratum that measures the quality, safety, and performance of system components against defined criteria. It sits alongside training and inference, drawing on their outputs to produce judgements that feed governance and research strata. It contains benchmarks, metrics, test harnesses, and the scoring procedures that quantify behaviour.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:evaluation-layer", "labels": [ "Evaluation Layer" ], "is_subclass_of": [ "AI Governance and Ethics", "owl:Thing" ], "wikilinks": [ "Model Layer", "Inference Layer", "Research Layer", "Governance Layer", "Benchmark", "Statistical Significance", "owl:Thing" ] }, { "id": "evaluation-metric", "title": "Evaluation Metric", "domain": "ai", "domain_name": "Ai", "definition": "An evaluation metric is a quantitative or qualitative measure used to assess the performance, quality, or behaviour of a machine learning model, algorithm, or system against a defined objective. Metrics are computed over held-out test data or through human judgement protocols and provide the empirical basis for model comparison, selection, and deployment decisions. The choice of metric directly shapes what properties a model optimises for during training and what trade-offs are made between competing objectives such as accuracy, fairness, and calibration. Metric selection is therefore a first-class design decision whose consequences cascade from training dynamics through to safety, governance, and societal impact.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:evaluation-metric", "labels": [ "Evaluation Metric", "Evaluation Metrics" ], "is_subclass_of": [ "Performance Metrics", "AI Research Area", "Measurement Theory" ], "wikilinks": [] }, { "id": "evaluation-benchmarks-and-leaderboards", "title": "Evaluation benchmarks and leaderboards", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Ebenchmarks and leaderboards constitute the standardised measurement infrastructure of contemporary artificial intelligence, comprising curated test datasets, scoring protocols, and publicly comparable scoreboards that quantify model capability, safety, robustness, and alignment across narrowly s...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:evaluation-benchmarks-and-leaderboards", "labels": [ "Evaluation benchmarks and leaderboards", "Evaluation Benchmarks", "Evaluation Benchmarks and Leaderboards", "Leaderboards" ], "is_subclass_of": [ "AI Evaluation", "AI Research Area", "Measurement", "Standardised Testing", "Empirical Methodology", "Capability Assessment" ], "wikilinks": [ "A/B Testing", "Academic Peer Review", "Adversarial Generation", "AI Evaluation", "AI Procurement", "AISafetyDomain", "Alignment", "Austin et al. 2021 MBPP", "Bradley-Terry Aggregation", "Capability Assessment", "Capability Elicitation", "Capability Forecasting", "Capital Allocation", "Chen et al. 2021 HumanEval Codex", "Chiang et al. 2024 Chatbot Arena", "Chollet 2019 ARC-AGI", "Cobbe et al. 2021 GSM8K", "Construct Validity", "Contamination Detection", "Cross-Validation" ] }, { "id": "evapotranspiration-estimation", "title": "Evapotranspiration Estimation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:evapotranspiration-estimation", "labels": [ "Evapotranspiration Estimation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "event-driven-architecture", "title": "Event Driven Architecture", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A software architecture pattern built from decoupled services that publish, consume, and route events representing state changes, enabling real-time responsiveness, independent scaling, and resilient distributed systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:event-driven-architecture", "labels": [ "Event Driven Architecture", "Event-Driven Architecture" ], "is_subclass_of": [ "Software Engineering", "System Architecture" ], "wikilinks": [ "Scalable Metaverse Infrastructure", "metaverse", "System Architecture" ] }, { "id": "event-emission", "title": "Event Emission", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Event emission is the act by which a smart contract or software component publishes a structured log entry recording that something notable occurred during execution. On blockchains, emitted events are written to transaction logs that off-chain applications and indexers subscribe to, since contracts cannot push data outward directly. Events provide a cheap, queryable record of state changes and are the primary bridge between on-chain logic and external systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:event-emission", "labels": [ "Event Emission", "Event Emitter" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "event-log", "title": "Event Log", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Event Log is an append-only record of discrete events emitted by a system, used to capture state changes for auditing, indexing and downstream processing. On blockchains, smart contracts emit events that are written to transaction receipts and stored in the log structure of each block, where they can be efficiently queried by off-chain services. Event logs provide an immutable, ordered history that decentralised applications use to reconstruct state and trigger reactions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:event-log", "labels": [ "Event Log", "Event Logs" ], "is_subclass_of": [ "Append-Only Log" ], "wikilinks": [] }, { "id": "event-loop", "title": "Event Loop", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An event loop is a programming construct that waits for and dispatches events or messages within a single-threaded execution model, repeatedly polling a queue of pending tasks and invoking their associated handlers. It is the engine of asynchronous, non-blocking programming, allowing a program to perform I/O and respond to many concurrent events without spawning a thread per operation. Event loops underpin runtime environments, user-interface frameworks, and high-concurrency network servers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:event-loop", "labels": [ "Event Loop" ], "is_subclass_of": [ "Concurrency", "Execution Model" ], "wikilinks": [] }, { "id": "event-management", "title": "Event Management", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The planning, coordination, technical orchestration, and delivery of virtual or hybrid events within spatial computing and metaverse platforms, encompassing scheduling, access control, audience management, live streaming, and post-event analytics for concerts, conferences, and social gatherings.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:event-management", "labels": [ "Event Management" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "event-sourcing", "title": "Event Sourcing", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Event sourcing is an architectural pattern for data persistence in which the state of a system is stored not as a mutable current-state record but as an append-only, ordered log of discrete domain events \u2014 each representing a fact that occurred at a specific point in time. Current application state is derived by replaying the event log from the beginning (or from a periodic snapshot), making the full history of state transitions a first-class, queryable artefact. This contrasts with CRUD-oriented architectures where only the latest state is stored, discarding historical change information.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:event-sourcing", "labels": [ "Event Sourcing" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "event-streaming", "title": "Event Streaming", "domain": "data", "domain_name": "Data", "definition": "Event streaming is a data-processing paradigm in which records of events are captured, stored as an ordered, append-only log, and continuously delivered to consumers in real time. Platforms such as Apache Kafka and Pulsar implement it to decouple producers from consumers and enable scalable, replayable data pipelines. It underpins real-time analytics, event-driven microservices, and the ingestion side of modern data architectures.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:event-streaming", "labels": [ "Event Streaming", "Event Streaming Platform" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "event", "title": "Event", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A significant occurrence at a specific point in time that represents a change in system state, triggers a process, or carries information between components.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:event", "labels": [ "Event" ], "is_subclass_of": [ "Computing and Cloud", "Temporal Entity" ], "wikilinks": [ "Complex Event Processing", "Event Sourcing", "Reactive Systems", "Temporal Entity", "Event-Driven Architecture" ] }, { "id": "eventual-consistency", "title": "Eventual Consistency", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Eventual consistency is a consistency model for distributed data stores that guarantees that, in the absence of new updates, all replicas of a given data item will eventually converge to the same value. The model deliberately relaxes the requirement for immediate, global agreement in favour of higher availability and tolerance of network partitions, as described by the CAP theorem. Reads may transiently return stale data, and divergent replicas are reconciled through background propagation, gossip protocols, or explicit conflict resolution strategies such as last-write-wins or multi-version concurrency control. It is foundational to the design of large-scale internet-facing systems including DNS, distributed caches, and wide-area NoSQL databases.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:eventual-consistency", "labels": [ "Eventual Consistency", "Strong Eventual Consistency" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Distributed Computing", "Distributed Storage", "Vector Clocks" ] }, { "id": "everledger", "title": "Everledger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A technology company that built a blockchain-based registry for tracking the provenance of high-value physical assets such as diamonds and gemstones. The registry links a unique digital record to each physical item to support authenticity and ownership claims.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:everledger", "labels": [ "Everledger" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [ "Distributed Ledger Technology", "Provenance", "Permissioned Blockchain", "Distributed Ledger", "Supply Chain", "https://everledger.io/" ] }, { "id": "evidence-collection", "title": "Evidence Collection", "domain": "governance", "domain_name": "Governance", "definition": "Evidence collection is the systematic process of identifying, preserving, and documenting artefacts \u2014 digital or physical \u2014 in a manner that maintains their integrity and admissibility for legal, regulatory, or investigative proceedings. In digital contexts it encompasses forensic acquisition of disk images, memory dumps, network captures, and log files while maintaining strict chain-of-custody documentation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:evidence-collection", "labels": [ "Evidence Collection" ], "is_subclass_of": [ "Digital Forensics" ], "wikilinks": [] }, { "id": "evidence-lower-bound", "title": "Evidence Lower Bound", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The evidence lower bound (ELBO) is a tractable lower bound on the log marginal likelihood of observed data under a probabilistic model with latent variables. Maximising the ELBO is equivalent to minimising the divergence between an approximate posterior and the true posterior, making intractable inference tractable. It is the training objective of variational autoencoders and a cornerstone of variational inference.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:evidence-lower-bound", "labels": [ "Evidence Lower Bound" ], "is_subclass_of": [ "Variational Inference", "Objective Function", "Probabilistic Model" ], "wikilinks": [] }, { "id": "evidence-based-design", "title": "Evidence-Based Design", "domain": "data", "domain_name": "Data", "definition": "Evidence-based design is a design practice in which decisions about a product, space or interface are grounded in data gathered through empirical methods such as usability testing and user research, rather than on intuition or convention alone. It treats each design choice as a hypothesis to be validated against observed user behaviour and outcomes, iterating the design in response to findings. The approach originated in healthcare and architectural design and has since become standard practice in digital product and interaction design.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:evidence-based-design", "labels": [ "Evidence-Based Design", "Evidence Based Design" ], "is_subclass_of": [ "Usability Testing" ], "wikilinks": [] }, { "id": "evidence-based-medicine", "title": "Evidence-Based Medicine", "domain": "governance", "domain_name": "Governance", "definition": "The practice of making clinical decisions by conscientiously integrating the best available research evidence with clinical expertise and patient values. Formalised in the early 1990s, it ranks evidence by methodological rigour \u2014 systematic reviews and randomised controlled trials at the top, expert opinion at the bottom \u2014 and operationalises care through critical appraisal, clinical guidelines, and decision support, replacing tradition- and authority-based practice with explicit, auditable use of research findings.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:evidence-based-medicine", "labels": [ "Evidence-Based Medicine", "Evidence Based Medicine" ], "is_subclass_of": [ "Healthcare" ], "wikilinks": [ "Healthcare", "Randomised Controlled Trial", "Clinical Decision Support", "Precision Medicine" ] }, { "id": "evidence-based-policy", "title": "Evidence-Based Policy", "domain": "governance", "domain_name": "Governance", "definition": "Evidence-based policy is an approach to policymaking that grounds decisions in systematically gathered data, empirical evaluation, and rigorous analysis rather than ideology or precedent alone. It draws on decision-support tools, controlled trials, and outcome monitoring to test whether interventions achieve their intended effects before wider rollout. Mechanisms such as futarchy propose formalising this link between evidence and policy choice using prediction markets.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:evidence-based-policy", "labels": [ "Evidence-Based Policy" ], "is_subclass_of": [ "Public Policy" ], "wikilinks": [] }, { "id": "evolutionary-algorithm", "title": "Evolutionary Algorithm", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An evolutionary algorithm is a population-based, stochastic optimisation method inspired by biological evolution, in which a population of candidate solutions is iteratively improved through selection, recombination (crossover) and mutation guided by a fitness function. Because they require only the ability to evaluate a fitness score, evolutionary algorithms are derivative-free and well suited to non-differentiable, noisy, multimodal or black-box optimisation problems where gradient methods struggle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:evolutionary-algorithm", "labels": [ "Evolutionary Algorithm", "Evolutionary Algorithms" ], "is_subclass_of": [ "Optimisation", "Mathematical Optimisation", "Search Algorithm" ], "wikilinks": [] }, { "id": "excalidraw", "title": "Excalidraw", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Excalidraw is an open-source, browser-based collaborative whiteboard and diagramming application characterised by its distinctive hand-drawn visual aesthetic, which renders all shapes, lines, and text in a sketchy style designed to encourage low-fidelity ideation rather than polished production diagrams. The tool stores diagrams in a human-readable JSON format and supports real-time multiplayer collaboration, end-to-end encrypted sharing, and an extensive library of reusable shapes, making it widely used for software architecture sketching, workshop facilitation, and technical communication.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:excalidraw", "labels": [ "Excalidraw" ], "is_subclass_of": [ "Collaborative Whiteboard" ], "wikilinks": [] }, { "id": "exchange-custody", "title": "Exchange Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Exchange custody is the model in which a cryptocurrency exchange or platform holds and controls the private keys to users' digital assets on their behalf, rather than users self-custodying their own keys. The custodian operates wallets, manages security and reconciles internal ledgers crediting customers' balances. This arrangement simplifies user experience and trading but concentrates risk, since users rely on the exchange's solvency, controls and honesty rather than holding their keys directly.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:exchange-custody", "labels": [ "Exchange Custody" ], "is_subclass_of": [ "Institutional Custody" ], "wikilinks": [] }, { "id": "exchange-mechanism", "title": "Exchange Mechanism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Exchange Mechanism is a protocol or technical construct that governs the transfer of digital assets, tokens, or value between parties within a virtual economy or cross-platform environment. Exchange mechanisms specify the rules for matching buyers and sellers, executing atomic swaps, handling liquidity, and settling transactions, forming the economic infrastructure of metaverse marketplaces.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:exchange-mechanism", "labels": [ "Exchange Mechanism" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "exchange-rate", "title": "Exchange Rate", "domain": "finance", "domain_name": "Finance", "definition": "An exchange rate is the price of one currency expressed in terms of another, determining how much of one monetary unit is required to purchase a unit of the other. Exchange rates are set in foreign-exchange markets through supply and demand, influenced by interest rates, inflation, trade balances, and central-bank policy, and may float freely, be pegged, or be managed. In crypto and stablecoin contexts the same concept governs the peg between a token and a fiat reference. Exchange rates are fundamental to international trade, monetary policy transmission, and cross-border value transfer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:exchange-rate", "labels": [ "Exchange Rate" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "exchange-traded-fund", "title": "Exchange-Traded Fund", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An exchange-traded fund (ETF) is a pooled investment vehicle whose shares trade on a stock exchange throughout the day at market prices. It typically tracks an index, asset, or basket of assets and uses a create-redeem mechanism with authorised participants to keep its price close to net asset value. ETFs combine the diversification of funds with the intraday liquidity and accessibility of listed securities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:exchange-traded-fund", "labels": [ "Exchange-Traded Fund" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "execution-model", "title": "Execution Model", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An execution model is the abstract specification of how a computing system interprets, schedules, and carries out instructions or computations, defining the rules governing ordering, concurrency, memory access, and resource allocation. It forms the semantic foundation atop which programming languages, runtimes, and hardware platforms are designed, ensuring consistent and predictable behaviour across implementations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:execution-model", "labels": [ "Execution Model", "Dataflow Execution Model" ], "is_subclass_of": [ "Programming Paradigm" ], "wikilinks": [] }, { "id": "executive-sponsorship", "title": "Executive Sponsorship", "domain": "governance", "domain_name": "Governance", "definition": "Executive sponsorship is the active, visible backing of a major initiative by a senior leader who secures funding, removes organisational obstacles, and aligns the effort with strategy. It is widely identified as a critical success factor in technology adoption and transformation programmes because it confers authority, accountability, and resource commitment. Without credible executive sponsorship, large-scale initiatives commonly stall amid competing priorities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:executive-sponsorship", "labels": [ "Executive Sponsorship" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "existential-ai-risk", "title": "Existential AI Risk", "domain": "ai", "domain_name": "Ai", "definition": "Existential AI risk is the class of scenarios in which advanced artificial intelligence systems could cause human extinction or permanently and drastically curtail humanity's long-term potential, in ways that are irreversible at civilisational scale. These risks arise from misalignment between AI objectives and human values, from insufficient human oversight of increasingly capable systems, or from deliberate misuse enabling catastrophic outcomes. The concept motivates foundational research in AI alignment, corrigibility, and interpretability, as well as international governance frameworks aimed at preventing unrecoverable failure modes. It is distinguished from near-term harms by its emphasis on trajectories toward transformative, hard-to-reverse states rather than localised or recoverable damage.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:existential-ai-risk", "labels": [ "Existential AI Risk" ], "is_subclass_of": [ "Existential Risk", "Global Catastrophic Risk", "AI Risk" ], "wikilinks": [ "AI Safety", "AI Alignment", "Existential Risk" ] }, { "id": "existential-risk", "title": "Existential Risk", "domain": "governance", "domain_name": "Governance", "definition": "An existential risk is any risk that could permanently and drastically curtail humanity's long-run potential \u2014 including human extinction, irreversible civilisational collapse, or permanent totalitarian lock-in \u2014 as distinct from severe but recoverable catastrophes. The concept grounds a research agenda that prioritises preventing outcomes from which recovery is impossible, because their badness is unbounded by the loss of all future generations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:existential-risk", "labels": [ "Existential Risk" ], "is_subclass_of": [ "Catastrophic Risk Assessment" ], "wikilinks": [] }, { "id": "exoplanet-direct-imaging", "title": "Exoplanet Direct Imaging", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:exoplanet-direct-imaging", "labels": [ "Exoplanet Direct Imaging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "exoplanet-radial-velocity-method", "title": "Exoplanet Radial Velocity Method", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:exoplanet-radial-velocity-method", "labels": [ "Exoplanet Radial Velocity Method" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "exoplanet-transit-method", "title": "Exoplanet Transit Method", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:exoplanet-transit-method", "labels": [ "Exoplanet Transit Method" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "exoplanet", "title": "Exoplanet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:exoplanet", "labels": [ "Exoplanet" ], "is_subclass_of": [ "Planet" ], "wikilinks": [] }, { "id": "exoskeleton-control", "title": "Exoskeleton Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Exoskeleton control is the set of control strategies that coordinate a wearable robotic exoskeleton's actuators with the intent and movement of its human wearer. It fuses proprioceptive and biomechanical sensing with kinematic and dynamic models to provide assistive torque while preserving stability and safety. Effective control must adapt to gait phase, user effort, and varying loads in real time to augment strength or restore mobility.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:exoskeleton-control", "labels": [ "Exoskeleton Control" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "exoskeleton-robot", "title": "Exoskeleton Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Exoskeleton Robot - A wearable robotic framework that augments human strength and endurance by providing motorised Joint Support and force amplification, reducing musculoskeletal strain during heavy lifting, hazardous material handling, or prolonged repetitive tasks.", "entityType": "Class", "qualityScore": 0.57, "maturity": "draft", "iri": "urn:ngm:class:exoskeleton-robot", "labels": [ "Exoskeleton Robot" ], "is_subclass_of": [ "Robot Type", "Robotics", "Wearable Robotics" ], "wikilinks": [ "Accessibility in Harsh Environments", "Human Augmentation System", "Injury Prevention", "Joint Support", "Occupational Safety Equipment", "Real-time Control", "Wearable Robotics", "Worker Productivity", "Motion Capture", "Power Supply", "Robotics", "RoboticsDomain", "Telecollaboration" ] }, { "id": "exoskeleton", "title": "Exoskeleton", "domain": "robotics", "domain_name": "Robotics", "definition": "An exoskeleton is a wearable robotic structure that augments or supports human movement by applying forces in parallel with the wearer's limbs, used for assistance, rehabilitation, and load support.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:exoskeleton", "labels": [ "Exoskeleton", "Exoskeleton Design", "Powered Exoskeleton" ], "is_subclass_of": [ "Exoskeleton Robot" ], "wikilinks": [ "Actuator", "Sensor Fusion", "Assistive Robotics", "Human-Robot Collaboration", "Human Robot Interaction", "Exoskeleton Robot" ] }, { "id": "exosphere", "title": "Exosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:exosphere", "labels": [ "Exosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "expectation-maximisation", "title": "Expectation Maximisation", "domain": "ai", "domain_name": "Ai", "definition": "Expectation Maximisation (EM) is an iterative algorithm for finding maximum-likelihood or maximum-a-posteriori estimates of parameters in statistical models with latent (unobserved) variables. It alternates between an E-step, which computes the expected value of the complete-data log-likelihood given current parameters, and an M-step, which maximises that expectation to update the parameters. EM is guaranteed to monotonically increase the likelihood at each iteration and is widely used for Gaussian mixture models, hidden Markov models, and missing-data problems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:expectation-maximisation", "labels": [ "Expectation Maximisation", "Expectation Maximization", "Expectation-Maximisation" ], "is_subclass_of": [ "Machine Learning Technique", "Maximum Likelihood Estimation", "Inference Algorithm", "Iterative Algorithm" ], "wikilinks": [] }, { "id": "expected-utility-theory", "title": "Expected Utility Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Expected utility theory is a formal framework for decision-making under uncertainty in which a rational agent chooses the action that maximises the probability-weighted sum of the utilities of its possible outcomes. Originating with von Neumann and Morgenstern, it provides the axiomatic basis for treating preferences over uncertain outcomes as numerically comparable. It underlies Bayesian decision theory and the formulation of reward and value in Markov decision processes used throughout reinforcement learning.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:expected-utility-theory", "labels": [ "Expected Utility Theory" ], "is_subclass_of": [ "Utility Theory" ], "wikilinks": [] }, { "id": "experience-layer", "title": "Experience Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User-facing architectural layer responsible for rendering immersive content, managing user interactions, and delivering cohesive UX/UI across metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:experience-layer", "labels": [ "Experience Layer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Content Delivery", "Display Technology", "Graphics Pipeline", "Input System", "Interaction Manager", "MSF Taxonomy 2025", "Natural Interaction", "Presence System", "User Engagement", "UX Framework", "Audio System", "Avatar System", "Compute Layer", "Haptic Feedback", "Immersive Experience", "Immersive Interface", "InteractionDomain", "Presence", "Rendering Engine", "Spatial Computing" ] }, { "id": "experiential-learning", "title": "Experiential Learning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Experiential learning is an approach in which knowledge and skills are acquired through direct activity and reflection on concrete experience rather than passive instruction. In spatial computing it is realised through immersive simulations where learners act, observe outcomes and iterate. The cycle of acting, reflecting and conceptualising makes it well suited to embodied, situated training.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:experiential-learning", "labels": [ "Experiential Learning" ], "is_subclass_of": [ "Immersive Learning" ], "wikilinks": [] }, { "id": "empirical-experimental-design-tracking", "title": "Experiment Tracking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Experiment tracking is the practice of recording the configuration, code, data, and results of machine learning experiments so they can be compared and reproduced, enabling teams to audit, iterate, and roll back to prior model states. It is a core discipline within MLOps.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:empirical-experimental-design-tracking", "labels": [ "Experiment Tracking" ], "is_subclass_of": [ "MLOps", "AI Infrastructure (Artificial Intelligence)", "Scientific Method", "Metadata Management" ], "wikilinks": [ "Model Training", "Machine Learning", "Open Source", "MLOps", "https://mlflow.org/docs/latest/tracking.html", "https://en.wikipedia.org/wiki/MLOps" ] }, { "id": "experimental-design", "title": "Experimental Design", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Experimental design is the systematic planning of investigations so that the resulting data can support valid, efficient, and unbiased inferences about cause and effect. It specifies the treatments, controls, randomisation, replication, and blocking that isolate the effect of manipulated variables from confounders and noise. Grounded in the scientific method and statistical theory, it governs how hypotheses are tested, how sample sizes and power are determined, and how variability is controlled, and it underpins disciplined experimentation from laboratory trials to large-scale online A/B tests.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:experimental-design", "labels": [ "Experimental Design" ], "is_subclass_of": [ "Scientific Method", "AI Research Area" ], "wikilinks": [] }, { "id": "expert-parallelism", "title": "Expert Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Expert parallelism is a distributed-training strategy for mixture-of-experts models that places different expert sub-networks on different accelerators and routes each token to its selected experts via all-to-all communication. Because only a sparse subset of experts is activated per token, expert parallelism scales total parameter count without a proportional rise in per-token compute. It is typically combined with data, tensor, and pipeline parallelism, and its efficiency hinges on balanced routing and low-latency interconnects.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:expert-parallelism", "labels": [ "Expert Parallelism" ], "is_subclass_of": [ "Model Parallelism", "Distributed Training", "AI Infrastructure" ], "wikilinks": [] }, { "id": "expert-systems", "title": "Expert Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Expert Systems are rule-based AI programs that encode domain-specific knowledge in a knowledge base and apply logical inference to emulate the decision-making capability of human experts, enabling decision support, diagnosis, and configuration in specialised domains.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:expert-systems", "labels": [ "Expert Systems", "Expert System", "Rule-Based Expert Systems" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Decision Support", "Autonomous Robot", "Blockchain", "Knowledge Representation" ] }, { "id": "explainability-oecd", "title": "Explainability (OECD)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The OECD AI Principle (1.3) requiring that people affected by AI-based outcomes are able to understand how and why particular decisions or recommendations were reached. Explanations must be contextually appropriate, enable meaningful contestation, and illuminate causal factors\u2014operationalised through LIME, SHAP, attention visualisation, and inherently interpretable model architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:explainability-oecd", "labels": [ "Explainability (OECD)" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "explainability", "title": "Explainability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The extent to which an AI system's decision-making processes, outputs, and behaviors can be understood and articulated in human-comprehensible terms, enabling stakeholders to grasp how and why specific outcomes were produced.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:explainability", "labels": [ "Explainability", "AI Explainability", "Explainability (AI-0063)", "Explainability (AI-0064)", "Model Explainability" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Accountability (AI-0068)", "Contestability (AI-0043)", "MetaverseDomain" ] }, { "id": "explainable-ai", "title": "Explainable AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI systems designed to provide clear, understandable explanations of their decision-making processes, enabling stakeholders to comprehend how and why specific outputs are generated.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:explainable-ai", "labels": [ "Explainable AI", "ExplainableAI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Algorithmic Transparency", "IEEE P2976 (XAI)", "LIME", "SHAP", "XAI Methods", "Artificial Intelligence", "Interpretable AI", "Intrinsic Interpretability", "Machine Learning", "MetaverseDomain", "Model Interpretability", "Post Hoc Explanation" ] }, { "id": "explanation", "title": "Explanation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Explanation in the context of AI systems is a human-interpretable account of why a model produced a particular output, which features influenced the prediction, and how the system would behave differently under alternative inputs. Explanations may be global (describing overall model behaviour across the input space), local (accounting for a single prediction), or contrastive (answering why outcome A rather than outcome B). They are the primary instrument through which AI Transparency and Explainability is operationalised for stakeholders including regulators, domain experts, and affected individuals.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:explanation", "labels": [ "Explanation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Artificial Intelligence", "Blockchain", "Digital Twin" ] }, { "id": "exploit", "title": "Exploit", "domain": "security", "domain_name": "Security", "definition": "An exploit is a piece of code, data, or sequence of actions that takes advantage of a vulnerability to cause unintended behaviour, such as gaining unauthorised access, escalating privileges, or executing arbitrary code. Exploits turn a latent weakness into a concrete attack and are studied both offensively, in penetration testing, and defensively, to prioritise remediation. A zero-day exploit targets a vulnerability for which no patch yet exists.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:exploit", "labels": [ "Exploit" ], "is_subclass_of": [ "Vulnerability" ], "wikilinks": [] }, { "id": "exploration-exploitation-tradeoff", "title": "Exploration Exploitation Tradeoff", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The exploration-exploitation tradeoff is the fundamental dilemma faced by a learning agent that must choose between exploiting known rewarding actions and exploring uncertain actions that may yield greater long-term return. Over-exploitation locks the agent into suboptimal behaviour, while over-exploration squanders opportunity on unrewarding choices. Balancing the two is central to reinforcement learning, multi-armed bandit problems and sequential decision making, with strategies ranging from epsilon-greedy selection to upper confidence bounds and posterior sampling.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:exploration-exploitation-tradeoff", "labels": [ "Exploration Exploitation Tradeoff", "Exploration-Exploitation Trade-off", "Exploration-Exploitation Tradeoff" ], "is_subclass_of": [ "Reinforcement Learning", "Decision Making", "Machine Learning" ], "wikilinks": [ "Reinforcement Learning", "Decision Making", "Markov Decision Process", "Reward Function", "Q Learning", "Monte Carlo Tree Search", "Active Learning", "Multi-Armed Bandit", "Thompson Sampling", "Upper Confidence Bound", "Epsilon-Greedy", "Policy Gradient Methods", "Temporal Difference Learning", "Value Function", "Deep Reinforcement Learning", "Bellman Equation", "Intrinsic Motivation", "Curiosity-Driven Exploration", "Regret Minimisation", "Bayesian Optimisation" ] }, { "id": "exploratory-concepts-seed-space", "title": "Exploratory Concepts Seed Space", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A collection of exploratory concepts and prototype application sketches within the AI and metaverse infrastructure domain, serving as a seed space for tooling, workflows, and agentic system designs that may be formalised into distinct ontology nodes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:exploratory-concepts-seed-space", "labels": [ "Exploratory Concepts Seed Space", "Ideas" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "AI Agent System" ] }, { "id": "exploratory-data-analysis", "title": "Exploratory Data Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Exploratory data analysis (EDA) is the practice of summarising, visualising, and interrogating a dataset to understand its structure, distributions, relationships, and anomalies before formal modelling. Introduced as a discipline by John Tukey, it emphasises graphical methods and descriptive statistics to generate hypotheses rather than confirm them. EDA is an early, iterative phase of the data-science workflow that informs data cleaning, feature engineering, and model selection.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:exploratory-data-analysis", "labels": [ "Exploratory Data Analysis" ], "is_subclass_of": [ "Data Analysis" ], "wikilinks": [] }, { "id": "exponential-moving-average", "title": "Exponential Moving Average", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An exponential moving average (EMA) is a weighted moving average that applies exponentially decreasing weights to successive observations in a time series, giving greater significance to recent data than to older data. It is computed recursively as a convex combination of the current observation and the previous EMA value, governed by a smoothing factor derived from a chosen window length. In blockchain and decentralised finance contexts the EMA is widely used to smooth on-chain price feeds, dampen oracle noise, and drive technical-analysis signals and adaptive parameters in automated trading and risk systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:exponential-moving-average", "labels": [ "Exponential Moving Average" ], "is_subclass_of": [ "Statistical Analysis" ], "wikilinks": [] }, { "id": "export-controls", "title": "Export Controls", "domain": "governance", "domain_name": "Governance", "definition": "Export controls are government-imposed legal restrictions on the cross-border transfer of specified goods, software, technology, and technical knowledge, typically for national-security, foreign-policy, or non-proliferation reasons. In the AI domain they increasingly govern advanced semiconductors, high-performance compute, and associated design tools, restricting which jurisdictions and entities may receive cutting-edge hardware. Compliance regimes such as the US EAR and the Wassenaar Arrangement define controlled-item lists, licensing requirements, and end-use screening.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:export-controls", "labels": [ "Export Controls", "BIS Export Controls", "Semiconductor Export Controls", "Technology Export Regulation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "extended-kalman-filter", "title": "Extended Kalman Filter", "domain": "robotics", "domain_name": "Robotics", "definition": "The Extended Kalman Filter is a recursive state estimator that applies the Kalman filter to non-linear systems by linearising the process and measurement models about the current estimate using first-order Taylor expansion via Jacobian matrices.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:extended-kalman-filter", "labels": [ "Extended Kalman Filter" ], "is_subclass_of": [ "Kalman Filter" ], "wikilinks": [ "State Estimation", "Bayesian Inference", "Localization", "Sensor Fusion", "Probabilistic Robotics", "Kalman Filter" ] }, { "id": "extended-producer-responsibility", "title": "Extended Producer Responsibility", "domain": "governance", "domain_name": "Governance", "definition": "Extended Producer Responsibility (EPR) is an environmental policy principle that makes manufacturers financially and operationally responsible for the entire lifecycle of their products, especially the take-back, recycling, and final disposal stages. By internalising end-of-life costs, EPR incentivises producers to design for durability, repairability, and recyclability. It is implemented through schemes such as packaging levies, electronic-waste recovery obligations, and deposit-return systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:extended-producer-responsibility", "labels": [ "Extended Producer Responsibility" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "extended-reality-xr", "title": "Extended Reality (XR)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Umbrella term encompassing all immersive technologies including Augmented Reality (AR), Virtual Reality (VR), and Mixed Reality (MR), representing the full spectrum from entirely physical to entirely virtual environments and all hybrid states between.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:extended-reality-xr", "labels": [ "Extended Reality (XR)", "Extended Reality Xr", "ExtendedRealityXR" ], "is_subclass_of": [ "Platform and Environment", "Immersive Technology" ], "wikilinks": [ "ACM Glossary", "Cross-Reality Transitions", "Graphics Processing", "Head-Mounted Display", "Input Device", "ISO 9241-940", "Augmented Reality (AR)", "ComputeLayer", "Computer Vision", "Human-Computer Interaction", "Immersive Experiences", "Immersive Technology", "InteractionDomain", "Mixed Reality (MR)", "NetworkLayer", "Presence", "Real-Time Rendering", "Reality-Virtuality Continuum", "Sensor Fusion", "Spatial Computing" ] }, { "id": "extended-reality", "title": "Extended Reality", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Extended Reality (XR) is an umbrella term encompassing Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR) \u2014 technologies that collectively span the full Reality-Virtuality Continuum first formalised by Milgram and Kishino (1994). XR systems alter or extend a user's perception of the physical environment through head-mounted displays, inside-out spatial tracking, real-time 3D rendering, and multi-modal input (gaze, gesture, voice, haptics). As the primary experiential layer of the spatial computing stack, XR bridges physical and digital environments across enterprise training, healthcare simulation, collaborative design, and consumer entertainment. Standardisation through Khronos OpenXR and W3C WebXR enables cross-vendor application portability and browser-based delivery.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:extended-reality", "labels": [ "Extended Reality", "Extended Reality System" ], "is_subclass_of": [ "Spatial Computing", "Extended Reality (XR)" ], "wikilinks": [] }, { "id": "external-ai-harness", "title": "External AI Harness", "domain": "ai", "domain_name": "Ai", "definition": "An external AI harness is an out-of-process orchestration framework that manages AI model inference via network APIs, message queues, or service meshes, providing process-boundary isolation, horizontal scalability, multi-model routing, fault tolerance, and multi-tenant governance at the cost of additional serialisation latency and inter-process communication overhead, making it the preferred architecture for enterprise-scale agentic deployments requiring auditability, model versioning, and independent component scaling.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:external-ai-harness", "labels": [ "External AI Harness" ], "is_subclass_of": [ "Agent Harness", "Evaluation Harness", "AI Infrastructure", "Distributed Systems" ], "wikilinks": [ "Agent Harness", "Internal AI Harness", "Model Serving", "API Gateway", "Microservices Architecture", "Container Orchestration", "Kubernetes", "AI Agent Coordination", "Multi-Agent Orchestration", "LLM Orchestration", "Inference Serving", "AI Inference", "Model Inference", "Large Language Models", "Agent Frameworks", "Agent Execution Sandboxes", "Model Context Protocol", "Agent-to-Agent Protocol", "REST API", "gRPC" ] }, { "id": "externally-owned-account", "title": "Externally Owned Account", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Externally Owned Account (EOA) is a type of Ethereum account controlled by a private key held outside the blockchain \u2014 typically by an end user via a wallet \u2014 as opposed to a contract account controlled by smart contract code. EOAs can initiate transactions, sign messages, and hold ether and tokens; they have no associated code. Every Ethereum transaction must originate from an EOA, making them the fundamental actor type in the Ethereum account model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:externally-owned-account", "labels": [ "Externally Owned Account" ], "is_subclass_of": [ "Blockchain Wallet" ], "wikilinks": [] }, { "id": "exteroceptive-sensor", "title": "Exteroceptive Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "Exteroceptive sensors are robot perception transducers that measure information about the external environment surrounding the robot rather than its internal kinematic, dynamic, or energetic state, providing the raw signals from which obstacle maps, semantic scene representations, object poses, t...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:exteroceptive-sensor", "labels": [ "Exteroceptive Sensor", "ExteroceptiveSensor" ], "is_subclass_of": [ "Perception and Sensing", "Sensor", "Robot Sensor", "Environmental Sensor", "Perception Transducer" ], "wikilinks": [ "3D Perception", "Acoustic Sensing", "Agricultural Robotics", "Autonomous Driving", "Bayesian Inference", "Bird's-Eye-View Perception", "Brohan et al 2023 RT-2", "Caesar et al 2020 nuScenes", "Calibration", "Campos et al 2021 ORB-SLAM3", "Computational Processing", "Coordinate Frame", "Data Interface", "Davison et al 2007 MonoSLAM", "Deep Neural Network", "Driver Layer", "Drone Navigation", "Embodied AI", "Environmental Properties", "Environmental Sensor" ] }, { "id": "extragalactic-astronomy", "title": "Extragalactic Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:extragalactic-astronomy", "labels": [ "Extragalactic Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "extreme-programming", "title": "Extreme Programming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Extreme Programming (XP) is an agile software-development methodology that emphasises short iterations, continuous feedback, and engineering discipline to deliver high-quality software responsive to changing requirements. Its core practices include test-driven development, pair programming, continuous integration, collective code ownership, and frequent small releases. XP treats good practices to an 'extreme' degree, for example applying continuous code review through constant pairing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:extreme-programming", "labels": [ "Extreme Programming", "Extreme Programming Explained (Beck 1999)" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "eye-contact-correction", "title": "Eye Contact Correction", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Eye Contact Correction is a computational technique that synthetically redirects a participant's gaze in a video stream so that they appear to be looking directly into the camera even when their eyes are focused on a screen display. It uses machine learning models to detect and synthesise eye and facial regions, overcoming the geometric offset between camera and display positions. This technology is critical for preserving natural eye contact as a social signal in remote video communication.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:eye-contact-correction", "labels": [ "Eye Contact Correction" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "eye-tracking", "title": "Eye Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical sensor hardware that measures gaze direction, pupil dilation, and eye movements to enable foveated rendering, attention analytics, and natural interaction in XR devices.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:eye-tracking", "labels": [ "Eye Tracking", "Eye Tracking Hardware", "Eye-Tracking Sensor" ], "is_subclass_of": [ "Interaction Technology", "Infrared Camera" ], "wikilinks": [ "ACM", "Attention Analytics", "Calibration System", "EdgeLayer", "ETSI GR ARF 010", "Eye Gesture Control", "Foveated Rendering", "Gaze-Based Interaction", "Graphics Processing Unit", "Head-Mounted Display", "High-Speed Camera", "Hot Mirror", "Human-Computer Interaction Framework", "Image Sensor", "Low-Latency Data Bus", "Perceptual Computing System", "Pupil Detection Algorithm", "Real-Time Processing Unit", "Vergence-Accommodation Matching", "Infrared Camera" ] }, { "id": "f1-score", "title": "F1 Score", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A classification performance metric representing the harmonic mean of precision and recall, providing a single score that balances a model's ability to avoid false positives (precision) with its ability to avoid false negatives (recall), calculated to give equal weight to both metrics whilst penalising extreme imbalances, particularly useful for comparing models or setting decision thresholds when both prediction reliability and completeness are important and when class distributions are imbalanced.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:f1-score", "labels": [ "F1 Score" ], "is_subclass_of": [ "Evaluation Metric" ], "wikilinks": [ "F-beta Score", "Macro F1", "Micro F1", "Precision-Recall Curve", "ROC-AUC", "Accuracy", "Confusion Matrix", "MetaverseDomain", "Model Performance", "Precision", "Recall" ] }, { "id": "f-2-pool", "title": "F2Pool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "F2Pool is one of the oldest Bitcoin mining pools, allowing miners to combine hash power and share block rewards. It also supports mining for other cryptocurrencies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:f-2-pool", "labels": [ "F2Pool" ], "is_subclass_of": [ "Mining Pool" ], "wikilinks": [ "Bitcoin Mining", "Transaction Validation", "Bitcoin Network", "Mining Pool", "https://www.f2pool.com", "https://www.f2pool.com/help" ] }, { "id": "fair-data-principles", "title": "FAIR Data Principles", "domain": "data", "domain_name": "Data", "definition": "The FAIR Data Principles are a set of guidelines stating that scientific and research data should be Findable, Accessible, Interoperable, and Reusable by both humans and machines. They emphasise persistent identifiers, rich machine-readable metadata, standardised vocabularies, and clear usage licences to maximise the long-term value of data. FAIR is widely adopted in research-data management, open science, and knowledge-graph and ontology engineering.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fair-data-principles", "labels": [ "FAIR Data Principles", "FAIR Principles" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "faiss", "title": "FAISS", "domain": "ai", "domain_name": "Ai", "definition": "FAISS (Facebook AI Similarity Search) is an open-source library for efficient similarity search and clustering of dense vectors, providing exact and approximate nearest-neighbour algorithms that scale to billions of embeddings. It implements index structures such as inverted files and product quantisation, with GPU acceleration for high-throughput retrieval. FAISS is widely used as the vector index backend for semantic search and retrieval-augmented generation pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:faiss", "labels": [ "FAISS", "Faiss" ], "is_subclass_of": [ "Vector Index" ], "wikilinks": [] }, { "id": "fasb-asu-2023-08", "title": "FASB ASU 2023-08", "domain": "finance", "domain_name": "Finance", "definition": "FASB Accounting Standards Update 2023-08 is a US accounting standard issued by the Financial Accounting Standards Board that requires public companies to measure certain cryptocurrency assets at fair value and recognise gains and losses in net income each reporting period. The standard applies to fungible crypto assets listed on active markets, effective for fiscal years beginning after 15 December 2024. It represents a significant departure from the previous indefinite-lived intangible asset treatment that suppressed recognised gains.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:fasb-asu-2023-08", "labels": [ "FASB ASU 2023-08" ], "is_subclass_of": [ "Compliance Standards" ], "wikilinks": [] }, { "id": "fatf-40-recommendations", "title": "FATF 40 Recommendations", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The FATF 40 Recommendations are the international standards on combating money laundering, terrorist financing, and proliferation financing issued by the Financial Action Task Force. They define a comprehensive framework covering customer due diligence, beneficial-ownership transparency, suspicious-transaction reporting, supervision, and international cooperation. National regimes implement them to achieve compliance assessed through FATF mutual evaluations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fatf-40-recommendations", "labels": [ "FATF 40 Recommendations", "Forty Recommendations" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "fatf-guidance-on-virtual-assets", "title": "FATF Guidance on Virtual Assets", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The FATF Guidance on Virtual Assets is the Financial Action Task Force's risk-based interpretation of how its anti-money-laundering standards apply to virtual assets and virtual-asset service providers (VASPs). It defines key terms, requires VASP licensing or registration and supervision, and introduces the 'travel rule' obliging the transfer of originator and beneficiary information with transactions. The guidance shapes how exchanges, custodians, and DeFi arrangements are regulated worldwide.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fatf-guidance-on-virtual-assets", "labels": [ "FATF Guidance on Virtual Assets", "FATF DeFi Guidance" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "fatf-recommendation-16", "title": "FATF Recommendation 16", "domain": "finance", "domain_name": "Finance", "definition": "The specific Financial Action Task Force recommendation, known as the Travel Rule, requiring financial institutions and virtual asset service providers to obtain, hold, and transmit accurate originator and beneficiary information alongside wire transfers and virtual asset transfers, so that anti-money-laundering and counter-terrorist-financing authorities can trace funds across institutions and borders.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:fatf-recommendation-16", "labels": [ "FATF Recommendation 16" ], "is_subclass_of": [ "FATF Recommendations" ], "wikilinks": [ "FATF Recommendations", "Travel Rule", "Anti-Money Laundering" ] }, { "id": "fatf-recommendations", "title": "FATF Recommendations", "domain": "governance", "domain_name": "Governance", "definition": "The FATF Recommendations are international standards on combating money laundering, terrorist financing and proliferation financing issued by the Financial Action Task Force.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fatf-recommendations", "labels": [ "FATF Recommendations", "FATF AML CFT Standards", "FATF Recommendation 10", "FATF Recommendation 15" ], "is_subclass_of": [ "Anti-Money Laundering" ], "wikilinks": [ "Regulatory Frameworks", "Travel Rule", "Know Your Customer", "Anti-Money Laundering", "https://www.fatf-gafi.org/en/publications/Fatfrecommendations/Fatf-recommendations.html", "https://www.fatf-gafi.org/" ] }, { "id": "fatf-travel-rule", "title": "fatf travel rule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The FATF Travel Rule (Recommendation 16 extended to virtual assets) requires Virtual Asset Service Providers (VASPs) to collect, verify, and transmit originator and beneficiary identity information alongside cryptocurrency transactions that meet or exceed a jurisdictional threshold, typically USD/EUR 1,000. Issued by the Financial Action Task Force, it extends the longstanding wire-transfer obligation \u2014 rooted in the FATF 40 Recommendations and aligned with SWIFT messaging norms \u2014 to crypto-asset transfers, obligating sending VASPs to share identifying data with receiving VASPs before or simultaneously with the transfer. Compliance demands interoperability protocols, counterparty discovery infrastructure, and shared data standards such as IVMS 101 between exchanges, custodians, and wallet providers across different jurisdictions. The rule has been progressively transposed into national and regional law, including the EU Transfer of Funds Regulation, UK Money Laundering Regulations, and Singapore Payment Services Act.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:fatf-travel-rule", "labels": [ "FATF Travel Rule", "FATF Recommendation 16 Travel Rule" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "fatf", "title": "fatf", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Action Task Force (FATF) is an intergovernmental policy-making body established in 1989 by the G7 that sets internationally recognised standards for combating money laundering, terrorist financing, and proliferation financing. Its Forty Recommendations and Nine Special Recommendations constitute the normative AML/CFT/CPF framework that member jurisdictions implement through national legislation and regulatory action. FATF conducts mutual evaluations of members' technical compliance and effectiveness, publishes grey and black lists of high-risk jurisdictions, and issues binding guidance on emerging risks including virtual assets, decentralised finance, and central bank digital currencies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:fatf", "labels": [ "FATF", "FATF (Financial Action Task Force)" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "fatfrecommendations", "title": "FATFRecommendations", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The FATF Recommendations are the comprehensive international standards established by the Financial Action Task Force\u2014an intergovernmental body founded in 1989 comprising 40 member jurisdictions\u2014for combating money laundering, terrorist financing, and proliferation financing, covering customer due diligence, suspicious transaction reporting, record-keeping, and the regulation of virtual asset service providers through Recommendation 15 and the Travel Rule. They constitute the primary global anti-financial-crime framework that national regulators translate into domestic law, creating compliance obligations for financial institutions and cryptocurrency exchanges.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fatfrecommendations", "labels": [ "FATFRecommendations" ], "is_subclass_of": [ "Governance and Regulation", "AML KYC Compliance" ], "wikilinks": [ "Anti-Money Laundering", "BC-0456-virtual-asset-service-providers", "BC-0457-aml-kyc-compliance", "BC-0482-eu-mica-regulation", "BC-0483-us-regulatory-framework", "Binance", "Blockchain Analytics", "CBDC", "Coinbase", "ComplianceDomain", "Counter-Terrorist Financing", "Cross-Chain Bridges", "Cryptocurrency Exchange", "Cryptocurrency Wallets", "Dash", "DeFi", "Decentralised Finance", "EU Transfer of Funds Regulation June 2023", "FATF", "FATF 40 Recommendations February 2012" ] }, { "id": "fbx", "title": "FBX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "FBX is a proprietary 3D asset interchange format used to transfer geometry, materials, skeletal rigs and animation between digital content creation tools and game engines. Developed by Kaydara and now owned by Autodesk, it stores scene graphs, mesh data, skinning weights, keyframed animation tracks, cameras, and lights, making it the dominant interchange format across 3D production and real-time engine pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fbx", "labels": [ "FBX", "FBX Format" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Computer Graphics", "Skeletal Animation", "Game Engine", "glTF", "Universal Scene Description" ] }, { "id": "fca-consumer-duty", "title": "FCA Consumer Duty", "domain": "governance", "domain_name": "Governance", "definition": "The FCA Consumer Duty is a regulatory standard set by the UK Financial Conduct Authority requiring financial firms to deliver good outcomes for retail customers. It mandates that firms act in good faith, avoid foreseeable harm, support customers in pursuing their financial objectives, and demonstrate fair value, clear communications, and adequate support. The Duty raises the bar above prior 'treating customers fairly' expectations and applies across products, pricing, and service channels.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fca-consumer-duty", "labels": [ "FCA Consumer Duty", "Consumer Duty", "Consumer Duty Compliance", "Consumer Duty FCA" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "fca", "title": "fca", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Conduct Authority (FCA) is the UK's statutory conduct regulator for financial services and markets, established under the Financial Services and Markets Act 2000 and reconstituted by the Financial Services Act 2012. Operating independently of government and funded by industry fees, it authorises firms, supervises conduct across retail and wholesale markets, and enforces anti-money-laundering obligations. In the digital asset sphere the FCA registers crypto-asset businesses under the Money Laundering Regulations, regulates crypto financial promotions, and coordinates with the Bank of England and HM Treasury on emerging regulatory frameworks for stablecoins, custody, and tokenised finance.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:fca", "labels": [ "FCA", "FCA Regulatory Regime" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "fda", "title": "FDA", "domain": "governance", "domain_name": "Governance", "definition": "The Food and Drug Administration (FDA) is a United States federal regulatory agency within the Department of Health and Human Services responsible for protecting public health by ensuring the safety, efficacy, and security of human and veterinary drugs, biological products, medical devices, the nation's food supply, cosmetics, and products that emit radiation. The agency enforces the Federal Food, Drug, and Cosmetic Act alongside a portfolio of additional statutes, conducting pre-market review and post-market surveillance across its regulated domains. Its approval pathways \u2014 including New Drug Applications, Biologics License Applications, and Premarket Approval \u2014 establish globally influential benchmarks that shape international pharmaceutical regulation, medical-device standards, and increasingly the governance of AI-enabled diagnostics and digital health software.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:fda", "labels": [ "FDA" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Governance", "Synthetic Biology", "owl:Thing", "https://www.fda.gov", "https://www.fda.gov/about-fda" ] }, { "id": "fid-benchmark-protocol", "title": "FID Benchmark Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The FID benchmark protocol is the standardised methodology for evaluating generative image models using the Fr\u00e9chet Inception Distance, which compares the distribution of generated images with that of real images in the feature space of a pretrained Inception network. The protocol fixes the feature extractor, the number of samples, and preprocessing so that scores are comparable across models; lower FID indicates greater similarity to real data. It is the de facto standard for benchmarking GANs and diffusion models.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:fid-benchmark-protocol", "labels": [ "FID Benchmark Protocol" ], "is_subclass_of": [ "Evaluation Metric", "Benchmarks", "Benchmark", "Image Quality Assessment", "Distributional Similarity Metric" ], "wikilinks": [ "Diffusion Models", "Generative Adversarial Networks", "Evaluation Metric", "Inception v3", "Fr\u00e9chet Distance", "Image Synthesis", "Generative Model", "Deep Learning", "Convolutional Neural Networks", "Inception Score", "CLIP Encoder", "Clean-FID", "Precision-Recall", "Variational Autoencoder", "ImageNet", "CIFAR-10", "Fr\u00e9chet Audio Distance", "Fr\u00e9chet Video Distance", "CMMD", "Text-to-Image" ] }, { "id": "fido-alliance", "title": "FIDO Alliance", "domain": "security", "domain_name": "Security", "definition": "An industry consortium that develops open authentication standards\u2014UAF, U2F, FIDO2, and WebAuthn\u2014designed to replace passwords with phishing-resistant public-key cryptography, enabling strong authentication across browsers, devices, and online services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fido-alliance", "labels": [ "FIDO Alliance" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "Asymmetric Cryptography", "Passwordless Authentication", "Multi-Factor Authentication", "Authentication", "Standards Body" ] }, { "id": "fido2", "title": "FIDO2", "domain": "security", "domain_name": "Security", "definition": "FIDO2 is an open authentication standard developed by the FIDO Alliance and W3C that enables passwordless, phishing-resistant authentication using public-key cryptography. It consists of two components: the W3C Web Authentication API (WebAuthn), which defines the browser and platform interface for creating and using public key credentials, and the Client to Authenticator Protocol (CTAP2), which defines the communication between a platform and an external authenticator such as a hardware security key or passkey-capable device. FIDO2 credentials are bound to a specific relying party origin, making them immune to phishing, and private keys never leave the authenticator device, eliminating the credential theft risk associated with password databases.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fido2", "labels": [ "FIDO2", "FIDO2 WebAuthn" ], "is_subclass_of": [ "Authentication Standards" ], "wikilinks": [] }, { "id": "fipa-acl", "title": "FIPA ACL", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "FIPA ACL is the Agent Communication Language standardised by the Foundation for Intelligent Physical Agents, defining a message format and a library of communicative acts that autonomous software agents use to exchange information and coordinate behaviour. Each message carries a performative such as inform, request, or propose, drawn from speech-act theory, together with parameters identifying sender, receiver, content language, and ontology. It provides the interoperability layer that lets heterogeneous agents in a multi-agent system understand one another's intentions.", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:fipa-acl", "labels": [ "FIPA ACL", "FIPA-ACL" ], "is_subclass_of": [ "Agent Communication Language", "Communication Protocol", "Inter-Agent Communication" ], "wikilinks": [ "Agent Communication Language", "Inter-Agent Communication", "Multi-Agent System", "Ontology", "Speech Act Theory", "Autonomous Agent", "Multi-Agent Coordination", "Communication Protocol", "KQML", "Interoperability", "Contract Net Protocol", "Distributed Systems", "Agent Platform", "Semantic Web", "Message Passing", "Knowledge Interchange Format", "Model Context Protocol", "Agent2Agent Protocol", "CLI Multi-Agent Systems", "Distributed Decision Making" ] }, { "id": "fipa", "title": "FIPA", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "FIPA, the Foundation for Intelligent Physical Agents, is a standards body \u2014 now an IEEE Computer Society standards committee \u2014 that produced specifications for interoperable software agents and multi-agent systems. Its best-known output is FIPA-ACL, an agent communication language with formally defined performatives and interaction protocols. FIPA standards enable heterogeneous agents from different developers to discover, communicate, and coordinate with one another.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fipa", "labels": [ "FIPA" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "fips-140-3", "title": "FIPS 140-3", "domain": "standards", "domain_name": "Standards", "definition": "FIPS 140-3 specifies security requirements for cryptographic modules used to protect sensitive information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fips-140-3", "labels": [ "FIPS 140-3", "FIPS 140" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "fips-186-5", "title": "FIPS 186-5", "domain": "standards", "domain_name": "Standards", "definition": "Revision 5 of the NIST Digital Signature Standard, specifying approved digital signature algorithms including ECDSA and EdDSA. It defines requirements for generating and verifying digital signatures.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fips-186-5", "labels": [ "FIPS 186-5" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "flp-impossibility", "title": "FLP Impossibility", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The FLP Impossibility (Fischer, Lynch, Paterson 1985) is a foundational theorem in distributed computing proving that no deterministic protocol can solve the [[Consensus]] problem in a fully asynchronous message-passing system if even one process may crash. The result arises because message delays are unbounded, making it impossible to distinguish a crashed process from a slow one, so any protocol that always terminates can be manipulated into a state of permanent indecision. Consequently, every practical [[Consensus Algorithm]] must relax at least one of the FLP assumptions \u2014 typically by introducing partial synchrony, randomisation, or probabilistic termination.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:flp-impossibility", "labels": [ "FLP Impossibility" ], "is_subclass_of": [ "Distributed Consensus" ], "wikilinks": [ "Distributed Consensus", "Consensus Algorithm", "Fault Tolerance" ] }, { "id": "fmea", "title": "FMEA", "domain": "robotics", "domain_name": "Robotics", "definition": "Failure Mode and Effects Analysis (FMEA) is a structured, proactive engineering methodology for identifying potential failure modes in a product, process, or system, assessing their causes and effects, and prioritising them for mitigation. Each failure mode is rated by severity, occurrence, and detectability, often combined into a Risk Priority Number that ranks risks for corrective action. FMEA is widely used in reliability engineering, safety-critical design, and quality management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fmea", "labels": [ "FMEA", "Failure Mode and Effects Analysis" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "fpga", "title": "FPGA", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Field-Programmable Gate Array (FPGA) is a reconfigurable integrated circuit that can be configured after manufacturing to implement custom digital logic, offering a middle ground between flexible general-purpose processors and fixed-function ASICs. FPGAs are widely used for low-latency AI inference acceleration, edge computing, real-time signal processing, and hardware prototyping due to their reconfigurable architecture, energy efficiency, and deterministic execution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:fpga", "labels": [ "FPGA", "FPGA Acceleration" ], "is_subclass_of": [ "Inference Hardware" ], "wikilinks": [ "Custom Neural Architectures", "Autonomous Robot", "Digital Twin", "Inference Hardware" ] }, { "id": "fsa", "title": "FSA", "domain": "governance", "domain_name": "Governance", "definition": "A financial regulatory authority, commonly the abbreviation for a national financial services agency that supervises banks, securities, and other financial institutions. The name refers to several distinct national regulators.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fsa", "labels": [ "FSA" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Financial Regulation", "Regulatory Compliance", "Monetary Policy" ] }, { "id": "fsb-cross-border-payments-roadmap", "title": "FSB Cross-Border Payments Roadmap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The FSB Cross-Border Payments Roadmap is a coordinated programme led by the Financial Stability Board to make international payments faster, cheaper, more transparent, and more inclusive. Endorsed by the G20, it sets quantitative targets and prioritises work on interoperability, legal and regulatory frameworks, and emerging payment infrastructures including CBDCs. It frames how new settlement rails are evaluated against global policy goals.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fsb-cross-border-payments-roadmap", "labels": [ "FSB Cross-Border Payments Roadmap" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "fsb", "title": "FSB", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Stability Board (FSB) is an international body established in 2009 by the G20 that monitors and makes recommendations about the global financial system. It coordinates the work of national financial authorities and international standard-setting bodies to develop and promote effective regulatory, supervisory, and other financial sector policies. The FSB addresses vulnerabilities in financial markets, oversees the implementation of agreed reforms, and assesses systemic risks arising from entities, activities, and instruments across the global financial system, including emerging risks from crypto-assets, non-bank financial intermediation, and climate-related financial exposures.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:fsb", "labels": [ "FSB" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Financial Regulation", "Financial Stability", "Systemic Risk", "Financial System" ] }, { "id": "fsdp", "title": "FSDP", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Fully Sharded Data Parallel (FSDP) is a distributed training technique implemented natively in PyTorch that shards a model's parameters, gradients, and optimizer states across multiple GPUs or nodes, enabling training of models too large for a single device's memory. Each worker holds only a fraction of every parameter tensor and gathers the full tensor on demand via all-gather collective operations before each layer's forward computation, immediately discarding the assembled tensor after use so that only the local shard is persistently stored. FSDP implements the ZeRO Stage 3 memory partitioning algorithm as a first-class PyTorch primitive, and integrates with sharded Distributed Checkpointing (DCP) so that each rank writes and reads its own parameter shard independently, enabling checkpoint resharding across different cluster topologies at load time.", "entityType": "Class", "qualityScore": 0.89, "maturity": "established", "iri": "urn:ngm:class:fsdp", "labels": [ "FSDP", "PyTorch FSDP", "PyTorch FSDP API" ], "is_subclass_of": [ "Distributed Training", "AI Infrastructure", "Data Parallelism", "Machine Learning" ], "wikilinks": [ "Distributed Training", "Checkpoints", "Data Parallelism", "Model Parallelism", "ZeRO Redundancy Optimiser", "DeepSpeed", "GPU Compute", "Collective Communication", "Gradient Checkpointing", "Mixed Precision Training", "Large Language Models", "PyTorch", "Megatron-LM", "Pipeline Parallelism", "Tensor Parallelism", "Gradient Synchronisation", "NVIDIA NCCL", "Fine-tuning", "Foundation Model", "TorchTitan" ] }, { "id": "fsma-2023", "title": "FSMA 2023", "domain": "governance", "domain_name": "Governance", "definition": "The Financial Services and Markets Act 2023, a United Kingdom statute that updates the framework for financial services regulation after departure from the European Union. It grants regulators new powers, including over digital assets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fsma-2023", "labels": [ "FSMA 2023" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Financial Regulation", "Regulatory Compliance", "Bank of England" ] }, { "id": "ftx", "title": "FTX", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cryptocurrency exchange founded in 2019 that collapsed in November 2022 amid a liquidity crisis and revelations of misuse of customer funds. Its failure led to bankruptcy proceedings and criminal convictions of its leadership.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ftx", "labels": [ "FTX" ], "is_subclass_of": [ "DeFi and Economics", "owl:Thing" ], "wikilinks": [ "Coinbase", "owl:Thing" ] }, { "id": "face-recognition", "title": "Face Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Face recognition is a biometric identification and verification technology that locates human faces in images or video frames, extracts a compact numerical embedding of facial geometry and texture, and compares that embedding against one or more enrolled templates to establish identity. It is a specialised subfield of Computer Vision and Biometric Authentication that combines deep convolutional feature learning with metric-learning loss functions such as ArcFace and AdaFace to achieve sub-second, high-accuracy identification at scale.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:face-recognition", "labels": [ "Face Recognition", "Face Match", "Face Recognition Model" ], "is_subclass_of": [ "Biometric Authentication", "Computer Vision" ], "wikilinks": [ "Computer Vision", "Identity Verification", "Biometric Authentication" ] }, { "id": "face-swap", "title": "Face Swap", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Face Swap is the family of computer vision techniques that transfers the identity of a source face onto a target image, video frame, or live stream while preserving the target's pose, expression, illumination, occlusions, and surrounding scene context, formalised as the conditional generation pro...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:face-swap", "labels": [ "Face Swap", "Face Swapping", "FaceSwap" ], "is_subclass_of": [ "AI Application", "Computer Vision", "Generative AI", "Conditional Image Synthesis", "Identity Transfer", "Synthetic Media", "Face Manipulation" ], "wikilinks": [ "Adaptive Instance Normalisation", "AlgorithmLayer", "Anonymisation", "ArcFace", "ArcFace Embedding", "Attribute Encoder", "Attribute Preservation", "Avatar Generation", "Blending Module", "C2PA Content Credentials", "Celeb-DF Benchmark", "Chen et al. 2020 SimSwap", "Chesney & Citron 2019 Deep Fakes California Law Review", "CodeFormer", "ComputerVisionDomain", "Conditional Image Synthesis", "ControlNet", "Convolutional Neural Networks", "CosFace", "Cross-Attention Conditioning" ] }, { "id": "faceted-classification", "title": "Faceted Classification", "domain": "data", "domain_name": "Data", "definition": "Faceted classification is a method of organising information by decomposing it into multiple independent categories, or facets, such as subject, format, date and location, which can be combined in any order to describe or retrieve an item. Unlike a single hierarchical taxonomy, it allows an item to be reached through several different combinations of attributes, which better suits heterogeneous and evolving collections. It is a foundational technique in information architecture and underlies faceted search interfaces and controlled vocabularies.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:faceted-classification", "labels": [ "Faceted Classification" ], "is_subclass_of": [ "Classification" ], "wikilinks": [] }, { "id": "facial-action-coding-system", "title": "Facial Action Coding System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Facial Action Coding System (FACS) is a comprehensive, anatomically grounded taxonomy for objectively describing and encoding visible facial movements by decomposing expressions into discrete Action Units (AUs), each corresponding to the contraction of one or more specific facial muscles. Developed originally by Swedish anatomist Carl-Herman Hjortsj\u00f6 and systematised by psychologists Paul Ekman and Wallace V. Friesen in 1978, FACS provides a muscle-based, observer-independent vocabulary for facial behaviour that separates objective biomechanical measurement from subjective emotional interpretation. It is the foundational measurement framework for automated Emotion Recognition, avatar facial animation, clinical pain and depression assessment, and affective computing research.", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:facial-action-coding-system", "labels": [ "Facial Action Coding System", "Action Unit Encoding" ], "is_subclass_of": [ "Affective Computing", "Computer Vision" ], "wikilinks": [ "Emotion Recognition", "Affective Computing", "Computer Vision", "Face Recognition", "Empathetic AI", "Deep Learning", "Convolutional Neural Network", "Action Recognition", "Human-Computer Interaction" ] }, { "id": "facial-animation", "title": "Facial Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Facial animation is the discipline of generating expressive, lifelike movement of a digital character's face, including lip synchronisation, eye motion, and emotional expression. It commonly combines blend-shape (morph-target) deformation, facial rigs, and performance-capture data to drive realistic results. Facial animation is essential for believable avatars, virtual humans, and immersive experiences in spatial computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:facial-animation", "labels": [ "Facial Animation" ], "is_subclass_of": [ "Character Animation" ], "wikilinks": [] }, { "id": "facial-capture-system", "title": "Facial Capture System", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A facial capture system is hardware and software that records the detailed movements of a performer's face and transfers them onto a digital character or model. Such systems use head-mounted cameras, structured light, marker-based tracking, or markerless computer-vision pipelines to reconstruct expressions, lip movement, and micro-deformations in real time or in post-production. They are central to high-fidelity digital characters in film, games, and virtual production.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:facial-capture-system", "labels": [ "Facial Capture System" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "facial-recognition", "title": "Facial Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Facial Recognition is a computer vision technology that automatically detects, aligns, and identifies or verifies individuals by analysing facial features extracted from images or video frames, producing compact numerical embeddings that encode discriminative facial geometry and appearance. These embeddings are compared against a gallery of known identities using similarity metrics (cosine similarity or L2 distance), operating in one-to-one verification mode (confirming claimed identity) or one-to-many identification mode (searching against a database of enrolled individuals). Modern systems rely on deep convolutional neural networks trained on large-scale labelled datasets and are subject to increasing regulatory oversight relating to accuracy disparities across demographic groups, privacy obligations, and prohibition in high-risk contexts under frameworks such as the EU AI Act.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:facial-recognition", "labels": [ "Facial Recognition", "Facial Recognition Sensor", "Facial Recognition Technology" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "factor-graph", "title": "Factor Graph", "domain": "robotics", "domain_name": "Robotics", "definition": "A factor graph is a bipartite graphical model that factorises a global function into a product of local factors, connecting variable nodes to the factor nodes that constrain them. It makes the structure of an inference problem explicit and supports efficient message-passing algorithms. In robotics it is the dominant representation for state estimation problems such as SLAM and sensor fusion.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:factor-graph", "labels": [ "Factor Graph" ], "is_subclass_of": [ "Graphical Model", "Navigation and Planning" ], "wikilinks": [] }, { "id": "failover", "title": "Failover", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Failover is the automatic switching of operation to a standby system, server, or network path when the active component fails or becomes unreachable, minimising service disruption. It is a core mechanism for achieving high availability, typically implemented through health checks, heartbeat monitoring, and standby replicas that can assume traffic within seconds. Failover strategies range from active-passive, where a standby remains idle until needed, to active-active, where multiple nodes share load and absorb failures without a distinct switchover step. Anycast routing achieves failover at the network layer by withdrawing route advertisements for unreachable nodes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:failover", "labels": [ "Failover" ], "is_subclass_of": [ "High Availability" ], "wikilinks": [] }, { "id": "failure-mode-and-effects-analysis", "title": "Failure Mode And Effects Analysis", "domain": "standards", "domain_name": "Standards", "definition": "Failure Mode and Effects Analysis (FMEA) is a systematic, bottom-up technique for identifying potential failure modes of a system, product, or process, assessing their causes and effects, and prioritising corrective action. Each failure mode is rated for severity, occurrence, and detectability, often combined into a risk priority number that ranks issues for mitigation. FMEA is widely used in safety, reliability, and quality engineering, and is a core analysis within functional-safety lifecycles defined by standards such as IEC 61508 and ISO 26262.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:failure-mode-and-effects-analysis", "labels": [ "Failure Mode And Effects Analysis", "Failure Mode and Effects Analysis" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [] }, { "id": "fair-adjudication", "title": "Fair Adjudication", "domain": "governance", "domain_name": "Governance", "definition": "Fair adjudication is the principle and process by which disputes are decided impartially, with due process, transparent rules, and the absence of bias toward any party. In legal and decentralised-arbitration contexts it requires neutral decision-makers, equal opportunity to present evidence, and reasoned, consistent rulings. It is a foundational requirement for legitimate dispute-resolution and arbitration mechanisms, including those implemented on-chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fair-adjudication", "labels": [ "Fair Adjudication" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "fairness-oecd", "title": "Fairness (OECD)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The OECD AI Principle (1.2) requiring that AI systems do not create or reinforce unfair bias, discrimination, or disparate impacts on individuals or groups based on protected characteristics. Fairness mandates proactive safeguards\u2014pre-processing reweighting, in-processing constraints, post-processing calibration\u2014ensuring equitable treatment whilst respecting legitimate, evidence-based differentiation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:fairness-oecd", "labels": [ "Fairness (OECD)" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "fairness-accuracy-tradeoffs", "title": "Fairness Accuracy Tradeoffs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The fundamental tension in supervised machine learning between maximising predictive accuracy and satisfying fairness constraints, characterised by the Pareto frontier of achievable (accuracy, fairness) pairs. Imposing fairness constraints restricts the hypothesis space, excluding models that achieve accuracy through reliance on protected-attribute correlations. The magnitude of the accuracy cost depends on the chosen fairness criterion (demographic parity, equalised odds, calibration), the base rate differences between groups, and the flexibility of the model class.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fairness-accuracy-tradeoffs", "labels": [ "Fairness Accuracy Tradeoffs" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Chouldechova (2017)", "Corbett-Davies et al. (2017)", "Kleinberg et al. (2017)", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin" ] }, { "id": "fairness-assessment", "title": "Fairness Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Fairness assessment is the systematic evaluation of whether an AI or machine-learning system produces equitable outcomes across protected groups and individuals. It quantifies disparities using metrics such as demographic parity, equalised odds, and predictive parity, often computed from confusion-matrix statistics segmented by subgroup. The results inform mitigation, auditing, and governance decisions about model deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fairness-assessment", "labels": [ "Fairness Assessment", "Fairness Assessment Criteria" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "fairness-auditing-tools", "title": "Fairness Auditing Tools", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Fairness Auditing Tools are software libraries, platforms, and frameworks designed to detect, measure, and mitigate algorithmic bias in AI systems through automated analysis, visualisation, and intervention capabilities. They operationalise fairness metrics\u2014such as demographic parity, equalised odds, and predictive parity\u2014across protected attribute groups, supporting both pre-deployment model assessments and continuous production monitoring.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fairness-auditing-tools", "labels": [ "Fairness Auditing Tools" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "AIF360", "Fairlearn", "IEEE P7003-2021", "ISO/IEC TR 24027", "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer" ] }, { "id": "fairness-constraints", "title": "Fairness Constraints", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Fairness Constraints are mathematical formalizations of equitable treatment requirements in AI systems, expressed as conditions that model predictions must satisfy with respect to protected attributes such as race, gender, or age. The three canonical constraint families are Independence (demographic parity: predictions are statistically independent of protected attributes), Separation (equalized odds: predictions are independent of protected attributes conditional on the true label), and Sufficiency (calibration: true labels are independent of protected attributes conditional on predictions). These constraints are incorporated into model training as regularisation penalties or constrained optimisation objectives, and are subject to fundamental incompatibility theorems when base rates differ across protected groups.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fairness-constraints", "labels": [ "Fairness Constraints" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Barocas et al. (2019)", "Chouldechova (2017)", "Hardt et al. (2016)", "AIEthicsDomain", "Blockchain", "ConceptualLayer", "Digital Twin" ] }, { "id": "fairness-metrics", "title": "Fairness Metrics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantitative measures and mathematical frameworks used to evaluate whether an AI system produces equitable outcomes across demographic groups. Core metrics include demographic parity, equalized odds, equal opportunity, and predictive parity; selection among them depends on context and regulatory requirements, as metrics can conflict and no single criterion satisfies all fairness definitions simultaneously.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:fairness-metrics", "labels": [ "Fairness Metrics", "Algorithmic Fairness Metrics", "Fairness Metric" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "IEEE P7003-2021", "ISO/IEC TR 24027", "NIST SP 1270", "AIEthicsDomain", "Blockchain", "ConceptualLayer" ] }, { "id": "fairness-in-machine-learning", "title": "Fairness in Machine Learning", "domain": "ai", "domain_name": "Ai", "definition": "Fairness in machine learning is the study and practice of ensuring that learned models do not produce systematically disadvantageous outcomes for individuals or groups defined by protected attributes such as race, gender, or age. It encompasses formal fairness criteria \u2014 including demographic parity, equalised odds, and individual fairness \u2014 and the techniques used to measure and mitigate disparate impact before, during, or after training. The field is central to responsible and ethical AI, balancing accuracy with equitable treatment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:fairness-in-machine-learning", "labels": [ "Fairness in Machine Learning", "Fairness In Machine Learning" ], "is_subclass_of": [ "Responsible AI" ], "wikilinks": [] }, { "id": "fairness", "title": "Fairness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The property of an AI system whereby it produces equitable outcomes and avoids creating or reinforcing unjustifiable disparities across different demographic groups or individuals, measured through various mathematical definitions and ethical principles including demographic parity, equalized odds, equal opportunity, calibration, and individual fairness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:fairness", "labels": [ "Fairness", "Bias and Fairness Analysis", "Model Fairness", "Network Fairness" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Bias Detection", "Bias Mitigation", "Equal Treatment", "Non-discrimination", "MetaverseDomain" ] }, { "id": "fallback", "title": "Fallback", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A fault-tolerance strategy in which a system, upon detecting that its primary path has failed, timed out, or returned an unacceptable result, automatically switches to a predefined alternative path that provides reduced but still useful behaviour. In agent orchestration the alternative is typically a cheaper or more reliable model, a cached response, a simpler tool, or a deterministic default, invoked so that the overall workflow degrades gracefully rather than failing outright.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:fallback", "labels": [ "Fallback" ], "is_subclass_of": [ "Fault Tolerance", "FaultTolerance" ], "wikilinks": [ "FaultTolerance", "GracefulDegradation", "CircuitBreaker", "Resilience" ] }, { "id": "false-negative", "title": "False Negative", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A false negative is a classification error in which a model incorrectly predicts the negative class for an instance that actually belongs to the positive class. It is a fundamental cell of the confusion matrix, often denoted FN, and directly reduces recall (sensitivity). In high-stakes domains such as medical screening or fraud detection, false negatives represent missed true cases and frequently carry asymmetric cost relative to false positives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:false-negative", "labels": [ "False Negative" ], "is_subclass_of": [ "Model Evaluation", "Model Evaluation Results" ], "wikilinks": [] }, { "id": "false-positive", "title": "False Positive", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A false positive is an outcome in which a classifier or detection system reports the positive class for an instance that actually belongs to the negative class. It is one of the four cells of a confusion matrix and corresponds to a Type I error in statistical terms. The rate of false positives directly shapes precision and specificity and is traded off against false negatives when a decision threshold is tuned.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:false-positive", "labels": [ "False Positive" ], "is_subclass_of": [ "Confusion Matrix", "AI Technique" ], "wikilinks": [] }, { "id": "fan-out", "title": "Fan-Out", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A concurrency and orchestration pattern in which a single request or task is split into many independent sub-tasks that are dispatched simultaneously to multiple workers, agents, or services, then optionally recombined by a downstream fan-in step. Fan-out trades higher aggregate resource consumption for reduced wall-clock latency and throughput, and is the structural basis for parallel sub-agent execution, scatter-gather search, and map-style batch processing in agent systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:fan-out", "labels": [ "Fan-Out" ], "is_subclass_of": [ "Design Pattern", "DesignPattern" ], "wikilinks": [ "DesignPattern", "TaskDelegation", "MultiAgentOrchestration", "DistributedComputing" ] }, { "id": "farcaster", "title": "Farcaster", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Farcaster is a sufficiently decentralised social networking protocol built on Ethereum and Optimism that anchors user identity and account data on-chain while storing social graph content and messages off-chain across a peer-to-peer network of Hubs. It separates identity (Farcaster ID, FID) from the client application layer, enabling permissionless third-party clients such as Warpcast to build on shared social data without platform lock-in. The protocol specifies message encoding via a data availability layer called Hubs and enforces message ordering through a CRDT-based conflict resolution mechanism, making it a credibly neutral substrate for decentralised social applications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:farcaster", "labels": [ "Farcaster" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Decentralised Identity", "Web3", "Ethereum" ] }, { "id": "fashion", "title": "Fashion", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fashion, as an AI-application domain, encompasses the convergence of machine learning, computer vision, generative models, and blockchain supply-chain infrastructure with the global apparel, luxury, and textile industries, valued at approximately 1.7 and forecast to reach 2.1 by 2030 .", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:fashion", "labels": [ "Fashion", "Fashion Technology" ], "is_subclass_of": [ "Computing and Cloud", "Software Engineering", "Creative Industries", "Retail Technology", "Computer Vision", "Generative AI", "Supply Chain Management", "Digital Asset" ], "wikilinks": [ "AI Styling", "ApplicationDomain", "Aura Blockchain", "Body Estimation Networks", "CLIP", "Collaborative Filtering", "Creative Industries", "CreativeIndustriesDomain", "Digital Fashion", "EU Digital Product Passport", "Fast Fashion", "Generative Design", "GOTS Organic Textile Standard", "GRI Sustainability Standards", "ISO 14040 LCA", "Manual Trend Analysis", "Mass Customisation", "NeRF", "Neural Garment Warping", "NFT" ] }, { "id": "fast-fourier-transform", "title": "Fast Fourier Transform", "domain": "ai", "domain_name": "Ai", "definition": "The Fast Fourier Transform (FFT) is a family of efficient algorithms for computing the discrete Fourier transform and its inverse, reducing the cost from quadratic to log-linear time in the number of samples. By exploiting symmetry and recursive divide-and-conquer factorisation, such as the Cooley-Tukey scheme, the FFT makes spectral analysis of large signals computationally practical. It is a foundational primitive in digital signal processing, communications, numerical methods and many machine learning and scientific computing workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fast-fourier-transform", "labels": [ "Fast Fourier Transform" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "fast-ice", "title": "Fast Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:fast-ice", "labels": [ "Fast Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "fast-spatial-queries", "title": "Fast Spatial Queries", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Optimised algorithms and data structures enabling rapid retrieval and processing of three-dimensional location-based data in metaverse environments, supporting real-time collision detection, proximity searches, visibility calculations, and spatial indexing for interactive virtual world experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:fast-spatial-queries", "labels": [ "Fast Spatial Queries" ], "is_subclass_of": [ "Display and Rendering", "Spatial Computing Paradigm" ], "wikilinks": [ "Real-Time Metaverse Interactions", "metaverse", "Spatial Computing" ] }, { "id": "fault-detection-isolation-and-recovery", "title": "Fault Detection Isolation and Recovery", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:fault-detection-isolation-and-recovery", "labels": [ "Fault Detection Isolation and Recovery" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "fault-tolerance-system", "title": "Fault Tolerance System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Fault Tolerance System is an engineering design framework enabling a distributed system to continue correct operation despite component failures, network disruptions, or malicious behaviour by a subset of participants. These systems employ redundancy, error detection, and automatic recovery to mask failures from end users, with Byzantine fault tolerance being the gold standard for adversarial environments such as public blockchains, which require at least 3f+1 total nodes to tolerate f Byzantine failures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fault-tolerance-system", "labels": [ "Fault Tolerance System", "Fault Tolerant System", "Fault-Tolerant System", "Fault-Tolerant Systems" ], "is_subclass_of": [ "Network Component", "Blockchain Entity" ], "wikilinks": [ "Byzantine Fault Tolerant System", "Checkpoint System", "Crash Fault Tolerant System", "Dependable Computing Systems", "Fault-Tolerant Systems", "Blockchain", "BlockchainDomain", "Blockchain Entity", "Byzantine Fault Tolerance", "ConceptualLayer", "Consensus Mechanism", "Replication System", "State Machine Replication" ] }, { "id": "fault-tolerance", "title": "Fault Tolerance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "System property denoting the capability of a computing or distributed system to continue providing correct service in the presence of component failures, encompassing the formal failure-model taxonomy enumerated by Cristian (fail-stop, fail-silent, omission, crash-recovery, timing, Byzantine arbi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:fault-tolerance", "labels": [ "Fault Tolerance", "FaultTolerance", "Hardware Fault Tolerance", "Partial Failure Tolerance", "System Fault Tolerance" ], "is_subclass_of": [ "Computing and Cloud", "Software Engineering", "Dependability", "System Property", "Reliability Engineering", "Resilience" ], "wikilinks": [ "Aerospace Avionics", "ARP4754A", "Automatic Recovery", "Autonomous Vehicles", "Avizienis Laprie Randell Landwehr 2004 Dependability Taxonomy", "Best-Effort Retry", "Beyer Jones Petoff Murphy 2016 Google SRE Book", "Brewer 2000 PODC Keynote CAP", "Buchman Kwon Milosevic 2018 Tendermint", "Burrows 2006 Chubby OSDI", "Byzantine Generals Problem", "Candea Fox 2003 Crash-Only Software HotOS", "CAP Theorem", "Castro Liskov 1999 PBFT OSDI", "Chain Replication", "Chandra Toueg 1996 Unreliable Failure Detectors", "Chandy Lamport 1985 Distributed Snapshots", "Chaos Engineering", "Checkpointing", "Continued Operation" ] }, { "id": "fault-tree-analysis", "title": "Fault Tree Analysis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fault Tree Analysis (FTA) is a top-down, deductive reliability and safety method that models how combinations of component failures and events can lead to a defined undesired top-level event. Using Boolean logic gates to connect basic events, it identifies minimal cut sets \u2014 the smallest combinations of failures sufficient to cause the top event \u2014 and can be evaluated qualitatively or quantitatively with probabilities. Originating in aerospace and nuclear engineering, FTA is a cornerstone of system safety, risk assessment, and certification across safety-critical industries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fault-tree-analysis", "labels": [ "Fault Tree Analysis" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "fault-tolerant-control", "title": "Fault-Tolerant Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Fault-tolerant control is a control-system design approach that maintains acceptable performance, or degrades gracefully, in the presence of sensor, actuator or component failures. It combines fault detection and isolation with reconfiguration mechanisms, such as switching to redundant actuators or re-tuning controller gains, so that a system continues operating safely rather than failing outright. It is widely applied in autonomous robots, aircraft and industrial process control where uninterrupted operation is safety-critical.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:fault-tolerant-control", "labels": [ "Fault-Tolerant Control" ], "is_subclass_of": [ "Control Algorithm" ], "wikilinks": [] }, { "id": "feature-attribution", "title": "Feature Attribution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A family of explainable AI techniques that assign a contribution score to each input feature, indicating how much it influenced a model's particular prediction.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:feature-attribution", "labels": [ "Feature Attribution", "Additive Feature Attribution" ], "is_subclass_of": [ "Explainable AI", "Model Interpretability" ], "wikilinks": [ "Machine Learning", "Interpretability", "SHAP", "Feature Importance", "Explainable AI", "LIME", "Gradient Descent", "Backpropagation", "Neural Network", "Mechanistic Interpretability", "Model Interpretability", "Feature Engineering", "Deep Learning", "Convolutional Neural Network", "Attention Mechanism", "Loss Function", "AI Safety" ] }, { "id": "feature-detection", "title": "Feature Detection", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Feature detection is the programmatic technique of querying a runtime environment to determine whether a specific capability, API, or behaviour is available before invoking it, rather than inferring support from user-agent strings or version numbers. In computer vision, it also denotes the algorithmic identification of salient points, edges, or regions within images that carry discriminative information for downstream tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:feature-detection", "labels": [ "Feature Detection" ], "is_subclass_of": [ "Runtime Inspection" ], "wikilinks": [] }, { "id": "feature-engineering", "title": "Feature Engineering", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The process of using domain knowledge and statistical transformations to construct, select, and encode input variables from raw data so that they better represent the underlying predictive signal for machine learning models. Techniques include polynomial expansion, normalisation, temporal feature extraction, embedding of categorical variables, and dimensionality reduction via PCA or autoencoders.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:feature-engineering", "labels": [ "Feature Engineering", "Feature Engineering Automation", "FeatureEngineering", "Manual Feature Engineering" ], "is_subclass_of": [ "Data Preprocessing" ], "wikilinks": [ "Data Preprocessing", "Autonomous Robot", "Digital Twin" ] }, { "id": "feature-extraction", "title": "Feature Extraction", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Feature Extraction is the process of transforming raw data into a reduced set of meaningful representations that capture task-relevant information for machine learning models. Deep learning architectures perform hierarchical feature extraction automatically through successive layers, whilst classical techniques such as PCA or wavelet transforms require manual engineering. Feature extraction reduces dimensionality, improves computational efficiency, and determines the quality of downstream model predictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:feature-extraction", "labels": [ "Feature Extraction", "Acoustic Feature Extraction", "Audio Feature Extraction", "Descriptor Extraction", "Feature Descriptor Extraction", "Image Feature Extraction" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "feature-importance", "title": "Feature Importance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantitative measures indicating the relative contribution or influence of individual input features on a machine learning model's predictions, enabling identification of the most critical variables driving model outputs. Methods include permutation importance, SHAP (SHapley Additive exPlanations) values, and tree-based Gini impurity scores, each providing global or local views of feature influence that support model debugging, data selection, and regulatory explainability requirements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:feature-importance", "labels": [ "Feature Importance", "Global Feature Importance", "Permutation Feature Importance" ], "is_subclass_of": [ "Explainability" ], "wikilinks": [ "Feature Attribution", "Feature Selection", "modeling", "Partial Dependence Plot", "Permutation Importance", "SHAP", "user experience", "Dimensionality Reduction", "Global Explanation", "MetaverseDomain", "Model Interpretability" ] }, { "id": "feature-learning", "title": "Feature Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Feature learning is the automatic discovery of useful representations or features directly from raw data, replacing manual feature engineering with representations learned by a model during training. Non-linear activation functions and layered network architectures such as feed-forward networks are what allow feature learning to build increasingly abstract representations across depth. It is foundational to deep learning's success across vision, language and speech tasks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:feature-learning", "labels": [ "Feature Learning" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "feature-map", "title": "Feature Map", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A feature map (or activation map) is the output produced when a convolutional filter is applied across an input in a convolutional neural network, encoding the spatial response of a learned feature. Each channel of a feature map highlights where a particular pattern, such as an edge or texture, occurs in the input. Stacks of feature maps form the intermediate representations that deeper layers compose into higher-level concepts.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:feature-map", "labels": [ "Feature Map" ], "is_subclass_of": [ "Convolutional Neural Network", "Deep Neural Network" ], "wikilinks": [ "Convolutional Neural Network", "Filter Kernel", "Receptive Field", "Convolution", "Activation Function", "Feature Extraction", "Pooling Layer", "Backpropagation", "Image Recognition", "Object Detection", "Feature Pyramid Network", "Transfer Learning", "Attention Mechanism", "Semantic Segmentation", "Deep Learning", "Tensor" ] }, { "id": "feature-matching", "title": "Feature Matching", "domain": "ai", "domain_name": "Ai", "definition": "Feature Matching is a computer vision technique that identifies and associates corresponding salient regions\u2014keypoints and their descriptors\u2014across two or more images or point clouds, enabling geometric relationships such as homographies, fundamental matrices, or rigid-body transformations to be estimated. Classical detectors such as SIFT, SURF, and ORB extract rotation- and scale-invariant descriptors; modern deep learning approaches learn matched embeddings end-to-end from training data. Feature matching is a foundational step in Structure-from-Motion, visual odometry, SLAM, and image-based localisation pipelines. The accuracy and efficiency of matching directly determine downstream reconstruction quality and real-time performance in robotics and augmented reality applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:feature-matching", "labels": [ "Feature Matching" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "feature-pyramid-network", "title": "Feature Pyramid Network", "domain": "ai", "domain_name": "Ai", "definition": "A Feature Pyramid Network (FPN) is a convolutional neural-network architecture that builds a multi-scale feature hierarchy with strong semantics at all levels by combining a bottom-up pathway with a top-down pathway and lateral connections. This design lets detectors and segmenters recognise objects across a wide range of sizes using features that are simultaneously high-resolution and semantically rich. FPN is a standard backbone component in modern object-detection and instance-segmentation pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:feature-pyramid-network", "labels": [ "Feature Pyramid Network", "Feature Pyramid" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "feature-selection", "title": "Feature Selection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Feature selection is the process of identifying and retaining the subset of input variables most relevant to a predictive task while discarding redundant or uninformative ones. By reducing dimensionality it can improve model generalisation, lower computational cost, and enhance interpretability without altering the underlying feature values. Methods range from filter approaches based on statistical relevance, through wrapper approaches that evaluate subsets via model performance, to embedded approaches integrated into model training.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:feature-selection", "labels": [ "Feature Selection" ], "is_subclass_of": [ "Feature Engineering" ], "wikilinks": [] }, { "id": "feature-store", "title": "Feature Store", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A centralised repository for storing, managing, versioning, and serving machine learning features at scale. It ensures point-in-time consistency between training and online inference, prevents training-serving skew, and enables cross-team feature reuse so that features computed once can be shared across multiple models and pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:feature-store", "labels": [ "Feature Store", "Feature Stores" ], "is_subclass_of": [ "AI Infrastructure", "Artificial Intelligence" ], "wikilinks": [ "Data Engineering", "MLOps", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Blockchain", "Digital Twin" ] }, { "id": "federated-byzantine-fault-tolerance", "title": "Federated Byzantine Fault Tolerance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Federated Byzantine Fault Tolerance (FBFT) is a consensus mechanism in which each node independently selects a trusted subset of peers\u2014its quorum slice\u2014forming overlapping quorums that propagate agreement without requiring global participation or expensive proof-of-work computation. Unlike classic BFT protocols that demand a fixed, known validator set, FBFT allows open membership: any node may join by declaring its quorum slices, and safety is guaranteed as long as quorum intersections contain at least one correct node. FBFT underpins the Stellar Consensus Protocol (SCP), enabling low-latency, high-throughput cross-border payment settlement with safety and liveness properties derived from quorum-intersection analysis.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:federated-byzantine-fault-tolerance", "labels": [ "Federated Byzantine Fault Tolerance", "Federated Byzantine Agreement" ], "is_subclass_of": [ "Protocol and Consensus", "Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Byzantine Fault Tolerance" ] }, { "id": "federated-credential-exchange", "title": "Federated Credential Exchange", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cross-platform workflow process that enables secure sharing and translation of identity credentials between different identity providers using standardized protocols, attribute mapping, and user consent mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:federated-credential-exchange", "labels": [ "Federated Credential Exchange" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Attribute Schema", "Attribute Translation", "Consent Verification", "Credential Request", "Federated Identity System", "Federation Protocol", "OpenID Connect", "Privacy-Preserving Authentication", "SAML 2.0", "Secure Exchange Protocol", "Single Sign-On", "Trust Registry", "W3C Verifiable Credentials", "Attribute Sharing", "Consent Management", "Credential Format Standard", "Cross-Platform Identity", "Cryptographic Keys", "Identity Provider", "MiddlewareLayer" ] }, { "id": "federated-edge-learning", "title": "Federated Edge Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Federated Edge Learning combines distributed machine learning with edge computing, enabling collaborative model training across decentralised edge devices while keeping training data locally on-device. Participants train local models on their private datasets and securely aggregate only model updates\u2014rather than raw data\u2014into a shared global model, preserving data sovereignty whilst enabling collective intelligence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:federated-edge-learning", "labels": [ "Federated Edge Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "AIEthicsDomain", "Autonomous Robot", "Blockchain", "ConceptualLayer" ] }, { "id": "federated-identity-system", "title": "Federated Identity System", "domain": "security", "domain_name": "Security", "definition": "A federated identity system enables a user's digital identity and authentication to be recognised across multiple independent organisations or domains without each maintaining separate credentials. It relies on trust relationships between identity providers and relying parties, exchanging assertions via protocols such as SAML, OpenID Connect, and OAuth. Federation underpins single sign-on, cross-organisation access, and trust-framework governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:federated-identity-system", "labels": [ "Federated Identity System", "Federated Access" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "federated-identity", "title": "Federated Identity", "domain": "security", "domain_name": "Security", "definition": "Federated Identity is an architectural paradigm in which a user authenticates once with a trusted Identity Provider and receives a signed assertion that is accepted by multiple independent relying-party services across organisational or administrative boundaries, eliminating per-service credential stores. The federation relationship is governed by bilateral or multilateral trust agreements and implemented through standard protocols such as SAML 2.0, OpenID Connect, and OAuth 2.0, which define how authentication tokens are issued, transported, and cryptographically verified. Federated identity is foundational to enterprise single sign-on, cross-institutional academic collaboration, and consumer social-login ecosystems, and is actively converging with decentralised-identity models that replace central providers with holder-controlled cryptographic credentials.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:federated-identity", "labels": [ "Federated Identity" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "federated-learning", "title": "Federated Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Distributed machine learning paradigm enabling collaborative model training across decentralized data sources without centralizing sensitive information; model updates are aggregated from local computations whilst raw data remains on-device, preserving privacy and enabling cross-organizational learning.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:federated-learning", "labels": [ "Federated Learning", "Federated Machine Learning", "FederatedLearning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "coordinated By", "DecentralizedAI", "DistributedLearning", "dt:distributedVia", "dt:implementedVia", "dt:incentivizedBy", "dt:protectedBy", "dt:securedBy", "hasParticipant", "implementsPrivacy", "P2PNetwork", "PrivacyPreservingML", "SecureAggregation", "TokenReward", "usesAggregation", "BlockchainNetwork", "CollaborativeLearning", "Consensus Protocol", "MetaverseDomain", "SmartContract" ] }, { "id": "federated-query", "title": "Federated Query", "domain": "data", "domain_name": "Data", "definition": "A federated query is a single query that is decomposed and executed across multiple autonomous, distributed data sources, with partial results combined into a unified answer without first consolidating the data into one store. In the semantic-web context, SPARQL federation evaluates sub-queries against several remote endpoints, joining their bindings transparently to the requester. It enables integrated access to heterogeneous, independently governed datasets while leaving each source in place. Effective federation depends on source description, query planning, and distributed join optimisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:federated-query", "labels": [ "Federated Query" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [] }, { "id": "federated-social-networks", "title": "Federated Social Networks", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Federated social networks are distributed communication platforms composed of independently operated servers that interoperate via shared open protocols, allowing users on different instances to follow, mention, and exchange content across administrative boundaries without centralised ownership. The model contrasts with siloed proprietary networks by enabling user data portability, operator sovereignty, and community-driven moderation policies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:federated-social-networks", "labels": [ "Federated Social Networks", "Federated Social Network" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "federated-system", "title": "Federated System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A federated system is a distributed architecture composed of autonomous, independently operated nodes or servers that cooperate under shared protocols while retaining local control over their own data and policies. Unlike fully centralised systems, federation distributes authority across multiple operators, and unlike fully decentralised peer-to-peer networks, it relies on a finite set of identifiable, semi-trusted servers. This model underpins federated social networks, messaging, and consortium-style trust arrangements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:federated-system", "labels": [ "Federated System", "Federated Architecture", "Federated Systems" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "federation-protocol", "title": "Federation Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A federation protocol is a formal specification for enabling independent, autonomously administered servers or identity domains to interoperate and exchange data, messages, or authentication credentials without centralised control, such that users of one domain can communicate with or access resources from another domain governed by different administrators. Federation protocols typically define message formats, authentication and authorisation mechanisms, actor representations, and the semantics of cross-domain identity references, enabling loosely coupled networks of independently operated services to collectively deliver the capabilities of a unified system. They underpin decentralised social networks, federated identity management, and distributed messaging systems that resist single-point control and censorship.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:federation-protocol", "labels": [ "Federation Protocol" ], "is_subclass_of": [ "Distributed Protocol" ], "wikilinks": [] }, { "id": "federation-surface", "title": "Federation Surface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A standardised JSON-LD 1.1 encoding surface (S1\u2013S11) that exposes agent state, credentials, events, and work metadata in a queryable, linkable format, enabling federated consumption by heterogeneous external systems (monitoring dashboards, compliance audits, blockchain oracles, knowledge grap...", "entityType": "Class", "qualityScore": 0.86, "maturity": "established", "iri": "urn:ngm:class:federation-surface", "labels": [ "Federation Surface" ], "is_subclass_of": [ "Network and Communication", "Middleware Layer" ], "wikilinks": [ "ADR-008", "ADR-012", "Agent Bead", "AgenticSystemsDomain", "APILayer", "Automated Monitoring", "Blockchain Oracle Integration", "Cross-System Querying", "DataIntegrationDomain", "Data Portal", "HTTP Endpoint", "JSON-LD 1.1", "JSON-LD 1.1 Spec", "JSON-LD 1.1 Standard", "JSON-LD Context", "JSON-LD Context", "Knowledge Graph Federation", "Linked Data Consumption", "PRD-006", "SemanticWebDomain" ] }, { "id": "federation", "title": "Federation", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "An architecture in which autonomous systems or organisations interoperate through agreed protocols and trust relationships while retaining independent control of their own resources, enabling cross-domain data and service exchange without centralised authority.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:federation", "labels": [ "Federation", "Federation Pattern", "Federation Server", "Server-to-Server Federation" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Communication Protocol", "Authentication", "Interoperability", "Identity Management", "Distributed Systems" ] }, { "id": "fedimint", "title": "Fedimint", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Fedimint is an open-source federated protocol for community-custodied Bitcoin Chaumian e-cash mints, in which a threshold of guardians collectively hold Bitcoin reserves and issue blinded bearer tokens redeemable for satoshis, providing privacy-preserving custody without requiring any single trusted party. The federation model distributes trust across a small, known set of guardians using Byzantine-fault-tolerant consensus, while the blind-signature scheme cryptographically prevents guardians from linking redemptions to issuances. Fedimint acts as a Bitcoin Layer 2 through embedded Lightning Network gateway integration, enabling community mints to interoperate with the broader Bitcoin payment ecosystem. It targets the gap between self-custody complexity and centralised custodial risk by enabling accountable, trust-minimised community banking at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fedimint", "labels": [ "Fedimint" ], "is_subclass_of": [ "ecash" ], "wikilinks": [] }, { "id": "fee-market", "title": "Fee Market", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Fee Market is the mechanism by which blockchain participants competitively bid transaction fees to have their transactions included in blocks, with miners or validators selecting transactions that maximise their revenue given limited block capacity. Fee markets emerge from the interplay between fixed block-space supply and variable transaction demand, producing dynamic price discovery that signals network congestion. Ethereum's EIP-1559 introduced a protocol-level base fee that adjusts algorithmically each block, partially burning fees to reduce token supply and adding a tip mechanism for priority inclusion.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fee-market", "labels": [ "Fee Market", "Fee Estimation" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "fee-tier", "title": "Fee Tier", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fee tier is a discrete level of trading or swap fee applied to a market or liquidity pool, allowing different markets to charge fees commensurate with their volatility and liquidity risk. On automated-market-maker decentralized exchanges, each pool is created under a chosen fee tier (for example 0.05%, 0.30%, or 1.00%), and the collected fees accrue to liquidity providers. Tiering lets stable pairs use low fees while volatile or exotic pairs use higher fees to compensate providers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fee-tier", "labels": [ "Fee Tier" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "feed-forward-network", "title": "Feed Forward Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A feed-forward network (FFN) is a class of artificial neural network in which information propagates strictly in one direction \u2014 from input nodes through one or more hidden layers to output nodes \u2014 with no feedback cycles, recurrent connections, or lateral synapses between units at the same layer.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:feed-forward-network", "labels": [ "Feed Forward Network", "Feed-Forward Network" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network", "Machine Learning Discipline", "Deep Learning", "Parametric Model", "Differentiable Programming" ], "wikilinks": [ "AlgorithmLayer", "Automatic Differentiation", "Bias Vector", "Calculus", "DeepLearningDomain", "Differentiable Programming", "Diffusion Model", "Feature Learning", "GELU", "Hidden Layer", "Hyperparameter Tuning", "ICLR", "ICML", "IEEE", "Information Theory", "Input Layer", "Large Language Model", "Linear Algebra", "MachineLearningDomain", "ModelArchitectureLayer" ] }, { "id": "feedback-control", "title": "Feedback Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Feedback Control - A closed-loop control mechanism in which Sensor measurements of actual system state are continuously compared against desired Setpoints, and control actions are adjusted in real time to minimise error and maintain stable, accurate Robot Behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:feedback-control", "labels": [ "Feedback Control", "Feedback Control Architecture", "Feedback Control Loop", "FeedbackControl" ], "is_subclass_of": [ "Control Theory", "Robotics" ], "wikilinks": [ "Actuator Response", "Disturbance Rejection", "Error Measurement", "Motion Control System", "Robot Behaviour", "Robust Operation", "Setpoints", "Stabilisation Architecture", "Trajectory Tracking", "AI Agent System", "Control Algorithm", "Control Theory", "Robotics", "RoboticsDomain", "Sensor" ] }, { "id": "feedback-loop", "title": "Feedback Loop", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Feedback Loop is a cybernetic and control-theoretic structure in which the output of a system is routed back as input to influence its subsequent behaviour, producing closed-loop regulation, amplification, learning, or instability depending on the loop's sign, gain, delay, and phase characteristi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:feedback-loop", "labels": [ "Feedback Loop", "Feedback Loops" ], "is_subclass_of": [ "Software Engineering", "Control Mechanism", "Circular Causality", "Closed-Loop System", "System Mechanism", "Cybernetic Process" ], "wikilinks": [ "Actuator", "Adaptation", "AlgorithmLayer", "Armstrong McKay et al 2022 Climate Tipping Points Science", "Ashby 1956 Introduction to Cybernetics", "Astrom Hagglund 2006 Advanced PID Control", "Astrom Murray 2020 Feedback Systems", "Autopoiesis", "Bai et al 2022 Constitutional AI", "Bail 2021 Breaking the Social Media Prism", "Bang-Bang Control", "Batch Process", "Bateson 1972 Steps to an Ecology of Mind", "Beer 1972 Brain of the Firm", "Bode 1945 Network Analysis and Feedback Amplifier Design", "Bode Plot", "Cannon 1932 The Wisdom of the Body", "Causal Loop Diagram", "Christiano et al 2017 Deep RL from Human Preferences", "Circular Causality" ] }, { "id": "feedback-mechanism", "title": "Feedback Mechanism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mod providing sensory response to user actions through haptic, audio, and visual channels to enhance interaction fidelity and user experience in immersive environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:feedback-mechanism", "labels": [ "Feedback Mechanism" ], "is_subclass_of": [ "Interaction Technology", "Immersive Experience Pipeline" ], "wikilinks": [ "Audio Response Module", "Enhanced Presence", "ETSI ARF 010", "Event Processing", "Haptic Feedback System", "Input Detection", "Low Latency Network", "Natural Interaction", "Sensor Data", "Sensory Immersion", "Sensory Integration Controller", "State Management", "User Feedback Loop", "User Interaction System", "Visual Feedback Renderer", "Immersive Experience Pipeline", "InteractionDomain", "NetworkLayer", "Rendering Engine", "User Context Awareness" ] }, { "id": "feedback-sensor", "title": "Feedback Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A feedback sensor is a transducer that measures the actual state of an actuator or controlled system, such as position, velocity, force, or current, and reports it back to a controller to close the control loop. By comparing the measured value with the commanded reference, the controller can correct errors and reject disturbances. Encoders, resolvers, tachometers, and load cells are common feedback sensors in servo and motion-control systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:feedback-sensor", "labels": [ "Feedback Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "feedforward-compensation", "title": "Feedforward Compensation", "domain": "robotics", "domain_name": "Robotics", "definition": "Feedforward compensation is a control technique that uses a model of the system or known disturbances to compute corrective control action in advance, rather than waiting for an error to appear at the output. By anticipating required effort, for example to overcome inertia, friction, or measurable load disturbances, it improves tracking and disturbance rejection beyond what feedback alone provides. It is typically combined with feedback control to handle modelling errors and unmeasured disturbances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:feedforward-compensation", "labels": [ "Feedforward Compensation" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "feedforward-control", "title": "Feedforward Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Feedforward control is a control strategy in which the controller acts on a reference command or a measured disturbance before it affects the output, rather than reacting to output error after the fact. Because it does not rely on output measurement, pure feedforward cannot correct for unmodelled effects and is therefore usually paired with feedback. It is widely used in motion and process control to improve command tracking and pre-empt known disturbances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:feedforward-control", "labels": [ "Feedforward Control" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "feedforward-neural-network", "title": "Feedforward Neural Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network architecture in which connections between nodes do not form cycles, with information flowing unidirectionally from input through hidden layers to output. The simplest and most foundational artificial neural network type, trained via backpropagation with gradient descent, and proven by the universal approximation theorem to model arbitrary continuous functions given sufficient width.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:feedforward-neural-network", "labels": [ "Feedforward Neural Network", "Feed-Forward Neural Network" ], "is_subclass_of": [ "AI Model Architecture", "Network Architecture" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "feeless-blockchain", "title": "Feeless Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A feeless blockchain is a distributed ledger that allows transactions to be submitted without an explicit per-transaction fee paid to validators. Such designs replace fee-based economics with alternative mechanisms \u2014 for example directed acyclic graph structures where each sender contributes a small proof-of-work to confirm prior transactions, or networks where validators are compensated through other means. Feeless models aim to enable micropayments and machine-to-machine value transfer that fee-bearing chains render uneconomic. They must still solve spam prevention and resource accounting without using fees as the deterrent.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:feeless-blockchain", "labels": [ "Feeless Blockchain" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "few-shot-examples", "title": "Few-Shot Examples", "domain": "ai", "domain_name": "Ai", "definition": "Few-shot examples are a small set of input-output demonstrations placed within a language model's prompt to illustrate the desired task, format, or reasoning pattern. By conditioning on these in-context examples, the model can perform the task without weight updates, leveraging in-context learning. The number, quality, ordering, and representativeness of the examples strongly influence output accuracy and consistency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:few-shot-examples", "labels": [ "Few-Shot Examples" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "few-shot-learning", "title": "Few-Shot Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A learning setting in which a model is required to generalise to a new task or class from only a small number of labelled examples, typically by leveraging prior knowledge or inductive biases learned across many related tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "growing", "iri": "urn:ngm:class:few-shot-learning", "labels": [ "Few-Shot Learning" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique" ], "wikilinks": [ "Transfer Learning", "Meta-Learning", "In-Context Learning", "Foundation Models", "Machine Learning" ] }, { "id": "few-shot-prompting", "title": "Few-Shot Prompting", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Few-shot prompting is a prompt engineering technique in which a small number of input-output demonstration examples are included directly in the context provided to a large language model, guiding the model to produce outputs conforming to the demonstrated pattern without any parameter updates. The approach exploits the in-context learning capability of transformer-based models, allowing task specification through examples rather than task-specific fine-tuning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:few-shot-prompting", "labels": [ "Few-Shot Prompting" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "fiat-currency", "title": "Fiat Currency", "domain": "finance", "domain_name": "Finance", "definition": "Fiat currency is money issued and declared legal tender by a sovereign government or central bank, whose value derives from state authority, institutional trust, and collective public acceptance rather than from any intrinsic commodity backing. Unlike commodity money, fiat currency is not convertible into a fixed quantity of gold, silver, or other physical assets at a legally guaranteed rate. Its purchasing power is maintained through monetary policy instruments \u2014 interest rates, reserve requirements, open-market operations \u2014 and is subject to inflation, deflation, and exchange-rate dynamics driven by macroeconomic conditions. Fiat currency forms the foundational settlement layer for modern banking systems, international trade, and the reserve benchmarks against which digital assets are priced.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:fiat-currency", "labels": [ "Fiat Currency", "Fiat Currency Exchange" ], "is_subclass_of": [ "Money" ], "wikilinks": [ "Stablecoin", "USD", "Central Bank Digital Currency", "Money" ] }, { "id": "fiat-on-ramp", "title": "Fiat On-Ramp", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fiat on-ramp is a service or integration that lets users convert government-issued currency into cryptocurrency, typically via bank transfer, card payment, or other traditional payment rails connected to an exchange or wallet. It handles identity verification, payment processing, and currency conversion so that users without existing crypto holdings can enter blockchain ecosystems. Fiat on-ramps are commonly embedded in custodial wallets and centralised exchanges such as Kraken to lower the barrier to first-time cryptocurrency acquisition.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:fiat-on-ramp", "labels": [ "Fiat On-Ramp" ], "is_subclass_of": [ "Fiat Currency" ], "wikilinks": [] }, { "id": "fiat-shamir-heuristic", "title": "Fiat Shamir Heuristic", "domain": "security", "domain_name": "Security", "definition": "The Fiat-Shamir heuristic is a cryptographic technique that transforms an interactive public-coin proof or identification protocol into a non-interactive one by replacing the verifier's random challenges with the output of a cryptographic hash function applied to the prover's messages. This removes the need for live interaction, allowing proofs and signatures to be generated and verified offline. It is foundational to many digital signature schemes and non-interactive zero-knowledge proofs, with security analysed in the random oracle model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fiat-shamir-heuristic", "labels": [ "Fiat Shamir Heuristic", "Fiat-Shamir Heuristic" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "fiber-optics", "title": "Fiber Optics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A technology for transmitting data as light pulses through glass or plastic fibers, serving as the primary physical medium for high-bandwidth networking and telecommunications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:fiber-optics", "labels": [ "Fiber Optics" ], "is_subclass_of": [ "Networking" ], "wikilinks": [] }, { "id": "fibre-optic-network", "title": "Fibre Optic Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A communications network that transmits data as pulses of light through strands of ultra-pure glass or plastic fibre, achieving terabit-class bandwidth, low attenuation over tens of kilometres, and immunity to electromagnetic interference. Fibre optic networks form the backbone of global telecommunications \u2014 submarine cables, metro rings, data-centre interconnects, and fibre-to-the-premises access \u2014 and set the bandwidth and latency envelope within which cloud, streaming, and real-time immersive applications operate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:fibre-optic-network", "labels": [ "Fibre Optic Network", "Fibre-Optic Network" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [ "Network Infrastructure", "Telecommunications Infrastructure", "Satellite Communication", "Data Centre", "Bandwidth", "Latency" ] }, { "id": "fiducial-marker", "title": "Fiducial Marker", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A fiducial marker is a designed visual pattern placed in a scene to serve as a reliable reference point for computer-vision systems. Its known geometry and high-contrast, machine-readable encoding allow algorithms to detect it robustly, recover camera pose, and assign a unique identifier. Fiducial markers such as ArUco and AprilTag families are widely used for camera calibration, augmented-reality registration, and robot localisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fiducial-marker", "labels": [ "Fiducial Marker" ], "is_subclass_of": [ "Optical Calibration Target" ], "wikilinks": [] }, { "id": "fiduciary-duty", "title": "Fiduciary Duty", "domain": "governance", "domain_name": "Governance", "definition": "A fiduciary duty is a legal and ethical obligation requiring one party, the fiduciary, to act in the best interests of another, the beneficiary, with loyalty, prudence, and good faith. It typically encompasses a duty of loyalty that prohibits self-dealing and conflicts of interest, and a duty of care that demands competent, diligent management of the beneficiary's affairs. Fiduciary duties arise in relationships such as those between directors and shareholders, trustees and beneficiaries, and asset managers and clients.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fiduciary-duty", "labels": [ "Fiduciary Duty" ], "is_subclass_of": [ "Corporate Governance" ], "wikilinks": [] }, { "id": "field-programmable-gate-array", "title": "Field-Programmable Gate Array", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A field-programmable gate array (FPGA) is an integrated circuit whose internal logic and interconnect can be reconfigured by the user after manufacture to implement arbitrary digital circuits. It comprises a fabric of programmable logic blocks, embedded memories, and routing that is configured from a hardware description language. FPGAs deliver hardware-level parallelism and low latency while remaining reprogrammable, sitting between fixed ASICs and general-purpose processors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:field-programmable-gate-array", "labels": [ "Field-Programmable Gate Array", "Field Programmable Gate Array" ], "is_subclass_of": [ "Hardware Accelerator" ], "wikilinks": [] }, { "id": "fieldbus", "title": "Fieldbus", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fieldbus is the family of digital, serial, multi-drop communication protocols \u2014 standardised in IEC 61158 and including PROFIBUS, FOUNDATION Fieldbus, Modbus, CAN-based networks, and DeviceNet \u2014 that connect field devices such as sensors, actuators, and drives to programmable logic controllers and distributed control systems. Introduced from the mid-1980s to replace point-to-point 4\u201320 mA and RS-232 wiring, a single shared cable carries deterministic cyclic process data and device diagnostics, cutting cabling cost and enabling intelligent field instruments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:fieldbus", "labels": [ "Fieldbus" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "Communication Protocol", "IndustrialAutomation", "Industrial Ethernet", "CAN Bus", "Programmable Logic Controller", "SCADA", "Time-Sensitive Networking" ] }, { "id": "fig-jam", "title": "FigJam", "domain": "metaverse", "domain_name": "Metaverse", "definition": "FigJam is an online collaborative whiteboard application from Figma designed for real-time ideation, brainstorming, diagramming, and workshop facilitation. It provides multi-user infinite canvases with sticky notes, shapes, connectors, stamps, and templates, synchronising edits across participants for distributed teams. FigJam complements Figma's design tooling by focusing on the early, divergent phases of product and design work.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fig-jam", "labels": [ "FigJam" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "figma", "title": "Figma", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Figma is a browser-based collaborative interface-design and prototyping platform that lets multiple designers work simultaneously on the same files in real time. It provides vector editing, component and design-system management, interactive prototyping, and developer hand-off, all stored in the cloud for seamless multiplayer collaboration. Figma has become a standard tool for UI/UX design and cross-functional product teams.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:figma", "labels": [ "Figma" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "file-storage", "title": "File Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "File storage is a storage model that organises data as named files within a hierarchical directory tree, accessed through file-system semantics such as open, read, write and seek. It presents a familiar path-based namespace and is typically shared over network protocols for concurrent access. File storage contrasts with block storage, which exposes raw volumes, and object storage, which uses a flat namespace of objects addressed by key.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:file-storage", "labels": [ "File Storage" ], "is_subclass_of": [ "Storage Infrastructure" ], "wikilinks": [] }, { "id": "file-system", "title": "File System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A file system is the data structure and associated management software that an operating system uses to organise, store, retrieve, and manage data on storage media. It defines how data is logically partitioned into named files and directories, governs access permissions, maintains metadata such as timestamps and ownership, and translates logical file operations into physical block I/O against underlying storage hardware. File systems range from local single-disk formats (NTFS, ext4, APFS) to distributed and network-attached systems (NFS, HDFS, GFS) that span many physical storage nodes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:file-system", "labels": [ "File System", "Cloud File System", "File System Access", "File Systems" ], "is_subclass_of": [ "Operating System" ], "wikilinks": [] }, { "id": "filecoin", "title": "Filecoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Filecoin is a decentralised storage network and blockchain protocol developed by Protocol Labs that creates a peer-to-peer marketplace for file storage and retrieval, using cryptographic proofs\u2014Proof of Replication (PoRep) and Proof of Spacetime (PoSt)\u2014to verifiably demonstrate that storage providers are dedicating physical disk space to client data over time. Storage providers earn FIL tokens by fulfilling storage deals and continuously proving their commitments on-chain, while clients pay FIL to store data with economic guarantees backed by the provider's staked collateral. Built on top of IPFS (InterPlanetary File System) for content-addressed data retrieval, Filecoin provides the economic incentive layer designed to make decentralised storage commercially viable, and has since expanded via the Filecoin Virtual Machine (FVM) to support programmable storage deals and decentralised autonomous organisations governing shared datasets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:filecoin", "labels": [ "Filecoin" ], "is_subclass_of": [ "Decentralized Storage" ], "wikilinks": [] }, { "id": "fill-value", "title": "Fill Value", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:fill-value", "labels": [ "Fill Value" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "film-production", "title": "Film Production", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The end-to-end creative and technical process of making motion picture content, encompassing development, pre-production, principal photography, visual effects, post-production, and distribution. Modern film production increasingly integrates spatial computing technologies such as virtual production workflows, LED volume stages, real-time game engine rendering, and AI-assisted tools that compress timelines and reduce location dependency.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:film-production", "labels": [ "Film Production", "Film Production Studio", "Film and Television Production" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "film-vfx", "title": "Film VFX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Film VFX (Visual Effects) encompasses the full suite of techniques used to create, augment, or manipulate imagery in cinematic productions that cannot be practically achieved through conventional photography, including computer-generated imagery (CGI), digital compositing, motion capture-driven character animation, photorealistic simulation of natural phenomena, and digital environment creation. Modern film VFX pipelines integrate real-time rendering engines, deep learning-based tools for rotoscoping, denoising, and face replacement, and virtual production stages using LED volume displays to blend physical and digital environments in-camera rather than in post-production. VFX is distinct from practical effects (pyrotechnics, prosthetics) and from colour grading, though all three coexist within the broader post-production discipline.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:film-vfx", "labels": [ "Film VFX", "Film VFX Production" ], "is_subclass_of": [ "Visual Effects" ], "wikilinks": [] }, { "id": "filter-bubble", "title": "Filter Bubble", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A state of informational isolation in which algorithmic personalisation \u2014 recommendation systems, personalised search, and engagement-optimised feeds \u2014 progressively narrows the content a person encounters to material predicted to match their existing preferences and beliefs, reducing exposure to disconfirming viewpoints without the person's awareness or consent; coined by Eli Pariser in 2011, the concept names a structural harm of personalised media that feeds polarisation, reinforces bias, and erodes the shared factual ground of public discourse.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:filter-bubble", "labels": [ "Filter Bubble" ], "is_subclass_of": [ "Digital Society Harms" ], "wikilinks": [ "Digital Society Harms", "Recommendation Systems", "Death of the Internet" ] }, { "id": "filtration", "title": "Filtration", "domain": "data", "domain_name": "Data", "definition": "In probability theory, a filtration is an increasing family of sigma-algebras indexed by time that formally represents the accumulation of information available up to each instant. It models what is knowable at each point as a stochastic process unfolds, with each sigma-algebra containing all events whose outcomes are determined by then. Filtrations are fundamental to defining adapted processes, martingales, and conditional expectations in stochastic analysis and mathematical finance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:filtration", "labels": [ "Filtration" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "fin-cen", "title": "FinCEN", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Crimes Enforcement Network (FinCEN) is a bureau of the US Department of the Treasury established in 1990 that serves as the primary US financial intelligence unit (FIU). FinCEN administers the Bank Secrecy Act (BSA), collects financial transaction reports from financial institutions (SARs, CTRs), analyses this data for patterns of money laundering, terrorist financing, and financial fraud, and shares intelligence with law enforcement and foreign FIUs through the Egmont Group. FinCEN also issues regulatory guidance and rulemaking for digital asset service providers including money service businesses (MSBs) and, since 2019, cryptocurrency exchanges.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fin-cen", "labels": [ "FinCEN" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "finality-gadget", "title": "Finality Gadget", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Finality Gadget is a protocol component layered onto a block-producing (typically probabilistically-final) blockchain to provide periodic deterministic finality for checkpointed epochs, operating as a BFT overlay that requires a supermajority of validator stake to attest to a canonical chain prefix before that prefix is considered irreversible. The archetypal finality gadget is Casper the Friendly Finality Gadget (Casper FFG), which Ethereum uses in its Gasper consensus construction alongside LMD-GHOST fork-choice to combine liveness (blocks always added) with periodic safety (epochs finalised every ~12.8 minutes). Finality gadgets decouple the latency-optimised block-production layer from the safety-optimised finality layer, allowing chains to maintain high throughput while providing applications with a provable point after which reorganisation is cryptoeconomically infeasible.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:finality-gadget", "labels": [ "Finality Gadget" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "finality", "title": "Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The property of a blockchain transaction or block whereby it becomes irreversible and cannot be reverted by any future state of the network. Probabilistic finality\u2014as in proof-of-work chains\u2014increases with confirmation depth, while deterministic finality\u2014as in BFT-based protocols\u2014is achieved at the point of commitment, with strong consequences for payment settlement, cross-chain interoperability, and user trust.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:finality", "labels": [ "Finality", "Block Finality", "Blockchain Finality", "Consensus Finality", "Finality Mechanism", "Instant Finality", "Network Finality", "Source Chain Finality" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "finance", "title": "Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Finance is the discipline concerned with the management, creation, and study of money, investments, and other financial instruments across individuals, organisations, and markets. It encompasses how capital is raised, allocated, priced, and risk-managed over time, spanning corporate finance, public finance, and personal finance. In the context of distributed systems, finance increasingly intersects with blockchain-based instruments and decentralised protocols that reimagine settlement, custody, and intermediation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:finance", "labels": [ "Finance" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "financial-action-task-force", "title": "Financial Action Task Force", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Action Task Force (FATF) is an intergovernmental body that sets international standards to combat money laundering, terrorist financing and proliferation financing. It issues the FATF Recommendations, conducts mutual evaluations of member jurisdictions, and maintains lists of high-risk and non-cooperative jurisdictions. Its guidance \u2014 including the Travel Rule for virtual assets \u2014 strongly shapes national anti-money-laundering and know-your-customer regimes worldwide.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-action-task-force", "labels": [ "Financial Action Task Force" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "financial-conduct-authority", "title": "Financial Conduct Authority", "domain": "governance", "domain_name": "Governance", "definition": "The Financial Conduct Authority (FCA) is the independent conduct and prudential regulator for financial services firms and markets in the United Kingdom, established under the Financial Services Act 2012 as successor to the Financial Services Authority. It is responsible for protecting consumers from harm, maintaining the integrity of UK financial markets, and promoting effective competition in the interests of consumers. The FCA authorises and supervises approximately 50,000 financial services firms, sets binding conduct rules through the FCA Handbook, and exercises enforcement powers including fines, prohibition orders, and market bans. Its remit has expanded to include cryptoasset registration, sustainability-related disclosure requirements, and Consumer Duty obligations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:financial-conduct-authority", "labels": [ "Financial Conduct Authority" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Securities Regulation", "Financial Regulation", "FCA" ] }, { "id": "financial-crime-compliance", "title": "Financial Crime Compliance", "domain": "finance", "domain_name": "Finance", "definition": "Financial crime compliance is the discipline by which regulated firms detect, prevent, and report illicit financial activity such as money laundering, terrorist financing, fraud, bribery, and sanctions evasion. It combines customer due diligence, transaction monitoring, sanctions screening, and suspicious-activity reporting under a risk-based framework mandated by regulators. The function protects the integrity of the financial system and exposes firms to significant penalties when controls fail.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-crime-compliance", "labels": [ "Financial Crime Compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "financial-crime-detection", "title": "Financial Crime Detection", "domain": "finance", "domain_name": "Finance", "definition": "Financial crime detection is the application of data analysis, machine learning, and rule-based systems to identify patterns indicative of money laundering, fraud, terrorist financing, bribery, market manipulation, and related illicit activities within financial transaction data. It operates at the intersection of compliance obligation and risk management, producing alerts that human investigators triage and escalate to regulatory or law-enforcement bodies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:financial-crime-detection", "labels": [ "Financial Crime Detection", "Financial Crime" ], "is_subclass_of": [ "Fraud Detection", "AI Application" ], "wikilinks": [] }, { "id": "financial-crime-prevention", "title": "Financial Crime Prevention", "domain": "finance", "domain_name": "Finance", "definition": "Financial Crime Prevention is the set of controls, processes and technologies that detect, deter and disrupt illicit financial activity such as money laundering, fraud, terrorist financing and sanctions evasion. It combines customer due diligence, transaction monitoring, screening and reporting obligations to satisfy regulatory expectations and protect the integrity of the financial system. The field draws on analytics and risk-based methodologies to focus scarce investigative resources on the highest-risk activity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-crime-prevention", "labels": [ "Financial Crime Prevention" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "financial-inclusion", "title": "financial inclusion", "domain": "finance", "domain_name": "Finance", "definition": "Financial inclusion is the policy objective and market condition in which individuals and businesses \u2014 particularly those in underserved, low-income, or geographically remote populations \u2014 have affordable, reliable access to a full range of financial services encompassing payments, credit, savings, insurance, and investment. It is pursued through regulatory reform, technology-led delivery (mobile money, digital wallets, open banking APIs), and identity infrastructure, aiming to integrate marginalised groups into the formal economy. Key enablers include tiered Know Your Customer frameworks, agent banking networks, interoperable payment rails, and programmable money mechanisms such as Central Bank Digital Currencies and regulated stablecoins. Progress is measured through account ownership rates, transaction volume among previously unbanked populations, and composite indices such as the World Bank Global Findex.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:financial-inclusion", "labels": [ "Financial Inclusion" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "financial-infrastructure", "title": "Financial Infrastructure", "domain": "finance", "domain_name": "Finance", "definition": "Financial Infrastructure comprises the foundational systems, networks, institutions, and regulatory frameworks that enable the creation, transfer, settlement, and custody of financial value at scale. It encompasses payment clearing and settlement networks, central securities depositories, central counterparty clearinghouses, correspondent banking rails, central bank digital currency platforms, and blockchain-based settlement layers. These components collectively underpin both legacy financial systems and emerging decentralised finance ecosystems, ensuring liquidity, finality, and systemic resilience. The integrity of financial infrastructure is a prerequisite for efficient capital allocation, monetary policy transmission, and macroeconomic stability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:financial-infrastructure", "labels": [ "Financial Infrastructure", "Global Financial Infrastructure", "Traditional Financial Infrastructure" ], "is_subclass_of": [ "Economic Layer" ], "wikilinks": [ "owl:Thing" ] }, { "id": "financial-instruments", "title": "Financial Instruments", "domain": "finance", "domain_name": "Finance", "definition": "Financial instruments are contractual agreements or documents that convey a monetary claim, obligation, or ownership right between parties, encompassing equities, debt securities, derivatives, currencies, and commodities. In contemporary digital finance they extend to programmable on-chain instruments encoded as smart contracts, including decentralised lending protocols, synthetic assets, and tokenised representations of traditional financial claims. Their valuation, transferability, and settlement properties are governed by both legal frameworks and, increasingly, algorithmic rules embedded in blockchain protocols. International Accounting Standard 32 defines them as any contract that gives rise to a financial asset of one entity and a financial liability or equity instrument of another.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:financial-instruments", "labels": [ "Financial Instruments", "Financial Instrument", "Regulated Financial Instrument" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "financial-intelligence", "title": "Financial Intelligence", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Financial intelligence is the collection, analysis, and dissemination of information about financial flows and transactions to detect, investigate, and disrupt illicit activity such as money laundering, terrorist financing, and sanctions evasion. In the blockchain context it applies graph analytics and transaction tracing across public ledgers to attribute addresses, cluster wallets, and surface suspicious patterns. It feeds compliance functions, law-enforcement investigations, and regulatory reporting, bridging on-chain data with off-chain identity and risk signals. Financial intelligence units and private analytics providers are its principal practitioners.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-intelligence", "labels": [ "Financial Intelligence" ], "is_subclass_of": [ "Blockchain Analytics" ], "wikilinks": [] }, { "id": "financial-market-infrastructure", "title": "Financial Market Infrastructure", "domain": "finance", "domain_name": "Finance", "definition": "Financial market infrastructure refers to the multilateral systems that record, clear and settle payments, securities, derivatives and other financial transactions among participating institutions. It encompasses payment systems, central counterparties, central securities depositories, securities settlement systems and trade repositories. By concentrating and standardising the plumbing of financial markets, these systems underpin market stability while concentrating operational and systemic risk that must be carefully governed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:financial-market-infrastructure", "labels": [ "Financial Market Infrastructure" ], "is_subclass_of": [ "Financial System" ], "wikilinks": [] }, { "id": "financial-modelling", "title": "Financial Modelling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Financial modelling is the practice of constructing abstract, quantitative representations of an organisation's financial performance, typically as interlinked spreadsheets or programmatic models that project future cash flows, valuations and outcomes under varying assumptions. Models combine historical data, accounting logic and forward-looking drivers to support decision-making, investment analysis and risk assessment. Common forms include three-statement models, discounted cash-flow valuations and scenario-based projections.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-modelling", "labels": [ "Financial Modelling" ], "is_subclass_of": [ "Quantitative Finance" ], "wikilinks": [] }, { "id": "financial-privacy", "title": "Financial Privacy", "domain": "finance", "domain_name": "Finance", "definition": "Financial Privacy refers to the rights, mechanisms, and technical controls that protect individuals and organisations from unwanted disclosure of their financial transactions, holdings, and economic behaviour. It encompasses both legal frameworks\u2014such as bank secrecy laws and GDPR financial data provisions\u2014and cryptographic techniques such as zero-knowledge proofs, stealth addresses, and confidential transactions. The tension between financial privacy and regulatory transparency requirements (AML, KYC) is a defining challenge of the digital payments era, particularly as public blockchain ledgers make transaction histories permanently visible. Financial privacy is increasingly recognised as a precondition for personal autonomy, political freedom, and protection against targeted financial censorship.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:financial-privacy", "labels": [ "Financial Privacy", "Financial Anonymity" ], "is_subclass_of": [ "Cryptography Security and Privacy" ], "wikilinks": [] }, { "id": "financial-regulation", "title": "Financial Regulation", "domain": "finance", "domain_name": "Finance", "definition": "Financial Regulation comprises the body of statutory rules, supervisory frameworks, licencing regimes, and oversight institutions that govern the conduct of financial markets, intermediaries, and participants. It encompasses prudential regulation (capital adequacy, liquidity, systemic risk), conduct regulation (market integrity, consumer protection, disclosure), and increasingly the oversight of digital asset ecosystems including tokens, stablecoins, and decentralised protocols. Regulatory mandates are administered by national and supranational authorities that establish binding standards, enforce compliance, and coordinate across jurisdictions to address cross-border capital flows, financial crime, and systemic interconnectedness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:financial-regulation", "labels": [ "Financial Regulation", "EU Financial Regulation", "G20 Financial Regulation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "financial-reporting", "title": "Financial Reporting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Financial reporting is the structured disclosure of an organisation's financial position, performance, and cash flows to stakeholders through standardised statements and supporting notes. It rests on recognised accounting frameworks and audit assurance to provide a faithful, comparable, and verifiable view of economic activity. In blockchain and decentralised contexts, financial reporting extends to on-chain transparency, proof-of-reserves attestations, and the accounting treatment of digital assets and tokenised value.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:financial-reporting", "labels": [ "Financial Reporting" ], "is_subclass_of": [ "Blockchain", "Governance and Regulation" ], "wikilinks": [] }, { "id": "financial-services", "title": "Financial Services", "domain": "finance", "domain_name": "Finance", "definition": "Financial Services is the broad sector comprising economic services provided by the finance industry \u2014 including banking, credit, investment management, insurance, payment systems, and capital markets intermediation. These services facilitate the allocation of capital, management of risk, transfer of funds, and exchange of financial instruments across individuals, institutions, and governments. In digital and decentralised contexts, financial services extend to blockchain-based instruments, smart-contract-mediated lending, tokenised asset management, and programmable cross-border payment rails that operate without traditional intermediaries. The sector is governed by a dense regulatory framework spanning prudential oversight, consumer protection, anti-money-laundering requirements, and market integrity rules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:financial-services", "labels": [ "Financial Services", "Financial Services Onboarding" ], "is_subclass_of": [ "Economic Systems" ], "wikilinks": [] }, { "id": "financial-sovereignty", "title": "Financial Sovereignty", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Financial Sovereignty is the capacity of an individual, community, or nation to exercise autonomous control over their financial resources, transactions, and monetary decisions without dependence on or interference from external authorities, intermediaries, or censorship mechanisms. At the individual level it encompasses self-custody of assets, privacy in transactions, and access to financial services irrespective of geographic or political constraints; at the national level it encompasses independent monetary policy and control over reserve currencies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:financial-sovereignty", "labels": [ "Financial Sovereignty" ], "is_subclass_of": [ "Monetary Sovereignty" ], "wikilinks": [] }, { "id": "financial-stability-board", "title": "Financial Stability Board", "domain": "finance", "domain_name": "Finance", "definition": "The Financial Stability Board (FSB) is an international body established in 2009 by the G20 that monitors and makes recommendations about the global financial system to promote financial stability. It coordinates the work of national financial authorities and international standard-setting bodies, and develops and promotes effective regulatory, supervisory, and other financial sector policies to reduce systemic risk.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:financial-stability-board", "labels": [ "Financial Stability Board" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "financial-stability", "title": "financial stability", "domain": "finance", "domain_name": "Finance", "definition": "Financial stability is the condition in which the financial system \u2014 encompassing banks, capital markets, payment infrastructure, insurers, and non-bank intermediaries \u2014 can absorb shocks, maintain the orderly allocation of capital, and continue to perform its core functions of credit intermediation, risk transfer, and payment settlement without requiring extraordinary public support. It is a macroprudential objective distinct from the microprudential health of individual institutions, focusing instead on the resilience and interconnectedness of the system as a whole. Threats to financial stability include excessive leverage, maturity mismatches, procyclical asset valuations, contagion through counterparty networks, and confidence crises that trigger self-fulfilling runs. International bodies such as the Financial Stability Board (FSB), the Bank for International Settlements (BIS), the International Monetary Fund (IMF), and national central banks monitor systemic risk indicators and coordinate macroprudential policy responses.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:financial-stability", "labels": [ "Financial Stability" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "financial-system", "title": "Financial System", "domain": "finance", "domain_name": "Finance", "definition": "The network of institutions, markets, instruments, and infrastructure that facilitate the creation, transfer, and management of financial value. In the context of the metaverse and blockchain ecosystems, financial systems include both traditional payment rails and decentralised protocols enabling programmable value exchange, digital asset custody, and cross-border settlement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:financial-system", "labels": [ "Financial System", "Global Financial System", "Incumbent Financial System" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "financial-technology", "title": "Financial Technology", "domain": "finance", "domain_name": "Finance", "definition": "Financial Technology (FinTech) is the application of digital innovation \u2014 including mobile computing, cloud infrastructure, cryptographic protocols, AI, and distributed ledger technology \u2014 to the design, delivery, and automation of financial products and services. It spans consumer-facing layers such as digital payments, neobanking, and robo-advisory, and infrastructure layers such as real-time payment rails, open banking APIs, regulatory technology, and central bank digital currency systems. FinTech fundamentally restructures the value chain of financial intermediation by reducing friction, lowering costs, and enabling programmable, data-driven financial interactions at global scale.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:financial-technology", "labels": [ "Financial Technology", "FinancialTechnology" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [ "owl:Thing" ] }, { "id": "financial-trading", "title": "Financial Trading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The exchange of financial instruments\u2014including equities, derivatives, cryptocurrencies, and digital assets\u2014within regulated or decentralised markets. In the metaverse and spatial computing context, financial trading encompasses algorithmic and AI-driven trading of virtual assets, NFTs, and tokenised real-world assets through smart-contract-enabled marketplaces and decentralised exchanges.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:financial-trading", "labels": [ "Financial Trading", "Financial Trading Systems" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "financial-transactions", "title": "Financial Transactions", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Financial Transactions are records of the transfer of monetary value or digital assets between parties, executed through centralised payment systems, blockchain networks, or Layer-2 protocols such as the Lightning Network. Within spatial computing and metaverse contexts, financial transactions underpin virtual economies, in-world purchases, NFT trades, and cross-platform value exchange, requiring robust settlement, fraud detection, and regulatory compliance mechanisms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:financial-transactions", "labels": [ "Financial Transactions" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "fine-tuning", "title": "Fine Tuning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The process of adapting a pre-trained model to a specific downstream task by continuing training on task-specific data, typically with a lower learning rate. Fine-tuning leverages knowledge acquired during pre-training whilst specialising the model for particular applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:fine-tuning", "labels": [ "Fine Tuning", "Adapter Fine-tuning", "Fine-Tuning", "Fine-tuning", "LLM Fine-Tuning", "Model Fine-Tuning" ], "is_subclass_of": [ "Transfer Learning", "AI Technique" ], "wikilinks": [ "ComfyWorkFlows", "flux", "visionflow", "ArtificialIntelligenceDomain", "ComfyUI", "ControlNet and Similar Spatial Conditioning Systems", "Death of the Internet", "Face Swap", "Fashion", "Flux.1", "Large Language Models", "LoRA", "LoRA DoRA etc", "Model Optimisation and Performance", "Multimodal", "Open Webui and Pipelines", "style transfer", "Training and fine tuning" ] }, { "id": "finite-element-analysis", "title": "Finite Element Analysis", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Finite Element Analysis (FEA) is a numerical method for approximating solutions to physical field problems by subdividing a continuous domain into a mesh of small, simple elements and assembling their local equations into a global system. It is widely used in engineering to predict structural stress, deformation, heat transfer, vibration and electromagnetic behaviour. FEA is a core technique in computer-aided engineering, enabling virtual testing of designs before physical prototypes are built.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:finite-element-analysis", "labels": [ "Finite Element Analysis", "Finite Element Method" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [] }, { "id": "finite-field-arithmetic", "title": "Finite Field Arithmetic", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Finite Field Arithmetic is the set of operations - addition, subtraction, multiplication and inversion - defined over a finite field (Galois field), a mathematical structure with a finite number of elements in which every non-zero element has a multiplicative inverse. Computations stay closed within the field and behave consistently under modular reduction by a prime or irreducible polynomial. It is foundational to error-correcting codes, cryptography and many digital signal-processing algorithms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:finite-field-arithmetic", "labels": [ "Finite Field Arithmetic" ], "is_subclass_of": [ "Modular Arithmetic" ], "wikilinks": [] }, { "id": "finite-field", "title": "Finite Field", "domain": "security", "domain_name": "Security", "definition": "A finite field, also called a Galois field, is an algebraic structure containing a finite number of elements on which addition, subtraction, multiplication, and division (excluding by zero) are defined and obey the field axioms. Every finite field has a number of elements equal to a prime power, and fields of a given size are unique up to isomorphism. Finite fields are central to cryptography, error-correcting codes, and many computational algorithms because they support exact arithmetic over bounded sets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:finite-field", "labels": [ "Finite Field" ], "is_subclass_of": [ "Modular Arithmetic" ], "wikilinks": [] }, { "id": "finite-state-machine", "title": "Finite State Machine", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A finite state machine (FSM) is an abstract computational model consisting of a finite set of states, a set of input events, and a transition function that maps a current state and input to a next state. At any moment the machine occupies exactly one state, and its behaviour is fully determined by its current state and the inputs it receives. FSMs are widely used to specify control logic, protocol behaviour, and reactive systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:finite-state-machine", "labels": [ "Finite State Machine", "Finite-State Machine" ], "is_subclass_of": [ "State Machine" ], "wikilinks": [] }, { "id": "fintech", "title": "Fintech", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fintech is the application of modern software, networked infrastructure, and data-driven techniques to the delivery and reinvention of financial services. It spans digital payments, mobile and neobanking, online lending, wealth and investment automation, insurance technology, and the regulatory technology that supports them. Fintech firms typically compete with or augment incumbent institutions by lowering cost, broadening access, and improving user experience, while operating within evolving regulatory frameworks that govern consumer protection, data privacy, and financial stability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fintech", "labels": [ "Fintech", "FinTech", "Financial Technology" ], "is_subclass_of": [ "Financial Technology" ], "wikilinks": [] }, { "id": "fireblocks", "title": "Fireblocks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Fireblocks is a company that provides digital asset custody and transfer infrastructure for institutions, using secure key management technology.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:fireblocks", "labels": [ "Fireblocks", "Fireblocks MPC" ], "is_subclass_of": [ "Institutional Custody" ], "wikilinks": [ "Key Management", "Institutional Custody", "Digital Asset" ] }, { "id": "firewall", "title": "Firewall", "domain": "security", "domain_name": "Security", "definition": "A firewall is a network security control that monitors and filters incoming and outgoing traffic according to a defined rule set, allowing or blocking packets and connections to enforce a security boundary. Firewalls range from stateless packet filters to stateful inspection devices and next-generation appliances that perform deep packet inspection and application awareness. They are a foundational component for segmenting trusted and untrusted networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:firewall", "labels": [ "Firewall" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "firmware", "title": "Firmware", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Firmware is the low-level software stored in non-volatile memory that provides the control, monitoring and data-handling logic for a hardware device. It sits between the physical hardware and higher-level software, initialising components, exposing device functions and often forming the only software a simple device runs. Firmware is typically tightly coupled to specific hardware and is updated through controlled mechanisms such as over-the-air updates. It is foundational to embedded systems, peripherals and connected devices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:firmware", "labels": [ "Firmware" ], "is_subclass_of": [ "Embedded Systems" ], "wikilinks": [] }, { "id": "first-party-data", "title": "First Party Data", "domain": "data", "domain_name": "Data", "definition": "First party data is information a company collects directly from its own customers, users or audience through owned channels such as websites, apps, purchase records and CRM systems, with the subject's knowledge or consent. It is distinguished from second-party data (another party's first-party data, shared by agreement) and third-party data (aggregated from external sources without a direct relationship). Its direct provenance makes it more accurate, more compliant with privacy regulation, and increasingly central to marketing and personalisation as third-party cookies are phased out.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:first-party-data", "labels": [ "First Party Data", "First-Party Data" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "first-order-logic", "title": "First-Order Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "First-Order Logic (FOL), also called predicate logic, is a formal system that extends propositional logic with quantifiers, variables, predicates and functions, allowing statements about objects and their relationships. It can express assertions such as \"every X has some Y\" through universal and existential quantification over a domain of discourse. FOL provides a precise syntax and model-theoretic semantics that underpin automated reasoning, knowledge representation and the foundations of mathematics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:first-order-logic", "labels": [ "First-Order Logic" ], "is_subclass_of": [ "Formal Logic" ], "wikilinks": [] }, { "id": "fiscal-policy", "title": "Fiscal Policy", "domain": "finance", "domain_name": "Finance", "definition": "Fiscal policy is the use of government spending and taxation to influence aggregate demand, employment, inflation and economic growth within an economy. It is enacted by a government's treasury or finance ministry through budgets that adjust expenditure programmes, tax rates and public borrowing, and it is the principal counterpart to the monetary policy operated by a central bank. Expansionary fiscal policy raises spending or cuts taxes to stimulate a weak economy, while contractionary policy does the reverse to restrain overheating or reduce public debt.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fiscal-policy", "labels": [ "Fiscal Policy" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "fitting", "title": "Fitting", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:fitting", "labels": [ "Fitting", "Fit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "fixed-supply-monetary-policy", "title": "Fixed Supply Monetary Policy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fixed supply monetary policy is a rule embedded in a cryptocurrency protocol that caps the total quantity of units that will ever exist, removing discretionary issuance. By making the maximum supply algorithmically predetermined and credibly enforced by consensus, it produces verifiable scarcity and predictable issuance schedules. Bitcoin's 21-million-coin cap is the canonical example, underpinning narratives of digital scarcity and store of value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fixed-supply-monetary-policy", "labels": [ "Fixed Supply Monetary Policy" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "fixed-supply-token", "title": "Fixed Supply Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Fixed Supply Token is a fungible blockchain token whose total issuance is capped at a hard-coded maximum that the protocol cannot exceed, regardless of future governance decisions. Distribution of the supply over time is governed by a predetermined emission schedule, and no additional tokens can be minted once the cap is reached, conferring deflationary scarcity properties analogous to precious metals.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:fixed-supply-token", "labels": [ "Fixed Supply Token" ], "is_subclass_of": [ "Token and Asset", "Fungible Token" ], "wikilinks": [ "Blockchain", "Fungible Token" ] }, { "id": "flash-attention", "title": "flash attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Flash Attention is an IO-aware, exact attention algorithm that exploits the GPU memory hierarchy by tiling query, key, and value matrices to keep intermediate activations in fast on-chip SRAM rather than the much slower high-bandwidth DRAM (HBM), thereby eliminating the O(N\u00b2) memory materialisation of standard scaled dot-product attention. By applying an online softmax recomputation strategy, it achieves bitwise-identical output to vanilla attention while reducing HBM reads and writes proportionally to the SRAM tile size. Originally introduced by Dao et al. in 2022, it has been extended through FlashAttention-2 (improved thread-block partitioning) and FlashAttention-3 (Hopper-native asynchrony and FP8 support), and is now embedded in PyTorch, JAX, vLLM, and virtually every frontier model training pipeline.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:flash-attention", "labels": [ "Flash Attention", "FlashAttention", "FlashAttention Kernel" ], "is_subclass_of": [ "Attention Mechanism" ], "wikilinks": [] }, { "id": "flash-loan", "title": "Flash Loan", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A flash loan is an uncollateralised loan mechanism in decentralised finance that exists entirely within a single blockchain transaction: the borrower receives an arbitrary amount of an asset, executes arbitrary on-chain operations with it, and repays the loan plus a fee within the same atomic transaction, with the entire sequence reverting if repayment fails. Because atomicity guarantees that funds never leave the lending pool without being returned, no collateral is needed; default is technically impossible since a failed repayment causes the transaction to revert as if the loan never occurred. Flash loans enable capital-efficient arbitrage, liquidation, collateral swaps, and self-liquidation operations that would otherwise require significant upfront capital.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:flash-loan", "labels": [ "Flash Loan" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "flat-panel-display", "title": "Flat Panel Display", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A flat panel display is a thin, lightweight electronic display that produces images on a flat surface using technologies such as liquid crystal, organic light-emitting diode or micro-LED panels. It contrasts with bulky cathode-ray-tube displays and forms the dominant display category for monitors, mobile devices and near-eye optics in head-mounted hardware. Key characteristics include resolution, refresh rate, contrast and pixel density.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:flat-panel-display", "labels": [ "Flat Panel Display" ], "is_subclass_of": [ "Display Technology" ], "wikilinks": [] }, { "id": "fleet-management", "title": "Fleet Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fleet management is the coordinated administration, monitoring, and optimisation of a collection of mobile assets \u2014 vehicles, drones, robots, or edge devices \u2014 across their operational lifecycle, encompassing real-time telemetry collection, route and task planning, maintenance scheduling, regulatory compliance, driver or operator management, and fuel or energy consumption optimisation. Modern fleet management systems integrate GPS tracking, telematics sensors, AI-driven analytics, and communications networks to provide operators with situational awareness and control over large numbers of assets dispersed across wide geographic areas.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:fleet-management", "labels": [ "Fleet Management", "Robot Fleet Management" ], "is_subclass_of": [ "Asset Management" ], "wikilinks": [] }, { "id": "flexible-manufacturing", "title": "Flexible Manufacturing", "domain": "robotics", "domain_name": "Robotics", "definition": "Flexible manufacturing is a production approach in which automated, reconfigurable equipment and material-handling systems can adapt rapidly to changes in product mix and volume with minimal downtime. It combines computer numerical control machines, robots, and automated transport under coordinated supervisory control so that the same line can produce varied parts on demand. The approach targets responsiveness and customisation while retaining the efficiency of automation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:flexible-manufacturing", "labels": [ "Flexible Manufacturing" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "flight-control-system", "title": "Flight Control System", "domain": "robotics", "domain_name": "Robotics", "definition": "A flight control system is the set of sensors, actuators, and control-law software that stabilises an aircraft or aerial robot and translates pilot or autopilot commands into control-surface or motor movements. It fuses data from inertial and other sensors to estimate attitude and rate, then computes corrective outputs at high frequency to maintain stable flight. In unmanned aerial robots it typically also implements guidance and navigation loops on top of the core stabilisation control.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:flight-control-system", "labels": [ "Flight Control System" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "flight-software", "title": "Flight Software", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:flight-software", "labels": [ "Flight Software" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "floating-point-arithmetic", "title": "Floating-Point Arithmetic", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Floating-point arithmetic is a method of representing and computing with real numbers on digital hardware using a sign, a fixed-precision significand, and an exponent, most commonly standardised by IEEE 754. It trades exactness for a wide dynamic range, so operations introduce rounding error, and properties such as associativity no longer hold exactly. Understanding its precision limits, rounding modes, and special values is essential for numerically reliable simulation, graphics, and machine learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:floating-point-arithmetic", "labels": [ "Floating-Point Arithmetic", "FP4 Arithmetic", "Fixed-Point Arithmetic", "Floating-Point Precision" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "flood-bank", "title": "Flood Bank", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:flood-bank", "labels": [ "Flood Bank" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "flood-mapping", "title": "Flood Mapping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:flood-mapping", "labels": [ "Flood Mapping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "flow-control", "title": "Flow Control", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Flow control is the mechanism by which a data communication protocol prevents a fast sender from overwhelming a slower receiver. It regulates the rate or volume of data in transit using techniques such as sliding windows, credit schemes and backpressure, so that the receiver's buffers are not exceeded. Distinct from congestion control, which protects the shared network, flow control is an end-to-end concern that protects the individual receiving endpoint.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:flow-control", "labels": [ "Flow Control" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "flow-matching", "title": "flow matching", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Flow Matching is a generative modelling framework that trains a neural network to parameterise the vector field of a continuous normalising flow by directly regressing on analytically tractable target vector fields defined along conditional flow paths between a source distribution (typically Gaussian noise) and a data distribution. Unlike score-based diffusion models, which require simulating a stochastic differential equation during training, flow matching uses deterministic conditional flow paths that can be computed in closed form via Conditional Flow Matching (CFM), yielding stable and efficient training without ODE simulation at training time. At inference the learned vector field is integrated with an ODE solver to transport samples from noise to data, enabling high-quality generation with fewer neural function evaluations than diffusion alternatives. The framework unifies and generalises prior work on continuous normalising flows, score matching, and diffusion probabilistic models under a single regression objective.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:flow-matching", "labels": [ "Flow Matching" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "flow-state", "title": "Flow State", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Flow state is a psychological condition of complete absorption in an activity, characterised by intense focus, loss of self-consciousness, distortion of the sense of time, and a balance between perceived challenge and skill. In spatial and immersive computing it is a target experiential outcome, where well-designed interaction and feedback sustain deep engagement. Cultivating flow is central to the design of immersive learning, games, and presence-driven environments because it maximises both performance and intrinsic satisfaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:flow-state", "labels": [ "Flow State" ], "is_subclass_of": [ "Immersion" ], "wikilinks": [] }, { "id": "fluid-power-device", "title": "Fluid Power Device", "domain": "robotics", "domain_name": "Robotics", "definition": "A fluid power device is a component that transmits or controls power through a pressurised fluid, either an incompressible liquid (hydraulic) or a compressible gas (pneumatic). Such devices convert fluid pressure and flow into mechanical motion and force, or regulate that flow, and include actuators, valves, pumps, and compressors. They are valued in robotics and industrial automation for high force-to-weight ratios and robust operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fluid-power-device", "labels": [ "Fluid Power Device" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "fluid-simulation", "title": "Fluid Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Fluid Simulation is the computational modelling of liquid and gas dynamics using numerical methods \u2014 notably computational fluid dynamics (CFD) and Navier-Stokes solvers \u2014 to produce physically plausible representations of water, smoke, fire, and atmospheric phenomena in real-time or offline rendering pipelines. In spatial computing and metaverse environments it contributes to environmental immersion and interactive physics, running on GPU compute shaders for performance-critical applications.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:fluid-simulation", "labels": [ "Fluid Simulation" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "CFD", "Metaverse", "MetaverseDomain", "Physics Engine" ] }, { "id": "flux-1", "title": "Flux.1", "domain": "ai", "domain_name": "Ai", "definition": "Flux.1 is a family of open-weight text-to-image generative models developed by Black Forest Labs, founded by former Stability AI researchers including Robin Rombach, the co-creator of Latent Diffusion. Released in August 2024, Flux.1 employs a hybrid architecture combining multimodal and parallel diffusion transformer (DiT) blocks, achieving state-of-the-art image quality and prompt adherence that surpasses earlier diffusion models on benchmarks such as GenEval and T2I-CompBench. The family offers three variants \u2014 Flux.1 [pro], Flux.1 [dev], and Flux.1 [schnell] \u2014 spanning commercial API, open-weights research, and fast inference use cases respectively.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:flux-1", "labels": [ "Flux.1" ], "is_subclass_of": [ "Diffusion Model", "Generative AI" ], "wikilinks": [] }, { "id": "fog-computing", "title": "Fog Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Fog computing is a distributed architecture that places compute, storage, and networking resources in an intermediate tier between end devices and the cloud, typically in gateways, routers, and local servers near the data source. It extends cloud capabilities toward the network edge to reduce latency, conserve bandwidth, and improve resilience and privacy for geographically dispersed Internet-of-Things deployments. Distinct from edge computing's focus on the device itself, fog computing emphasises a coordinated, hierarchical layer of regional nodes orchestrating many edges.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:fog-computing", "labels": [ "Fog Computing", "Fog Computing Layer" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "food-safety-blockchain", "title": "Food Safety Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Food Safety Blockchain is the application of distributed ledger technology to establish immutable, multi-party farm-to-fork audit trails that enable near-instantaneous contamination source identification and surgical product recalls. By recording each Critical Tracking Event\u2014harvest, cooling, packing, shipment, receipt\u2014as a cryptographically linked on-chain entry, the approach reduces trace-back from days to seconds, minimises the volume of food unnecessarily recalled, and provides regulators and consumers with verifiable provenance data.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:food-safety-blockchain", "labels": [ "Food Safety Blockchain", "BC-0443-food-safety-blockchain" ], "is_subclass_of": [ "Blockchain Application" ], "wikilinks": [ "BC-0013-smart-contracts", "BC-0029-permissioned-blockchain", "BC-0044-supply-chain-management", "BC-0067-hyperledger-fabric", "BC-0214-environmental-sustainability", "BC-0432-consortium-blockchain", "BC-0434-blockchain-as-a-service", "BC-0441-provenance-tracking", "BC-0442-pharmaceutical-traceability", "Cosmos", "Grass Roots Farmers' Cooperative", "GS1", "GS1 Blockchain Working Group", "IBM Food Trust", "Internet of Things", "OriginTrail", "Polkadot", "Provenance", "Ripe.io", "SafeTraces" ] }, { "id": "food-safety", "title": "Food Safety", "domain": "governance", "domain_name": "Governance", "definition": "Food Safety is the discipline concerned with preventing foodborne illness and injury through systematic controls applied across the entire food production and distribution chain, from farm to consumer. It encompasses hazard analysis, regulatory compliance, traceability, temperature monitoring, and the technical and institutional practices that ensure food products are safe for consumption.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:food-safety", "labels": [ "Food Safety" ], "is_subclass_of": [ "Compliance" ], "wikilinks": [] }, { "id": "fooocus", "title": "Fooocus", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Fooocus is an open-source desktop image-generation interface for Stable Diffusion XL released in August 2023 by Lvmin Zhang (GitHub handle lllyasviel, originator of ControlNet, IC-Light, Forge, Paints-Undo, FramePack and OmniControl), conceived as a deliberate philosophical and ergonomic reaction...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:fooocus", "labels": [ "Fooocus" ], "is_subclass_of": [ "AI Application", "Stable Diffusion Frontend", "Generative AI Application", "Open Source Software", "Local Image Generation Interface", "Gradio Application" ], "wikilinks": [ "ACM CHI 2024 Generative AI Interfaces Studies", "Adobe Firefly", "Automatic FreeU Module", "Automatic Refiner Scheduler", "AUTOMATIC1111 Stable Diffusion WebUI Repository", "AUTOMATIC1111 WebUI", "CLIP Skip", "CLIP Text Encoder", "ComfyUI Repository", "Concept Art Workflows", "ControlNet", "ControlNet Conditioning", "CreativeML Open RAIL-M Licence", "Creative Tools", "CUDA Toolkit", "DALL-E 3", "DPM Plus Plus 2M Karras Sampler", "Educational AI Demonstrations", "Faceswap", "Faceswap Module" ] }, { "id": "force-control", "title": "Force Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Force Control is the family of robotic control paradigms that regulate the contact force and/or torque exerted by a manipulator, end-effector, joint or whole-body system on its environment rather than (or in addition to) regulating Cartesian or joint position, formalised through a closed-loop rel...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:force-control", "labels": [ "Force Control", "Compliant Force Control", "Force Closure", "ForceControl" ], "is_subclass_of": [ "Actuation and Control", "Robotic Control", "Closed Loop Control", "Interaction Control", "Cyber-Physical Control" ], "wikilinks": [ "ActuationLayer", "Albu-Sch\u00e4ffer Ott Hirzinger 2007 Passivity Flexible Joint", "AlgorithmLayer", "Anti-Windup", "Archive Market Research 2025 Integrated Force Controller", "Bipedal Balance", "Brohan et al. 2023 RT-2 Vision Language Action", "Carpentier et al. 2019 Pinocchio C++ Library", "Chi et al. 2023 Diffusion Policy", "Collaborative Robotics", "Compliance Frame", "Compliant Manipulation", "Computed Torque Control", "Control Law", "ControlLayer", "ControlSystemsDomain", "Coordinate Transformation", "Cyber-Physical Control", "Deburring", "Differential Dynamic Programming" ] }, { "id": "force-feedback", "title": "Force Feedback", "domain": "robotics", "domain_name": "Robotics", "definition": "Force feedback is a class of haptic technology that renders mechanical forces \u2014 including resistance, weight, texture, and impact \u2014 directly to a user's body through a controlled actuator system, enabling the sense of touch and proprioception to convey information about virtual or remote physical environments. Distinguished from simpler vibrotactile feedback by its ability to generate directional, grounded forces (requiring a mechanical linkage to the user), force feedback systems are used in surgical simulators, teleoperation of remote robots, vehicle simulation, and advanced XR interfaces. The fidelity of the rendered force field is constrained by the bandwidth, peak force, backdrivability, and transparency of the underlying actuator mechanism.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:force-feedback", "labels": [ "Force Feedback", "ForceFeedback" ], "is_subclass_of": [ "Haptics" ], "wikilinks": [] }, { "id": "force-sensor", "title": "Force Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A force sensor is a transducer that measures the magnitude (and sometimes direction) of an applied mechanical force or torque, typically by converting strain in an elastic element into an electrical signal. Single-axis load cells and multi-axis force-torque sensors give robots haptic awareness of contact and interaction forces. Such measurements are essential for compliant manipulation, assembly, and safe physical human-robot interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:force-sensor", "labels": [ "Force Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "force-torque-control", "title": "Force Torque Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Force-torque control is a robot control strategy that regulates the contact forces and moments a manipulator exerts on its environment, rather than commanding position alone. It uses force-torque sensing, typically at the wrist or in the joints, within a feedback loop so that the robot can maintain a desired contact force or yield compliantly to external loads. This is essential for tasks involving physical contact, such as assembly, polishing, and safe interaction with people.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:force-torque-control", "labels": [ "Force Torque Control", "Force-Torque Control" ], "is_subclass_of": [ "Force Control" ], "wikilinks": [] }, { "id": "force-torque-sensor", "title": "Force Torque Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "Force-Torque Sensor - A multi-axis transducer mounted on the Robot Wrist that measures three-dimensional forces and torques (6-DoF) exerted during interaction with objects or humans, enabling Force Feedback, Contact Detection, and Compliance Control in precision manipulation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:force-torque-sensor", "labels": [ "Force Torque Sensor", "Force Torque Estimation", "Force Torque Sensing", "Force-Torque Sensing", "Force-Torque Sensor", "Force/Torque Sensor", "Six-Axis Force Torque Sensor" ], "is_subclass_of": [ "Sensor", "Robotics" ], "wikilinks": [ "Calibration", "Contact Detection", "Force Feedback", "Haptic Feedback System", "Precision Assembly", "Robot Wrist", "Signal Amplification", "Soft Robotic Grasping", "Wrist Assembly", "Compliance Control", "Computer Vision", "Data Processing", "Impedance Control", "Robotics", "RoboticsDomain", "Sensor" ] }, { "id": "forecast-horizon", "title": "Forecast Horizon", "domain": "data", "domain_name": "Data", "definition": "The forecast horizon is the length of time into the future over which a predictive model generates estimates, measured as the number of time steps ahead from the last observation. It is a key design parameter in time-series forecasting that trades off relevance against uncertainty, since predictive error generally grows with the horizon. Choice of horizon shapes model selection, evaluation strategy, and the operational decisions a forecast supports.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:forecast-horizon", "labels": [ "Forecast Horizon" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "foreign-exchange-market", "title": "Foreign Exchange Market", "domain": "finance", "domain_name": "Finance", "definition": "The foreign exchange market is the global, decentralised marketplace in which national currencies are bought, sold and exchanged at floating or managed rates. It is the largest and most liquid financial market in the world, operating continuously across major trading centres and enabling international trade, investment and the conversion of one fiat currency into another. Exchange rates set in this market influence cross-border payments, monetary policy transmission and international price competitiveness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:foreign-exchange-market", "labels": [ "Foreign Exchange Market" ], "is_subclass_of": [ "Monetary System" ], "wikilinks": [] }, { "id": "foreign-exchange", "title": "Foreign Exchange", "domain": "finance", "domain_name": "Finance", "definition": "The global, decentralised market and settlement activity in which one currency is exchanged for another at agreed exchange rates, spanning spot transactions, forwards, swaps, and options. Trading around nine and a half trillion US dollars daily (BIS 2025 Triennial Survey), it is the largest financial market in the world, providing the currency conversion that underlies international trade, investment, and cross-border payments, with settlement risk managed through mechanisms such as CLS payment-versus-payment and liquidity intermediated by dealer banks and correspondent networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:foreign-exchange", "labels": [ "Foreign Exchange" ], "is_subclass_of": [ "Financial Instruments" ], "wikilinks": [ "Financial Instruments", "Exchange Rate", "Correspondent Banking", "Cross-Border Payments" ] }, { "id": "forensic-analysis", "title": "Forensic Analysis", "domain": "security", "domain_name": "Security", "definition": "Forensic analysis is the disciplined investigation of digital systems and data to reconstruct events, attribute actions and preserve evidence to an evidentiary standard. In security it follows an incident to determine how a breach occurred, what was affected and who was responsible, maintaining a defensible chain of custody throughout. It draws on log analysis, memory and disk examination, and timeline reconstruction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:forensic-analysis", "labels": [ "Forensic Analysis" ], "is_subclass_of": [ "Incident Response" ], "wikilinks": [] }, { "id": "forensic-evidence-court-framework", "title": "Forensic Evidence Court Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A UK-focused jurisdictional practice framework for integrating crime scene reconstruction with court proceedings, combining forensic evidence management, chain-of-custody protocols, and digital evidence presentation standards to support admissible expert testimony.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:forensic-evidence-court-framework", "labels": [ "Forensic Evidence Court Framework", "arpana" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "crime uk", "law" ] }, { "id": "forensic-investigation", "title": "Forensic Investigation", "domain": "security", "domain_name": "Security", "definition": "Forensic investigation is the systematic collection, preservation, examination, and analysis of evidence \u2014 digital, physical, or financial \u2014 using scientifically validated methods that maintain legal admissibility and chain of custody, with the purpose of reconstructing events, attributing responsibility, and supporting judicial or regulatory proceedings. In digital contexts it encompasses disk imaging, memory acquisition, network traffic analysis, and log correlation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:forensic-investigation", "labels": [ "Forensic Investigation" ], "is_subclass_of": [ "Digital Forensics" ], "wikilinks": [] }, { "id": "fork-choice-rule", "title": "Fork Choice Rule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Algorithm by which blockchain nodes select the canonical chain head when multiple competing branches exist, resolving temporary forks deterministically. Bitcoin uses longest-chain (most cumulative proof-of-work) while Ethereum post-Merge uses the LMD-GHOST fork-choice weighted by validator attestation stake, determining network-wide finality.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:fork-choice-rule", "labels": [ "Fork Choice Rule" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "fork", "title": "Fork", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Fork in the context of blockchain and distributed systems is a divergence in the protocol rules or chain history that results in two or more distinct execution paths from a common ancestor state. Hard forks introduce backward-incompatible rule changes requiring all participants to upgrade, potentially creating a permanently divergent chain. Soft forks introduce backward-compatible tightening of rules. Forks can be planned governance events (protocol upgrades) or unintended consequences of network partitions or competing miner/validator behaviour.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fork", "labels": [ "Fork" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "formal-language", "title": "Formal Language", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A formal language is a set of strings of symbols drawn from a finite alphabet, defined precisely by formal rules such as a grammar or automaton rather than by usage or convention. The Chomsky hierarchy classifies formal languages by the generative power required to describe them, ranging from regular languages recognised by finite automata to recursively enumerable languages recognised by Turing machines. Formal languages provide the mathematical foundation for specifying syntax in programming languages, parsers, logic, and knowledge representation. They are studied in automata theory and underpin compilers and ontology languages.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:formal-language", "labels": [ "Formal Language" ], "is_subclass_of": [ "Automata Theory" ], "wikilinks": [] }, { "id": "formal-logic", "title": "Formal Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Formal logic is the study of inference using precisely defined symbolic languages and rules, abstracting valid reasoning patterns from the content of particular arguments. It specifies syntax for well-formed formulae and semantics that assign truth conditions, enabling proofs to be checked mechanically. In artificial intelligence it underpins knowledge representation, automated reasoning and the verification of systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:formal-logic", "labels": [ "Formal Logic" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "formal-methods", "title": "Formal Methods", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Formal methods are mathematically rigorous techniques for the specification, development, and verification of software and hardware systems. They use formal logic, set theory, and automata to express system requirements precisely and to prove that an implementation satisfies them. In blockchain and smart-contract engineering, formal methods are applied to prove the absence of classes of bugs and to guarantee that contract behaviour matches its intended specification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:formal-methods", "labels": [ "Formal Methods" ], "is_subclass_of": [ "Formal Verification" ], "wikilinks": [] }, { "id": "formal-proof", "title": "Formal Proof", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A formal proof is a rigorous, mathematically precise derivation that establishes a proposition follows from a set of axioms and inference rules, without reliance on informal argument or empirical testing. In distributed systems theory, formal proofs are used to establish results such as impossibility theorems and the correctness of consensus protocols under stated failure and timing assumptions. They provide the strongest available assurance of correctness, distinguishing rigorously proven properties from properties merely tested or believed.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:formal-proof", "labels": [ "Formal Proof" ], "is_subclass_of": [ "Formal Verification" ], "wikilinks": [] }, { "id": "formal-specification", "title": "Formal Specification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A formal specification is a precise, mathematically grounded description of the intended behaviour or structure of a system, written in a language with well-defined syntax and semantics. It allows properties of the system to be stated unambiguously and reasoned about or verified mechanically. Formal specifications underpin formal methods, model checking, and the construction of provably correct software and ontologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:formal-specification", "labels": [ "Formal Specification" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "formal-verification", "title": "formal verification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Formal verification is the application of mathematical proof techniques \u2014 including model checking, theorem proving, abstract interpretation, and satisfiability-modulo-theories (SMT) solving \u2014 to rigorously establish that a hardware or software system satisfies a specified set of correctness, safety, or security properties under all possible inputs and execution paths. Unlike testing, which can only expose defects in sampled executions, formal verification provides exhaustive, machine-checkable guarantees by reasoning over the entire state space or by constructing logical proofs relative to a formal specification. The discipline encompasses hardware circuit verification, operating-system kernel correctness proofs, smart-contract auditing, neural-network robustness certification, and protocol security analysis, and is increasingly mandated in safety-critical and regulated industries where defects carry life-critical or large financial consequences.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:formal-verification", "labels": [ "Formal Verification" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "format-compliance", "title": "Format Compliance", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Format compliance is the property of a digital asset or file conforming exactly to the structural, encoding, and schema rules of a defined file or data format specification. Verifying compliance ensures that assets can be reliably parsed, exchanged, and rendered by conformant tools across the supply chain. It is a precondition for interoperability between content producers and consumers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:format-compliance", "labels": [ "Format Compliance" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "format-migration", "title": "Format Migration", "domain": "data", "domain_name": "Data", "definition": "Format migration is the process of converting data or digital assets from one file format to another while preserving meaning, structure, and fidelity. It is central to digital preservation, where obsolete formats are migrated to current ones to keep content accessible over time. Migration must manage information loss, metadata mapping, and validation of the converted output.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:format-migration", "labels": [ "Format Migration" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "format-parser", "title": "Format Parser", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software components that interpret and convert various 3D asset file formats into internal representations for metaverse platforms, enabling interoperability between content creation tools and runtime environments whilst preserving geometry, materials, animations, and metadata.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:format-parser", "labels": [ "Format Parser" ], "is_subclass_of": [ "Standards and Interoperability", "Data Processing" ], "wikilinks": [ "Cross-Platform Content", "Data Processing", "metaverse" ] }, { "id": "formation-flying", "title": "Formation Flying", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:formation-flying", "labels": [ "Formation Flying" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "formative-assessment", "title": "Formative Assessment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Formative assessment is an ongoing, low-stakes evaluation practice used by educators and learners to monitor understanding and progress during the learning process rather than at its conclusion, with the explicit purpose of informing instructional adjustments and learner self-regulation. Unlike summative assessment, formative assessment results are used to guide immediate pedagogical decisions \u2014 providing feedback, identifying misconceptions, and adapting content difficulty \u2014 rather than to assign grades or certifications. In spatial computing and XR learning environments, formative assessment can be embedded invisibly in simulation tasks, analysing performance traces in real time.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:formative-assessment", "labels": [ "Formative Assessment" ], "is_subclass_of": [ "Spatial Computing", "Learning Analytics" ], "wikilinks": [] }, { "id": "forward-chaining", "title": "Forward Chaining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A data-driven inference strategy for rule-based systems that starts from known facts and repeatedly applies rules whose conditions are satisfied, asserting their conclusions as new facts until no further rules fire or a goal is derived. Formalised as repeated application of modus ponens over a working memory, it is the recognise-act cycle at the heart of production systems, complete for definite-clause knowledge bases, and efficiently implemented by pattern-matching algorithms such as Rete.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:forward-chaining", "labels": [ "Forward Chaining" ], "is_subclass_of": [ "Inference" ], "wikilinks": [ "Inference", "Rule-Based Systems", "Backward Chaining", "Expert Systems", "Inference Engine" ] }, { "id": "forward-error-correction", "title": "Forward Error Correction", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Forward error correction (FEC) is a channel-coding technique in which redundant information is added to a transmitted data stream so that the receiver can detect and correct errors introduced by the channel without requesting retransmission. FEC codes trade bandwidth or storage overhead for improved reliability over noisy or lossy channels, eliminating the round-trip latency penalty of automatic repeat request (ARQ) schemes.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:forward-error-correction", "labels": [ "Forward Error Correction" ], "is_subclass_of": [ "Channel Coding" ], "wikilinks": [] }, { "id": "forward-guidance", "title": "Forward Guidance", "domain": "finance", "domain_name": "Finance", "definition": "Forward guidance is a monetary policy communication tool through which a central bank signals its likely future path of interest rates or other policy actions, in order to shape market expectations and financial conditions ahead of actual policy changes. It works by influencing longer-term interest rates and asset prices today, since those prices reflect expectations of future short-term rates. Forward guidance is a core instrument of monetary policy transmission, particularly when policy rates are near their effective lower bound.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:forward-guidance", "labels": [ "Forward Guidance" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "forward-kinematics", "title": "Forward Kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "The process of determining the position and orientation of a robot's end-effector in Cartesian space given the joint parameters (angles or displacements). It maps from joint space to task space using geometric and trigonometric relationships, producing a unique closed-form solution via sequential homogeneous transformation matrices.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:forward-kinematics", "labels": [ "Forward Kinematics", "RB-1005-forward-kinematics" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [ "Computability", "Denavit-Hartenberg Parameters", "End-Effector Pose", "Homogeneous Coordinates", "Joint Parameters", "Kinematic Model", "Motion Visualization", "RB-0003-manipulator", "RB-1006-inverse-kinematics", "Robot Simulation", "Transformation Matrix", "Uniqueness", "Kinematics", "Robotics", "Spatial Computing" ] }, { "id": "forward-pass", "title": "Forward Pass", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The forward pass is the computation that propagates input data through the layers of a neural network to produce an output, applying weighted sums, biases and activation functions in sequence. It evaluates the network's current function and, during training, produces the predictions against which the loss is measured. The intermediate activations it computes are retained so that the subsequent backward pass can calculate gradients.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:forward-pass", "labels": [ "Forward Pass" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "forward-secrecy", "title": "Forward Secrecy", "domain": "security", "domain_name": "Security", "definition": "Forward secrecy, also called perfect forward secrecy, is a property of key-agreement protocols ensuring that the compromise of long-term private keys does not allow an attacker to decrypt previously recorded session traffic. It is achieved by deriving ephemeral session keys for each connection through a fresh key exchange and discarding them afterwards, so that no single persistent secret can retroactively unlock past communications. The property is a cornerstone of modern transport-layer security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:forward-secrecy", "labels": [ "Forward Secrecy" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "foundation-model-layer", "title": "Foundation Model Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Foundation Model Layer is the stratum that holds large, broadly pretrained models intended for adaptation to many downstream tasks. It sits above the Training Layer that produced it and below the Model and Inference Layers that specialise and serve it. It contains base model weights, pretraining configurations, and adaptation interfaces.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:foundation-model-layer", "labels": [ "Foundation Model Layer" ], "is_subclass_of": [ "AI Model Architecture", "owl:Thing" ], "wikilinks": [ "Training Layer", "Model Architecture Layer", "Model Layer", "Inference Layer", "Transfer Learning", "Self-Supervised Learning", "owl:Thing" ] }, { "id": "foundation-model", "title": "Foundation Model", "domain": "ai", "domain_name": "Ai", "definition": "A Foundation Model is a large-scale machine learning model trained on broad, diverse datasets using self-supervised or semi-supervised objectives, producing a general-purpose parametric representation that can be adapted to a wide range of downstream tasks through fine-tuning, prompting, or retrieval augmentation without retraining from scratch. Introduced as a conceptual category by the Stanford HAI Centre for Research on Foundation Models in 2021, the term emphasises the homogenising role such models play across AI research and application domains: a single pre-trained artefact serves as the foundation for specialised systems in natural language processing, computer vision, speech, multimodal reasoning, scientific discovery, and robotics. Foundation models are characterised by emergent capabilities that arise from scale \u2014 properties not present in smaller models that appear as parameter count and training-data volume increase \u2014 and by their fundamentally transferable representations, which dramatically lower the cost of building capable task-specific systems. The category encompasses large language models, vision-language models, diffusion models, and cross-modal architectures such as GPT-4, BERT, DALL-E, Stable Diffusion, and Segment Anything.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:foundation-model", "labels": [ "Foundation Model", "Foundation Model Race", "Foundation Models" ], "is_subclass_of": [ "Machine Learning Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "foundation-models", "title": "Foundation Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Foundation models are large neural networks pretrained on broad, diverse data at scale, producing general-purpose representations that can be adapted to many downstream tasks through fine-tuning or prompting. They underpin most modern large language models, vision models and multimodal systems, and their training typically relies on vast web-scale corpora such as Common Crawl. Their broad capability comes at the cost of high training compute and emergent, sometimes unpredictable, behaviour.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:foundation-models", "labels": [ "Foundation Models" ], "is_subclass_of": [ "Pretrained Model" ], "wikilinks": [] }, { "id": "foundry", "title": "Foundry", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Foundry is a fast, modular, and portable Ethereum application development toolkit written in Rust, comprising four core tools: Forge (test framework), Cast (EVM interaction CLI), Anvil (local testnet node), and Chisel (Solidity REPL). It enables developers to write, compile, fuzz-test, and deploy Solidity and Vyper smart contracts entirely from the command line, with tests written directly in Solidity rather than JavaScript. Foundry has become the dominant professional-grade smart-contract development environment on Ethereum-compatible chains, replacing earlier JavaScript-based toolchains such as Hardhat and Truffle for many teams. Its architecture emphasises speed through native compilation and parallelised test execution, deterministic reproducibility via pinned dependencies, and deep EVM-level inspection through cheatcodes and traces.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:foundry", "labels": [ "Foundry", "Semiconductor Foundry" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Solidity", "Smart Contract", "Ethereum" ] }, { "id": "fourier-analysis", "title": "Fourier Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Fourier analysis is the branch of mathematics that decomposes functions or signals into sums of sinusoidal components, representing them in terms of frequency rather than time or space. Its central tool, the Fourier transform, maps a signal to its spectrum, revealing periodicities and enabling operations such as filtering and convolution to be performed efficiently in the frequency domain. It is foundational to signal processing, communications, and many numerical and learning methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fourier-analysis", "labels": [ "Fourier Analysis" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "fourier-transform", "title": "Fourier Transform", "domain": "ai", "domain_name": "Ai", "definition": "The Fourier Transform is a mathematical operation that decomposes a function of time or space into its constituent frequency components, expressing the function as a sum of sinusoids weighted by complex amplitudes. It establishes a bijective mapping between the time domain and the frequency domain, enabling analysis, filtering, compression, and convolution of signals. The Discrete Fourier Transform (DFT) and its fast algorithm (FFT) are the computational workhorses of digital signal processing, image analysis, and frequency-domain machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fourier-transform", "labels": [ "Fourier Transform" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "fourth-industrial-revolution", "title": "Fourth Industrial Revolution", "domain": "governance", "domain_name": "Governance", "definition": "The Fourth Industrial Revolution (Industry 4.0) is the contemporary wave of technological change characterised by the fusion of physical, digital, and biological systems through AI, robotics, the Internet of Things, and cyber-physical systems. Popularised by the World Economic Forum, it describes how automation and data exchange reshape manufacturing, economies, and society. It is distinguished from prior revolutions by the speed, scope, and systemic impact of convergent technologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fourth-industrial-revolution", "labels": [ "Fourth Industrial Revolution" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "foveated-rendering", "title": "foveated rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Foveated rendering is a real-time graphics technique that exploits the non-uniform spatial acuity of the human visual system by rendering the region around the viewer's gaze point at full resolution whilst progressively reducing shading quality, texture resolution, and geometric detail in the peripheral visual field. Eye-tracking hardware continuously locates the foveal fixation point so that the high-quality region follows the gaze with sub-frame latency, making quality degradation imperceptible. The technique produces substantial GPU workload reductions that are critical for achieving high frame rates and thermal sustainability on power-constrained XR head-mounted displays, and is increasingly extended by neural super-resolution to reconstruct peripheral detail at low cost.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:foveated-rendering", "labels": [ "Foveated Rendering" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [] }, { "id": "fractional-ownership", "title": "Fractional Ownership", "domain": "finance", "domain_name": "Finance", "definition": "Fractional ownership is a model in which a high-value asset is divided into discrete shares that multiple parties hold simultaneously, each acquiring proportional economic rights and, in some structures, governance rights over the underlying asset. Historically applied to aircraft, real estate, and fine art through legal syndication, it has been radically simplified by blockchain-based tokenisation, which encodes shares as on-chain tokens transferable without traditional intermediaries. Fractional ownership democratises access to asset classes previously restricted to institutional or wealthy investors. It raises important questions of regulatory classification, valuation, and liquidity management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fractional-ownership", "labels": [ "Fractional Ownership" ], "is_subclass_of": [ "Asset Tokenization" ], "wikilinks": [] }, { "id": "fractional-reserve-banking", "title": "Fractional Reserve Banking", "domain": "finance", "domain_name": "Finance", "definition": "Fractional reserve banking is the standard banking system in which banks hold only a fraction of their deposit liabilities as reserves and lend out the remainder. Because lending creates new deposits that can be re-lent, the system expands the broad money supply far beyond the monetary base through the money multiplier. It enables credit creation and maturity transformation but exposes individual banks to liquidity risk and the possibility of bank runs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:fractional-reserve-banking", "labels": [ "Fractional Reserve Banking" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "fractionalized-nft", "title": "Fractionalized NFT", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A non-fungible token whose ownership has been divided into multiple fungible token shares, enabling collective ownership and enhanced liquidity of high-value unique digital assets.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:fractionalized-nft", "labels": [ "Fractionalized NFT" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Collective Ownership", "Custody System", "ERC-1155", "Fractional Trading", "Fractionalization Contract", "Fractionalization Protocol", "MSF Use Cases", "NFT", "NFT Liquidity", "OMA3", "Ownership Registry", "Tokenization System", "Blockchain", "Fungible Token", "Middleware Layer", "Price Discovery", "Shared Ownership Model", "Smart Contract", "Token Standard", "VirtualEconomyDomain" ] }, { "id": "fragment-shading", "title": "Fragment Shading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The programmable stage of the graphics pipeline in which per-fragment computations determine the final colour and depth of each candidate pixel produced by rasterisation. Fragment shaders evaluate lighting models, sample and filter textures, and apply material properties in parallel across thousands of GPU cores, consuming interpolated vertex attributes and producing the shaded values that are blended into the framebuffer, making this stage the dominant cost in most real-time rendering workloads.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:fragment-shading", "labels": [ "Fragment Shading" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [ "Rasterization", "Real-Time Rendering Pipeline", "Vertex Processing", "GPU", "Texture Mapping" ] }, { "id": "frame-reference-epoch", "title": "Frame Reference Epoch", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:frame-reference-epoch", "labels": [ "Frame Reference Epoch" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "framebuffer", "title": "Framebuffer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A framebuffer is a region of memory that holds the pixel data representing a complete frame to be displayed or further processed, typically organised as colour, depth, and stencil buffers. In a graphics pipeline, rendering operations write their results into a framebuffer, which the display hardware then scans out to a screen or which subsequent passes read as input. Framebuffers are central to double buffering, post-processing, off-screen rendering, and the multi-pass techniques that underpin modern real-time graphics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:framebuffer", "labels": [ "Framebuffer" ], "is_subclass_of": [ "Rendering Pipeline" ], "wikilinks": [] }, { "id": "frand-licensing", "title": "Frand Licensing", "domain": "standards", "domain_name": "Standards", "definition": "FRAND licensing refers to the commitment by holders of standard-essential patents to license those patents on fair, reasonable and non-discriminatory terms. It is a mechanism used by standards bodies to balance patent holders' rights with the need for broad, equitable access to technologies required to implement a standard. FRAND commitments aim to prevent hold-up and ensure interoperability across implementers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:frand-licensing", "labels": [ "Frand Licensing", "FRAND Licensing" ], "is_subclass_of": [ "Licensing" ], "wikilinks": [] }, { "id": "fraud-detection", "title": "fraud detection", "domain": "security", "domain_name": "Security", "definition": "Fraud detection is the automated or semi-automated identification of deceptive, unauthorised, or anomalous activities\u2014such as payment fraud, account takeover, synthetic identity creation, and insurance claim manipulation\u2014using statistical models, machine learning classifiers, graph analytics, and rule-based engines applied to transactional, behavioural, and network data. The field operates under severe class-imbalance constraints where fraudulent events are rare relative to legitimate activity, demanding specialised sampling strategies such as SMOTE and evaluation metrics including precision-recall curves and F1 scores. Modern production systems combine ensemble methods such as gradient-boosted trees for rapid inference with deep learning approaches including graph neural networks and LSTM sequence models to capture both point-in-time anomalies and temporal patterns indicative of coordinated fraud schemes. Explainability of model decisions is increasingly mandated by financial regulators to support human review of adverse determinations and compliance with consumer protection law.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fraud-detection", "labels": [ "Fraud Detection", "Anti-Fraud Verification" ], "is_subclass_of": [ "Anomaly Detection", "AI Application" ], "wikilinks": [] }, { "id": "fraud-prevention", "title": "Fraud Prevention", "domain": "security", "domain_name": "Security", "definition": "Fraud prevention is the set of controls, processes, and technologies designed to detect and stop deceptive activity intended to obtain money, data, or access illegitimately. In digital identity and finance it combines authentication, behavioural analytics, anomaly detection, and proof-of-personhood mechanisms. Effective fraud prevention reduces financial loss and protects the integrity of identity and payment systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fraud-prevention", "labels": [ "Fraud Prevention" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "fraud-proof", "title": "Fraud Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fraud proof is a cryptographic mechanism that allows any observer (a challenger) to demonstrate on-chain that a previously published off-chain state transition is invalid. It is the primary dispute mechanism for optimistic rollups, where sequencers post state roots optimistically and a challenge window allows watchers to submit a fraud proof if they detect an incorrect computation. Upon successful verification of a fraud proof by the base layer, the invalid state root is reverted and the sequencer's bonded stake is slashed as a penalty.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fraud-proof", "labels": [ "Fraud Proof", "Fraud Proofs", "Interactive Fraud Proof" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [ "Data Availability", "Scalability", "Rollup", "Layer 2 Networks" ] }, { "id": "frazil-ice", "title": "Frazil Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:frazil-ice", "labels": [ "Frazil Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "free-software", "title": "Free Software", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Free software is software distributed under a licence that guarantees users the freedom to run, study, modify, and redistribute it, as codified by the Free Software Foundation's four essential freedoms. The term is distinct from gratis pricing; a programme is free software because of the rights it grants, not because it costs nothing. Copyleft licences such as the GNU GPL operationalise this by requiring derivative works to be distributed under equivalent terms, distinguishing free software from permissively licensed open-source software.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:free-software", "labels": [ "Free Software" ], "is_subclass_of": [ "Open Source Software" ], "wikilinks": [] }, { "id": "free-troposphere", "title": "Free Troposphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:free-troposphere", "labels": [ "Free Troposphere" ], "is_subclass_of": [ "Troposphere" ], "wikilinks": [] }, { "id": "free-viewpoint-video", "title": "Free-Viewpoint Video", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Free-viewpoint video is video content that lets a viewer choose an arbitrary virtual camera position and angle at playback time, rather than being locked to the camera positions used during capture. It is typically produced from volumetric video captured by a multi-camera rig, using novel view synthesis techniques to interpolate or render views that were never directly recorded. It enables immersive playback experiences such as walking around a captured scene in virtual reality.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:free-viewpoint-video", "labels": [ "Free-Viewpoint Video" ], "is_subclass_of": [ "Volumetric Video" ], "wikilinks": [] }, { "id": "free-space-path-loss", "title": "Free-space Path Loss", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:free-space-path-loss", "labels": [ "Free-space Path Loss" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "freedom-of-expression", "title": "Freedom Of Expression", "domain": "governance", "domain_name": "Governance", "definition": "Freedom of expression is the right to hold opinions and to seek, receive and impart information and ideas without unwarranted interference. Recognised as a fundamental human right, it underpins democratic participation, accountability and open discourse, while being subject to narrowly defined limits. In digital governance it intersects with content moderation, censorship resistance and platform accountability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:freedom-of-expression", "labels": [ "Freedom Of Expression", "Freedom of Expression" ], "is_subclass_of": [ "Human Rights" ], "wikilinks": [] }, { "id": "frequentist-statistics", "title": "Frequentist Statistics", "domain": "data", "domain_name": "Data", "definition": "The school of statistical inference that defines probability as long-run relative frequency over repeated sampling, treats parameters as fixed unknowns rather than random variables, and evaluates procedures \u2014 estimators, hypothesis tests, confidence intervals \u2014 by their repeated-sampling operating characteristics such as bias, error rates, and coverage, in explicit contrast to Bayesian inference over posterior beliefs.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:frequentist-statistics", "labels": [ "Frequentist Statistics" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [ "Statistics", "Hypothesis Testing", "Bayesian Inference", "Conformal Prediction" ] }, { "id": "friction", "title": "Friction", "domain": "robotics", "domain_name": "Robotics", "definition": "Friction - The resistive force generated when surfaces slide or attempt to slide relative to one another, characterised by Coulomb friction (kinetic and static coefficients) and viscous damping, significantly impacting Joint Efficiency, Motor Performance, and Motion Accuracy in roboti...", "entityType": "Class", "qualityScore": 0.54, "maturity": "draft", "iri": "urn:ngm:class:friction", "labels": [ "Friction" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Robot Dynamics", "Physics Modelling" ], "wikilinks": [ "Energy Dissipation", "Energy Efficiency Calculation", "Friction Coefficient Estimation", "Joint Efficiency", "Joint Mechanics", "Lubrication Management", "Motion Accuracy", "Motor Performance", "Physics Modelling", "Wear Prediction", "AI Agent System", "Model-based Control", "Robot Dynamics", "RoboticsDomain" ] }, { "id": "front-running", "title": "Front-Running", "domain": "finance", "domain_name": "Finance", "definition": "Front-running is a form of market manipulation in which an actor with advance knowledge of a pending transaction executes their own trade ahead of it to profit from the anticipated price impact. In blockchain systems this typically takes the form of maximal extractable value, where searchers or validators observe pending transactions in the mempool and insert their own transactions ahead of them. Front-running erodes trust in fair transaction ordering and has motivated countermeasures such as private mempools and fair-ordering protocols.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:front-running", "labels": [ "Front-Running", "Front Running" ], "is_subclass_of": [ "Market Manipulation" ], "wikilinks": [] }, { "id": "frontier-ai-competition", "title": "Frontier AI Competition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The strategic and technological rivalry among leading AI organizations to develop the most capable general-purpose foundation models, characterized by rapid iteration, massive capital deployment, and aggressive talent acquisition.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:frontier-ai-competition", "labels": [ "Frontier AI Competition" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "frontier-ai-governance", "title": "Frontier AI Governance", "domain": "governance", "domain_name": "Governance", "definition": "The set of policies, regulatory frameworks, and institutional practices designed to manage the risks and ensure the responsible development of the most advanced and capable artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:frontier-ai-governance", "labels": [ "Frontier AI Governance" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "frontier-ai", "title": "Frontier AI", "domain": "ai", "domain_name": "Ai", "definition": "Frontier AI denotes the most capable and computationally intensive artificial intelligence systems at the leading edge of current technical progress, typically characterised by unprecedentedly large training compute budgets, novel emergent capabilities, and performance that approaches or exceeds human expert level across diverse cognitive tasks. These systems, predominantly large language models and multimodal foundation models, exhibit qualitatively new behaviours not present in smaller predecessors, including in-context learning, chain-of-thought reasoning, and cross-domain generalisation. Their development is concentrated among a small number of resource-rich organisations, raising distinct safety, governance, and geopolitical considerations absent from earlier AI generations. Regulatory frameworks such as the EU AI Act and the UK AI Safety Institute use compute-threshold criteria to demarcate frontier systems from other AI, typically above 10^26 training FLOPs for general-purpose AI.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:frontier-ai", "labels": [ "Frontier AI" ], "is_subclass_of": [ "Large-Scale Pretrained Foundation Model", "Foundation Model Layer" ], "wikilinks": [] }, { "id": "frontier-labs", "title": "Frontier Labs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Leading AI research organizations that develop the most advanced and capable large language models and AI systems, often setting the pace for the entire industry.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:frontier-labs", "labels": [ "Frontier Labs" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "frontier-model-evaluation", "title": "Frontier Model Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "Frontier model evaluation is the systematic assessment of the capabilities, limitations, and risks of the most advanced AI systems, including dangerous-capability and safety testing. It uses benchmarks, red-teaming, and threat-model-driven evaluations to inform deployment decisions and regulation. Such evaluation is increasingly required by AI governance regimes before high-capability models are released.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:frontier-model-evaluation", "labels": [ "Frontier Model Evaluation", "Frontier AI Evaluation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "frontier-model-forum", "title": "frontier model forum", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Frontier Model Forum (FMF) is an industry consortium founded in July 2023 by Anthropic, Google, Microsoft, and OpenAI to advance the safe and responsible development of frontier AI models \u2014 the most capable general-purpose AI systems at the frontier of performance. It operates as a collaborative, pre-competitive body focused on technical AI safety research, shared safety evaluations, red-teaming methodologies, and engagement with policymakers and civil society. The Forum functions as a bridge between frontier AI developers and regulatory bodies, aiming to establish shared technical standards and best practices without constituting a formal standards organisation or regulatory authority.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:frontier-model-forum", "labels": [ "Frontier Model Forum" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "frontier-model-regulation", "title": "Frontier Model Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Frontier model regulation is the emerging body of law and policy that targets the most capable AI systems specifically, imposing obligations such as pre-deployment risk assessment, capability evaluation, and incident reporting on developers of models above defined compute or capability thresholds. It is a subset of AI regulation distinguished by its focus on systemic and catastrophic risk rather than narrow, application-specific harms. Frontier model regulation has been advanced through instruments such as the EU AI Act's general-purpose AI provisions and voluntary commitments negotiated with leading labs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:frontier-model-regulation", "labels": [ "Frontier Model Regulation" ], "is_subclass_of": [ "AI Regulation" ], "wikilinks": [] }, { "id": "frontier-model-training", "title": "Frontier Model Training", "domain": "ai", "domain_name": "Ai", "definition": "Frontier model training refers to the end-to-end process of constructing the largest and most capable AI systems at the current performance frontier, encompassing data curation at web scale, distributed pretraining across thousands of accelerators, supervised fine-tuning, and reinforcement learning from human feedback, at compute costs exceeding tens of millions of US dollars per run. These training pipelines push the boundaries of achievable capability and introduce novel safety, governance, and infrastructure challenges not present in smaller-scale model development.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:frontier-model-training", "labels": [ "Frontier Model Training" ], "is_subclass_of": [ "Large Language Model Training" ], "wikilinks": [] }, { "id": "frontier-models", "title": "Frontier Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Frontier models are large-scale AI systems trained at the leading edge of compute, data, and capability, exhibiting emergent behaviours not observed in smaller models and presenting both transformative societal potential and novel safety risks. The term typically refers to the most capable foundation models available at any given time, assessed across benchmarks spanning reasoning, coding, science, and multimodal tasks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:frontier-models", "labels": [ "Frontier Models", "Frontier Model" ], "is_subclass_of": [ "Large-Scale Pretrained Foundation Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "frozen-ground", "title": "Frozen Ground", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:frozen-ground", "labels": [ "Frozen Ground" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "frustum-culling", "title": "Frustum Culling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Frustum culling is a real-time rendering visibility optimisation technique that discards scene objects whose bounding volumes lie entirely outside the camera's view frustum\u2014the truncated pyramid defined by the near and far clipping planes and the four side planes corresponding to the viewport edges. By testing object bounding spheres or axis-aligned bounding boxes against the six frustum planes before submitting draw calls, the GPU receives only geometry that could potentially contribute to the final image, dramatically reducing vertex processing and rasterisation work. It is a foundational stage in all production scene management pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:frustum-culling", "labels": [ "Frustum Culling", "View Frustum Culling" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "full-duplex-communication", "title": "Full Duplex Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Full-duplex communication is a mode of data exchange in which both endpoints can transmit and receive simultaneously over a single connection. It contrasts with half-duplex, where only one side may send at a time. Full-duplex channels enable low-latency, bidirectional interaction and are central to technologies such as web sockets and real-time messaging.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:full-duplex-communication", "labels": [ "Full Duplex Communication", "Full-Duplex Communication" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "full-fine-tuning", "title": "Full Fine Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A fine-tuning approach that updates all parameters of a pre-trained model during adaptation to a downstream task, requiring approximately four times the model's memory footprint to store weights, gradients, and optimiser states. Full fine-tuning provides maximum task-specific flexibility and sets the performance ceiling against which parameter-efficient alternatives such as LoRA are benchmarked, but creates a separate full-sized model copy per task.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:full-fine-tuning", "labels": [ "Full Fine Tuning", "Full Fine-Tuning" ], "is_subclass_of": [ "AI Technique", "Fine Tuning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "full-node", "title": "Full Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain node that independently downloads, validates, and stores the complete transaction history of the chain from the genesis block, enforcing all consensus rules without trusting external parties. Full nodes are the gold standard for trustless participation and are the backbone of decentralisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:full-node", "labels": [ "Full Node", "Bitcoin Full Node", "Full Node Infrastructure" ], "is_subclass_of": [ "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "full-text-search", "title": "Full-Text Search", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Full-text search is a technique for locating documents or records that contain specified words or phrases by matching against an inverted index built over tokenised text content, rather than scanning raw text linearly. It supports ranking by relevance, stemming, and fuzzy matching, and is provided natively by systems such as PostgreSQL and dedicated engines such as Elasticsearch. Full-text search underpins query engines and log aggregation systems that must retrieve relevant records from large unstructured text corpora.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:full-text-search", "labels": [ "Full-Text Search" ], "is_subclass_of": [ "Search Index" ], "wikilinks": [] }, { "id": "function-calling", "title": "Function Calling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Function Calling (also termed tool use or tool invocation) is the capability of large language models to emit structured requests selecting and parameterising external functions described to them via JSON-Schema tool definitions, with the application layer executing the selected tools and returni...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:function-calling", "labels": [ "Function Calling", "OpenAI Function Calling API" ], "is_subclass_of": [ "AI Technique", "AI Agent System", "Tool Use", "LLM Capability", "Structured Output Generation", "Agentic Capability", "Language Model Augmentation" ], "wikilinks": [ "Agent-to-Agent Protocol", "Agentic AI", "Agentic Capability", "AgenticSystemsDomain", "Anthropic 2024 Model Context Protocol Specification", "Anthropic 2024 Tool Use Documentation", "Anthropic Tool Use API", "Application Runtime", "Argument Parser", "Argument Validator", "Beurer-Kellner et al 2024 LMQL", "Browser Automation by LLMs", "Chain-of-Thought Prompting", "Code Execution by LLMs", "Code Generation Agents", "Code Interpreter", "Computer Use Agents", "Constrained Decoding", "Conversation Context", "Conversation Loop" ] }, { "id": "function-schemas", "title": "Function Schemas", "domain": "ai", "domain_name": "Ai", "definition": "Function schemas are structured, machine-readable declarations that describe the name, parameters, and types of tools or functions that a large language model may call. Typically expressed as JSON Schema, they let a model select an appropriate tool and emit validly structured arguments. Function schemas are the contract that enables reliable tool use and agentic workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:function-schemas", "labels": [ "Function Schemas" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "functional-analysis", "title": "Functional Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Functional analysis is the branch of mathematics studying vector spaces with a notion of limit, such as Banach and Hilbert spaces, and the linear operators acting on them. It underlies much of modern analysis and its applications.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:functional-analysis", "labels": [ "Functional Analysis" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Convex Optimisation", "Probability Theory", "owl:Thing" ] }, { "id": "functional-safety", "title": "Functional Safety", "domain": "robotics", "domain_name": "Robotics", "definition": "Functional Safety - The discipline of designing, implementing, and verifying safety-critical control systems (per IEC 61508, ISO 26262) to ensure systems fail safely and prevent hazardous failures that could harm humans, equipment, or processes through systematic risk assessment, safety integrity levels, and redundancy in electrical, electronic, and programmable electronic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:functional-safety", "labels": [ "Functional Safety", "Functional Safety Management" ], "is_subclass_of": [ "Safety Engineering", "Robotics" ], "wikilinks": [ "Fault Tree Analysis", "Hazard Analysis", "Human-Robot Safety", "Liability Mitigation", "Safe Operation Certification", "Safety Engineering", "Testing & Validation", "AI Agent System", "Regulatory Compliance", "Risk Management", "Robotics", "RoboticsDomain" ] }, { "id": "fundamental-rights-impact-assessment", "title": "Fundamental Rights Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A fundamental rights impact assessment is a structured process for evaluating, before and during deployment, how a system, policy, or AI application may affect people's fundamental rights such as dignity, non-discrimination, privacy, and freedom of expression. It documents the context of use, identifies affected groups and potential harms, assesses risks, and defines mitigation and oversight measures. It has become a governance instrument for high-risk AI, complementing data protection impact assessments and broader risk assessment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:fundamental-rights-impact-assessment", "labels": [ "Fundamental Rights Impact Assessment" ], "is_subclass_of": [ "Impact Assessment" ], "wikilinks": [] }, { "id": "fundamental-rights", "title": "Fundamental Rights", "domain": "governance", "domain_name": "Governance", "definition": "The basic rights and freedoms guaranteed to individuals by constitutional orders and supranational charters \u2014 dignity, privacy, freedom of expression, equality, non-discrimination, and effective remedy among them \u2014 which bind public authorities and increasingly shape technology governance, serving as the normative benchmark for instruments such as the EU Charter of Fundamental Rights, the GDPR, and the EU AI Act.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:fundamental-rights", "labels": [ "Fundamental Rights" ], "is_subclass_of": [ "Human Rights" ], "wikilinks": [ "Human Rights", "Rule of Law", "EU HLEG AI" ] }, { "id": "fungibility", "title": "Fungibility", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Fungibility is the property of a good or asset whereby individual units are mutually interchangeable, each unit being indistinguishable from and equal in value to any other. It is a defining characteristic of money and of fungible blockchain tokens, in which any one token of a given type can substitute for another. Fungibility contrasts with non-fungibility, where each unit is unique and not interchangeable, as in non-fungible tokens.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:fungibility", "labels": [ "Fungibility" ], "is_subclass_of": [ "Tokenization" ], "wikilinks": [] }, { "id": "fungible-token", "title": "Fungible Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain token where each unit is identical and fully interchangeable with any other unit of the same type, analogous to traditional fiat currency. Fungibility is enforced at the protocol level via standards such as ERC-20, ensuring uniform value and seamless divisibility across all holders.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:fungible-token", "labels": [ "Fungible Token", "ICS-20 Fungible Token Transfer", "Omni Fungible Token" ], "is_subclass_of": [ "Token" ], "wikilinks": [ "Blockchain", "Token" ] }, { "id": "futarchy", "title": "Futarchy", "domain": "governance", "domain_name": "Governance", "definition": "Futarchy is a governance mechanism proposed by economist Robin Hanson in which policy decisions are made by first defining measurable societal welfare metrics and then adopting whichever policy candidate is predicted by speculative prediction markets to maximise those metrics. Under futarchy, democratic institutions vote on the values or welfare criteria to optimise, while market prices \u2014 which aggregate information from many participants through financial incentives \u2014 determine the means by which those values are pursued, on the premise that markets are better information aggregators than committees for empirical questions about causal consequences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:futarchy", "labels": [ "Futarchy" ], "is_subclass_of": [ "Voting Mechanism" ], "wikilinks": [] }, { "id": "future-of-work", "title": "Future Of Work", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Future of Work is the study and anticipation of how technological change, especially artificial intelligence and automation, reshapes jobs, skills, organisational structures, and the broader labour economy. It examines task automation and augmentation, the emergence of new roles, shifts toward distributed and hybrid working, and the policy responses needed for equitable transitions. The field blends economics, organisational behaviour, and technology forecasting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:future-of-work", "labels": [ "Future Of Work", "Future of Work" ], "is_subclass_of": [ "Artificial Intelligence", "Employment Social Contract Under Automation" ], "wikilinks": [] }, { "id": "future-of-humanity-institute", "title": "Future of Humanity Institute", "domain": "ai", "domain_name": "Ai", "definition": "The Future of Humanity Institute (FHI) was a multidisciplinary research centre at the University of Oxford, founded in 2005 and directed by Nick Bostrom, focused on big-picture questions about humanity's long-term prospects. It produced influential work on existential risk, AI safety, and the governance of transformative technologies. The institute closed in 2024, but its research substantially shaped the fields of AI safety and longtermism.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:future-of-humanity-institute", "labels": [ "Future of Humanity Institute" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "fuzzy-logic", "title": "Fuzzy Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Fuzzy Logic is a artificial intelligence concept and a type of Artificial Intelligence. that enables Control Systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:fuzzy-logic", "labels": [ "Fuzzy Logic", "Fuzzy Matching" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "Control Systems", "Artificial Intelligence" ] }, { "id": "g20", "title": "G20", "domain": "governance", "domain_name": "Governance", "definition": "The G20 (Group of Twenty) is an intergovernmental forum of nineteen major economies plus the European Union and African Union that coordinates international economic and financial policy. It convenes heads of state, finance ministers and central bank governors to address global financial stability, trade, development and systemic risk. Through its leaders' summits and working groups, the G20 sets the political agenda that bodies such as the Financial Stability Board, IMF and Basel Committee translate into standards and recommendations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:g20", "labels": [ "G20" ], "is_subclass_of": [ "Economic Governance" ], "wikilinks": [] }, { "id": "gan-virtual-landscape-art", "title": "GAN Virtual Landscape Art", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Virscapes are AI-generated virtual landscape images produced by training generative adversarial networks (StyleGAN) on a curated dataset of digital game landscapes collected over seven years, exploring latent space to create emotionally resonant synthetic environments. The practice investigates questions of landscape identity, simulacra, and the distinction between human aesthetic curation and algorithmic generation in digital art.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:gan-virtual-landscape-art", "labels": [ "GAN Virtual Landscape Art", "Virscapes" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "gan", "title": "GAN", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Generative Adversarial Network (GAN) is a deep-learning framework in which two neural networks \u2014 a generator and a discriminator \u2014 are trained simultaneously in a minimax game: the generator maps random latent vectors to synthetic data samples, while the discriminator learns to distinguish real training samples from generated ones. Through adversarial feedback propagated via backpropagation, the generator progressively improves at synthesising realistic outputs while the discriminator improves at detection. GANs have achieved state-of-the-art results in image synthesis, style transfer, super-resolution, and conditional generation, but are prone to training instabilities such as mode collapse and vanishing gradients.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:gan", "labels": [ "GAN" ], "is_subclass_of": [ "Generative Model", "Deep Generative Model" ], "wikilinks": [ "Backpropagation", "Image Generation", "Generative Model", "Generative Adversarial Network" ] }, { "id": "gdpr-article-22-compliance", "title": "GDPR Article 22 Compliance", "domain": "governance", "domain_name": "Governance", "definition": "GDPR Article 22 compliance concerns adherence to the EU General Data Protection Regulation provision that grants individuals the right not to be subject to decisions based solely on automated processing, including profiling, that produce legal or similarly significant effects. Compliance requires lawful bases, human oversight, the ability to contest decisions, and meaningful information about the logic involved. It is a key constraint on automated decision-making and policy enforcement systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gdpr-article-22-compliance", "labels": [ "GDPR Article 22 Compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "gdpr-article-25", "title": "GDPR Article 25", "domain": "metaverse", "domain_name": "Metaverse", "definition": "GDPR Article 25 sets out the obligation of data protection by design and by default, requiring controllers to embed privacy safeguards into processing systems and to minimise data collection from the outset.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gdpr-article-25", "labels": [ "GDPR Article 25" ], "is_subclass_of": [ "Data Protection" ], "wikilinks": [ "GDPR", "Data Minimisation", "Privacy", "Regulatory Compliance", "Data Protection" ] }, { "id": "gdpr-compliance", "title": "gdpr compliance", "domain": "governance", "domain_name": "Governance", "definition": "GDPR Compliance is the totality of organisational, technical, and procedural measures an entity must implement to satisfy obligations under the EU General Data Protection Regulation (Regulation 2016/679), including identifying a lawful basis for each processing activity, fulfilling data subject rights, performing Data Protection Impact Assessments for high-risk processing, notifying supervisory authorities of breaches within 72 hours, and appointing a Data Protection Officer where required. The regulation applies extraterritorially to any organisation processing personal data of EU data subjects regardless of where processing occurs. Compliance programmes encompass privacy-by-design architecture, data minimisation, consent lifecycle management, Records of Processing Activities, demonstrable accountability mechanisms, and contractual safeguards for cross-border data transfers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gdpr-compliance", "labels": [ "GDPR Compliance", "Compliance with GDPR" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "gdpr", "title": "GDPR", "domain": "governance", "domain_name": "Governance", "definition": "The General Data Protection Regulation (GDPR) is a comprehensive EU legal framework (Regulation 2016/679) that governs the collection, processing, storage, and transfer of personal data belonging to individuals in the European Union and European Economic Area. It establishes six lawful bases for processing, grants data subjects extensive individual rights, and imposes organisational obligations including Data Protection Impact Assessments and appointment of Data Protection Officers. Enforced by national supervisory authorities coordinated by the European Data Protection Board, GDPR carries penalties of up to \u20ac20 million or 4% of global annual turnover, making it the most consequential data protection law globally and a template for privacy legislation worldwide.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:gdpr", "labels": [ "GDPR", "GDPR Accountability", "GDPR Article 89" ], "is_subclass_of": [ "Data Protection Regulation" ], "wikilinks": [] }, { "id": "gdpvala-benchmark", "title": "GDPvala Benchmark", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A standardized evaluation metric designed to assess the economic productivity and value-generating capabilities of large language models in real-world professional tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:gdpvala-benchmark", "labels": [ "GDPvala Benchmark" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "genius-act", "title": "GENIUS Act", "domain": "governance", "domain_name": "Governance", "definition": "The GENIUS Act is United States legislation establishing a federal regulatory framework for the issuance of payment stablecoins.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:genius-act", "labels": [ "GENIUS Act", "GENIUS Act 2025" ], "is_subclass_of": [ "Crypto Regulation" ], "wikilinks": [ "Regulatory Frameworks", "Investor Protection", "Stablecoin", "Crypto Regulation", "https://www.congress.gov/bill/119th-congress/senate-bill/394", "https://en.wikipedia.org/wiki/GENIUS_Act" ] }, { "id": "gfpgan", "title": "GFPGAN", "domain": "ai", "domain_name": "Ai", "definition": "GFPGAN (Generative Facial Prior GAN) is a practical deep-learning model for blind face restoration that leverages the rich facial priors encoded in a pretrained StyleGAN to recover realistic detail from degraded portraits. It restores resolution, removes artefacts, and reconstructs plausible facial features in a single forward pass. It is widely used in photo restoration and as a face-enhancement component in image and video pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gfpgan", "labels": [ "GFPGAN", "GFPGAN Face Restoration" ], "is_subclass_of": [ "Generative Model", "Generative Adversarial Networks" ], "wikilinks": [] }, { "id": "gguf-format", "title": "GGUF Format", "domain": "ai", "domain_name": "Ai", "definition": "GGUF (GPT-Generated Unified Format) is a binary file format for storing large language model weights, metadata, and tokenizer data in a single self-contained file optimised for fast loading and local inference. Developed in the llama.cpp ecosystem as a successor to GGML, it supports a range of quantization schemes and embeds the metadata needed to run a model without external configuration. It is the de facto format for running quantized LLMs on consumer hardware.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:gguf-format", "labels": [ "GGUF Format", "GGUF", "GGUF Specification" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "ghg-protocol-corporate-standard", "title": "GHG Protocol Corporate Standard", "domain": "governance", "domain_name": "Governance", "definition": "The GHG Protocol Corporate Accounting and Reporting Standard is the most widely used international framework for measuring and reporting an organisation's greenhouse-gas emissions. Developed by the World Resources Institute and the World Business Council for Sustainable Development, it organises emissions into Scope 1 (direct), Scope 2 (purchased energy), and Scope 3 (value-chain) categories, establishing consistent boundaries, accounting principles, and disclosure rules. It underpins corporate climate reporting, science-based targets, and most national and regional emissions-disclosure regulations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ghg-protocol-corporate-standard", "labels": [ "GHG Protocol Corporate Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ghg-protocol", "title": "ghg protocol", "domain": "governance", "domain_name": "Governance", "definition": "The GHG Protocol is the globally dominant greenhouse gas accounting and reporting framework, co-developed by the World Resources Institute (WRI) and the World Business Council for Sustainable Development (WBCSD) and first published as the Corporate Standard in 2001. It establishes methodologies for measuring and categorising greenhouse gas emissions across three scopes: Scope 1 (direct emissions from owned or controlled sources), Scope 2 (indirect emissions from purchased energy), and Scope 3 (all other upstream and downstream value-chain emissions). The framework underpins virtually all major corporate carbon reporting mandates, voluntary disclosure programmes, and science-based target-setting initiatives worldwide, making it the de facto lingua franca of corporate climate accounting.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:ghg-protocol", "labels": [ "GHG Protocol" ], "is_subclass_of": [ "Environmental Standards" ], "wikilinks": [] }, { "id": "glsl", "title": "GLSL", "domain": "metaverse", "domain_name": "Metaverse", "definition": "GLSL (OpenGL Shading Language) is a high-level C-like programming language for writing shaders that execute on the GPU within the OpenGL and WebGL graphics pipelines. It lets developers program the programmable stages, principally vertex and fragment shaders, to control geometry transformation, lighting, and per-pixel colour. GLSL is a foundational tool for real-time rendering and visual effects.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:glsl", "labels": [ "GLSL", "GLSL ES", "GLSL Shaders" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "glue-benchmark", "title": "GLUE Benchmark", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The General Language Understanding Evaluation, a 2018 suite of nine English sentence- and sentence-pair tasks \u2014 spanning acceptability, sentiment, paraphrase, similarity, and natural language inference \u2014 with a public leaderboard and diagnostic set, which became the standard yardstick for pretrained language models such as BERT and RoBERTa until model performance surpassed human baselines and evaluation moved to harder successors.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:glue-benchmark", "labels": [ "GLUE Benchmark" ], "is_subclass_of": [ "Benchmarking" ], "wikilinks": [ "Benchmarking", "Natural Language Understanding", "BERT", "BEIR Benchmark" ] }, { "id": "gmx", "title": "GMX", "domain": "blockchain", "domain_name": "Blockchain", "definition": "GMX is a decentralised spot and perpetual-futures exchange deployed on the Arbitrum and Avalanche networks. It allows traders to take leveraged positions against a shared multi-asset liquidity pool rather than a traditional order book, with prices supplied by external oracles. Liquidity providers deposit assets into the pool and earn a share of trading fees while acting as the counterparty to traders.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gmx", "labels": [ "GMX" ], "is_subclass_of": [ "Decentralised Finance", "Decentralised Finance Domain" ], "wikilinks": [ "Oracle", "Arbitrum", "Liquidity Pool", "Perpetual Futures", "Leveraged Trading", "dYdX", "Automated Market Maker", "Decentralised Finance Domain" ] }, { "id": "gnss-almanac", "title": "GNSS Almanac", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-almanac", "labels": [ "GNSS Almanac" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-ambiguity-resolution", "title": "GNSS Ambiguity Resolution", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-ambiguity-resolution", "labels": [ "GNSS Ambiguity Resolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-antenna-phase-centre", "title": "GNSS Antenna Phase Centre", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-antenna-phase-centre", "labels": [ "GNSS Antenna Phase Centre" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-antenna", "title": "GNSS Antenna", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-antenna", "labels": [ "GNSS Antenna" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-augmentation", "title": "GNSS Augmentation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-augmentation", "labels": [ "GNSS Augmentation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-availability", "title": "GNSS Availability", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-availability", "labels": [ "GNSS Availability" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-carrier-phase", "title": "GNSS Carrier Phase", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-carrier-phase", "labels": [ "GNSS Carrier Phase" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-code-measurement", "title": "GNSS Code Measurement", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-code-measurement", "labels": [ "GNSS Code Measurement" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-continuity", "title": "GNSS Continuity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-continuity", "labels": [ "GNSS Continuity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-cycle-slip", "title": "GNSS Cycle Slip", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-cycle-slip", "labels": [ "GNSS Cycle Slip" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-doppler-measurement", "title": "GNSS Doppler Measurement", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-doppler-measurement", "labels": [ "GNSS Doppler Measurement" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-ephemeris", "title": "GNSS Ephemeris", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-ephemeris", "labels": [ "GNSS Ephemeris" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-integrity", "title": "GNSS Integrity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-integrity", "labels": [ "GNSS Integrity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-interference", "title": "GNSS Interference", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-interference", "labels": [ "GNSS Interference" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-jamming", "title": "GNSS Jamming", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-jamming", "labels": [ "GNSS Jamming" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-multipath", "title": "GNSS Multipath", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-multipath", "labels": [ "GNSS Multipath" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-navigation-message", "title": "GNSS Navigation Message", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-navigation-message", "labels": [ "GNSS Navigation Message" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-positioning-accuracy", "title": "GNSS Positioning Accuracy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-positioning-accuracy", "labels": [ "GNSS Positioning Accuracy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-positioning-error", "title": "GNSS Positioning Error", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-positioning-error", "labels": [ "GNSS Positioning Error" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-pseudorange", "title": "GNSS Pseudorange", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-pseudorange", "labels": [ "GNSS Pseudorange" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-radio-occultation", "title": "GNSS Radio Occultation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-radio-occultation", "labels": [ "GNSS Radio Occultation" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "gnss-receiver", "title": "GNSS Receiver", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-receiver", "labels": [ "GNSS Receiver" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "gnss-reflectometry", "title": "GNSS Reflectometry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-reflectometry", "labels": [ "GNSS Reflectometry" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "gnss-satellite", "title": "GNSS Satellite", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-satellite", "labels": [ "GNSS Satellite" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "gnss-signal-acquisition", "title": "GNSS Signal Acquisition", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-signal-acquisition", "labels": [ "GNSS Signal Acquisition" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-signal-tracking", "title": "GNSS Signal Tracking", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-signal-tracking", "labels": [ "GNSS Signal Tracking" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-signal", "title": "GNSS Signal", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-signal", "labels": [ "GNSS Signal" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-spoofing", "title": "GNSS Spoofing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-spoofing", "labels": [ "GNSS Spoofing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-time-transfer", "title": "GNSS Time Transfer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-time-transfer", "labels": [ "GNSS Time Transfer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gnss-time-to-first-fix", "title": "GNSS Time to First Fix", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gnss-time-to-first-fix", "labels": [ "GNSS Time to First Fix" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gov-uk-one-login", "title": "GOV.UK One Login", "domain": "security", "domain_name": "Security", "definition": "GOV.UK One Login is the UK government's single sign-on and identity-verification service that lets people access central-government digital services with one account. It combines authentication with identity proofing to a defined assurance level, replacing a fragmented landscape of departmental logins. It is a national digital-identity infrastructure aligned with emerging standards for verifiable credentials and digital identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:gov-uk-one-login", "labels": [ "GOV.UK One Login" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "gpt-4", "title": "GPT 4", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A large multimodal language model released by OpenAI in 2023, capable of accepting text and image inputs and producing text outputs, and a successor to the GPT-3.5 series.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gpt-4", "labels": [ "GPT 4", "GPT-4" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [ "Transformer", "OpenAI", "In-Context Learning", "Multimodal AI", "Large Language Models" ] }, { "id": "gpt-engineer", "title": "GPT Engineer", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "GPT Engineer is an open-source agentic software-development framework initiated by Anton Osika in June 2023 that autonomously transforms a natural-language specification into a working multi-file codebase through a sequential loop of specification elicitation, file-level planning, code generation...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:gpt-engineer", "labels": [ "GPT Engineer" ], "is_subclass_of": [ "AI Application", "AI Adoption", "CLI Multi-Agent Systems", "Agent Frameworks", "Agents", "Agentic Internet", "Cognitive AI" ], "wikilinks": [ "AgentLayer", "AgenticAIDomain", "Agentic Loop", "AI-Assisted Software Engineering", "Aider", "Automated Software Scaffolding", "Chain of Thought", "Cline", "Cursor", "Devin", "File Planning Agent", "Full-Stack Application Generation", "GitHub", "GitHub Copilot", "Hackathon Development", "HumanEval", "MBPP Benchmark", "Multi-File Code Generation", "No-Code Development", "Non-Engineer Development" ] }, { "id": "gpt-5-3-codex-spark", "title": "GPT-5.3 Codex Spark", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A specific version of OpenAI's GPT model series optimized for code generation and designed to run on non-Nvidia hardware such as Cerebras chips.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:gpt-5-3-codex-spark", "labels": [ "GPT-5.3 Codex Spark" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "gpt", "title": "GPT", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative Pre-trained Transformer: an autoregressive language model that uses transformer decoder architecture and is pre-trained on large text corpora using next-token prediction. GPT models learn rich representations through unsupervised pre-training and are subsequently fine-tuned for diverse downstream NLP tasks, demonstrating that scale and the language modelling objective yield powerful transfer learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:gpt", "labels": [ "GPT", "GPT 3", "OpenAI GPT" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "gpts-and-custom-assistants", "title": "GPTs and Custom Assistants", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "GPTs and Custom Assistants are a family of user-configured task-specific applications built on top of foundation large language models in which a vendor exposes a no-code or low-code authoring surface that lets a user, an enterprise administrator, or a third-party developer compose a persistent a...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:gpts-and-custom-assistants", "labels": [ "GPTs and Custom Assistants" ], "is_subclass_of": [ "AI Application", "LLM Application", "Conversational AI System", "Retrieval-Augmented Generation System", "No-Code Platform", "AI Productivity Tool" ], "wikilinks": [ "AI Productivity Tool", "Anthropic Agent Skills October 2025", "Anthropic Agent Skills Open Standard", "Anthropic Claude Projects June 2024", "ApplicationLayerDomain", "Asai et al. 2023 Self-RAG", "Authentication", "BM25", "Bot Marketplaces", "Chunking", "Conversation Memory", "Conversation Starter", "Conversational AI System", "Customer Support", "Document Q&A", "Education", "Embedding Model", "Embeddings", "Enterprise Knowledge Surfacing", "Fine-Tuned Models" ] }, { "id": "gpu-acceleration", "title": "GPU Acceleration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "GPU Acceleration is the use of graphics processing units to perform general-purpose computational workloads in a massively parallel fashion, exploiting thousands of shader cores arranged in a single-instruction-multiple-data (SIMD) architecture to achieve throughput orders of magnitude beyond conventional CPUs for data-parallel tasks such as matrix multiplication and tensor contraction. It is the dominant execution paradigm for training and inference in modern deep learning, scientific simulation, and real-time rendering pipelines. The programming model is exposed through vendor APIs such as CUDA and ROCm, as well as cross-platform standards including OpenCL and SYCL. Hierarchical parallelism \u2014 threads grouped into warps, warps into thread blocks, blocks into grids dispatched across streaming multiprocessors \u2014 enables fine-grained exploitation of data parallelism at every level of the memory hierarchy.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:gpu-acceleration", "labels": [ "GPU Acceleration", "Gpu Acceleration" ], "is_subclass_of": [ "Hardware Acceleration" ], "wikilinks": [] }, { "id": "gpu-architecture", "title": "GPU Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "GPU architecture describes the design of graphics processing units as massively parallel processors built around thousands of simple shader cores, wide high-bandwidth memory interfaces, and dedicated fixed-function units for texturing, rasterisation, ray tracing, and tensor computation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gpu-architecture", "labels": [ "GPU Architecture" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [ "Memory Hierarchy", "Real-Time Rendering", "Parallel Computing", "GPU Computing", "Graphics Pipeline", "Computer Hardware" ] }, { "id": "gpu-cluster", "title": "GPU Cluster", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A GPU cluster is a group of interconnected computers each equipped with graphics processing units, used together for parallel computation. Such clusters are central to training large AI models and high-performance computing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gpu-cluster", "labels": [ "GPU Cluster" ], "is_subclass_of": [ "GPU Computing" ], "wikilinks": [ "GPU", "NVIDIA H100", "Deep Learning", "GPU Computing" ] }, { "id": "gpu-compute", "title": "gpu compute", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "GPU Compute (General-Purpose GPU computing, GPGPU) is the practice of executing data-parallel numerical workloads on Graphics Processing Units originally designed for rasterising 3D geometry, exploiting their thousands of shader cores and high-bandwidth memory to perform tensor operations \u2014 principally matrix multiplications and convolutions \u2014 at throughputs far exceeding those achievable on CPUs. Programming models such as NVIDIA CUDA and AMD ROCm expose the GPU's Single Instruction Multiple Data (SIMD) execution model to software, enabling general-purpose scientific, engineering, and machine-learning workloads to run directly on the GPU die. GPU compute is the dominant hardware substrate for training and inference of deep neural networks, large language models, and diffusion models, with specialised Tensor Core and matrix-engine hardware units providing mixed-precision acceleration orders of magnitude beyond scalar compute. The field now extends to heterogeneous cluster computing, where thousands of GPUs are interconnected via NVLink and high-speed fabrics to train models at scales that would be infeasible on any single device.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:gpu-compute", "labels": [ "GPU Compute", "GPU Compute API" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "gpu-computing", "title": "GPU Computing", "domain": "ai", "domain_name": "Ai", "definition": "GPU Computing is the use of graphics processing units as massively parallel co-processors to accelerate general-purpose computational workloads beyond rendering. Modern GPUs contain thousands of shader cores organised into streaming multiprocessors capable of executing thousands of threads simultaneously, making them ideal for data-parallel algorithms in deep learning training, scientific simulation, and signal processing. Frameworks such as CUDA and OpenCL expose this parallelism to application developers through a hierarchical thread and memory model. GPU computing has become the primary accelerator substrate for large-scale machine learning, high-performance computing, and increasingly for inference serving at cloud scale.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:gpu-computing", "labels": [ "GPU Computing", "Accelerated Computing", "GPU Accelerated Computing", "GPU-Accelerated Computing" ], "is_subclass_of": [ "Hardware Acceleration" ], "wikilinks": [] }, { "id": "gpu-driver", "title": "GPU Driver", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A GPU driver is the system software that mediates between an operating system or application graphics API and the physical graphics processing unit, translating high-level rendering and compute commands into hardware-specific instructions. It manages GPU memory, command submission queues, context switching, and synchronisation, and exposes standard interfaces so that applications need not target individual hardware models. Driver quality and versioning directly affect performance, feature availability, and stability of graphics and GPU-compute workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gpu-driver", "labels": [ "GPU Driver" ], "is_subclass_of": [ "Hardware Abstraction Layer" ], "wikilinks": [] }, { "id": "gpu-infrastructure", "title": "GPU Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The specialized hardware clusters and data center architectures designed to support the high-throughput parallel processing required for training and deploying large-scale artificial intelligence models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:gpu-infrastructure", "labels": [ "GPU Infrastructure" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "gpu-knowledge-graph-platform", "title": "GPU Knowledge Graph Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "VisionFlow is a GPU-accelerated, Rust-backed collaborative platform for real-time 3D knowledge graph visualisation and human-AI agent orchestration on private datasets, integrating Claude Flow MCP for spawning specialised agents. Junkie Jarvis refers to the integrated AI orchestration layer enabling multi-agent task delegation within the VisionFlow environment.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:gpu-knowledge-graph-platform", "labels": [ "GPU Knowledge Graph Platform", "VisionFlow and Junkie Jarvis" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "VisionFlow Client" ] }, { "id": "gpu-rendering", "title": "GPU Rendering", "domain": "metaverse", "domain_name": "Metaverse", "definition": "GPU rendering is the computation of images from 2D or 3D scene descriptions using the massively parallel processing units of a graphics processing unit rather than the CPU. By executing shading, rasterisation, and ray-tracing workloads across thousands of cores, it delivers real-time interactive graphics and accelerates offline rendering. It is foundational to game engines, visualisation, and digital-content creation tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gpu-rendering", "labels": [ "GPU Rendering", "GPU Accelerated Rendering" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "gpu-resources", "title": "GPU Resources", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "GPU resources refer to the pool of graphics processing unit capacity \u2014 including VRAM, streaming multiprocessors, tensor cores, and associated interconnect bandwidth \u2014 that is provisioned, allocated, and managed as a computational resource for parallel workloads such as AI model training and inference, scientific simulation, computer graphics rendering, and high-performance computing. In cloud and data centre contexts, GPU resources are typically accessed through virtualisation or direct hardware passthrough, managed by schedulers that partition capacity across competing workloads and billed on per-hour or per-token consumption models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gpu-resources", "labels": [ "GPU Resources" ], "is_subclass_of": [ "Compute Resources", "Computing and Cloud" ], "wikilinks": [] }, { "id": "gpu-supply-chain", "title": "GPU Supply Chain", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The GPU supply chain is the network of design, fabrication, packaging and distribution processes that produce graphics processing units used for AI training and inference, spanning fabless designers, foundries such as TSMC, memory suppliers and system integrators. Its capacity constraints and export controls directly shape which organisations can access the compute needed for frontier model development. It is a specific, compute-focused segment of the broader semiconductor supply chain.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:gpu-supply-chain", "labels": [ "GPU Supply Chain" ], "is_subclass_of": [ "Semiconductor Supply Chain" ], "wikilinks": [] }, { "id": "gpu", "title": "GPU", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A GPU (Graphics Processing Unit) is a highly parallel processor optimised for the throughput-oriented computation required to render images and to accelerate data-parallel workloads, and increasingly used as the primary compute substrate for deep learning, scientific simulation, and XR rendering.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gpu", "labels": [ "GPU", "GPU Hardware", "GPU Support", "General-Purpose GPU", "Gpu", "NVIDIA RTX GPU" ], "is_subclass_of": [ "Graphics Processing", "Computing and Cloud" ], "wikilinks": [ "Shader", "Real-Time Rendering", "Rasterization", "Ray Tracing", "Graphics Processing Unit", "Graphics Processing" ] }, { "id": "gri-standards", "title": "GRI Standards", "domain": "governance", "domain_name": "Governance", "definition": "The GRI Standards are a globally adopted framework published by the Global Reporting Initiative for organisations to disclose their economic, environmental, and social impacts. They provide a modular set of universal, sector, and topic standards that standardise sustainability and ESG reporting for comparability across firms. Widely referenced by regulators and investors, they form a backbone of corporate non-financial disclosure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gri-standards", "labels": [ "GRI Standards", "GRI", "GRI Supply Chain Standards", "GRI Sustainability Standards", "GRI Universal Standards 2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "gs1-epcis", "title": "GS1 EPCIS", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "GS1 EPCIS (Electronic Product Code Information Services) is a global standard that defines a common data model and interface for capturing and sharing supply-chain visibility events, answering what, when, where, why, and how for product movements. It standardises event-based traceability across trading partners using GS1 identifiers. EPCIS is widely deployed in pharmaceuticals, food safety, and logistics to enable end-to-end provenance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gs1-epcis", "labels": [ "GS1 EPCIS", "EPCIS", "EPCIS Standard", "GS1 EPCIS Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "gs-1-standards", "title": "GS1 Standards", "domain": "standards", "domain_name": "Standards", "definition": "GS1 Standards constitute a globally adopted, open framework of identification keys, data carriers (barcodes and RFID tags), and messaging formats that uniquely and unambiguously identify physical and digital entities \u2014 products, logistics units, locations, assets, and services \u2014 across entire supply chain ecosystems. Governed by GS1, a neutral non-profit organisation with national member organisations in over 115 countries, the standards define how identifiers are structured (e.g. GTIN, GLN, SSCC), how they are encoded for machine reading (EAN/UPC, GS1-128, QR, DataMatrix, EPC/RFID), and how master and transactional data are shared between trading partners via EDI and web services. GS1 Standards form the interoperability backbone of global retail, healthcare, logistics, foodservice, and increasingly digital product passport and IoT applications.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:gs-1-standards", "labels": [ "GS1 Standards" ], "is_subclass_of": [ "Standards" ], "wikilinks": [ "Standards", "Supply Chain", "Open Standards", "Internet of Things" ] }, { "id": "gs1", "title": "GS1", "domain": "standards", "domain_name": "Standards", "definition": "GS1 is the global not-for-profit standards organisation responsible for designing, administering, and maintaining the world's most widely deployed open supply chain identification system, used by over two million companies in more than 150 countries. Its core standards define globally unique identification schemes for trade items (GTIN), locations (GLN), logistic units (SSCC), and assets (GIAI), encoding them in barcodes (EAN-13, ITF-14, GS1-128), two-dimensional symbols (GS1 DataMatrix, GS1 QR), and digital web URIs via GS1 Digital Link. GS1 also governs EPCIS (Electronic Product Code Information Services), the event-data exchange standard that captures What, Where, When, Why, and How across the supply chain, and underpins regulatory traceability mandates in pharmaceuticals, food safety, and sustainable products across the EU, US, and beyond.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:gs1", "labels": [ "GS1", "GS1 Organisation" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "gain-tuning", "title": "Gain Tuning", "domain": "robotics", "domain_name": "Robotics", "definition": "Gain tuning is the process of selecting the proportional, integral and derivative gain values of a feedback control loop so that the controlled system responds with the desired speed, stability and overshoot characteristics. Gains are set through analytical methods such as Ziegler-Nichols, model-based optimisation, or empirical trial-and-error adjustment on the physical system. Poorly tuned gains can cause sluggish response, excessive overshoot or instability, making gain tuning a critical step in commissioning any PID-controlled actuator, including derivative-only stages of the loop.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:gain-tuning", "labels": [ "Gain Tuning" ], "is_subclass_of": [ "PID Control" ], "wikilinks": [] }, { "id": "galactic-astronomy", "title": "Galactic Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:galactic-astronomy", "labels": [ "Galactic Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "galaxy-cluster", "title": "Galaxy Cluster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:galaxy-cluster", "labels": [ "Galaxy Cluster" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "galaxy-group", "title": "Galaxy Group", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:galaxy-group", "labels": [ "Galaxy Group" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "galaxy-supercluster", "title": "Galaxy Supercluster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:galaxy-supercluster", "labels": [ "Galaxy Supercluster" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "galaxy", "title": "Galaxy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:galaxy", "labels": [ "Galaxy" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "game-ai", "title": "Game AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Game AI is the branch of artificial intelligence applied to interactive entertainment, encompassing the design and implementation of non-player character behaviour, opponent strategy, pathfinding, procedural content generation, and narrative decision systems within game engines. It prioritises perceived intelligence, responsiveness, and entertainment value over computational optimality, distinguishing it from classical AI research focused on provably correct or globally optimal solutions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:game-ai", "labels": [ "Game AI", "Game Playing AI", "Game-Playing AI", "GameAI", "Video Game AI" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Application" ], "wikilinks": [] }, { "id": "game-asset-generation", "title": "Game Asset Generation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Game asset generation is the automated or semi-automated production of digital resources \u2014 including 3D meshes, textures, animations, sound effects, and narrative content \u2014 used in interactive entertainment, employing procedural algorithms, machine learning models, or generative AI to reduce manual authoring costs, accelerate iteration cycles, and enable content scale impossible through purely human workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:game-asset-generation", "labels": [ "Game Asset Generation", "Game Asset Marketplace", "Indie Game Asset Creation" ], "is_subclass_of": [ "Procedural Content Generation" ], "wikilinks": [] }, { "id": "game-development", "title": "game development", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Game development is the multidisciplinary engineering and creative practice of designing, building, and publishing interactive software experiences, spanning the full production pipeline from concept and pre-production through asset creation, engine programming, quality assurance, and live operations. It integrates disciplines including game design, real-time rendering, audio engineering, narrative scripting, physics simulation, and network programming into a unified production workflow centred on a game engine runtime. The field increasingly leverages machine learning for procedural content generation, NPC behaviour training via reinforcement learning, and generative AI for accelerating art and dialogue pipelines. Game development techniques have generalised beyond entertainment into spatial computing, robotics simulation, digital twins, and serious-games applications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:game-development", "labels": [ "Game Development", "Video Game Development" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "game-engine", "title": "Game Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software framework providing core functionality for rendering, physics, and interaction in real-time 3D environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:game-engine", "labels": [ "Game Engine", "GameEngine" ], "is_subclass_of": [ "Platform and Environment", "Software Platform" ], "wikilinks": [ "3D Engine", "Asset Management System", "Audio Engine", "Development Infrastructure", "Graphics Driver", "Interactive Experience", "Metaverse 101", "Multiplayer Gameplay", "OMA3 Media WG", "Scripting Runtime", "SIGGRAPH Pipeline WG", "Compute Infrastructure", "CreativeMediaDomain", "Graphics API", "Hardware Acceleration", "InfrastructureDomain", "Operating System", "Physics Engine", "PlatformLayer", "Procedural Content Generation" ] }, { "id": "game-mechanics", "title": "Game Mechanics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The formal rules, feedback systems, and interaction patterns that govern player agency, emergent behaviour, and progression within a game, simulation, or interactive virtual environment. Game mechanics define the space of possible player actions and their consequences through structures such as reward loops, resource management, collision and physics constraints, NPC behaviour trees, and win or loss conditions. In metaverse and blockchain-enabled contexts, mechanics are increasingly encoded in smart contracts, enabling transparent, tamper-resistant enforcement and player-owned economies. The discipline draws on systems theory, behavioural psychology, and human-computer interaction to produce engaging, balanced, and culturally durable play experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:game-mechanics", "labels": [ "Game Mechanics" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "game-narratives", "title": "Game Narratives", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interactive storytelling frameworks within metaverse gaming environments featuring branching storylines, player-driven plot progression, and adaptive narratives that evolve based on collective player decisions, creating dynamic and personalised gaming experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:game-narratives", "labels": [ "Game Narratives" ], "is_subclass_of": [ "Content and Assets", "Interactive Storytelling" ], "wikilinks": [ "Immersive Gaming Experiences", "Interactive Storytelling", "metaverse" ] }, { "id": "game-playing", "title": "Game Playing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Game playing is the AI task of selecting actions within the formal rules of a game -- board, card or video game -- in order to win, draw favourably or otherwise optimise an objective against one or more opponents or the environment. It has long served as a benchmark domain for artificial intelligence because games provide clear rules, measurable outcomes and tunable difficulty while still requiring planning, adversarial reasoning or pattern recognition. Search algorithms, which explore possible move sequences to evaluate their consequences, are a core technique for building game-playing agents.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:game-playing", "labels": [ "Game Playing" ], "is_subclass_of": [ "Game AI" ], "wikilinks": [] }, { "id": "game-theory", "title": "Game Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Game theory is the mathematical study of strategic interaction among rational agents, providing formal models for analysing decisions when outcomes depend on the choices of multiple actors with potentially conflicting interests. Founded by von Neumann and Morgenstern (1944) and extended by Nash's equilibrium concept (1950), it encompasses non-cooperative and cooperative game theory, mechanism design, and evolutionary dynamics. Its core solution concepts \u2014 Nash equilibrium, dominant strategy equilibrium, subgame-perfect equilibrium, and correlated equilibrium \u2014 are applied across economics, computer science, evolutionary biology, AI alignment research, and blockchain protocol design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:game-theory", "labels": [ "Game Theory" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "game-tree-search", "title": "Game Tree Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Game tree search is a family of algorithms that explore the tree of possible future game states branching from the current position, evaluating outcomes to choose the move that optimises a player's expected result against an adversary. Classical approaches such as minimax and alpha-beta pruning traverse the tree exhaustively or with heuristic bounds, while Monte Carlo tree search samples promising branches statistically to scale to games with very large branching factors. It is a foundational technique in adversarial game-playing AI, from classical board games to modern reinforcement-learning agents.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:game-tree-search", "labels": [ "Game Tree Search" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [] }, { "id": "game-fi", "title": "GameFi", "domain": "blockchain", "domain_name": "Blockchain", "definition": "GameFi refers to the combination of gaming with decentralised finance, where players can earn cryptocurrency or tokenised assets through gameplay. It overlaps with blockchain gaming and play-to-earn models.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:game-fi", "labels": [ "GameFi" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Token", "Smart Contract", "gaming", "NFT", "DeFi", "https://en.wikipedia.org/wiki/Play-to-earn", "https://ethereum.org/en/gaming/" ] }, { "id": "gamification", "title": "Gamification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Gamification is the application of game-design elements and play mechanics to non-game contexts in order to increase motivation, engagement, and desired behaviours. Typical elements include points, badges, leaderboards, levels, challenges, and progress feedback, often grounded in theories of intrinsic and extrinsic motivation. When well designed it leverages goal-setting, immediate feedback, and a sense of progression; when poorly designed it can produce superficial extrinsic incentives that erode lasting engagement. It is applied across learning, health, productivity, and immersive experiences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gamification", "labels": [ "Gamification" ], "is_subclass_of": [ "Interaction Design" ], "wikilinks": [] }, { "id": "gamma-ray-astronomy", "title": "Gamma-ray Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gamma-ray-astronomy", "labels": [ "Gamma-ray Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "garbled-circuits", "title": "Garbled Circuits", "domain": "security", "domain_name": "Security", "definition": "Garbled Circuits is a cryptographic technique, introduced by Yao in 1986, that enables two-party secure computation by encoding a Boolean circuit such that one party (the garbler) produces an encrypted representation of the circuit and the other party (the evaluator) can compute the output without learning the garbler's private inputs. Each gate of the circuit is replaced by a garbled truth table consisting of four ciphertexts, and the evaluator decrypts exactly one row per gate using wire labels obtained through [[Oblivious Transfer]]. Modern optimisations \u2014 including Free XOR, Half Gates, and Three Halves \u2014 reduce the communication and computation overhead to practical levels. Garbled circuits are foundational to general-purpose [[Multi-Party Computation]] and form the basis of many practical secure function evaluation protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:garbled-circuits", "labels": [ "Garbled Circuits" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "gartner-prediction", "title": "Gartner Prediction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A formal technology forecast published by Gartner Research, typically quantifying the adoption timeline, business impact, or market penetration of an emerging technology within a stated confidence window. Gartner Predictions appear in annual research reports and are contextualised within the Hype Cycle framework, providing enterprise IT strategists and technology leaders with actionable guidance on when and whether to invest in specific capabilities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:gartner-prediction", "labels": [ "Gartner Prediction" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Artificial Intelligence", "Edge Computing" ] }, { "id": "gas-fee-market", "title": "Gas Fee Market", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Gas Fee Market is the economic mechanism governing blockchain transaction costs, where users bid gas fees to prioritise transaction processing. Governed by EIP-1559 on Ethereum, it separates a burned base fee from an optional priority tip, creating a market-based congestion control system essential for NFT trades, smart contract execution, and virtual asset transfers in spatial computing platforms.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:gas-fee-market", "labels": [ "Gas Fee Market" ], "is_subclass_of": [ "Blockchain Economics" ], "wikilinks": [ "Blockchain Economics", "metaverse" ] }, { "id": "gas-fee", "title": "Gas Fee", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A gas fee is the payment a user must make to compensate the network validators or miners for the computational resources consumed when processing a transaction or executing a smart contract on a blockchain. On Ethereum-compatible networks, 'gas' is an abstract unit measuring the computational effort required by an operation; the fee is calculated as gas units consumed multiplied by a price per unit (the gas price), denominated in the network's native currency. Following EIP-1559 on Ethereum, gas fees split into a protocol-set base fee \u2014 which is burned, permanently removing supply \u2014 and an optional priority tip paid directly to the block proposer. Gas fees serve the dual purpose of economically compensating validators and acting as a spam-prevention mechanism by making resource-intensive computation costly.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:gas-fee", "labels": [ "Gas Fee", "Gas Fee Payment", "Gas Fees" ], "is_subclass_of": [ "Transaction Fee" ], "wikilinks": [ "Smart Contract", "Ethereum", "Transaction Fee" ] }, { "id": "gas-giant", "title": "Gas Giant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gas-giant", "labels": [ "Gas Giant" ], "is_subclass_of": [ "Planet" ], "wikilinks": [] }, { "id": "gas-limit", "title": "Gas Limit", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Gas Limit is the maximum amount of gas \u2014 the unit measuring computational effort \u2014 that a sender authorises for a blockchain transaction or that a block may contain in aggregate. It serves as a hard cap preventing unbounded resource consumption, protecting network nodes from denial-of-service attacks and ensuring predictable block processing times. On Ethereum, each transaction carries a user-set gas limit and each block carries a protocol-enforced block gas limit that validators adjust over time.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:gas-limit", "labels": [ "Gas Limit", "Block Gas Limit" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "gas-mechanism", "title": "Gas Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The system in a blockchain such as Ethereum that meters and prices computation, requiring users to pay fees denominated in units of gas for the operations their transactions perform. It limits resource use and prioritises transactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gas-mechanism", "labels": [ "Gas Mechanism", "Gas Fee Mechanism" ], "is_subclass_of": [ "Ethereum Smart Contract Platform" ], "wikilinks": [ "Ethereum Virtual Machine", "Smart Contract", "Ethereum" ] }, { "id": "gas-metering", "title": "Gas Metering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gas metering is the mechanism by which a blockchain virtual machine accounts for the computational, storage and bandwidth resources consumed by executing a transaction or smart contract. Each low-level operation is assigned a gas cost, and execution proceeds only while the sender's prepaid gas budget remains, halting deterministically when the budget is exhausted. By pricing computation, gas metering deters denial-of-service abuse, bounds execution and forms the basis of transaction fees.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gas-metering", "labels": [ "Gas Metering" ], "is_subclass_of": [ "Smart Contract Platform", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "gas-optimization", "title": "Gas Optimization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gas optimization is the systematic reduction of computational resources required for smart contract execution on blockchain networks, achieved through efficient storage patterns, opcode selection, data structure design, and batching strategies to minimise transaction costs and improve economic vi...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:gas-optimization", "labels": [ "Gas Optimization", "Gas Optimisation" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain" ], "wikilinks": [ "EVM (Ethereum Virtual Machine)", "Scalability Solutions", "Smart Contract Development", "Blockchain", "BlockchainDomain", "Gas", "Smart Contract", "Transaction Fee" ] }, { "id": "gas-price", "title": "Gas Price", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gas Price is the amount of cryptocurrency (denominated in gwei on Ethereum) that a transaction sender is willing to pay per unit of gas consumed during execution. It serves as the primary market mechanism for prioritising transactions within a block and compensating validators or miners for computational work. Gas price interacts with the gas limit and base fee to determine total transaction cost, and is subject to dynamic adjustment under fee-market protocols such as EIP-1559.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:gas-price", "labels": [ "Gas Price" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "gas", "title": "Gas", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gas is the unit of computational work measurement within EVM-compatible blockchains. Every opcode executed by the Ethereum Virtual Machine consumes a defined gas quantity; users pay gas_price \u00d7 gas_used to validators, preventing denial-of-service through unbounded computation and creating a market for block space.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gas", "labels": [ "Gas" ], "is_subclass_of": [ "Economic Mechanism", "Blockchain Entity", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "gasless-transaction", "title": "Gasless Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Gasless Transaction is a blockchain transaction in which the end user does not directly pay the network's gas fee, because a relayer, paymaster, or sponsoring contract covers the cost on their behalf, typically in exchange for an off-chain signature authorising the action. It removes the requirement that a wallet hold the native gas token before it can transact, lowering the barrier to onboarding new users. Account abstraction standards such as ERC-4337 provide the infrastructure that makes gasless transactions practical at scale.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:gasless-transaction", "labels": [ "Gasless Transaction" ], "is_subclass_of": [ "Account Abstraction" ], "wikilinks": [] }, { "id": "gasper-consensus", "title": "Gasper Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gasper Consensus is Ethereum's Proof-of-Stake consensus protocol, combining Casper FFG (Friendly Finality Gadget) as a finality mechanism with LMD GHOST (Latest Message Driven Greedy Heaviest Observed Subtree) as the fork choice rule. Validators stake ETH directly and attest to blocks in each epoch; Casper FFG provides economic finality by requiring two-thirds supermajority agreement, while LMD GHOST guides validators towards the heaviest chain, resolving short-lived forks without full finalisation delay.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:gasper-consensus", "labels": [ "Gasper Consensus" ], "is_subclass_of": [ "Protocol and Consensus", "Hybrid Consensus" ], "wikilinks": [ "Blockchain", "Hybrid Consensus" ] }, { "id": "gauge-voting", "title": "Gauge Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gauge voting is a mechanism in DeFi protocols such as Curve where token holders allocate weights to liquidity pool gauges, determining how reward emissions are distributed across pools each epoch. Voting power is typically proportional to vote-escrowed (veToken) balances, creating direct economic incentives for bribe markets and liquidity direction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gauge-voting", "labels": [ "Gauge Voting", "Gauge Weight Voting", "Gauge Weighting" ], "is_subclass_of": [ "Governance Token" ], "wikilinks": [ "Curve Finance", "Votium", "Hidden Hand", "Tokenomics", "Governance Token" ] }, { "id": "gaussian-distribution", "title": "Gaussian Distribution", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A continuous probability distribution, also called the normal distribution, characterised by a symmetric bell-shaped density defined by its mean and variance. It arises naturally via the central limit theorem and is foundational to statistics, machine learning, and signal processing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:gaussian-distribution", "labels": [ "Gaussian Distribution" ], "is_subclass_of": [ "AI Technique", "Probability Theory" ], "wikilinks": [ "Probability Theory", "Statistics", "Bayesian Inference", "Markov Chain Monte Carlo" ] }, { "id": "gaussian-mechanism", "title": "Gaussian Mechanism", "domain": "security", "domain_name": "Security", "definition": "A differential privacy noise mechanism that releases a numeric query result after adding noise drawn from a normal distribution calibrated to the query's L2 sensitivity, providing approximate (epsilon, delta)-differential privacy; its light tails, per-coordinate efficiency on high-dimensional vectors, and tight composition under R\u00e9nyi and zero-concentrated accounting make it the mechanism of choice for iterative computations, most prominently gradient perturbation in differentially private stochastic gradient descent (DP-SGD).", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:gaussian-mechanism", "labels": [ "Gaussian Mechanism" ], "is_subclass_of": [ "Noise Mechanisms" ], "wikilinks": [ "Noise Mechanisms", "Differential Privacy", "Laplace Mechanism" ] }, { "id": "gaussian-mixture-model", "title": "Gaussian Mixture Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Gaussian mixture model is a probabilistic model that represents a population as a weighted combination of several Gaussian distributions, each describing a latent subpopulation or cluster. Its parameters \u2014 the mixing weights, means, and covariance matrices \u2014 are typically estimated by the expectation-maximisation algorithm, which iteratively assigns soft responsibilities to data points and updates the component parameters. As a generative latent-variable model, it supports soft clustering, density estimation, and probabilistic classification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gaussian-mixture-model", "labels": [ "Gaussian Mixture Model" ], "is_subclass_of": [ "Latent Variable Model" ], "wikilinks": [] }, { "id": "gaussian-process-regression", "title": "Gaussian Process Regression", "domain": "ai", "domain_name": "Ai", "definition": "Gaussian process regression is a non-parametric Bayesian method that models an unknown function as a distribution over functions defined by a mean and a covariance (kernel) function. Given observations, it produces a posterior that yields predictions together with calibrated uncertainty estimates. It is widely used where data is scarce and quantified uncertainty matters, such as Bayesian optimisation and surrogate modelling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gaussian-process-regression", "labels": [ "Gaussian Process Regression" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "gaussian-process", "title": "Gaussian Process", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Gaussian Process (GP) is a non-parametric Bayesian model that defines a probability distribution over functions, fully characterised by a mean function and a covariance (kernel) function, such that any finite collection of function evaluations follows a joint Gaussian distribution. Conditioning the GP prior on observed data yields a closed-form posterior distribution over functions that simultaneously provides point predictions and principled uncertainty estimates. GPs are widely used for regression, classification, surrogate modelling, and as acquisition-function models in Bayesian optimisation, with exact inference scaling cubically in the number of observations and sparse inducing-point approximations enabling scalable variants.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:gaussian-process", "labels": [ "Gaussian Process", "Gaussian Processes" ], "is_subclass_of": [ "Bayesian Inference" ], "wikilinks": [] }, { "id": "gaussian-splatting", "title": "Gaussian Splatting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Novel view synthesis and 3D scene representation technique introduced by Kerbl, Kopanas, Leimk\u00fchler and Drettakis at SIGGRAPH 2023 (INRIA Sophia Antlis), representing scenes as explicit collections of s of anisotropic 3D Gaussian primitives \u2014 each defined by a 3D mean position \u03bc \u2208 \u211d\u00b3, a 3\u00d73 covar...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:gaussian-splatting", "labels": [ "Gaussian Splatting", "4D Gaussian Splatting", "Gaussian Splatting Rasterisation" ], "is_subclass_of": [ "Display and Rendering", "3D Scene Reconstruction", "Neural Rendering", "Novel View Synthesis", "Differentiable Rendering", "Computer Vision" ], "wikilinks": [ "3D Scene Reconstruction", "ACM SIGGRAPH", "Adam Optimiser", "Adaptive Densification", "AlgorithmLayer", "Alpha Blending", "Alpha Compositing", "Camera Calibration", "COLMAP", "Covariance Decomposition", "CUDA", "Depth Sorting", "Differentiable Rasterizer", "Differentiable Rendering", "EWA Splatting", "Gaussian Distribution", "Gaussian Primitive", "GPU Compute", "IEEE CVPR", "IEEE ICCV" ] }, { "id": "gaze-awareness", "title": "Gaze Awareness", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Gaze Awareness is the capability of a collaborative system to detect, track, and communicate where participants are directing their visual attention during a shared session. It enables collaborators to perceive mutual attention states \u2014 such as who is looking at whom, or what shared artefact is being observed \u2014 without explicit verbal indication. Gaze awareness is a foundational non-verbal cue that supports turn-taking, joint attention, and coordinated action in distributed teams.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:gaze-awareness", "labels": [ "Gaze Awareness" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "gaze-contingent-telepresence-display", "title": "Gaze Contingent Telepresence Display", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Telethrone is a research concept for a hyper-personal display and telepresence installation that resolves the challenge of rendering a spatially aware, photorealistic metahuman avatar for a single collocated viewer. It combines personalised 3D reconstruction from 2D imagery, gaze-contingent rendering, and situated display technology to achieve high-fidelity presence without requiring the viewer to wear head-mounted displays.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:gaze-contingent-telepresence-display", "labels": [ "Gaze Contingent Telepresence Display", "Telethrone" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MUST" ] }, { "id": "gaze-control", "title": "Gaze Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Gaze control regulates robot eye and head movement to establish, maintain, and redirect visual attention toward objects and people, conveying robot intent and facilitating natural Human-Robot Interaction.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:gaze-control", "labels": [ "Gaze Control" ], "is_subclass_of": [ "Human-Robot Interaction", "Robotics", "Attention Control" ], "wikilinks": [ "Attention Control", "Attention Redirection", "Camera Actuators", "Eye Movement Controller", "Gaze Target Estimator", "Head Movement Mechanism", "HRI", "Joint Attention", "Kinematics Computation", "Natural Interaction", "Pan-Tilt Unit", "Servo Motors", "Social Signal Transmission", "Social Understanding", "Target Detection", "Target Tracking", "Visual Attention Model", "Visual Perception", "Computer Vision", "Convolutional Neural Network" ] }, { "id": "gaze-tracking", "title": "Gaze Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Gaze tracking is the measurement of where a person is looking by estimating the direction and point of regard of the eyes, typically using cameras and infrared illumination to locate pupil and corneal reflections. In spatial computing it provides a hands-free input modality and a signal for attention, enabling interfaces that respond to where the user looks. Gaze tracking is foundational to foveated rendering, intent inference, and accessible interaction in head-mounted displays and immersive systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:gaze-tracking", "labels": [ "Gaze Tracking" ], "is_subclass_of": [ "Eye Tracking" ], "wikilinks": [] }, { "id": "gazebo-simulator", "title": "Gazebo Simulator", "domain": "robotics", "domain_name": "Robotics", "definition": "Gazebo Simulator (now branded Gz Sim following the Open Robotics / Intrinsic rebranding) is an open-source, physics-accurate 3D robotics simulator providing rigid-body dynamics (via ODE, Bullet, DART, or Simbody), sensor simulation (cameras, LiDAR, IMU, GPS), and a plugin architecture for custom ...", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:gazebo-simulator", "labels": [ "Gazebo Simulator", "Gazebo", "Gazebo Simulation" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Physics Simulator", "Robotics Development Tool" ], "wikilinks": [ "AI Domain", "colcon Build System", "Gazebo Plugin API", "Gz Physics Plugin", "Gz Rendering", "Gz Sensors", "Gz Sim Documentation", "Gz Transport", "gz-transport Message Protocol", "Hardware-in-the-Loop Testing", "Ignition Fuel Model Database", "IROS 2004 Gazebo Paper", "Multi-Robot Simulation", "OGRE3D Rendering Engine", "Physics Simulator", "Reinforcement Learning Training", "Robotics Development Tool", "ros2_control Hardware Interface", "SDF Robot Description", "SDF World Model" ] }, { "id": "gazetteer", "title": "Gazetteer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gazetteer", "labels": [ "Gazetteer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gemini-multimodal-language-model", "title": "Gemini Multimodal Language Model", "domain": "ai", "domain_name": "Ai", "definition": "Gemini is Google DeepMind's family of natively multimodal large language models, announced in December 2023 as the successor to PaLM 2, designed from the ground up to reason across text, images, audio, video, and code within a single unified architecture. The Gemini family spans Ultra, Pro, Flash, and Nano capability tiers, enabling deployment from data-centre scale to on-device inference. It directly competes with OpenAI's GPT-4 family and Anthropic Claude as one of the three dominant frontier model lineages.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gemini-multimodal-language-model", "labels": [ "Gemini Multimodal Language Model", "Gemini", "Gemini API", "Gemini for Workspace" ], "is_subclass_of": [ "Multimodal AI" ], "wikilinks": [] }, { "id": "general-purpose-ai-model", "title": "General Purpose AI Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An AI model that displays significant generality and is capable of competently performing a wide range of distinct tasks regardless of the way the model is placed on the market, as defined by EU AI Act Article 3(63). GPAI models face specific transparency and documentation obligations, with enhanced requirements for those meeting systemic risk thresholds based on training compute or high-impact capability benchmarks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:general-purpose-ai-model", "labels": [ "General Purpose AI Model", "General-Purpose AI" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "generalisation", "title": "Generalisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Generalisation is the capacity of a machine-learning model to perform accurately on previously unseen data drawn from the same distribution as its training set, rather than merely memorising training examples. It is the central objective of supervised learning and is quantified by the gap between training and held-out performance. Good generalisation reflects a model that has captured the underlying regularities of a task instead of spurious correlations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:generalisation", "labels": [ "Generalisation" ], "is_subclass_of": [ "Machine Learning", "AI Research Area" ], "wikilinks": [] }, { "id": "generative-ai-development-methodology", "title": "Generative AI Development Methodology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A practitioner's phased methodology for building generative AI applications, progressing from proof-of-concept (using the best available models and rapid prototyping tools) through stakeholder validation to a robust production build, with explicit guidance on iterative evaluation, legal risk deferral, and avoidance of premature fine-tuning.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:generative-ai-development-methodology", "labels": [ "Generative AI Development Methodology", "Advice for developing GenAI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "qLoRA", "Runway", "Vercel", "AI Agent System", "AI Video", "LoRA DoRA etc", "Midjourney", "Music and Audio", "OpenAI Whisper" ] }, { "id": "generative-ai-engineering", "title": "Generative AI Engineering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative AI Engineering is the applied discipline concerned with designing, developing, fine-tuning, and deploying generative AI systems\u2014such as large language models, diffusion models, and multimodal transformer architectures\u2014to create novel artefacts including text, images, audio, code, and synthetic data. It encompasses the full engineering lifecycle from model selection and prompt engineering through retrieval-augmented generation, evaluation, and production observability. The field bridges machine learning research and software engineering, requiring competence in model architecture, infrastructure, and responsible AI practices.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:generative-ai-engineering", "labels": [ "Generative AI Engineering" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "generative-ai-near-term-forecasts", "title": "Generative AI Near-Term Forecasts", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated set of near-term forecasts for the generative AI video and creative-tool landscape, covering model aggregators, story-level tooling, template ecosystems, real-time trend integration, open-source convergence, inference cost reduction, specialised models, and the social/legal challenges of AI-generated content. These predictions frame strategic planning for platforms operating in the AI creative economy.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:generative-ai-near-term-forecasts", "labels": [ "Generative AI Near-Term Forecasts", "Predictions" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "generative-ai", "title": "Generative AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative AI encompasses Machine Learning systems capable of creating new content across modalities including text, images, audio, video, and code through Neural Networks trained on large datasets.", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:generative-ai", "labels": [ "Generative AI", "Generative AI Boom", "Generative AI Products", "Generative Ai", "GenerativeAI" ], "is_subclass_of": [ "AI Application", "Deep Learning", "Artificial Intelligence", "Machine Learning Discipline" ], "wikilinks": [ "AI Art", "AlgorithmicLayer", "Anthropic", "Art", "Audio Generation", "Automated Design", "Code Synthesis", "Content Creation", "Creative AI", "Design", "GAN", "GPU Compute", "Large-Scale Datasets", "Music Generation", "Neural Network Training", "Neural Networks", "NFT", "NIST AI Standards", "Stability AI", "Synthetic Media" ] }, { "id": "generative-adversarial-network", "title": "Generative Adversarial Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Generative Adversarial Network (GAN) is a deep learning architecture in which a generator network and a discriminator network are trained simultaneously in an adversarial min-max game: the generator learns to produce synthetic samples indistinguishable from real data, while the discriminator learns to detect fakes. GANs underpin high-fidelity image synthesis, video generation, data augmentation, and synthetic data creation across domains including healthcare, finance, and computer vision.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:generative-adversarial-network", "labels": [ "Generative Adversarial Network" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "\ud83e\udd16", "Diffusion Models", "Generative Adversarial Networks", "MetaverseDomain", "Research Tools" ] }, { "id": "generative-adversarial-networks", "title": "Generative Adversarial Networks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative Adversarial Networks (GANs) are a class of deep generative models introduced by Ian Goodfellow and colleagues at the University of Montreal framing density estimation as a two-player zero-sum minimax game between a generator network G:Z\u2192X mapping samples from a low-dimensional latent p...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:generative-adversarial-networks", "labels": [ "Generative Adversarial Networks" ], "is_subclass_of": [ "AI Model Architecture", "Generative Model", "Deep Generative Model", "Neural Network", "Unsupervised Learning", "Implicit Density Model", "Adversarial Learning System" ], "wikilinks": [ "Adam Optimiser", "Adversarial Learning System", "Adversarial Loss", "AlgorithmLayer", "Anomaly Detection", "Arjovsky et al. 2017 Wasserstein GAN", "Autoregressive Model", "Brock et al. 2019 BigGAN", "Convolution", "Convolutional Neural Networks", "Creative Tools", "CVPR", "Data Augmentation", "De Cao & Kipf 2018 MolGAN", "Deep Generative Model", "Deepfake Detection", "Differentiable Architecture", "Diffusion Model", "Discriminator Network", "Domain Adaptation" ] }, { "id": "generative-content", "title": "Generative Content", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative Content is digital media \u2014 text, imagery, audio, video, 3D assets, or code \u2014 produced algorithmically by generative AI models rather than created directly by human authors. It encompasses outputs of large language models, diffusion models, generative adversarial networks, variational autoencoders, and multimodal neural synthesis systems. The concept carries significant implications for provenance tracking, copyright attribution, content authenticity, and quality assurance, particularly in high-throughput pipelines for spatial computing, metaverse environments, and educational platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:generative-content", "labels": [ "Generative Content" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "owl:Thing" ] }, { "id": "generative-design-tool", "title": "Generative Design Tool", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI-assisted software application that produces optimized 3D designs from functional constraints using machine learning and computational algorithms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:generative-design-tool", "labels": [ "Generative Design Tool" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "3D Generator", "AI Engine", "AI Model", "Autodesk Design ML", "Automated Design", "Constraint Solver", "Content Creation Tool", "Design Database", "Design Optimization", "Design Optimizer", "SIGGRAPH AI Design WG", "Authoring Tool", "ComputationAndIntelligenceDomain", "Compute Infrastructure", "ComputeLayer", "Constraint-Based Design", "CreativeMediaDomain", "DataLayer", "Graphics API", "Machine Learning Infrastructure" ] }, { "id": "generative-design", "title": "Generative Design", "domain": "ai", "domain_name": "Ai", "definition": "Generative Design is a computational design methodology in which algorithms autonomously explore a defined design space\u2014bounded by performance constraints, manufacturing requirements, and material properties\u2014to generate and evaluate large numbers of design candidates, surfacing options that meet objectives a human designer specifies but did not hand-craft. Implemented through topology optimisation, evolutionary algorithms, and increasingly through deep generative models, it produces geometrically complex structures\u2014often resembling organic forms\u2014that achieve material efficiency or performance targets unattainable through conventional manual design. The output is typically a ranked set of design alternatives that engineers evaluate and refine.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:generative-design", "labels": [ "Generative Design", "GenerativeDesign" ], "is_subclass_of": [ "Parametric Modeling" ], "wikilinks": [] }, { "id": "generative-engine-optimization", "title": "Generative Engine Optimization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The practice of structuring content and data to improve visibility and ranking within generative AI search engines and large language model responses.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:generative-engine-optimization", "labels": [ "Generative Engine Optimization" ], "is_subclass_of": [ "Enterprise AI" ], "wikilinks": [] }, { "id": "generative-model", "title": "Generative Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A class of machine learning models that learn the underlying probability distribution of training data and can sample novel instances from that distribution. Architectures include generative adversarial networks, variational autoencoders, diffusion models, normalising flows, and autoregressive transformers, covering domains such as text, image, audio, and code generation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:generative-model", "labels": [ "Generative Model", "Generative Modelling", "GenerativeModel", "Probabilistic Generative Model" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Model Architecture" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning" ] }, { "id": "generative-models", "title": "Generative Models", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Generative models are machine learning models that learn the underlying distribution of data so they can produce new samples resembling the training data. Major families include generative adversarial networks, variational autoencoders, autoregressive models and diffusion models, underpinning image, text, audio and video generation across AI applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:generative-models", "labels": [ "Generative Models", "Generative Image Models" ], "is_subclass_of": [ "Generative Model", "Generative AI" ], "wikilinks": [ "Neural Network", "Generative AI", "Image Generation", "Diffusion Model", "Generative Adversarial Network", "Generative Model" ] }, { "id": "generative-world-models", "title": "Generative World Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A class of AI models capable of simulating and generating interactive, three-dimensional environments or physical realities, often used for training agents or creating synthetic data.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:generative-world-models", "labels": [ "Generative World Models" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "generator-network", "title": "Generator Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A generator network is the synthesis component of a generative deep learning architecture \u2014 most prominently within Generative Adversarial Networks (GANs) \u2014 that learns to map samples from a low-dimensional latent space into high-dimensional data outputs (images, audio, video, 3D shapes) whose statistical distribution matches that of a training dataset. The generator is trained in adversarial competition with a discriminator network, receiving gradient signal not from direct comparison with target samples but from the discriminator's attempt to distinguish generated from real samples, forcing the generator to produce increasingly realistic outputs. Generator networks are the conceptual precursors to the decoder components of VAEs and the denoising networks of diffusion models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:generator-network", "labels": [ "Generator Network" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "genesis-block", "title": "Genesis Block", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The genesis block is the first block in a blockchain, hardcoded into the client software and serving as the immutable anchor from which the entire chain of subsequent blocks descends. It establishes the initial chain state, encodes the founding parameters of the network, and contains no reference to a previous block hash. Because it cannot be altered without invalidating all subsequent blocks, the genesis block is the ultimate root of trust for the distributed ledger.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:genesis-block", "labels": [ "Genesis Block" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Autonomous Robot", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "genetic-algorithm", "title": "Genetic Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A genetic algorithm is a population-based metaheuristic for optimisation and search inspired by Darwinian natural selection, in which candidate solutions are encoded as chromosomes and evolved across generations through selection, crossover, and mutation. A fitness function ranks individuals so that fitter solutions are preferentially recombined, gradually steering the population towards high-quality regions of the search space without requiring gradient information. Genetic algorithms are well suited to combinatorial, non-convex, and black-box problems where the objective is rugged, discontinuous, or expensive to differentiate.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:genetic-algorithm", "labels": [ "Genetic Algorithm" ], "is_subclass_of": [ "Evolutionary Algorithm" ], "wikilinks": [] }, { "id": "genomics", "title": "Genomics", "domain": "ai", "domain_name": "Ai", "definition": "Genomics is the study of the complete set of genetic material in organisms, encompassing the structure, function, evolution, and editing of genomes. It combines high-throughput DNA sequencing with computational analysis to interpret vast quantities of genetic data. Genomics underpins precision medicine, evolutionary biology, agriculture, and the increasing application of machine learning to biological sequence interpretation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:genomics", "labels": [ "Genomics" ], "is_subclass_of": [ "Computational Biology" ], "wikilinks": [] }, { "id": "geocentric-latitude", "title": "Geocentric Latitude", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geocentric-latitude", "labels": [ "Geocentric Latitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geocode", "title": "Geocode", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geocode", "labels": [ "Geocode" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodesy", "title": "Geodesy", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodesy", "labels": [ "Geodesy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodetic-control-point", "title": "Geodetic Control Point", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-control-point", "labels": [ "Geodetic Control Point" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodetic-coordinate-reference-system", "title": "Geodetic Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-coordinate-reference-system", "labels": [ "Geodetic Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "geodetic-datum", "title": "Geodetic Datum", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-datum", "labels": [ "Geodetic Datum" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodetic-latitude", "title": "Geodetic Latitude", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-latitude", "labels": [ "Geodetic Latitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodetic-longitude", "title": "Geodetic Longitude", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-longitude", "labels": [ "Geodetic Longitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geodetic-reference-frame", "title": "Geodetic Reference Frame", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geodetic-reference-frame", "labels": [ "Geodetic Reference Frame" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geographic-coordinate-reference-system", "title": "Geographic Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geographic-coordinate-reference-system", "labels": [ "Geographic Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "geographic-distribution", "title": "Geographic Distribution", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Geographic distribution is the placement of computing resources, data, or services across physically separate locations and regions. In distributed systems it reduces latency for global users, improves fault tolerance against regional outages, and helps satisfy data-residency requirements. It introduces trade-offs around consistency, replication, and inter-region network costs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:geographic-distribution", "labels": [ "Geographic Distribution" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "geographic-information-system", "title": "Geographic Information System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Geographic Information System (GIS) is an integrated framework for capturing, storing, managing, analysing and visualising data that is referenced to locations on the Earth's surface. It combines spatial databases, coordinate systems and analytical operations to model geographic phenomena and answer questions about place, distance, pattern and relationship. GIS underpins applications ranging from urban planning and environmental monitoring to logistics and emergency response.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:geographic-information-system", "labels": [ "Geographic Information System" ], "is_subclass_of": [ "Spatial Computing", "Remote Sensing" ], "wikilinks": [] }, { "id": "geoid-undulation", "title": "Geoid Undulation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geoid-undulation", "labels": [ "Geoid Undulation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geoid", "title": "Geoid", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geoid", "labels": [ "Geoid" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geolocation-accuracy", "title": "Geolocation Accuracy", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geolocation-accuracy", "labels": [ "Geolocation Accuracy" ], "is_subclass_of": [ "Measurement Accuracy" ], "wikilinks": [] }, { "id": "geolocation", "title": "Geolocation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geolocation", "labels": [ "Geolocation", "Geolocate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geomagnetic-field", "title": "Geomagnetic Field", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geomagnetic-field", "labels": [ "Geomagnetic Field" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geomagnetic-storm", "title": "Geomagnetic Storm", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geomagnetic-storm", "labels": [ "Geomagnetic Storm" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geomagnetic-substorm", "title": "Geomagnetic Substorm", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geomagnetic-substorm", "labels": [ "Geomagnetic Substorm" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geometric-calibration", "title": "Geometric Calibration", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geometric-calibration", "labels": [ "Geometric Calibration" ], "is_subclass_of": [ "Calibration" ], "wikilinks": [] }, { "id": "geometric-dilution-of-precision", "title": "Geometric Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geometric-dilution-of-precision", "labels": [ "Geometric Dilution of Precision" ], "is_subclass_of": [ "Dilution of Precision" ], "wikilinks": [] }, { "id": "geometry-processing", "title": "Geometry Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Geometry processing is the field concerned with acquiring, representing, analysing and manipulating geometric models, typically triangle meshes and point clouds. It encompasses operations such as smoothing, simplification, parameterisation, remeshing and surface reconstruction, drawing on differential geometry and linear algebra. Geometry processing underpins 3D content pipelines, simulation and rendering by transforming raw or noisy geometric data into clean, structured representations suitable for downstream use.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:geometry-processing", "labels": [ "Geometry Processing" ], "is_subclass_of": [ "Computational Geometry" ], "wikilinks": [] }, { "id": "geometry-shader", "title": "Geometry Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A geometry shader is a programmable stage of the graphics pipeline that operates on whole primitives, such as points, lines or triangles, after vertex processing and before rasterisation. Unlike vertex shaders, it can create or discard primitives, emitting zero or more output primitives for each input. This capability supports effects such as silhouette extrusion, point-sprite expansion and dynamic level-of-detail generation directly on the GPU.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:geometry-shader", "labels": [ "Geometry Shader" ], "is_subclass_of": [ "Shader Language" ], "wikilinks": [] }, { "id": "geometry", "title": "Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The branch of mathematics concerned with the properties of space and of figures within it \u2014 points, lines, curves, surfaces, and solids \u2014 and with the notions of distance, angle, area, volume, and transformation that relate them. From Euclid's axiomatic plane geometry through analytic, projective, differential, and computational branches, geometry supplies the formal language in which spatial computing describes the world: coordinate frames and transformations in robotics, projective camera models in computer vision, meshes and parametric surfaces in graphics, and spatial subdivisions such as Voronoi diagrams in computational geometry.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:geometry", "labels": [ "Geometry" ], "is_subclass_of": [ "Mathematical Science" ], "wikilinks": [ "Mathematical Science", "Computational Geometry", "Scene Geometry", "Voronoi Diagram" ] }, { "id": "geopolitics", "title": "Geopolitics", "domain": "governance", "domain_name": "Governance", "definition": "Geopolitics is the study and practice of how geography, resources, and national power shape relations and competition among states. In technology it frames how control over compute, semiconductors, data, and AI capability translates into strategic advantage and influences export controls, alliances, and standards-setting. It is a core lens for understanding international competition in AI.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:geopolitics", "labels": [ "Geopolitics" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "georeferenced-data-state", "title": "Georeferenced Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:georeferenced-data-state", "labels": [ "Georeferenced Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "georeferencing", "title": "Georeferencing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:georeferencing", "labels": [ "Georeferencing", "Georeference" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-buffering", "title": "Geospatial Buffering", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-buffering", "labels": [ "Geospatial Buffering" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-clipping", "title": "Geospatial Clipping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-clipping", "labels": [ "Geospatial Clipping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-data-layer", "title": "Geospatial Data Layer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-data-layer", "labels": [ "Geospatial Data Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-data", "title": "Geospatial Data", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Geospatial data is information explicitly tied to a location on or near the Earth's surface, expressed through coordinates, geometries, or raster grids. It encompasses vector features (points, lines, polygons), satellite and aerial imagery, elevation models, point clouds, and attributes associated with places. Such data underpins mapping, navigation, remote sensing, logistics, urban planning, and spatial analysis across many domains.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:geospatial-data", "labels": [ "Geospatial Data", "Geospatial Data Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "geospatial-engine", "title": "Geospatial Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software platforms that stream, render, and process large volumes of geographic and 3D spatial data in real-time for metaverse applications, enabling digital twins of real-world locations, location-based experiences, and integration between physical geography and virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:geospatial-engine", "labels": [ "Geospatial Engine" ], "is_subclass_of": [ "Platform and Environment", "Spatial Computing Paradigm" ], "wikilinks": [ "Geospatially-Accurate Metaverse", "metaverse", "Spatial Computing" ] }, { "id": "geospatial-feature-extraction", "title": "Geospatial Feature Extraction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-feature-extraction", "labels": [ "Geospatial Feature Extraction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-information", "title": "Geospatial Information", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Geospatial Information is data that describes objects, events, or phenomena with reference to a location on or near the Earth's surface. It integrates coordinate systems, geographic data layers, and spatial analytics to support location-aware applications including digital twins, augmented reality overlays, and urban simulation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:geospatial-information", "labels": [ "Geospatial Information" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "geospatial-interface-protocol", "title": "Geospatial Interface Protocol", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-interface-protocol", "labels": [ "Geospatial Interface Protocol" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-metadata", "title": "Geospatial Metadata", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-metadata", "labels": [ "Geospatial Metadata" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-overlay", "title": "Geospatial Overlay", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-overlay", "labels": [ "Geospatial Overlay" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-service", "title": "Geospatial Service", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geospatial-service", "labels": [ "Geospatial Service" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "geospatial-technology", "title": "Geospatial Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Geospatial Technology encompasses the integrated suite of tools, systems, and methodologies used to acquire, store, process, analyse, and visualise data that is referenced to locations on or near the Earth's surface. It combines hardware systems such as GPS receivers, LiDAR scanners, and remote sensing satellites with software platforms including Geographic Information Systems (GIS), geospatial databases, and mapping engines to produce authoritative spatial representations. In spatial computing and immersive environments, geospatial technology provides the world-scale coordinate infrastructure required for digital twin construction, precise AR/VR anchoring, autonomous navigation, and location-aware services. The discipline spans geodesy, photogrammetry, cartography, and real-time positioning, forming a foundational layer beneath smart city infrastructure, environmental monitoring, and logistics optimisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:geospatial-technology", "labels": [ "Geospatial Technology" ], "is_subclass_of": [ "Spatial Computing", "Geospatial Engine" ], "wikilinks": [ "owl:Thing" ] }, { "id": "geostationary-orbit", "title": "Geostationary Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geostationary-orbit", "labels": [ "Geostationary Orbit" ], "is_subclass_of": [ "Geosynchronous Orbit" ], "wikilinks": [] }, { "id": "geosynchronous-orbit", "title": "Geosynchronous Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:geosynchronous-orbit", "labels": [ "Geosynchronous Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gesture-recognition", "title": "Gesture Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Gesture recognition is the computational process of identifying and interpreting human physical movements\u2014typically of the hands, arms, head, or full body\u2014as meaningful symbolic inputs to a computing system, enabling touchless, natural interaction paradigms. The field encompasses multiple sensing modalities including optical cameras, depth sensors, inertial measurement units, and electromagnetic field detectors, combined with machine learning pipelines that extract skeletal or surface features and classify movement sequences into semantic categories. Gesture recognition underpins human-computer interaction in extended reality, robotics, automotive interfaces, and accessibility technology, replacing or augmenting physical input devices with embodied motion vocabulary.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gesture-recognition", "labels": [ "Gesture Recognition", "Gesture Control", "GestureRecognition" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "gibbs-sampling", "title": "Gibbs Sampling", "domain": "ai", "domain_name": "Ai", "definition": "Gibbs sampling is a Markov chain Monte Carlo algorithm that draws samples from a multivariate distribution by iteratively sampling each variable from its full conditional distribution given the current values of all others. It is a special case of Metropolis-Hastings in which every proposal is accepted, and it requires the conditionals to be tractable. Gibbs sampling is widely used for posterior inference in hierarchical and graphical models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gibbs-sampling", "labels": [ "Gibbs Sampling" ], "is_subclass_of": [ "Markov Chain Monte Carlo" ], "wikilinks": [] }, { "id": "gig-economy", "title": "Gig Economy", "domain": "finance", "domain_name": "Finance", "definition": "The gig economy is a labour market structured around short-term, task-based, and on-demand work mediated largely by digital platforms, rather than long-term salaried employment. Workers are typically engaged as independent contractors, gaining flexibility and autonomy while bearing income volatility and reduced access to traditional employment protections. It is enabled by platform technology that matches supply and demand at scale and managed through algorithmic coordination, raising ongoing questions about worker classification and social protection.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gig-economy", "labels": [ "Gig Economy" ], "is_subclass_of": [ "Platform Economy" ], "wikilinks": [] }, { "id": "gini-coefficient", "title": "gini coefficient", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Gini coefficient is a scalar summary statistic derived from the Lorenz curve that measures the degree of inequality in a distribution, yielding 0 for perfect equality and 1 for maximum concentration in a single entity. Originally developed by statistician Corrado Gini to measure income and wealth inequality, it has been applied in blockchain analytics to quantify the concentration of token holdings, staking power, or validator stake across on-chain addresses. High Gini values in a proof-of-stake network signal plutocratic tendencies that may undermine decentralisation claims.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:gini-coefficient", "labels": [ "Gini Coefficient", "Gini Coefficient Analysis" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "git-hub-actions", "title": "Git Hub Actions", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A CI/CD automation platform integrated into GitHub that executes workflow pipelines triggered by repository events, schedules, or manual dispatch. GitHub Actions enables ML model training, data versioning with DVC, experiment tracking with MLflow, fairness testing, Docker container builds, and scheduled retraining, making it the standard orchestration layer for MLOps pipelines on GitHub-hosted or self-hosted runners.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:git-hub-actions", "labels": [ "Git Hub Actions", "GitHub Actions" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "git-mark", "title": "Git Mark", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Git Mark is a Block Trails Profile that anchors a Git history to Bitcoin by using each commit hash as the tweak that advances a trail, so the sequence of commits becomes a single-use-seal chain whose ordering and uniqueness are enforced by Bitcoin's UTXO model. The Blocktrails verifier checks a git-mark trail against the chain, confirming that each marked commit was timestamped and is tamper-evident on Bitcoin \u2014 it proves the history's immutability and temporal anchoring, not the correctness of the code itself. This gives a Bitcoin-secured provenance log for source repositories without storing any repository data on-chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:git-mark", "labels": [ "Git Mark", "Git-mark", "Gitmark", "git-mark" ], "is_subclass_of": [ "Block Trails" ], "wikilinks": [] }, { "id": "git", "title": "Git", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Git is a distributed version control system that tracks changes to files and coordinates work across multiple contributors. It was created by Linus Torvalds for Linux kernel development and has become the dominant source-control system in modern software engineering.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:git", "labels": [ "Git" ], "is_subclass_of": [ "Version Control" ], "wikilinks": [ "GitHub", "Software Development", "Open Source", "Software Engineering", "Version Control", "https://git-scm.com", "https://git-scm.com/book" ] }, { "id": "git-hub", "title": "GitHub", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "GitHub is a web-based platform for hosting Git repositories that provides version control, code review, issue tracking, project management, and collaborative software development features. Acquired by Microsoft in 2018, it serves as the primary distribution hub for open-source software and integrates CI/CD via GitHub Actions, package registries, and an extensible API and webhook ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:git-hub", "labels": [ "GitHub", "GitHub Integration" ], "is_subclass_of": [ "Version Control" ], "wikilinks": [ "Git", "Open Source", "Software Development", "Software Engineering", "Version Control", "https://github.com", "https://docs.github.com" ] }, { "id": "gitops", "title": "GitOps", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "GitOps is an operational model in which the desired state of infrastructure and applications is declared in version-controlled repositories and continuously reconciled into running systems by automated agents. Git becomes the single source of truth, so changes flow through pull requests and merges while reconcilers detect and correct drift. It applies software-delivery practices, review, audit and rollback, to operations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gitops", "labels": [ "GitOps" ], "is_subclass_of": [ "Infrastructure as Code" ], "wikilinks": [] }, { "id": "gitcoin-grants", "title": "Gitcoin Grants", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gitcoin Grants is a decentralised crowdfunding programme for open-source and public goods projects in the Web3 ecosystem, which implements quadratic funding as its core matching mechanism to amplify contributions from a broad community of small donors relative to a smaller number of large funders. In quadratic funding, the matching pool allocation for each project is proportional to the square of the sum of the square roots of individual contributions, incentivising broad participation over concentrated giving and making grant outcomes resistant to plutocratic capture. Gitcoin Grants operates in periodic rounds with a central matching pool funded by protocols, DAOs, and foundations, and uses Gitcoin Passport as a Sybil resistance layer to prevent fake identity manipulation of the quadratic formula.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gitcoin-grants", "labels": [ "Gitcoin Grants" ], "is_subclass_of": [ "Public Goods Funding" ], "wikilinks": [] }, { "id": "gitcoin-passport", "title": "Gitcoin Passport", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gitcoin Passport is a decentralised identity aggregation and sybil-resistance protocol that collects verifiable credential stamps from diverse identity providers\u2014including Web2 social platforms, biometric services, on-chain activity records, and professional attestation networks\u2014into a composable trust score attesting to the humanness and uniqueness of a wallet address. The system is built on W3C Decentralised Identifier and Verifiable Credential standards, storing credentials on the Ceramic Network to maintain user sovereignty over personal data without centralised custody. Its primary use case is protecting quadratic funding rounds and other public-goods allocation mechanisms from sybil attacks, where a single actor creates many accounts to multiply their influence. The stamp-based, configurable-weight architecture allows applications to set their own scoring thresholds according to their specific risk tolerance, and the open API enables Gitcoin Passport to function as a composable trust primitive across the broader Web3 ecosystem.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:gitcoin-passport", "labels": [ "Gitcoin Passport" ], "is_subclass_of": [ "Decentralized Identity (DID)" ], "wikilinks": [] }, { "id": "gitcoin", "title": "Gitcoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gitcoin is a blockchain-based public goods funding platform founded in 2017 that finances open-source software development through decentralised grant programmes, bounties, and hackathons, most notably employing quadratic funding to allocate matching pools democratically. The platform uses the Ethereum ecosystem as its primary coordination layer, with GTC governance tokens enabling decentralised community control via its DAO. Gitcoin Passport, a credential aggregation system layering Verifiable Credentials from multiple identity providers, addresses Sybil resistance within open grant rounds and serves as broader Web3 identity infrastructure. Together these mechanisms position Gitcoin as a reference implementation of public goods coordination, bridging open-source software sustainability with decentralised finance and digital identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gitcoin", "labels": [ "Gitcoin", "Gitcoin DAO" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "glacial-region", "title": "Glacial Region", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:glacial-region", "labels": [ "Glacial Region" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "glacier-monitoring", "title": "Glacier Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:glacier-monitoring", "labels": [ "Glacier Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "glacier-terminus", "title": "Glacier Terminus", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:glacier-terminus", "labels": [ "Glacier Terminus", "Snout" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "glacier", "title": "Glacier", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:glacier", "labels": [ "Glacier" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "glen-weyl", "title": "Glen Weyl", "domain": "finance", "domain_name": "Finance", "definition": "Glen Weyl is an economist known for work on mechanism design, quadratic voting and quadratic funding, and the ideas in the book Radical Markets. He founded the RadicalxChange movement.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:glen-weyl", "labels": [ "Glen Weyl" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [ "Quadratic Funding", "Economics", "Mechanism Design" ] }, { "id": "global-catastrophic-risk", "title": "Global Catastrophic Risk", "domain": "governance", "domain_name": "Governance", "definition": "Global Catastrophic Risk is a category of risk capable of inflicting severe, worldwide damage to human civilisation, potentially including mass casualties, societal collapse, or permanent loss of capability, without necessarily causing human extinction. It is distinguished from existential risk, which specifically threatens humanity's long-term potential or survival, though the two categories overlap substantially. Advanced AI, engineered pandemics, and nuclear conflict are commonly cited sources of global catastrophic risk.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:global-catastrophic-risk", "labels": [ "Global Catastrophic Risk" ], "is_subclass_of": [ "Existential Risk" ], "wikilinks": [] }, { "id": "global-digital-finance", "title": "Global Digital Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Global Digital Finance (GDF) is an industry membership association and self-regulatory organisation founded to develop voluntary codes of conduct, taxonomies, and governance frameworks for the digital asset and cryptocurrency sector. It convenes cross-sector working groups that produce best-practice guidance on topics such as crypto asset taxonomy, market conduct, custody, and consumer protection, while engaging with regulators, standards bodies, and central banks to inform policy. GDF acts as a liaison body between industry participants and public authorities including the Financial Stability Board, IOSCO, and national financial regulators, promoting responsible adoption of blockchain-based financial instruments. Its outputs serve as voluntary soft-law standards intended to accelerate convergence toward enforceable regulatory frameworks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:global-digital-finance", "labels": [ "Global Digital Finance" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "Digital Asset", "Cryptocurrency", "Standards Body" ] }, { "id": "global-explanation", "title": "Global Explanation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Interpretability techniques that characterise the overall behaviour, decision-making patterns, and feature importance of a machine learning model across its entire input space, rather than explaining individual predictions. Global explanations\u2014such as feature importance rankings, partial dependence plots, and surrogate model trees\u2014reveal systematic model tendencies and support auditing, debugging, and regulatory compliance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:global-explanation", "labels": [ "Global Explanation", "Global Model Explanation" ], "is_subclass_of": [ "AI Technique", "Explainability" ], "wikilinks": [ "caballero2008financial; @spiro2019hidden", "carney2019growing; @piffaretti2009reshaping", "grewal2020struggling", "Individual Conditional Expectation", "Partial Dependence Plot", "Permutation Importance", "SHAP", "Srinivasan2022", "stoeferle2018gold", "Surrogate Models", "tomlinson2003third", "Agents", "Explainable AI", "Feature Importance", "Local Explanation", "MetaverseDomain", "Model Interpretability", "Model Transparency" ] }, { "id": "global-illumination", "title": "Global Illumination", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Global Illumination (GI) is a rendering approach that simulates all light interactions within a scene, including both direct illumination from light sources and indirect illumination from light bouncing between surfaces. Techniques range from offline radiosity and photon mapping to real-time approximations such as voxel cone tracing, screen-space ambient occlusion, and hardware-accelerated ray tracing, producing physically plausible colour bleeding, soft shadows, and caustics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:global-illumination", "labels": [ "Global Illumination", "Realistic Illumination" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "Light Probe", "Compute Shader", "Metaverse", "Physically-Based Rendering", "Rasterization", "Ray Tracing" ] }, { "id": "global-inequality", "title": "Global Inequality", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Global Inequality is the systematic, multi-dimensional divergence in life chances, material resources, political power, and technological access experienced by individuals, households, communities, and nations across the world economic order \u2014 encompassing income inequality (the Gini coefficient ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:global-inequality", "labels": [ "Global Inequality" ], "is_subclass_of": [ "AI Governance and Ethics", "Distributive Justice", "Political Economy", "Development Economics", "Social Stratification", "Economic Inequality" ], "wikilinks": [ "AI Labour Displacement", "Atkinson Index", "Austerity", "booth2020price", "Climate Change", "Climate Justice", "Compute Divide", "Data Science", "Development Economics", "Development Finance", "Digital Divide", "Digital Infrastructure Investment", "DigitalSocietyDomain", "Distributional Data", "Distributive Justice", "Economic Inequality", "EconomicsDomain", "Efficient Market Hypothesis", "EmpiricalResearchLayer", "EthicsAndSocietyDomain" ] }, { "id": "global-localisation", "title": "Global Localisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Global localisation is the problem of estimating a robot's pose within a known map without any prior knowledge of its starting position, often called the kidnapped-robot problem. Unlike pose tracking, it must resolve ambiguity across the entire map, typically by maintaining and refining multiple pose hypotheses from sensor observations. It is fundamental to robot navigation, autonomous vehicles, and recovery from localisation failures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:global-localisation", "labels": [ "Global Localisation" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "global-metaverse-operations", "title": "Global Metaverse Operations", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The international coordination and management of metaverse platforms across jurisdictions, encompassing cross-border infrastructure deployment, regulatory compliance, interoperability standards, and unified user experiences that enable seamless virtual world access regardless of geographic location.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:global-metaverse-operations", "labels": [ "Global Metaverse Operations" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Infrastructure" ], "wikilinks": [ "Unified Virtual Experience", "metaverse", "Metaverse Infrastructure" ] }, { "id": "global-navigation-satellite-system", "title": "Global Navigation Satellite System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:global-navigation-satellite-system", "labels": [ "Global Navigation Satellite System" ], "is_subclass_of": [ "Navigation System" ], "wikilinks": [] }, { "id": "global-trade", "title": "Global Trade", "domain": "finance", "domain_name": "Finance", "definition": "Global trade is the world-wide system of exchange of goods, services, and intangibles across national borders, viewed as an aggregate: the sum of all international trade flows together with the networks of shipping, finance, standards, and agreements that carry them. Worth roughly one third of world GDP, it is organised increasingly through global value chains in which production stages are distributed across many countries, and it depends on shared technical standards, interoperable logistics, and multilateral rules to keep transaction costs low.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:global-trade", "labels": [ "Global Trade" ], "is_subclass_of": [ "International Trade" ], "wikilinks": [ "International Trade", "Supply Chain", "Logistics", "ISO", "Economic Growth" ] }, { "id": "glossary-index", "title": "Glossary Index", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A centralized terminology reference system that aggregates, defines, and cross-references all metaverse concepts with their synonyms, abbreviations, and semantic relationships, serving as the human-readable interface to the formal ontology schema.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:glossary-index", "labels": [ "Glossary Index" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Consistency Checking", "Cross References", "Documentation System", "Dublin Core", "ISO 25964 Thesaurus Standard", "Knowledge Management Infrastructure", "Multi-Language Support", "Semantic Relations", "Synonym Mappings", "Term Definitions", "Terminology Lookup", "Thesaurus Structure", "Usage Examples", "ApplicationLayer", "Category Hierarchies", "Controlled Vocabulary", "InfrastructureDomain", "Learning Resources", "Metaverse Ontology Schema", "Semantic Search" ] }, { "id": "gltf-standard", "title": "Gltf Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Graphics Library Transmission Format (glTF) is a royalty-free, open interoperable 3D asset format developed by Khronos Group, designed for efficient runtime transmission and loading of 3D scenes across native and web-based engines, serving as a foundational standard for metaverse asset interoperability. Standardised as ISO/IEC 12113:2022, glTF 2.0 supports geometry, PBR materials, skeletal animation, morph targets, and a modular extension system enabling progressive capability extension.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:gltf-standard", "labels": [ "Gltf Standard", "glTF", "glTF Specification", "glTF/GLB" ], "is_subclass_of": [ "Standards and Interoperability", "3D Asset Standard" ], "wikilinks": [ "3D Asset Standard", "Cross-Platform 3D Content", "metaverse" ] }, { "id": "gnosis-chain", "title": "Gnosis Chain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum-compatible blockchain secured by proof of stake, designed for low-cost transactions and used for payments, governance tooling, and application deployment.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gnosis-chain", "labels": [ "Gnosis Chain" ], "is_subclass_of": [ "Blockchain", "Sidechain" ], "wikilinks": [ "EVM", "Proof of Stake", "Gnosis Safe", "Smart Contract", "Blockchain" ] }, { "id": "gnosis-safe", "title": "Gnosis Safe", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gnosis Safe, later rebranded as Safe, is a smart-contract wallet for Ethereum and compatible networks that requires multiple signatures to authorise transactions. Rather than relying on a single private key, it enforces a configurable threshold, such as three of five owners, before funds move or contract calls execute. It is widely used by decentralised autonomous organisations, projects and individuals to manage treasuries and reduce the risk of a single compromised key.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gnosis-safe", "labels": [ "Gnosis Safe", "Safe Multi-Sig", "Safe Multisig", "Safe Smart Account" ], "is_subclass_of": [ "Network Component", "Blockchain Domain" ], "wikilinks": [ "Smart Contract", "Ethereum", "Multisignature Wallet", "Treasury Management", "Snapshot", "Decentralised Autonomous Organisation", "Blockchain Domain" ] }, { "id": "gnosis", "title": "Gnosis", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum-based platform and chain known for prediction markets and the widely used Safe multisignature smart contract wallet.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gnosis", "labels": [ "Gnosis" ], "is_subclass_of": [ "Ethereum Smart Contract Platform" ], "wikilinks": [ "Smart Contract", "Multisignature", "Ethereum Smart Contracts", "Ethereum" ] }, { "id": "gnss", "title": "Gnss", "domain": "robotics", "domain_name": "Robotics", "definition": "A Global Navigation Satellite System (GNSS) is a constellation of satellites that broadcast timed signals enabling receivers to compute their absolute position, velocity and time anywhere on Earth. GPS, Galileo, GLONASS and BeiDou are the principal systems, with receivers trilaterating position from signal travel times across multiple satellites. In robotics, GNSS provides global geo-referenced localisation that anchors local sensor-based estimates to an absolute coordinate frame.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:gnss", "labels": [ "Gnss", "GNSS" ], "is_subclass_of": [ "Localization" ], "wikilinks": [] }, { "id": "go-quorum", "title": "GoQuorum", "domain": "blockchain", "domain_name": "Blockchain", "definition": "GoQuorum is an open-source, enterprise-focused Ethereum client originally developed by J.P. Morgan and forked from go-ethereum to support permissioned consortium networks. It adds private transactions and contracts, permissioning, and pluggable consensus algorithms such as IBFT, QBFT, and Raft suited to known-validator settings. It is widely used to build private and consortium blockchains for financial and supply-chain applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:go-quorum", "labels": [ "GoQuorum" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "goal-configuration", "title": "Goal Configuration", "domain": "robotics", "domain_name": "Robotics", "definition": "A goal configuration is the desired target state of a robot or articulated system, expressed in its configuration space as a set of joint angles or a pose. Motion planners search for a collision-free path from the start configuration to this goal. Specifying it precisely is a prerequisite for path planning and trajectory generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:goal-configuration", "labels": [ "Goal Configuration" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "goal-specification", "title": "Goal Specification", "domain": "ai", "domain_name": "Ai", "definition": "Goal specification is the formal description of what an autonomous agent or planner is meant to achieve, expressed as target states, conditions, objectives, or reward functions. It translates high-level intent into a representation that planning and reasoning systems can evaluate and pursue. Clear goal specification is essential for task planning and avoiding misaligned or unsafe agent behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:goal-specification", "labels": [ "Goal Specification" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "goal", "title": "Goal", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A desired future state that an agent aims to realise through planned action sequences. Goals are declarative, future-oriented, and action-guiding; they encompass achievement, maintenance, optimisation, and avoidance types, and are managed in hierarchies that decompose complex objectives into sub-goals. In AI systems, goal specification is central to alignment: misspecified goals produce unintended consequences regardless of capability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:goal", "labels": [ "Goal" ], "is_subclass_of": [ "AI Technique", "Objective", "Artificial Intelligence" ], "wikilinks": [ "Action", "Agent Concept", "Plan", "State", "Task", "Agent", "Artificial Intelligence", "Autonomy Level", "BDI Model", "Objective", "Value Alignment" ] }, { "id": "gold-standard", "title": "Gold Standard", "domain": "finance", "domain_name": "Finance", "definition": "A monetary system in which a currency's value is legally fixed to and redeemable in a defined quantity of gold, constraining the total money supply to the size of gold reserves held by the issuing authority. Under this regime, exchange rates between participating countries are effectively fixed because each currency is independently pegged to gold at a known rate, enabling automatic balance-of-payments adjustment through the price-specie-flow mechanism. The classical gold standard operated broadly from the 1870s until the outbreak of World War I in 1914, was briefly restored in modified forms during the interwar period, and gave way to the Bretton Woods system of dollar-gold convertibility in 1944, which itself ended in 1971 when the United States suspended gold convertibility.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:gold-standard", "labels": [ "Gold Standard", "Gold Standard Foundation" ], "is_subclass_of": [ "Monetary System" ], "wikilinks": [ "Monetary Policy", "Money", "Central Bank", "Inflation" ] }, { "id": "gold", "title": "Gold", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Gold (chemical symbol Au, atomic number 79, ISO currency code XAU) is a dense, corrosion-resistant transition metal that has served as the pre-eminent monetary commodity for at least years and \u20132026 occupies a unique dual role as both the world's third-largest reserve asset by value (approximatel...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:gold", "labels": [ "Gold" ], "is_subclass_of": [ "Token and Asset", "Hard Money", "Digital Asset", "Commodity", "Reserve Asset", "Tangible Asset", "Monetary Metal" ], "wikilinks": [ "Algorand", "Allocated Account", "Assay Certificate", "BIS Basel III CRE20", "BIS Basel III HQLA Classification", "Central Bank Gold Reserve", "Chain of Custody", "COMEX Contract Specifications", "COMEX Futures Contract Standard", "COMEX Gold Futures", "Commodity", "CommodityMarketDomain", "Cross-Border Settlement", "Currency Diversification", "Custodian Bank", "DigitalAssetDomain", "EconomicsDomain", "ERC-20 Token Standard", "FATF Precious Metals Guidance", "Fiat Currency" ] }, { "id": "golden-set", "title": "Golden Set", "domain": "data", "domain_name": "Data", "definition": "A small, carefully curated and human-verified collection of input-output examples that serves as the authoritative reference against which a model, prompt, agent, or pipeline is repeatedly evaluated. Each item pairs a representative input with an expected or ideal output, and the set is deliberately kept stable across versions so that regression can be measured, quality can be tracked over time, and changes can be accepted or rejected on the basis of a consistent benchmark rather than anecdote.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:golden-set", "labels": [ "Golden Set" ], "is_subclass_of": [ "Test Dataset", "TestDataset" ], "wikilinks": [ "TestDataset", "GroundTruth", "EvaluationHarness", "LLMEvaluation" ] }, { "id": "google-ai-technology-corporation", "title": "Google AI Technology Corporation", "domain": "ai", "domain_name": "Ai", "definition": "Google AI Technology Corporation (operating publicly as Google LLC, a subsidiary of Alphabet Inc.) is a multinational technology company whose research and product divisions constitute one of the world's leading forces in artificial intelligence, encompassing large language models, foundation models, and applied machine-learning infrastructure. The company develops and deploys AI across Search, Ads, Cloud, Android, and dedicated AI-native products such as Gemini, while its research arm \u2014 Google DeepMind \u2014 pursues frontier science in reinforcement learning, protein structure prediction, and multi-modal reasoning. Google's open and proprietary frameworks, including TensorFlow and JAX, underpin a significant portion of the global AI research ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:google-ai-technology-corporation", "labels": [ "Google AI Technology Corporation", "Google", "Google LLC" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Machine Learning", "Cloud Computing", "Generative AI", "Artificial Intelligence" ] }, { "id": "google-cloud", "title": "Google Cloud", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Google Cloud (formally Google Cloud Platform, GCP) is a suite of public cloud computing services operated by Google LLC, providing infrastructure-as-a-service (IaaS), platform-as-a-service (PaaS), and software-as-a-service (SaaS) offerings spanning compute, storage, networking, data analytics, and artificial intelligence. It is built on the same global infrastructure that powers Google Search, YouTube, and Google Workspace, spanning a worldwide network of data centres connected by Google's private fibre backbone. Google Cloud competes directly with Amazon Web Services and Microsoft Azure as one of the three dominant hyperscale cloud providers, and differentiates through deep integration of AI/ML capabilities via Vertex AI, TPU accelerator hardware, and pre-trained foundation models such as Gemini.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:google-ai-technology-corporation-cloud", "labels": [ "Google Cloud", "Google Cloud Platform", "Google Cloud Translation API" ], "is_subclass_of": [ "Cloud Platform" ], "wikilinks": [ "Cloud Computing", "Machine Learning", "Google", "Cloud Platform" ] }, { "id": "google-deep-mind", "title": "Google DeepMind", "domain": "ai", "domain_name": "Ai", "definition": "Google DeepMind is an AI research and development division of Alphabet Inc., formed in April 2023 by the merger of Google Brain (founded 2011) and DeepMind (founded 2010, acquired by Google in 2014). It is responsible for foundational breakthroughs in reinforcement learning, protein structure prediction (AlphaFold), and large-scale multimodal AI (Gemini), and pursues both long-term fundamental research and product integration across Google's services. Operating from London, Mountain View, and additional global sites, it is one of the largest and most influential AI research institutions in the world, with a stated mission of advancing artificial intelligence for the benefit of humanity whilst maintaining safety-centred development practices.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:google-ai-technology-corporation-deep-mind", "labels": [ "Google DeepMind" ], "is_subclass_of": [ "AI Companies" ], "wikilinks": [] }, { "id": "google-gemini", "title": "Google Gemini", "domain": "ai", "domain_name": "Ai", "definition": "Google Gemini is a family of natively multimodal large language models developed by Google DeepMind, capable of reasoning across text, images, audio, video, and code. Released in tiers (such as Ultra, Pro, Flash, and Nano) it spans data-centre to on-device deployment and powers Google's assistant and developer APIs. It is a leading frontier model used for chat assistants, agents, and integrated productivity tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:google-ai-technology-corporation-gemini", "labels": [ "Google Gemini", "Gemini Model Family", "Google Gemini Function Declarations" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "gossip-protocol", "title": "Gossip Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A peer-to-peer communication protocol in which nodes periodically exchange state information with randomly selected neighbours, enabling eventual consistency and fault-tolerant information dissemination across large-scale distributed systems without centralised coordination.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gossip-protocol", "labels": [ "Gossip Protocol", "Gossip Consensus", "GossipSub", "P2P Gossip Protocol", "Universe Gossip Protocol" ], "is_subclass_of": [ "Network and Communication", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "governance-architecture", "title": "Governance Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The structural framework defining decision-making processes, authority distribution, and rule enforcement mechanisms within metaverse platforms, encompassing both centralised corporate governance models and decentralised blockchain-based systems using DAOs and smart contracts for community-driven...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:governance-architecture", "labels": [ "Governance Architecture", "G20 Governance Architecture" ], "is_subclass_of": [ "AI Governance and Ethics", "Platform Governance" ], "wikilinks": [ "Decentralised Decision Making", "metaverse", "Platform Governance" ] }, { "id": "governance-attack", "title": "Governance Attack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A governance attack is an exploit in which an adversary acquires or temporarily controls sufficient voting power within a decentralised governance system to pass malicious proposals against the interests of the wider community. Such attacks often combine economic mechanisms, such as borrowing governance tokens via flash loans, with the on-chain execution semantics of decentralised autonomous organisations. The aim is typically to drain a treasury, alter protocol parameters, or seize privileged contract roles.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:governance-attack", "labels": [ "Governance Attack" ], "is_subclass_of": [ "Voting Mechanism" ], "wikilinks": [] }, { "id": "governance-board", "title": "Governance Board", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An oversight body responsible for strategic decision-making and policy enforcement within metaverse organisations, evolving from traditional corporate board structures to decentralised autonomous organisation (DAO) models where token holders collectively govern through transparent on-chain voting...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:governance-board", "labels": [ "Governance Board" ], "is_subclass_of": [ "AI Governance and Ethics", "Governance Architecture" ], "wikilinks": [ "Collective Decision Making", "Governance Architecture", "metaverse" ] }, { "id": "governance-framework", "title": "Governance Framework", "domain": "governance", "domain_name": "Governance", "definition": "A governance framework is a structured system of policies, processes, procedures, and controls that organisations use to align their technology resources and operations with business objectives, providing the foundation for strategic decision-making, risk management, resource optimisation, performance measurement, and compliance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:governance-framework", "labels": [ "Governance Framework", "DeFi Governance Framework", "Ethical Governance Framework", "Governance Framework (AI-0035)", "Legal Governance Framework" ], "is_subclass_of": [ "Legal and Regulatory", "Organisational Framework" ], "wikilinks": [ "Core Technology", "Decision Rights", "Organisational Framework", "Performance Management", "Strategic Alignment", "Accountability", "Policy Enforcement", "Risk Management" ] }, { "id": "governance-frameworks", "title": "Governance Frameworks", "domain": "governance", "domain_name": "Governance", "definition": "Governance frameworks are structured sets of policies, roles, and processes that define how decisions are made and oversight is exercised within an organisation or system.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:governance-frameworks", "labels": [ "Governance Frameworks" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "AI Governance", "Governance" ] }, { "id": "governance-infrastructure", "title": "Governance Infrastructure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Governance infrastructure is the set of tools, smart contracts, and systems that enable decentralised organisations to propose, deliberate, vote on, and execute collective decisions. It encompasses voting modules, treasury management, delegation, and on-chain execution that turn community decisions into enforced actions. It is the operational backbone of DAOs and on-chain governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:governance-infrastructure", "labels": [ "Governance Infrastructure" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "governance-layer", "title": "Governance Layer", "domain": "governance", "domain_name": "Governance", "definition": "The Governance Layer is the cross-cutting stratum where human and institutional intent over a system is decided, recorded, and amended. It sits above the Policy Layer, which it parameterises, and draws on identity and economic structures to allocate decision rights. It contains decision processes, voting mechanisms, charters, and the records of who may change what.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:governance-layer", "labels": [ "Governance Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Policy Layer", "Identity Layer", "Institutional Layer", "Compliance Layer", "Decentralised Governance", "Mechanism Design", "owl:Thing" ] }, { "id": "governance-model", "title": "Governance Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Framework of rules and decision-making processes defining authority and accountability within a metaverse ecosystem.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:governance-model", "labels": [ "Governance Model", "Non-Profit Governance Model", "Tiered Governance Model" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance Framework" ], "wikilinks": [ "Accountability Mechanism", "Decision Structure", "ETSI GR ARF 010", "MSF", "Policy Framework", "Self-Regulation", "Stakeholder Agreement", "Access Control", "AI Governance Framework", "Community Governance", "Decentralized Governance", "Ethical Framework", "Identity Management", "Legal Framework", "Metaverse Architecture", "MiddlewareLayer", "Platform Governance", "Regulatory Compliance", "TrustAndGovernanceDomain" ] }, { "id": "governance-proposal", "title": "Governance Proposal", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A governance proposal is a formally submitted, votable item that requests a change to a decentralised protocol, treasury allocation or organisational parameter, typically within a DAO. It encapsulates a description, an executable payload or off-chain intent, and voting parameters such as quorum and threshold. Token holders or delegates vote on the proposal, and on approval it may be executed automatically by smart contracts.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:governance-proposal", "labels": [ "Governance Proposal" ], "is_subclass_of": [ "On-Chain Governance" ], "wikilinks": [] }, { "id": "governance-risk-compliance", "title": "Governance Risk Compliance", "domain": "governance", "domain_name": "Governance", "definition": "An integrated organisational discipline that aligns corporate governance, enterprise risk management, and regulatory compliance into one coordinated capability, so that objectives are set and overseen reliably, uncertainty is addressed within appetite, and the organisation acts with integrity within legal and regulatory boundaries. Formalised by OCEG's 'principled performance' model, GRC replaces siloed oversight functions with shared taxonomies of risks, controls, policies, and obligations, typically operationalised through dedicated platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:governance-risk-compliance", "labels": [ "Governance Risk Compliance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance", "Risk Management", "Regulatory Compliance", "Compliance Monitoring", "Grc Platform", "Operational Risk" ] }, { "id": "governance-structure", "title": "Governance Structure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A governance structure is the formal arrangement of authority, accountability, decision-making rights, and control mechanisms within an organisation, system, protocol, or jurisdiction that determines how strategic direction is set, resources are allocated, obligations are enforced, and stakeholde...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:governance-structure", "labels": [ "Governance Structure" ], "is_subclass_of": [ "Legal and Regulatory", "Governance Framework", "Compliance Framework", "Institutional Design", "Accountability Mechanism", "Trust Framework" ], "wikilinks": [ "Accountability Mechanism", "Anarchic Coordination", "Audit Committee", "Board of Directors", "COBIT Framework", "Companies Act", "CorporateGovernanceDomain", "DistributedCollaborationDomain", "Enforcement Mechanism", "Informal Norm", "Institutional Design", "InstitutionalLayer", "ISO 42001", "ITIL Framework", "NIST AI RMF", "OECD Corporate Governance Principles", "OECD Guidelines", "Principal Agent Problem", "RegulatoryDomain", "Self-Regulation" ] }, { "id": "governance-system", "title": "Governance System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A governance system is the structured set of rules, roles, and processes by which a blockchain protocol or organisation makes collective decisions about its parameters, upgrades, and resource allocation. It defines who can propose changes, how they are ratified, and how outcomes are enforced and recorded. In blockchains it is integral to protocol evolution and to the auditability of decisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:governance-system", "labels": [ "Governance System" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "governance-token", "title": "Governance Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Governance Token is a class of ERC-20 (or equivalent chain-native) fungible cryptographic asset that confers programmable voting rights and protocol decision-making authority over a Decentralised Autonomous Organisation or smart-contract protocol, unbundling political control (one-token-one-v...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:governance-token", "labels": [ "Governance Token", "BC-0463-governance-token", "GovernanceToken", "Single Token Governance", "UNI Governance Token" ], "is_subclass_of": [ "Token and Asset", "Fungible Token", "ERC-20 Token", "Digital Asset", "Crypto-Asset", "Coordination Primitive", "Voting Instrument" ], "wikilinks": [ "1Hive", "1inch", "a16z", "Aave", "Aave", "Aerodrome Finance", "Aragon", "Aragon Association", "Arbitrum", "Arbitrum DAO", "Arbitrum Foundation", "Argent", "Aura Finance", "Axelar", "Aztec Network", "Balancer", "BanklessDAO", "Basel Committee 2022 Cryptoasset Prudential Treatment", "Basel Committee Crypto Asset Standards", "BAT" ] }, { "id": "governance-voting", "title": "Governance Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Governance voting is the mechanism by which holders of governance tokens or members of a decentralised organisation cast votes on proposals to direct a protocol's decisions. Voting power is typically proportional to token holdings or delegated stake, and outcomes can execute automatically on-chain. It is the primary means by which token holders exercise collective control over protocol changes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:governance-voting", "labels": [ "Governance Voting" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "governance", "title": "Governance", "domain": "governance", "domain_name": "Governance", "definition": "Governance is the set of processes, institutions, norms, rules, and practices by which authority is exercised and decisions are made within an organisation, system, or society. It encompasses the mechanisms through which stakeholders participate in, constrain, and are held accountable by decision-making structures, balancing competing interests through legitimate frameworks. In technical and sociotechnical contexts, governance determines who controls shared resources, how policies are formed and enforced, and how disputes are resolved. Effective governance integrates regulatory compliance, ethical accountability, risk management, and participatory design to achieve legitimate and adaptive coordination.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:governance", "labels": [ "Governance", "Bicameral Governance", "Compound Governance", "Effective governance", "Governance Institution", "Governance Module", "Governance Process", "Plutocratic Governance" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "owl:Thing" ] }, { "id": "government-ai-modernization", "title": "Government AI Modernization", "domain": "governance", "domain_name": "Governance", "definition": "The strategic initiative by public sector entities to adopt, integrate, and scale artificial intelligence technologies to enhance operational efficiency, service delivery, and infrastructure management.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:government-ai-modernization", "labels": [ "Government AI Modernization" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "government-digital-identity", "title": "Government Digital Identity", "domain": "security", "domain_name": "Security", "definition": "Government digital identity is a state-issued or state-recognised electronic identity that citizens and residents use to prove who they are when accessing public and private services online. It provides identity proofing and authentication backed by authoritative government records, often within a national trust framework. It increasingly draws on verifiable credentials and decentralised identity standards to give users portable, privacy-preserving credentials.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:government-digital-identity", "labels": [ "Government Digital Identity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "gps-navigation", "title": "Gps Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "GPS navigation is the use of the Global Positioning System \u2014 a satellite-based radio navigation system operated by the United States government \u2014 to determine the precise position, velocity, and time of a receiver anywhere on or near Earth. A GPS receiver calculates its location by measuring the time of arrival of signals from at least four satellites and applying trilateration. It is widely used in autonomous robots, vehicles, aircraft, and mobile devices as a primary or complementary localisation sensor.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:gps-navigation", "labels": [ "Gps Navigation", "GPS Navigation" ], "is_subclass_of": [ "Robotics", "Navigation System" ], "wikilinks": [] }, { "id": "gps", "title": "Gps", "domain": "robotics", "domain_name": "Robotics", "definition": "GPS (Global Positioning System) - A satellite-based Navigation System that determines the absolute geographical location of a robot, providing latitude, longitude, and altitude data for large-scale autonomous navigation, outdoor delivery, and trajectory planning with typical accuracy of 5-15 ...", "entityType": "Class", "qualityScore": 0.51, "maturity": "draft", "iri": "urn:ngm:class:gps", "labels": [ "Gps", "GPS", "GPS Positioning", "GPS Reference" ], "is_subclass_of": [ "Navigation and Planning", "Navigation System", "Robotics" ], "wikilinks": [ "Antenna Array", "Geofencing", "Global Localisation", "Long-range Navigation", "Map Integration", "Mission Planning", "Multi-robot Coordination", "Navigation System", "Robotics", "RoboticsDomain", "Signal Processing", "Spatial Computing" ] }, { "id": "gpu-driven-rendering", "title": "Gpu Driven Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "GPU-driven rendering is an architecture in which the GPU itself determines what to draw and issues its own draw commands, rather than relying on the CPU to traverse the scene and submit each object individually. Using compute shaders for culling and indirect, multi-draw commands, it minimises CPU overhead and draw-call cost, enabling scenes with very large object counts. It builds on compute capability, indirect drawing, and modern graphics APIs to keep the GPU saturated and scalable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:gpu-driven-rendering", "labels": [ "Gpu Driven Rendering", "GPU-Driven Rendering" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [] }, { "id": "gpu-memory", "title": "Gpu Memory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "GPU memory is the high-bandwidth memory resident on or tightly coupled to a graphics processing unit that stores model weights, activations, gradients, and intermediate buffers during computation. Its capacity and bandwidth are frequently the binding constraint on the size of models that can be trained or served, motivating techniques such as quantisation, gradient checkpointing, and model parallelism. Efficient use of GPU memory directly determines achievable throughput and batch size.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gpu-memory", "labels": [ "Gpu Memory", "GPU Memory" ], "is_subclass_of": [ "GPU" ], "wikilinks": [] }, { "id": "gpu-programming", "title": "Gpu Programming", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "GPU programming is the practice of writing software that exploits the massively parallel architecture of graphics processing units to accelerate computation across thousands of concurrent threads. It encompasses both graphics pipelines, expressed through shaders, and general-purpose compute expressed through frameworks such as CUDA, OpenCL, and Vulkan compute. Effective GPU programming requires reasoning about memory hierarchies, thread divergence, and data parallelism to achieve high throughput on suitable workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gpu-programming", "labels": [ "Gpu Programming", "GPU Programming" ], "is_subclass_of": [ "High Performance Computing" ], "wikilinks": [] }, { "id": "graceful-degradation", "title": "Graceful Degradation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Graceful degradation is a design property whereby a system continues to provide reduced but acceptable functionality when some of its components fail or operate under stress, rather than failing completely. It prioritises essential services, sheds non-critical load, and offers fallback behaviours so that partial failure does not cascade into total outage. It is a cornerstone of resilient, fault-tolerant infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graceful-degradation", "labels": [ "Graceful Degradation" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [] }, { "id": "gradient-accumulation", "title": "Gradient Accumulation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Gradient accumulation is a training technique that sums the gradients computed over several consecutive mini-batches before performing a single parameter update, thereby simulating a larger effective batch size than fits in device memory. It allows training of large models on limited hardware by trading additional forward and backward passes for reduced peak memory usage. The optimiser step and gradient reset occur only after the configured number of accumulation steps has been reached.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gradient-accumulation", "labels": [ "Gradient Accumulation" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "gradient-aggregation", "title": "Gradient Aggregation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Gradient aggregation is the step in distributed machine learning where gradients computed independently on different workers or data shards are combined into a single update for the shared model. Typically realised by summing or averaging local gradients, it lets parallel workers train a consistent global model despite operating on disjoint data. The aggregation strategy and its communication pattern strongly influence training throughput, convergence and, in federated settings, privacy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gradient-aggregation", "labels": [ "Gradient Aggregation" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "gradient-boosted-trees", "title": "Gradient Boosted Trees", "domain": "ai", "domain_name": "Ai", "definition": "Gradient boosted trees are an ensemble learning method that builds a strong predictor by sequentially adding shallow decision trees, each fitted to the negative gradient of a differentiable loss with respect to the current model's predictions. By combining many weak learners in an additive, stage-wise manner, the method achieves high accuracy on structured and tabular data while controlling overfitting through regularisation, shrinkage and subsampling. It is among the most effective approaches for supervised regression and classification on heterogeneous features.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:gradient-boosted-trees", "labels": [ "Gradient Boosted Trees" ], "is_subclass_of": [ "Boosting" ], "wikilinks": [] }, { "id": "gradient-checkpointing", "title": "Gradient Checkpointing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Gradient checkpointing is a memory-efficient training technique for deep neural networks that reduces peak activation memory by storing only a strategically chosen subset of intermediate activations (checkpoints) during the forward pass, then recomputing the discarded activations on demand during backpropagation. The technique was formalised by Chen et al. (2016) under the name 'Training Deep Networks with Sublinear Memory Cost', achieving O(sqrt(N)) memory in the number of layers N at the cost of one additional forward pass per training step. It is now a foundational primitive in large-model training, implemented natively in PyTorch, JAX, TensorFlow, and most major deep-learning frameworks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:gradient-checkpointing", "labels": [ "Gradient Checkpointing", "Activation Checkpointing" ], "is_subclass_of": [ "Backpropagation" ], "wikilinks": [ "Backpropagation", "Deep Learning", "Gradient Descent" ] }, { "id": "gradient-clipping", "title": "Gradient Clipping", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A technique that limits the magnitude of gradients during backpropagation to prevent exploding gradients and training instability. Gradient clipping rescales gradients when their norm exceeds a threshold, enabling stable training of deep networks, especially recurrent architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:gradient-clipping", "labels": [ "Gradient Clipping" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "gradient-compression", "title": "Gradient Compression", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of communication-efficiency techniques for distributed and federated machine learning that reduce the volume of gradient data exchanged between workers during training, using quantisation to fewer bits, sparsification of small-magnitude entries, or low-rank decomposition, usually combined with error-feedback so the accumulated compression error is reapplied and convergence is preserved.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:gradient-compression", "labels": [ "Gradient Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [ "Distributed Training", "Gradient Aggregation", "Federated Learning" ] }, { "id": "gradient-descent", "title": "Gradient Descent", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Gradient Descent is an iterative first-order optimisation algorithm that minimises a differentiable loss function by repeatedly updating model parameters in the direction of the negative gradient. It is the foundational optimisation strategy for training machine learning models, with variants including batch, stochastic, and mini-batch gradient descent, as well as adaptive-rate methods such as Adam and RMSProp.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:gradient-descent", "labels": [ "Gradient Descent", "Gradient Descent Optimisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "gradient-synchronisation", "title": "Gradient Synchronisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Gradient Synchronisation is the process of aggregating and distributing gradient updates across multiple workers or devices during distributed training of a neural network, ensuring all replicas converge on a consistent set of model parameters. It typically uses all-reduce or parameter-server communication patterns, and is a major bottleneck in large-scale training due to network bandwidth constraints. Techniques such as gradient compression and asynchronous updates trade off consistency for throughput.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:gradient-synchronisation", "labels": [ "Gradient Synchronisation" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "gradient", "title": "Gradient", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A gradient is the vector of partial derivatives of a scalar-valued function with respect to each of its input variables, pointing in the direction of steepest ascent at a given point. In machine learning it is the quantity computed by backpropagation and consumed by optimisers to update model parameters. Its magnitude and direction drive first-order optimisation methods such as gradient descent and its momentum-based variants.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:gradient", "labels": [ "Gradient" ], "is_subclass_of": [ "Calculus" ], "wikilinks": [] }, { "id": "gradle", "title": "Gradle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Gradle is an open-source build-automation tool that uses a directed acyclic graph of tasks to compile, test, and package software, primarily for JVM languages such as Java, Kotlin, and Groovy. Its build scripts are written as a domain-specific language in Groovy or Kotlin, and it supports incremental builds, dependency management, and a build cache for speed. It is widely used in enterprise and Android development, including blockchain platforms built on the JVM.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gradle", "labels": [ "Gradle" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "grammar-system", "title": "Grammar System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Natural language processing components within metaverse platforms that interpret, process, and generate human language for user interactions, enabling AI-powered conversations with virtual avatars, real-time language translation, grammar correction, and adaptive dialogue systems that respond to u...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:grammar-system", "labels": [ "Grammar System" ], "is_subclass_of": [ "Content and Assets", "Natural Language Processing" ], "wikilinks": [ "Multilingual Metaverse Experience", "metaverse", "Natural Language Processing" ] }, { "id": "grant-programs", "title": "Grant Programs", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Grant Programs are structured resource allocation mechanisms through which DAOs and blockchain protocols deploy governance treasuries to ecosystem development, public goods, protocol research, and community initiatives. They range from prospective grants and quadratic funding (Gitcoin) to retroactive public goods funding (Optimism RetroPGF), and are administered by elected committees, independent foundations, or algorithmic allocation systems.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:grant-programs", "labels": [ "Grant Programs" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "1Hive", "Aave", "Allo Protocol", "Arbitrum", "Balancer", "BC-0410-decentralised-autonomous-organisations", "BC-0411-dao-governance-models", "BC-0462-treasury-management", "BC-0469-governance-tokens", "BC-0471-tokenomics-governance", "BC-0472-dao-tooling", "BC-0473-delegate-democracy", "BC-0475-dao-analytics", "Camelot", "Compound", "Ethereum Foundation", "EthereumJS", "Ethers.js", "Foundry", "Geth" ] }, { "id": "granular-consent-control", "title": "Granular Consent Control", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Privacy management mechanisms enabling users to selectively authorise specific types of data collection and processing within metaverse environments, allowing separate consent decisions for analytics, advertising, cross-device tracking, and data transfers while maintaining GDPR compliance and use...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:granular-consent-control", "labels": [ "Granular Consent Control" ], "is_subclass_of": [ "Governance and Safety", "Privacy Framework" ], "wikilinks": [ "User Data Autonomy", "Computer Vision", "metaverse", "Privacy Framework" ] }, { "id": "graph-algorithms", "title": "Graph Algorithms", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A family of computational methods that operate on graph structures\u2014nodes and edges\u2014to solve problems such as shortest-path finding, community detection, ranking, and traversal. Graph algorithms are foundational to knowledge graph querying, social network analysis, and AI reasoning over relational data.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:graph-algorithms", "labels": [ "Graph Algorithms", "Graph Traversal", "Graph Traversal Algorithm" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "owl:Thing" ] }, { "id": "graph-analytics", "title": "Graph Analytics", "domain": "ai", "domain_name": "Ai", "definition": "Graph Analytics is a computational discipline that applies graph-theoretic algorithms to structured relational data in order to discover patterns, measure importance, detect communities, and predict missing links. It operates on data modelled as nodes (entities) and edges (relationships), enabling analyses that are intractable with tabular data models when relationship structure is central to insight.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graph-analytics", "labels": [ "Graph Analytics" ], "is_subclass_of": [ "Network Analysis" ], "wikilinks": [] }, { "id": "graph-attention-network", "title": "Graph Attention Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A graph attention network is a graph neural network architecture that aggregates information from a node's neighbours using learned attention coefficients, allowing the model to weight each neighbour's contribution according to its relevance. By replacing fixed or degree-normalised aggregation with attention, it adapts to local structure without requiring knowledge of the full graph in advance. Multi-head attention stabilises learning and lets the model capture several relational patterns simultaneously.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graph-attention-network", "labels": [ "Graph Attention Network" ], "is_subclass_of": [ "Graph Neural Network" ], "wikilinks": [] }, { "id": "graph-classification", "title": "Graph Classification", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Graph classification is the machine learning task of assigning a label to an entire graph, such as a molecule, social network or program dependency graph, based on its structure and node or edge attributes, in contrast to node classification which labels individual vertices. It is typically performed by a graph neural network that repeatedly aggregates neighbourhood information through message passing before pooling node representations into a single graph-level embedding for classification. Graph classification is applied to problems such as molecular property prediction, protein function prediction and program analysis.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:graph-classification", "labels": [ "Graph Classification" ], "is_subclass_of": [ "Classification" ], "wikilinks": [] }, { "id": "graph-convolutional-network", "title": "Graph Convolutional Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A graph neural network architecture that generalises convolution to graph-structured data: each layer updates every node's feature vector by aggregating the degree-normalised features of its neighbours and transforming them with a shared learned weight matrix, so stacked layers propagate information across progressively larger neighbourhoods; formalised by Kipf and Welling in 2017 as a first-order approximation of spectral graph convolution, the GCN is the canonical baseline for node classification, link prediction, and graph-level learning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:graph-convolutional-network", "labels": [ "Graph Convolutional Network" ], "is_subclass_of": [ "Graph Neural Network" ], "wikilinks": [ "Graph Neural Network", "Graph Attention Network", "Neural Network" ] }, { "id": "graph-data-model", "title": "Graph Data Model", "domain": "data", "domain_name": "Data", "definition": "A graph data model represents information as nodes connected by edges, where edges carry the semantics of relationships between entities. It makes connections first-class, so traversing and querying relationships is direct rather than reconstructed through joins as in tabular models. The two dominant variants are the labelled property graph, which attaches key-value properties to nodes and edges, and the RDF triple model, which expresses facts as subject-predicate-object statements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graph-data-model", "labels": [ "Graph Data Model" ], "is_subclass_of": [ "Data Model" ], "wikilinks": [] }, { "id": "graph-database", "title": "Graph Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Graph Database is a specialised data management system representing data as a network of nodes (vertices) connected by relationships (edges) carrying their own properties and semantics, contrasted with relational databases that scatter relationships across foreign-key joins, organised aroun...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:graph-database", "labels": [ "Graph Database", "Plot Graph Database" ], "is_subclass_of": [ "Data Management", "Data Management System", "Database", "NoSQL Database", "Persistent Data Store" ], "wikilinks": [ "Adjacency List", "Angles 2018 The Property Graph Database Model", "Angles Arenas Barcelo Hogan Reutter Vrgoc 2017 Foundations of Modern Query Languages for Graph Databases", "Angles Gutierrez 2008 Survey of Graph Database Models", "Anti-Money Laundering", "Apache TinkerPop", "ArangoDB", "B-Tree", "B-Tree Indexing", "BenevolentAI", "Berners-Lee Hendler Lassila 2001 The Semantic Web", "Bizer Heath Berners-Lee 2009 Linked Data Story So Far", "Bloom Filter", "Bronson et al 2013 TAO Facebook Social Graph USENIX ATC", "Columnar Database", "Cybersecurity Analytics", "Cypher Query Language", "DataManagementDomain", "Data Management System", "Database" ] }, { "id": "graph-databases", "title": "Graph Databases", "domain": "data", "domain_name": "Data", "definition": "Graph databases are database management systems that represent and store data as a network of nodes (entities) and edges (relationships), with both capable of carrying named property sets. Unlike relational models that encode relationships through foreign-key joins, graph databases make adjacency a first-class storage primitive, enabling traversal-based queries that navigate multi-hop paths in near-constant time per hop. They implement the property-graph or RDF triple-store data models, and are queried via languages such as Cypher, Gremlin, or SPARQL. Their native graph storage and index-free adjacency make them especially suited to domains where relationships are as semantically rich as the entities themselves.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graph-databases", "labels": [ "Graph Databases" ], "is_subclass_of": [ "Database Systems" ], "wikilinks": [ "Knowledge Graphs", "Network Analysis", "Database Systems", "https://en.wikipedia.org/wiki/Graph_database", "https://neo4j.com/developer/graph-database/" ] }, { "id": "graph-embedding", "title": "Graph Embedding", "domain": "ai", "domain_name": "Ai", "definition": "Graph embedding is a family of representation-learning techniques that map the nodes, edges, or whole subgraphs of a graph into a continuous low-dimensional vector space while preserving structural and relational properties. The learned vectors place topologically or semantically similar elements close together, enabling machine-learning models to operate on graph-structured data. Methods range from random-walk approaches to neural graph encoders.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graph-embedding", "labels": [ "Graph Embedding" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "graph-neural-network", "title": "Graph Neural Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Graph Neural Network (GNN) is a deep learning architecture that operates directly on graph-structured data by iteratively propagating and aggregating feature information across node neighbourhoods. GNNs generalise convolutional and attention mechanisms to non-Euclidean domains, learning node, edge, and graph-level representations suitable for tasks including node classification, link prediction, and graph classification across domains such as knowledge graphs, social networks, molecular modelling, and recommendation systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:graph-neural-network", "labels": [ "Graph Neural Network" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Autogen", "fleeting \ud83e\udeb4", "github", "GraphRAG", "Immersive", "Llama", "Machine Vision", "Microsoft", "Notion", "Tom Smoker", "Agentic Metaverse for Global Creatives", "Agentic Mycelia", "agents", "Agents", "ChatGPT", "Could", "Decentralised Web", "Diagrams as Code", "Evaluation benchmarks and leaderboards", "GPT" ] }, { "id": "graph-neural-networks", "title": "Graph Neural Networks", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of neural network architectures that operate directly on graph-structured data, computing node, edge or graph representations by iteratively exchanging information along edges through message-passing schemes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "growing", "iri": "urn:ngm:class:graph-neural-networks", "labels": [ "Graph Neural Networks", "Graph Neural Network" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Message Passing", "Graph Theory", "Knowledge Graph", "Deep Learning Domain", "Neural Network" ] }, { "id": "graph-optimisation", "title": "Graph Optimisation", "domain": "ai", "domain_name": "Ai", "definition": "Graph optimisation is the set of compiler transformations applied to a model's computation graph to reduce latency, memory footprint and energy use without altering the model's semantics. Typical passes include operator fusion, constant folding, dead-node elimination, layout reordering and kernel selection. It is performed by inference runtimes and ahead-of-time compilers as a precursor to deployment on a target accelerator.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graph-optimisation", "labels": [ "Graph Optimisation" ], "is_subclass_of": [ "Model Optimisation and Performance", "Inference Optimisation" ], "wikilinks": [] }, { "id": "graph-query-language", "title": "Graph Query Language", "domain": "data", "domain_name": "Data", "definition": "A graph query language is a declarative language for expressing queries, traversals, and pattern matches over graph-structured data composed of nodes and edges. Rather than joining tables, it lets users describe paths and subgraph patterns directly, making relationship-centric questions concise. Examples include Cypher and Gremlin, with GQL emerging as an ISO standard that unifies property-graph querying alongside the W3C SPARQL language for RDF.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:graph-query-language", "labels": [ "Graph Query Language" ], "is_subclass_of": [ "Graph Database" ], "wikilinks": [] }, { "id": "graph-representation", "title": "Graph Representation", "domain": "data", "domain_name": "Data", "definition": "Graph representation is the encoding of a problem domain as a set of nodes (vertices) connected by edges, capturing entities and the relationships between them. It enables algorithmic reasoning over connectivity, distance, and structure, and underpins pathfinding, routing, and topological analysis. Common concrete forms include adjacency matrices, adjacency lists, and edge lists.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graph-representation", "labels": [ "Graph Representation" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "graph-search", "title": "Graph Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Graph search encompasses algorithms that systematically traverse graph-structured state spaces \u2014 sets of nodes (states) connected by edges (transitions) \u2014 to discover paths, optimal solutions, or reachable configurations satisfying given criteria. Classical uninformed methods (Breadth-First Search, Depth-First Search, Dijkstra) provide completeness and optimality guarantees on finite discrete graphs; informed heuristic methods (A*, IDA*, weighted A*) accelerate search using problem-specific cost estimates; sampling-based planners (RRT, PRM, RRT*) extend graph search to high-dimensional continuous configuration spaces by building implicit graphs from random samples. Graph search is the algorithmic backbone of robot motion planning, AI planning, knowledge graph querying, route navigation, and game-tree evaluation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:graph-search", "labels": [ "Graph Search" ], "is_subclass_of": [ "Graph Algorithms" ], "wikilinks": [] }, { "id": "graph-theory", "title": "Graph Theory", "domain": "data", "domain_name": "Data", "definition": "Graph Theory is the branch of mathematics that studies graphs, structures consisting of vertices connected by edges, used to model pairwise relationships between objects. It examines properties such as connectivity, paths, cycles, colourings, matchings and flows, and classifies graphs by structure (for example trees, bipartite and planar graphs). Originating with Euler's 1736 solution of the Seven Bridges of Konigsberg problem, it now underpins network analysis, optimisation and computer science. Graph algorithms are fundamental to routing, scheduling, social network analysis and the representation of knowledge.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graph-theory", "labels": [ "Graph Theory" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Tree", "Shortest Path", "Graph Colouring", "Network Analysis", "Knowledge Graph", "Linear Algebra", "Distributed Systems Domain", "owl:Thing" ] }, { "id": "graph-ql", "title": "GraphQL", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "GraphQL is a query language and runtime for APIs, developed by Facebook and open-sourced in 2015, that allows clients to specify precisely the data they need in a single request rather than consuming fixed-shape REST endpoints. It is defined by a strongly typed schema that describes the graph of types and fields the API exposes, and a runtime that resolves client queries against that schema by executing resolver functions. GraphQL eliminates the over-fetching and under-fetching problems inherent in REST by shifting data shape control to the client, and its introspection capability enables rich developer tooling and automatic documentation generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graph-ql", "labels": [ "GraphQL", "GraphQL API" ], "is_subclass_of": [ "API Standard" ], "wikilinks": [] }, { "id": "graph-rag", "title": "GraphRAG", "domain": "ai", "domain_name": "Ai", "definition": "GraphRAG (Graph Retrieval-Augmented Generation) is an architecture that extends standard retrieval-augmented generation by structuring the indexed knowledge corpus as a knowledge graph of entities and relationships rather than as a flat collection of text chunks, enabling the retrieval system to answer questions that require multi-hop reasoning over connected facts \u2014 such as 'what do entities A and B have in common' \u2014 which naive vector similarity search over disconnected chunks cannot reliably resolve. Microsoft Research's GraphRAG implementation, open-sourced in 2024, uses an LLM to extract entity-relationship triples from a document corpus, builds a community-detected hierarchical graph, generates community summaries at multiple granularities, and retrieves relevant subgraphs and summaries at query time to ground the LLM's response in structured relational context.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:graph-rag", "labels": [ "GraphRAG" ], "is_subclass_of": [ "Retrieval-Augmented Generation" ], "wikilinks": [] }, { "id": "graphical-model", "title": "Graphical Model", "domain": "ai", "domain_name": "Ai", "definition": "A graphical model is a probabilistic model that expresses the conditional dependence structure among random variables as a graph, where nodes are variables and edges encode statistical relationships. Bayesian networks use directed acyclic graphs while Markov random fields use undirected graphs, both enabling compact representation of joint distributions. They support efficient inference and learning by exploiting conditional independence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graphical-model", "labels": [ "Graphical Model", "Probabilistic Graphical Model", "Probabilistic Graphical Models" ], "is_subclass_of": [ "Machine Learning Discipline" ], "wikilinks": [] }, { "id": "graphical-user-interface", "title": "Graphical User Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A form of user interface that lets people interact with computing systems through visual elements \u2014 windows, icons, menus, buttons, and pointers \u2014 rendered on a two-dimensional display and manipulated directly with a pointing device or touch. By replacing memorised text commands with recognisable on-screen objects and immediate visual feedback, the GUI established the dominant interaction paradigm of personal computing from the 1980s onward and remains the baseline against which 3D, voice, and spatial interfaces are contrasted.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:graphical-user-interface", "labels": [ "Graphical User Interface" ], "is_subclass_of": [ "User Interface" ], "wikilinks": [ "User Interface", "3D User Interface", "User Interface Design", "Human Computer Interaction" ] }, { "id": "graphics-api", "title": "Graphics API", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Graphics API (Application Programming Interface) is a standardised software interface mediating between application code and Graphics Processing Unit (GPU) hardware, exposing primitives for command submission, shader compilation, resource allocation, synchronisation, and frame presentation acro...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:graphics-api", "labels": [ "Graphics API" ], "is_subclass_of": [ "Display and Rendering", "Hardware Interface", "Application Programming Interface", "Hardware Abstraction Layer", "System Software", "Rendering Interface" ], "wikilinks": [ "AI Inference", "Akenine-M\u00f6ller et al. 2018 Real-Time Rendering 4th ed", "Apple Metal Shading Language Specification v3.2", "Application Programming Interface", "Bailey 2024 Vulkan Programming Guide SIGGRAPH course", "CAD Software", "Command Buffer", "Command Submission Pattern", "CPU Rendering", "Cross-Platform Rendering", "Descriptor Set", "Device Driver", "Driver Compiler", "DriverLayer", "Driver Stack", "Fixed-Function Pipeline", "Foley et al. 2013 Computer Graphics Principles and Practice", "Frame Presentation Protocol", "Game Engine Framework", "Game Engines" ] }, { "id": "graphics-library", "title": "Graphics Library", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A software library that provides a standardised API for issuing draw calls, managing GPU resources, and configuring the graphics pipeline, abstracting hardware differences from application code. Canonical graphics libraries\u2014OpenGL, Vulkan, Metal, DirectX, WebGL\u2014enable portable, high-performance 2D and 3D rendering across diverse hardware platforms and are foundational to real-time spatial computing applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:graphics-library", "labels": [ "Graphics Library" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "graphics-pipeline", "title": "Graphics Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The graphics pipeline is the ordered sequence of stages that transforms a 3D scene description into a 2D raster image, encompassing vertex processing, primitive assembly, rasterisation, fragment shading, and output merging, executed on programmable GPU hardware and exposed through graphics APIs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graphics-pipeline", "labels": [ "Graphics Pipeline", "Real-Time Graphics Pipeline" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "GPU Architecture", "Real-Time Rendering", "Rasterization", "Shader", "Computer Graphics" ] }, { "id": "graphics-processing-unit", "title": "Graphics Processing Unit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Graphics Processing Unit is a parallel processor designed to accelerate the rendering of images and other data-parallel workloads through many concurrent execution units, and widely repurposed for general-purpose computing in machine learning, scientific simulation, and spatial computing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graphics-processing-unit", "labels": [ "Graphics Processing Unit" ], "is_subclass_of": [ "Graphics Processing", "Computing Infrastructure" ], "wikilinks": [ "Shader", "Real-Time Rendering", "Compute Shader", "GPU", "OpenGL", "Graphics Processing" ] }, { "id": "graphics-processing", "title": "Graphics Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Graphics processing is the computation that transforms scene descriptions into rendered images, covering geometry transformation, shading, rasterisation and output to a display.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:graphics-processing", "labels": [ "Graphics Processing" ], "is_subclass_of": [ "Spatial Computing", "Display and Rendering", "Spatial Computing Domain" ], "wikilinks": [ "GPU", "Rendering Pipeline", "3D Rendering", "Real-Time Rendering", "Graphics API", "Spatial Computing Domain" ] }, { "id": "grasp-planning", "title": "Grasp Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Grasp planning is the computational problem of determining stable contact configurations between a robotic end-effector and an object, such that the resulting grasp resists external disturbances and enables the desired manipulation task. It combines geometric modelling of object shape, force-closure analysis, kinematics constraints, and task-level objectives to synthesise executable grasp poses.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:grasp-planning", "labels": [ "Grasp Planning", "GraspPlanning" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "gravitational-microlensing", "title": "Gravitational Microlensing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gravitational-microlensing", "labels": [ "Gravitational Microlensing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gravitational-wave", "title": "Gravitational Wave", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gravitational-wave", "labels": [ "Gravitational Wave" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gravity-assist", "title": "Gravity Assist", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gravity-assist", "labels": [ "Gravity Assist" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "grc-platform", "title": "Grc Platform", "domain": "governance", "domain_name": "Governance", "definition": "A GRC platform is an integrated software system that unifies governance, risk management and compliance activities across an organisation. It maintains a common library of controls, policies and risks, maps them to regulatory frameworks, and automates assessment, evidence collection and reporting. By consolidating these functions it gives leadership a consistent view of risk posture and control effectiveness while reducing duplicated manual effort.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:grc-platform", "labels": [ "Grc Platform", "GRC Platform" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "greedy-decoding", "title": "Greedy Decoding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Greedy Decoding is a sequence generation strategy that, at each step of an autoregressive model, selects the single token with the highest predicted probability. It is the simplest decoding method, fully deterministic and computationally cheap, but it can be myopic and miss globally higher-probability sequences. It serves as the baseline against which beam search and sampling-based strategies are compared.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:greedy-decoding", "labels": [ "Greedy Decoding" ], "is_subclass_of": [ "Text Generation" ], "wikilinks": [] }, { "id": "green-blockchain-initiatives", "title": "Green Blockchain Initiatives", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Green Blockchain Initiatives encompass the coordinated technological, economic, and governance efforts aimed at eliminating or materially reducing the environmental footprint of distributed ledger systems, spanning four principal intervention vectors: (1) consensus mechanism replacement \u2014 transit...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:green-blockchain-initiatives", "labels": [ "Green Blockchain Initiatives" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Network", "ESG", "Sustainable Finance", "Environmental Technology", "Distributed Ledger Technology" ], "wikilinks": [ "Algorand", "Base Carbon Tonne Standard", "Biodiversity Credit Markets", "Bitcoin Lightning Network", "CapitalMarketsDomain", "Carbon Accounting Standards", "Carbon Credits", "Carbon Offset Programme", "Carbon Token Standard", "CCRI Methodology", "Celo", "Chainlink Oracles", "CleanSpark", "Climate Finance Innovation", "ConsensusLayer", "Core Carbon Principles", "Crypto Climate Accord", "dClimate Network", "DeFi", "Decentralised Autonomous Organisation" ] }, { "id": "green-bond-market", "title": "Green Bond Market", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The green bond market is the segment of fixed-income capital markets in which debt instruments are issued specifically to finance projects with environmental or climate benefits, such as renewable energy, clean transport, or sustainable infrastructure. Proceeds are ring-fenced and reported against recognised frameworks like the ICMA Green Bond Principles. It channels institutional capital toward decarbonisation while offering investors verifiable sustainability exposure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:green-bond-market", "labels": [ "Green Bond Market", "Green Bond" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "green-computing", "title": "Green Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Green computing is the design, manufacture, operation and disposal of computing systems in ways that minimise environmental impact across their lifecycle. It targets energy efficiency, reduced carbon emissions, sustainable materials and responsible electronic-waste management. Practices range from efficient hardware and data-centre design to carbon-aware scheduling of workloads onto low-carbon energy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:green-computing", "labels": [ "Green Computing" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "green-finance", "title": "green finance", "domain": "finance", "domain_name": "Finance", "definition": "Green finance is the broad domain of financial instruments, investment strategies, market mechanisms, and regulatory frameworks that direct capital towards environmentally sustainable economic activities with the explicit objectives of mitigating climate change, halting biodiversity loss, and accelerating the transition to a low-carbon circular economy. It encompasses green bonds, sustainability-linked loans and bonds, green funds, voluntary and compliance carbon markets, blended finance structures, and impact investment vehicles, all underpinned by taxonomic classification systems such as the EU Taxonomy for Sustainable Activities and the ICMA Green Bond Principles that define and verify environmental eligibility criteria. Green finance sits at the intersection of macroeconomic policy, capital markets, and environmental science, requiring robust ESG disclosure, lifecycle assessment methodologies, and increasingly blockchain-backed registries to prevent greenwashing and ensure traceability of environmental claims.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:green-finance", "labels": [ "Green Finance", "Green Finance Standards" ], "is_subclass_of": [ "Sustainable Finance" ], "wikilinks": [] }, { "id": "green-hydrogen", "title": "Green Hydrogen", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Green hydrogen is hydrogen gas produced by electrolysing water using electricity drawn entirely from renewable sources such as wind or solar, in contrast to grey hydrogen produced from fossil natural gas via steam methane reforming. It is used as a zero-carbon fuel and chemical feedstock for sectors that are difficult to electrify directly, including heavy industry, long-haul transport, and seasonal energy storage. Green hydrogen production scales with, and helps absorb curtailed output from, renewable energy generation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:green-hydrogen", "labels": [ "Green Hydrogen" ], "is_subclass_of": [ "Renewable Energy" ], "wikilinks": [] }, { "id": "green-mining-pool", "title": "Green Mining Pool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A green mining pool is a cooperative of cryptocurrency miners that aggregates hashing power while sourcing electricity predominantly from renewable or low-carbon generation, often coupling mining to surplus or curtailed grid energy. It rewards participants for verified sustainable energy use and reports the carbon intensity of the pooled hashrate. The model aims to reduce the environmental footprint of proof-of-work consensus.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:green-mining-pool", "labels": [ "Green Mining Pool" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "greenhouse-gas-emissions", "title": "Greenhouse Gas Emissions", "domain": "governance", "domain_name": "Governance", "definition": "Releases into the atmosphere of gases that trap outgoing infrared radiation \u2014 principally carbon dioxide, methane, nitrous oxide, and fluorinated gases \u2014 from fossil-fuel combustion, industrial processes, agriculture, and land-use change. They are the dominant anthropogenic driver of climate change, are quantified in tonnes of CO2-equivalent using global warming potentials, and are the object of measurement, reporting, and reduction regimes ranging from the GHG Protocol's scope 1\u20133 accounting to national inventories under the Paris Agreement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:greenhouse-gas-emissions", "labels": [ "Greenhouse Gas Emissions" ], "is_subclass_of": [ "Environment" ], "wikilinks": [ "Environment", "Climate Change", "GHG Protocol", "Energy Consumption", "Net Zero Targets" ] }, { "id": "greenhouse-gas-inventory", "title": "Greenhouse Gas Inventory", "domain": "governance", "domain_name": "Governance", "definition": "A greenhouse gas inventory is a systematic, periodic accounting of the greenhouse gas emissions and removals attributable to an organisation, activity or jurisdiction over a defined period, expressed in carbon-dioxide-equivalent units. It categorises emissions by source and by scope, applies emission factors to activity data, and forms the quantitative basis for target-setting, reporting and reduction strategies. Inventories underpin compliance, disclosure and progress toward net-zero commitments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:greenhouse-gas-inventory", "labels": [ "Greenhouse Gas Inventory" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "greenhouse-gas-protocol", "title": "Greenhouse Gas Protocol", "domain": "governance", "domain_name": "Governance", "definition": "The Greenhouse Gas Protocol is the most widely used set of standards for measuring and reporting greenhouse gas emissions across organisations and value chains. It defines the categorisation of emissions into Scope 1 direct emissions, Scope 2 purchased energy and Scope 3 value-chain emissions, and provides accounting and reporting principles. Developed by the World Resources Institute and the World Business Council for Sustainable Development, it underpins corporate carbon accounting and disclosure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:greenhouse-gas-protocol", "labels": [ "Greenhouse Gas Protocol" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "greenwashing-prevention", "title": "Greenwashing Prevention", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Greenwashing Prevention is the set of practices, controls and verification mechanisms that ensure environmental and sustainability claims are substantiated by auditable evidence rather than misleading marketing. In blockchain contexts it leverages tamper-evident records, on-chain provenance and independent attestation to make carbon and ESG claims falsifiable and traceable. It addresses the risk that organisations overstate climate benefits without verifiable backing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:greenwashing-prevention", "labels": [ "Greenwashing Prevention" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "greenwashing", "title": "Greenwashing", "domain": "governance", "domain_name": "Governance", "definition": "Greenwashing is the practice of making misleading, exaggerated, or unsubstantiated claims about the environmental or sustainability credentials of a product, service, organisation, or financial instrument in order to gain reputational, commercial, or regulatory advantage. It encompasses both deliberate deception and negligent miscommunication, ranging from vague marketing language ('eco-friendly', 'carbon neutral') to selective disclosure that conceals material environmental harms. Greenwashing undermines the integrity of sustainability frameworks, distorts capital allocation in ESG finance, and erodes public trust in genuine environmental initiatives. Regulatory bodies worldwide are introducing mandatory disclosure regimes, verification standards, and enforcement mechanisms to combat it.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:greenwashing", "labels": [ "Greenwashing" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [ "Energy Consumption", "Sustainability" ] }, { "id": "grid-infrastructure", "title": "Grid Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Grid infrastructure is the physical and control system of generation, transmission, distribution, and balancing assets that delivers electrical power from sources to consumers. It includes substations, transmission lines, transformers, and increasingly digital control layers for monitoring and demand response. Reliable grid infrastructure is a prerequisite for energy-intensive computing, mining, and data-centre operations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:grid-infrastructure", "labels": [ "Grid Infrastructure", "Electricity Grid", "Electricity Grid Infrastructure", "Power Grid Infrastructure" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "grid-search", "title": "Grid Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Grid search is a hyperparameter-optimisation method that exhaustively evaluates every combination of values drawn from a predefined discrete grid over the hyperparameter space. Each candidate configuration is trained and scored, typically using cross-validation, and the best-performing combination is selected. Grid search is simple and fully parallelisable but scales exponentially with the number of hyperparameters.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:grid-search", "labels": [ "Grid Search" ], "is_subclass_of": [ "Hyperparameter Optimisation", "AI Technique" ], "wikilinks": [] }, { "id": "gridded-data-state", "title": "Gridded Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gridded-data-state", "labels": [ "Gridded Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "gridding", "title": "Gridding", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:gridding", "labels": [ "Gridding" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "grokking", "title": "Grokking", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A training phenomenon in deep learning, first documented by Power et al. (2022) on algorithmic tasks, in which a network first memorises its training data \u2014 reaching perfect training accuracy while test accuracy stays at chance \u2014 and then, after continued training far beyond apparent convergence, abruptly transitions to near-perfect generalisation; interpreted as a delayed phase change from a memorising solution to a simpler, structured circuit, typically induced by regularisation such as weight decay.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:grokking", "labels": [ "Grokking" ], "is_subclass_of": [ "Generalisation" ], "wikilinks": [ "Generalisation", "Overfitting", "Emergent Capabilities" ] }, { "id": "gross-domestic-product", "title": "Gross Domestic Product", "domain": "finance", "domain_name": "Finance", "definition": "Gross domestic product (GDP) is the total monetary value of all final goods and services produced within a country's borders over a given period. It is the principal aggregate measure of economic activity, computed via the production, income, or expenditure approaches, and is reported in nominal and real (inflation-adjusted) terms. GDP underpins growth measurement, fiscal and monetary policy, and cross-country comparison.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:gross-domestic-product", "labels": [ "Gross Domestic Product" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "groth-16", "title": "Groth16", "domain": "security", "domain_name": "Security", "definition": "A pairing-based zk-SNARK proving system that produces constant-size proofs verifiable with a few elliptic curve pairing operations, at the cost of a per-circuit trusted setup.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:groth-16", "labels": [ "Groth16" ], "is_subclass_of": [ "ZK-SNARK" ], "wikilinks": [ "Elliptic Curve Cryptography", "ZK-SNARK", "Zero-Knowledge Proof", "Cryptographic Primitive" ] }, { "id": "ground-control-experiment", "title": "Ground Control Experiment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-control-experiment", "labels": [ "Ground Control Experiment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-control-point", "title": "Ground Control Point", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-control-point", "labels": [ "Ground Control Point" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-deformation-monitoring", "title": "Ground Deformation Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-deformation-monitoring", "labels": [ "Ground Deformation Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-robot", "title": "Ground Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Ground robot is a mobile robotic platform that operates on terrestrial surfaces using wheeled, tracked, legged, or hybrid locomotion systems to navigate structured and unstructured environments while executing purposeful tasks \u2014 including material transport, environmental inspection, search and r...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ground-robot", "labels": [ "Ground Robot", "Unmanned Ground Vehicle" ], "is_subclass_of": [ "Robot Type", "Autonomous System", "Mobile Robot", "Autonomous Systems", "Cyber-Physical Systems", "Embodied AI" ], "wikilinks": [ "5G Connectivity", "Agricultural Robotics", "ANSI/RIA R15.08", "Autonomous Systems", "Battery Management System", "Cartographer SLAM", "Communication Interface", "Control Layer", "Cyber-Physical Systems", "Deep Reinforcement Learning", "Disaster Response", "Embodied AI", "Fixed Robot Arm", "Gazebo Simulation", "IEC 62061", "IEEE 1872-2015", "Industrial Inspection", "Industry 4.0", "Inertial Measurement Unit", "Infrastructure Inspection" ] }, { "id": "ground-sample-distance", "title": "Ground Sample Distance", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-sample-distance", "labels": [ "Ground Sample Distance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-segment", "title": "Ground Segment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-segment", "labels": [ "Ground Segment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-station", "title": "Ground Station", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A ground station is the terrestrial radio or optical facility that communicates with a spacecraft. It provides telemetry, tracking and command, often abbreviated **TT&C**, and may also receive payload data. The station is one part of the wider ground segment, which includes mission-control systems, networks, storage, processing, user terminals and operations staff.[^1]", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-station", "labels": [ "Ground Station" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-support-equipment", "title": "Ground Support Equipment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-support-equipment", "labels": [ "Ground Support Equipment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-track", "title": "Ground Track", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-track", "labels": [ "Ground Track" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-truth-labels", "title": "Ground Truth Labels", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Ground truth labels are the authoritative, human-verified or empirically observed target values assigned to data instances, used to train and evaluate supervised machine-learning models. They represent the correct answer against which model predictions are compared, forming the basis for loss computation during training and accuracy measurement during evaluation. The quality, consistency, and coverage of ground truth labels directly bound the performance a learned model can achieve.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ground-truth-labels", "labels": [ "Ground Truth Labels", "Ground Truth", "Ground Truth Label" ], "is_subclass_of": [ "Data Annotation" ], "wikilinks": [] }, { "id": "ground-truth", "title": "Ground Truth", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Ground truth is the set of verified, correct labels or measurements against which a model's predictions are compared to assess accuracy. It is typically produced by expert annotation, direct measurement, or a trusted reference process, and underpins benchmark datasets used for training and evaluation. The reliability of any accuracy or error metric is bounded by the quality of the ground truth it is measured against.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ground-truth", "labels": [ "Ground Truth" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "ground-based-augmentation-system", "title": "Ground-based Augmentation System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-based-augmentation-system", "labels": [ "Ground-based Augmentation System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ground-based-remote-sensing", "title": "Ground-based Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ground-based-remote-sensing", "labels": [ "Ground-based Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "grounded-language-understanding", "title": "Grounded Language Understanding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Grounded language understanding is the capacity of an agent to connect linguistic meaning to perception and action in a physical or simulated environment, rather than treating language as a purely symbolic, text-only system. It requires linking words and phrases to sensory referents, spatial relations and executable actions so that an instruction such as 'pick up the red block' resolves to concrete perception and motor commands. It is a prerequisite for embodied AI and embodied-minds research, where an agent must interpret language in the context of its own body and surroundings.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:grounded-language-understanding", "labels": [ "Grounded Language Understanding" ], "is_subclass_of": [ "Natural Language Understanding" ], "wikilinks": [] }, { "id": "groundwater-level", "title": "Groundwater Level", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:groundwater-level", "labels": [ "Groundwater Level" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "group-chat-channel", "title": "Group Chat Channel", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A group chat channel is a persistent, named messaging space where multiple participants can send and receive messages collectively. Channels are typically organised by topic, project, or team, providing a shared context for ongoing conversation and record-keeping. They support both synchronous and asynchronous communication patterns within distributed organisations.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:group-chat-channel", "labels": [ "Group Chat Channel" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "group-vs-individual-fairness", "title": "Group vs Individual Fairness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Group vs Individual Fairness denotes two competing paradigms for defining and enforcing algorithmic fairness: group fairness requires statistical parity of outcomes or error rates across protected demographic cohorts, while individual fairness requires that similar individuals receive similar predictions regardless of group membership. The two paradigms are formally incompatible in general \u2014 satisfying demographic parity does not guarantee individual fairness and vice versa \u2014 representing a fundamental tension in fair machine learning that practitioners must resolve through context-specific policy choices. This distinction shapes the selection of fairness metrics, audit methodologies, and bias mitigation interventions in AI system design.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:group-vs-individual-fairness", "labels": [ "Group vs Individual Fairness", "Individual Fairness" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "Barocas et al. (2019)", "Dwork et al. (2012)", "Hardt et al. (2016)", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "grouped-query-attention", "title": "grouped query attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Grouped Query Attention (GQA) is a transformer attention variant that partitions the set of query heads into G groups, each group sharing a single pair of key and value heads, thereby interpolating between Multi-Head Attention (MHA, where each query head has its own KV head) and Multi-Query Attention (MQA, where all query heads share one KV head). GQA reduces the key-value cache memory footprint during autoregressive inference\u2014proportionally to the number of groups\u2014while preserving model quality closer to MHA than MQA. It has been adopted in production LLMs including Llama 2, Mistral, and Gemma.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:grouped-query-attention", "labels": [ "Grouped Query Attention" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "gs1-digital-link", "title": "Gs1 Digital Link", "domain": "standards", "domain_name": "Standards", "definition": "GS1 Digital Link is a standard that expresses GS1 identifiers, such as the Global Trade Item Number, as web URIs so that a single QR code or data carrier can connect a physical product to multiple online resources. A resolver service interprets the URI and routes requests to information such as product details, instructions, provenance, or recycling guidance. It bridges traditional barcodes with the web and semantic data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:gs1-digital-link", "labels": [ "Gs1 Digital Link", "GS1 Digital Link" ], "is_subclass_of": [ "GS1" ], "wikilinks": [] }, { "id": "guaranteed-bandwidth", "title": "Guaranteed Bandwidth", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Guaranteed Bandwidth is a network quality commitment that ensures a minimum throughput level is reserved for a specific application or user session, regardless of concurrent network load. In spatial computing and XR contexts, it is a prerequisite for low-latency immersive streaming, enabling consistent frame delivery without compression artefacts or stutter that would degrade presence. It is typically enforced through Quality of Service mechanisms, traffic prioritisation, or network slicing in 5G infrastructure.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:guaranteed-bandwidth", "labels": [ "Guaranteed Bandwidth" ], "is_subclass_of": [ "Network Quality Metric" ], "wikilinks": [ "Network Quality Metric" ] }, { "id": "guardrail", "title": "Guardrail", "domain": "security", "domain_name": "Security", "definition": "A programmable safety control placed around a language model or agent that inspects, constrains, or rewrites inputs and outputs to keep behaviour within a defined policy envelope. Guardrails operate at runtime as input filters, output validators, topic and PII detectors, schema or format enforcers, and tool-permission gates; they block, redact, or re-prompt when a violation is detected, providing an enforcement layer that is independent of, and complementary to, the alignment baked into the model itself.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:guardrail", "labels": [ "Guardrail" ], "is_subclass_of": [ "AI Safety", "AISafety" ], "wikilinks": [ "AISafety", "ContentModeration", "ConstrainedDecoding", "ConstitutionalAI" ] }, { "id": "gyroscope", "title": "Gyroscope", "domain": "robotics", "domain_name": "Robotics", "definition": "Gyroscope - An angular velocity sensor that detects rotation rates about three orthogonal axes, enabling Attitude Estimation, Roll/Pitch/Yaw Measurement, and Orientation Tracking for balance control and Inertial Navigation in aerial and mobile robots.", "entityType": "Class", "qualityScore": 0.56, "maturity": "draft", "iri": "urn:ngm:class:gyroscope", "labels": [ "Gyroscope", "MEMS Gyroscope" ], "is_subclass_of": [ "Perception and Sensing", "Robotics", "Inertial Sensor" ], "wikilinks": [ "Attitude Determination", "Attitude Estimation", "Balance Control", "Bias Compensation", "Drone Stabilisation", "Inertial Measurement Unit", "Inertial Navigation", "Inertial Sensor", "Orientation Reference", "Orientation Tracking", "Roll/Pitch/Yaw Measurement", "AI Agent System", "Robotics", "RoboticsDomain", "Sensor Fusion" ] }, { "id": "h-bridge", "title": "H-Bridge", "domain": "robotics", "domain_name": "Robotics", "definition": "An H-bridge is a power-electronics circuit of four switching elements arranged in an 'H' around a load, enabling voltage of either polarity to be applied from a single supply. It is the core stage of DC motor drivers: diagonal switch pairs drive the motor forward or reverse, pulse-width modulation of the switches regulates speed and torque, and shorting or opening both legs provides braking or coasting. Implemented with MOSFETs or IGBTs plus gate drivers and protection, H-bridges power robotics actuators, servo drives, inverters, and battery-powered traction.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:h-bridge", "labels": [ "H-Bridge" ], "is_subclass_of": [ "Power Electronics" ], "wikilinks": [ "Power Electronics", "Motor Driver", "Pulse Width Modulation", "Motor Control" ] }, { "id": "h-264", "title": "H.264", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "H.264, also known as AVC (Advanced Video Coding) or MPEG-4 Part 10, is a block-oriented, motion-compensated video compression standard jointly developed by the ITU-T and ISO/IEC and finalised in 2003. It achieves substantially better compression efficiency than its predecessors through advanced intra and inter prediction, variable block-size motion compensation, an in-loop deblocking filter, and context-adaptive entropy coding. For two decades it has been the most widely deployed video codec, used in streaming, broadcast, Blu-ray, video conferencing, and surveillance, with broad hardware acceleration.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:h-264", "labels": [ "H.264", "H.264 Screen Content Coding", "ITU-T H.264" ], "is_subclass_of": [ "Video Codec" ], "wikilinks": [] }, { "id": "hazop", "title": "HAZOP", "domain": "governance", "domain_name": "Governance", "definition": "HAZOP (Hazard and Operability Study) is a structured, systematic technique for identifying potential hazards and operability problems in a process or system by applying a set of guide words \u2014 such as 'more', 'less', 'no', and 'reverse' \u2014 to each design intention and examining the consequences of deviation. It is widely used in process industries and safety-critical engineering, and is referenced by functional-safety standards such as IEC 61508. A HAZOP study is typically conducted by a multidisciplinary team working systematically through a piping and instrumentation diagram or equivalent process representation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hazop", "labels": [ "HAZOP" ], "is_subclass_of": [ "Hazard Analysis" ], "wikilinks": [] }, { "id": "hd-maps", "title": "HD Maps", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "HD maps (high-definition maps) are centimetre-accurate digital representations of road infrastructure \u2014 including lane geometry, traffic signs, signal positions and road markings \u2014 used by autonomous vehicles to localise themselves precisely and to anticipate road features beyond the range of onboard sensors. They are built from surveyed lidar and camera data and are typically fused with real-time perception output to cross-check and supplement what the vehicle currently observes. HD maps trade off high localisation accuracy against the cost and staleness risk of maintaining a pre-built map as road conditions change.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:hd-maps", "labels": [ "HD Maps" ], "is_subclass_of": [ "Autonomous Driving" ], "wikilinks": [] }, { "id": "hevc", "title": "HEVC", "domain": "metaverse", "domain_name": "Metaverse", "definition": "HEVC (High Efficiency Video Coding), also known as H.265, is a video compression standard that roughly doubles the data compression ratio of its predecessor H.264 at the same visual quality. It introduces larger, flexible coding tree units and improved prediction and entropy coding to support 4K and 8K video. As a proprietary, patent-encumbered codec it carries licensing obligations that motivate royalty-free alternatives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hevc", "labels": [ "HEVC" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "hipaa", "title": "HIPAA", "domain": "governance", "domain_name": "Governance", "definition": "The Health Insurance Portability and Accountability Act, a United States federal law enacted in 1996 that establishes national standards for protecting the privacy and security of certain health information (protected health information) held by covered entities and their business associates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hipaa", "labels": [ "HIPAA", "HIPAA Compliance" ], "is_subclass_of": [ "Data Protection Law" ], "wikilinks": [ "Data Protection", "Information Security", "Privacy", "Data Privacy", "Data Protection Law" ] }, { "id": "hl7-fhir", "title": "HL7 FHIR", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "HL7 FHIR (Fast Healthcare Interoperability Resources) is a standard developed by Health Level Seven International for the electronic exchange of healthcare information. It models clinical and administrative data as modular Resources accessed through a RESTful API using JSON or XML representations, combining a defined data model with web-native interaction. It has become the dominant modern standard for healthcare interoperability, enabling exchange between electronic health records, apps and analytics systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hl7-fhir", "labels": [ "HL7 FHIR" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "hlsl", "title": "HLSL", "domain": "metaverse", "domain_name": "Metaverse", "definition": "HLSL (High-Level Shading Language) is Microsoft's C-like programming language for writing GPU shader programs targeting the Direct3D graphics API. It lets developers author vertex, pixel, geometry, and compute shaders that run on the programmable stages of the rendering pipeline. HLSL is a cornerstone of real-time graphics on Windows and Xbox platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hlsl", "labels": [ "HLSL" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "hm-treasury", "title": "HM Treasury", "domain": "finance", "domain_name": "Finance", "definition": "HM Treasury (His Majesty's Treasury) is the United Kingdom government's principal economics and finance ministry, responsible for setting and implementing fiscal policy, managing public expenditure, and overseeing the financial services regulatory framework. Led by the Chancellor of the Exchequer, it coordinates macroeconomic strategy, tax policy, and sovereign debt management through the Debt Management Office. HM Treasury acts as the primary interface between the UK government and major financial regulators \u2014 including the Bank of England, the Financial Conduct Authority, and the Prudential Regulation Authority \u2014 establishing the legislative framework within which those bodies operate. It also leads the UK government's policy position on emerging financial technologies including cryptoassets, stablecoins, and central bank digital currencies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:hm-treasury", "labels": [ "HM Treasury", "UK HM Treasury" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Financial Regulation", "owl:Thing" ] }, { "id": "hmac", "title": "HMAC", "domain": "security", "domain_name": "Security", "definition": "HMAC (hash-based message authentication code) is a construction that combines a cryptographic hash function with a secret key to produce a fixed-length tag verifying both the integrity and the authenticity of a message. It applies the underlying hash twice with key-derived inner and outer padding, providing security that does not depend on the hash being collision-resistant in the same way a plain hash would. HMAC is widely used to authenticate API requests, tokens, and protocol messages.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hmac", "labels": [ "HMAC" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "hmrc", "title": "HMRC", "domain": "governance", "domain_name": "Governance", "definition": "His Majesty's Revenue and Customs (HMRC) is the non-ministerial department of the UK government responsible for the administration and collection of taxes, payment of certain state benefits, and enforcement of customs regulations. Formed in 2005 by the merger of the Inland Revenue and HM Customs and Excise, HMRC operates the UK's tax self-assessment regime, Pay As You Earn (PAYE) system, and Making Tax Digital (MTD) programme. It also issues authoritative guidance on the taxation of cryptoassets, financial instruments, and cross-border transactions, making it a central regulatory reference for UK digital finance and compliance obligations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:hmrc", "labels": [ "HMRC", "HMRC CDS" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Body" ], "wikilinks": [ "Tax", "Financial Regulation", "Cryptocurrency", "Entity" ] }, { "id": "hnsw-index", "title": "HNSW Index", "domain": "data", "domain_name": "Data", "definition": "An HNSW (Hierarchical Navigable Small World) index is a graph-based data structure for approximate nearest-neighbour search over high-dimensional vectors. It builds a multi-layer proximity graph where greedy traversal from a sparse top layer down to a dense base layer locates close vectors in logarithmic time. HNSW is the standard index backing vector databases and semantic search at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hnsw-index", "labels": [ "HNSW Index", "HNSW" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "hodl-waves", "title": "HODL Waves", "domain": "blockchain", "domain_name": "Blockchain", "definition": "HODL Waves is an on-chain analytics visualisation that bands the entire Bitcoin supply by the age since each coin last moved, showing what proportion is held over various time horizons. The coloured bands reveal accumulation and distribution behaviour, distinguishing long-term holders from short-term speculators. It is a key tool for interpreting holder conviction and market cycles from blockchain data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hodl-waves", "labels": [ "HODL Waves" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "hri", "title": "HRI", "domain": "robotics", "domain_name": "Robotics", "definition": "Human-Robot Interaction is the study and design of how people and robots communicate and work together. It draws on robotics, psychology, and human-computer interaction.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hri", "labels": [ "HRI" ], "is_subclass_of": [ "Robotics", "Human-Robot Interaction" ], "wikilinks": [ "User Experience", "Robot Control", "Conversational AI", "Robotics", "https://humanrobotinteraction.org", "https://en.wikipedia.org/wiki/Human%E2%80%93robot_interaction" ] }, { "id": "hsbc-orion", "title": "HSBC Orion", "domain": "blockchain", "domain_name": "Blockchain", "definition": "HSBC Orion is HSBC's institutional tokenisation platform for issuing and managing digital bonds and other assets on a permissioned distributed ledger. It enables on-chain issuance, settlement, and custody of tokenised securities within a regulated banking framework, used for sovereign and corporate digital bond programmes. Orion exemplifies enterprise adoption of consortium blockchain for capital markets infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hsbc-orion", "labels": [ "HSBC Orion", "Orion" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "hsbc", "title": "HSBC", "domain": "finance", "domain_name": "Finance", "definition": "HSBC is a multinational banking and financial services group headquartered in London, providing retail, commercial, and investment banking across many countries.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hsbc", "labels": [ "HSBC" ], "is_subclass_of": [ "Traditional Banking" ], "wikilinks": [ "Financial Regulation", "Traditional Banking" ] }, { "id": "htlc", "title": "HTLC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Hashed Time-Locked Contract is a conditional payment construct that releases funds when a preimage is revealed before a deadline, otherwise refunding the sender. It combines a cryptographic hash commitment with an on-chain timeout to achieve trustless conditional transfer across untrusted intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:htlc", "labels": [ "HTLC" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Hash Function", "Lightning Network", "Payment Channel", "Bitcoin Script", "Smart Contract" ] }, { "id": "html", "title": "HTML", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "HTML (HyperText Markup Language) is the standard markup language for creating and structuring documents and applications on the World Wide Web. It uses a system of nested elements and attributes to describe the semantic structure of content such as headings, paragraphs, links, images, and interactive controls. Web browsers parse HTML into a document object model that is rendered visually and made accessible to assistive technologies and scripts.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:html", "labels": [ "HTML" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "htn-planning", "title": "HTN Planning", "domain": "ai", "domain_name": "Ai", "definition": "HTN (Hierarchical Task Network) planning is an automated planning approach that solves problems by recursively decomposing high-level compound tasks into ordered subtasks using a library of methods, until only directly executable primitive actions remain. Unlike classical goal-state planning, it encodes domain knowledge as task decompositions, yielding efficient, human-interpretable plans. HTN planning is widely used in robotics, game AI, and workflow automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:htn-planning", "labels": [ "HTN Planning", "HTN Decomposition" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "http-protocol", "title": "HTTP Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Hypertext Transfer Protocol (HTTP) is a stateless, application-layer request-response protocol that forms the foundation of data communication on the World Wide Web, defining the format and semantics of messages exchanged between clients (browsers, API consumers) and servers (web servers, API gateways) over TCP/IP connections. Each HTTP transaction consists of a request message specifying a method (GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS), a target URI, headers conveying metadata, and an optional body, followed by a response message containing a status code, headers, and an optional body. HTTP has evolved through versions 1.0, 1.1, 2, and 3, with each version improving multiplexing, compression, and connection management.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:http-protocol", "labels": [ "HTTP Protocol", "HTTP/3" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "http", "title": "HTTP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Hypertext Transfer Protocol, a stateless application-layer protocol for transferring hypertext and other resources between clients and servers that underpins the World Wide Web.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:http", "labels": [ "HTTP", "HTTP 402", "HTTP Authentication Scheme", "HTTP Long Polling", "HTTP Method", "HTTP/2", "HTTPS" ], "is_subclass_of": [ "Application Layer" ], "wikilinks": [ "Transport Layer", "API", "Network Protocol", "Communication Protocols", "Application Layer" ] }, { "id": "habitable-zone", "title": "Habitable Zone", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:habitable-zone", "labels": [ "Habitable Zone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "habitat-mapping", "title": "Habitat Mapping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:habitat-mapping", "labels": [ "Habitat Mapping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hall-effect-thruster", "title": "Hall-effect Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hall-effect-thruster", "labels": [ "Hall-effect Thruster" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "hallucination-rate", "title": "Hallucination Rate", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A quantitative metric measuring the frequency with which a generative AI model produces factually incorrect or fabricated information.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:hallucination-rate", "labels": [ "Hallucination Rate" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [] }, { "id": "hallucination-reduction", "title": "Hallucination Reduction", "domain": "ai", "domain_name": "Ai", "definition": "Hallucination reduction is the set of techniques used to decrease the rate at which large language models generate fluent but factually incorrect or unsupported output. Approaches include grounding generation in retrieved evidence, fact-checking against trusted sources, calibrated abstention, and fine-tuning for faithfulness. It is central to deploying generative AI in high-stakes domains where accuracy is critical.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hallucination-reduction", "labels": [ "Hallucination Reduction", "Hallucination Mitigation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "hallucination", "title": "Hallucination", "domain": "ai", "domain_name": "Ai", "definition": "Hallucination in artificial intelligence refers to the phenomenon whereby a generative model produces fluent, confident, and syntactically plausible output that is factually incorrect, unsupported by provided context, or entirely fabricated. It is a fundamental failure mode of large language models, vision-language models, and other neural generative systems that optimise for next-token probability rather than verifiable truth. Hallucinations range from subtle factual distortions to wholesale invention of citations, persons, dates, or events, and represent a critical safety and reliability concern for deployed AI systems. Mitigation strategies include retrieval-augmented generation, grounding with tool use, chain-of-thought prompting, and output verification pipelines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hallucination", "labels": [ "Hallucination", "Hallucination in Language Models" ], "is_subclass_of": [ "AI Risk" ], "wikilinks": [ "Tool-Augmented Reasoning", "Conversational AI", "AI Risk" ] }, { "id": "halo-orbit", "title": "Halo Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:halo-orbit", "labels": [ "Halo Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "halving-schedule", "title": "Halving Schedule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The halving schedule is the predetermined, code-enforced rule in proof-of-work cryptocurrencies that periodically cuts the block subsidy paid to miners by half. In Bitcoin this occurs every 210,000 blocks (roughly four years), progressively reducing new issuance until the 21 million supply cap is reached. It is the core monetary policy mechanism that makes the asset's supply predictable and disinflationary.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:halving-schedule", "labels": [ "Halving Schedule" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "halving", "title": "Halving", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A programmatic event in proof-of-work blockchains, most notably Bitcoin, whereby the block reward issued to miners is cut in half at a predetermined block height. Halvings enforce a fixed, disinflationary supply schedule that asymptotically approaches the maximum coin supply. By reducing new coin issuance, halvings create predictable scarcity and are a central component of Bitcoin's monetary policy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:halving", "labels": [ "Halving" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "hamiltonian-dynamics", "title": "Hamiltonian Dynamics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hamiltonian dynamics describes the evolution of a physical system through a Hamiltonian function that encodes total energy as a function of position and momentum, generating equations of motion that conserve energy and preserve phase-space volume. In machine learning it underlies Hamiltonian Monte Carlo, where a sampler's proposal is generated by simulating Hamiltonian trajectories through parameter space augmented with auxiliary momentum variables, enabling long, low-rejection-rate moves through complex posterior distributions. Its energy-conserving structure makes proposals far more efficient than random-walk methods for high-dimensional, correlated distributions. Numerical integrators such as the leapfrog method are used to simulate the trajectories while approximately preserving the conservation properties that make the method valid.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hamiltonian-dynamics", "labels": [ "Hamiltonian Dynamics" ], "is_subclass_of": [ "Dynamical Systems" ], "wikilinks": [] }, { "id": "hamiltonian-monte-carlo", "title": "Hamiltonian Monte Carlo", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hamiltonian Monte Carlo (HMC) is a Markov chain Monte Carlo sampling algorithm that uses Hamiltonian dynamics to generate distant, low-autocorrelation proposals in high-dimensional parameter spaces. By treating the negative log-posterior as a potential energy and augmenting with auxiliary momentum variables, HMC can traverse the posterior landscape far more efficiently than random-walk Metropolis methods.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hamiltonian-monte-carlo", "labels": [ "Hamiltonian Monte Carlo" ], "is_subclass_of": [ "Bayesian Inference" ], "wikilinks": [] }, { "id": "hand-tracking-telepresence", "title": "Hand Tracking Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Hand Tracking Telepresence is the real-time capture, transmission, and rendering of hand pose and gesture data from a remote participant into a local or shared virtual environment. It enables natural gestural communication \u2014 such as pointing, waving, or manipulating shared objects \u2014 between geographically distributed collaborators. By preserving expressive hand movements, this technology significantly enriches the non-verbal communication channel in remote collaboration systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:hand-tracking-telepresence", "labels": [ "Hand Tracking Telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "hand-tracking", "title": "Hand Tracking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hand Tracking is a real-time computational pipeline that continuously estimates the three-dimensional position, orientation, and articulation state of one or both human hands from sensor input, delivering a skeletal or parametric representation of all fingers and joints at interactive framerates ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hand-tracking", "labels": [ "Hand Tracking" ], "is_subclass_of": [ "AI Application", "Gesture Recognition", "Pose Estimation", "Computer Vision", "Human-Computer Interaction", "Spatial Computing Paradigm" ], "wikilinks": [ "Accessibility Technology", "Air Typing", "AlgorithmLayer", "Annotated Hand Dataset", "Apple visionOS Hand Input", "Apple Vision Pro", "Body Tracking", "Calibration Data", "Camera Sensor", "Confidence Estimator", "Controller-Based Input", "Data Glove", "Depth Sensor", "Electromyography", "Extended Reality", "Gesture Recognition", "Hand Skeleton Model", "HumanComputerInteractionDomain", "Infrared Illumination", "ISO 9241 Ergonomics of Human-System Interaction" ] }, { "id": "hanim-standard", "title": "Hanim Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Humanoid Animation (H-Anim) is an ISO/IEC approved international standard developed by the Web3D Consortium for interchangeable humanoid figures, defining specifications for articulated avatars, skeletal hierarchies, and animation systems that enable character portability across 3D games, simulat...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:hanim-standard", "labels": [ "Hanim Standard", "HAnim" ], "is_subclass_of": [ "Standards and Interoperability", "Animation Standard" ], "wikilinks": [ "Cross-Platform Avatars", "Animation Standard", "Blockchain", "metaverse" ] }, { "id": "haptic-feedback-system", "title": "Haptic Feedback System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Haptic Feedback System is a hardware-software assembly that generates controlled tactile and kinaesthetic sensations in response to digital events, using actuator arrays (vibrotactile motors, piezoelectric patches, pneumatic bladders, or shape-memory alloys), real-time rendering engines, and closed-loop sensorimotor control loops to simulate textures, resistances, impacts, and spatial forces for users of VR/AR headsets, surgical simulators, teleoperation systems, and mobile devices. The system comprises transducers that convert electrical signals into mechanical motion, driver electronics that modulate waveform parameters (frequency, amplitude, duration), rendering middleware that maps virtual-world physics to actuator commands, and perceptual models calibrated to human mechanoreceptor response characteristics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:haptic-feedback-system", "labels": [ "Haptic Feedback System", "Haptic Systems" ], "is_subclass_of": [ "Haptic Feedback" ], "wikilinks": [] }, { "id": "haptic-feedback-telepresence", "title": "Haptic Feedback Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Haptic Feedback Telepresence is the integration of tactile and kinesthetic feedback devices \u2014 such as force-feedback gloves, exoskeletons, and haptic controllers \u2014 into telepresence systems, enabling bidirectional touch sensation between a remote operator and a distant physical or virtual environment. It closes the sensorimotor loop beyond visual and auditory channels, allowing operators to feel forces, textures, weight, and vibration, which significantly enhances manipulation precision and immersive presence in teleoperation, surgical robotics, and VR collaboration contexts.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:haptic-feedback-telepresence", "labels": [ "Haptic Feedback Telepresence", "TELE-203-haptic-feedback-telepresence" ], "is_subclass_of": [ "Telepresence", "Haptic Technology" ], "wikilinks": [ "Haptic Technology", "TELE-020-virtual-reality-telepresence", "TELE-200-robotic-telepresence", "TELE-201-teleoperation-systems", "TELE-001-telepresence" ] }, { "id": "haptic-feedback", "title": "Haptic Feedback", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Tactile sensory technology that provides physical sensations to enhance virtual and augmented experiences by translating digital signals into mechanical vibrations, force responses, and pressure cues through actuators integrated into controllers, gloves, and body suits, enabling users to perceive touch, texture, and resistance when interacting with virtual objects.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:haptic-feedback", "labels": [ "Haptic Feedback", "HapticFeedback", "Tactile Feedback" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "dt:calibratedBy", "dt:controlledBy", "dt:enhances", "dt:integratedWith", "dt:optimizedBy", "enhancesExperience", "ForceFeedback", "integratsWith", "SensoryImmersion", "simulatesSensation", "TactileActuator", "TactileActuator", "TouchSimulation", "usesActuator", "WearableHaptics", "AISystem", "AugmentedReality", "Haptics", "MachineLearning", "MetaverseDomain" ] }, { "id": "haptics", "title": "Haptics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical hardware systems that simulate tactile sensations and force feedback within virtual environments through actuators and sensors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:haptics", "labels": [ "Haptics", "Wearable Haptics" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Force Feedback", "Force Feedback Actuators", "ISO 9241-960", "Low Latency Communication", "Physical Presence", "Piezoelectric Sensors", "Signal Processing Unit", "Tactile Actuators", "Tactile Feedback", "Texture Simulation", "Vibration Motors", "Driver Software", "Human Interface Device", "InteractionDomain", "PhysicalLayer", "Power Supply", "Real-time Processing" ] }, { "id": "hard-fork", "title": "Hard Fork", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Hard Fork is a backward-incompatible change to a blockchain protocol's consensus rules that permanently diverges the chain into two distinct networks if not universally adopted by all nodes. Because old software versions reject blocks produced under the new rules, a hard fork requires coordinated network-wide consensus to proceed without a chain split. Historical hard forks include the Ethereum/Ethereum Classic split in 2016 and the Bitcoin/Bitcoin Cash split in 2017.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:hard-fork", "labels": [ "Hard Fork" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "hard-money", "title": "Hard Money", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hard money refers to a form of money whose supply is difficult to expand, giving it a high stock-to-flow ratio and resistance to debasement. Historically embodied by gold, the concept extends to assets like Bitcoin whose issuance is algorithmically constrained. Hard money is valued as a store of value because its scarcity cannot be diluted by discretionary issuance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hard-money", "labels": [ "Hard Money", "Hard Money Property" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "hardhat", "title": "Hardhat", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hardhat is a professional Ethereum development environment that provides a comprehensive toolchain for compiling, deploying, testing, and debugging Solidity smart contracts. It ships with Hardhat Network, an in-process Ethereum node implementation designed for local development that supports forking mainnet state and emitting Solidity stack traces on failure. Hardhat's plugin architecture integrates tightly with ethers.js and Waffle, and its tasks system allows developers to automate bespoke deployment and verification workflows. It has become the dominant development framework in the Ethereum ecosystem, displacing earlier tools such as Truffle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hardhat", "labels": [ "Hardhat" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Solidity", "Smart Contract", "Ethereum" ] }, { "id": "hardware-abstraction-layer-hal", "title": "Hardware Abstraction Layer (HAL)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software interface that lets applications interact with hardware without device-specific code, providing a standardized abstraction between software and hardware components.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:hardware-abstraction-layer-hal", "labels": [ "Hardware Abstraction Layer (HAL)" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Abstraction Modules", "API Interfaces", "Device Drivers", "Device Portability", "MSF Taxonomy 2025", "Platform Independence", "Hardware Resources", "InfrastructureDomain", "Infrastructure Layer", "Network Layer", "Operating System", "Physical Layer", "Unified Hardware Access" ] }, { "id": "hardware-abstraction-layer", "title": "Hardware Abstraction Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Hardware Abstraction Layer (HAL) is a software layer that presents a uniform, hardware-independent interface to upper software layers \u2014 operating systems, middleware, or application code \u2014 while encapsulating the vendor-specific, register-level details of physical devices in the implementation beneath it. By isolating hardware dependencies behind a stable API, the HAL enables the same kernel or application binary to run on different processor architectures, microcontroller families, or peripheral configurations without source-code changes. HALs appear throughout the software stack: in embedded microcontroller SDKs (STM32 HAL, Arduino abstraction), in operating-system kernels (Windows HAL.dll), in robotic middleware (ROS Hardware Interface), and in graphics stacks (Vulkan's hardware abstraction over GPU vendors). The HAL pattern is a foundational principle of portable, maintainable system software.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:hardware-abstraction-layer", "labels": [ "Hardware Abstraction Layer", "HardwareAbstractionLayer" ], "is_subclass_of": [ "Hardware Abstraction" ], "wikilinks": [] }, { "id": "hardware-abstraction", "title": "Hardware Abstraction", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hardware Abstraction is the software engineering principle and architectural practice of interposing a uniform programmatic interface layer between high-level software and the physical characteristics of underlying hardware components, enabling software to operate independently of specific hardware implementations. The abstraction layer translates generic API calls into vendor-specific or device-specific commands, shielding operating systems, runtimes, and applications from the diversity of processor architectures, memory subsystems, I/O controllers, graphics units, and peripheral devices. This principle underpins portability, vendor independence, and long-term maintainability of system software stacks across the full spectrum of computing platforms, from embedded microcontrollers to cloud server farms to spatial-computing headsets. Hardware Abstraction Layers (HALs) are the canonical realisation of this principle in operating systems, device driver frameworks, and graphics APIs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:hardware-abstraction", "labels": [ "Hardware Abstraction" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "hardware-acceleration", "title": "Hardware Acceleration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The use of specialised hardware components, particularly GPUs and dedicated processors, to offload computationally intensive rendering, physics simulation, and AI workloads from the CPU, enabling real-time performance essential for immersive VR/AR metaverse experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:hardware-acceleration", "labels": [ "Hardware Acceleration" ], "is_subclass_of": [ "Display and Rendering", "Computing Infrastructure" ], "wikilinks": [ "Real-Time VR Performance", "Computing Infrastructure", "metaverse" ] }, { "id": "hardware-accelerator", "title": "Hardware Accelerator", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A hardware accelerator is a specialised electronic device designed to perform a particular class of computation far more efficiently than a general-purpose central processing unit. By dedicating silicon to highly parallel arithmetic such as matrix multiplication, accelerators dramatically increase throughput and energy efficiency for workloads like neural network training and inference. Common forms include graphics processing units, tensor processing units, field-programmable gate arrays and application-specific integrated circuits.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hardware-accelerator", "labels": [ "Hardware Accelerator" ], "is_subclass_of": [ "Hardware Acceleration" ], "wikilinks": [] }, { "id": "hardware-component", "title": "Hardware Component", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Hardware Component is a discrete physical element \u2014 such as a processor, memory module, sensor, display panel, network interface, or power management unit \u2014 that constitutes a functional building block within a computing or electronic system. Hardware components are characterised by their electrical specifications, mechanical form factor, thermal envelope, and interface standards, and their selection and integration collectively determine the throughput, latency, energy efficiency, and reliability of the host system. They are manufactured to industry or bespoke specifications and may be field-replaceable or permanently integrated at the board or package level. In the context of embedded, edge, and spatial-computing systems, the choice of hardware components directly constrains the achievable computational workloads, sensor modalities, and communication bandwidths.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:hardware-component", "labels": [ "Hardware Component" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "owl:Thing" ] }, { "id": "hardware-description-language", "title": "Hardware Description Language", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A hardware description language (HDL) is a specialised programming language used to describe the structure, behaviour and timing of digital electronic circuits at varying levels of abstraction. HDLs allow engineers to specify logic at the register-transfer level, which synthesis tools then translate into gate-level netlists for fabrication or for configuring reconfigurable devices such as FPGAs. The two dominant HDLs are Verilog and VHDL, complemented by higher-level and verification-oriented variants. By capturing concurrency and precise timing semantics, HDLs make digital designs simulatable, verifiable and reproducible before any silicon is committed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hardware-description-language", "labels": [ "Hardware Description Language" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "hardware-design", "title": "Hardware Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The engineering discipline of specifying, architecting, implementing, and verifying physical computing and electronic systems \u2014 from printed circuit boards and mechatronic assemblies to FPGAs and full-custom silicon \u2014 spanning requirements capture, architectural trade-off between performance, power, area, and cost, register-transfer-level description in hardware description languages, synthesis and physical implementation, and exhaustive pre-fabrication verification, since unlike software a shipped hardware error cannot be patched and a mask respin costs months and millions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:hardware-design", "labels": [ "Hardware Design", "HardwareDesign" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [ "Systems Engineering", "Hardware Description Language", "Formal Verification", "Semiconductor" ] }, { "id": "hardware-layer", "title": "Hardware Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Hardware Layer is the lowest stratum of the canonical stack, comprising the physical computing, storage, and signalling devices on which everything above runs. Nothing sits below it; immediately above it is the Network Layer, which connects discrete machines. It contains processors, memory, storage media, accelerators, and the physical transmission media that carry signals.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:hardware-layer", "labels": [ "Hardware Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Network Layer", "Compute Layer", "Computer Architecture", "Digital Signal Processing", "owl:Thing", "IEEE (Institute of Electrical and Electronics Engineers)" ] }, { "id": "hardware-platform-agnostic", "title": "Hardware Platform Agnostic", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A design principle and architectural quality ensuring that software, systems, or protocols can operate independently of specific hardware architectures or platform implementations. Hardware-agnostic systems achieve portability through abstraction layers\u2014such as virtual machines, containers, or cross-platform runtimes\u2014that decouple application logic from underlying physical or operating-system constraints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:hardware-platform-agnostic", "labels": [ "Hardware Platform Agnostic", "Hardware-/Platform-Agnostic", "Hardware-Agnostic XR Development" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "hardware-resources", "title": "Hardware Resources", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The physical computing infrastructure required for metaverse access and operation, encompassing VR/AR headsets, computing devices, display technologies, tracking sensors, and connectivity hardware that collectively enable immersive virtual experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:hardware-resources", "labels": [ "Hardware Resources" ], "is_subclass_of": [ "Platform and Environment", "Computing Infrastructure" ], "wikilinks": [ "Metaverse Access", "Computing Infrastructure", "metaverse" ] }, { "id": "hardware-root-of-trust", "title": "Hardware Root of Trust", "domain": "security", "domain_name": "Security", "definition": "A Hardware Root of Trust (HRoT) is an immutable, hardware-anchored set of functions and keys that a system inherently trusts and from which all higher-level security properties are derived. Implemented in silicon or a dedicated security chip, it provides the foundation for secure boot, measured boot, attestation and key protection by establishing a starting point that cannot be modified by software. Because every subsequent trust decision chains back to it, the integrity of the HRoT determines the trustworthiness of the entire platform.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hardware-root-of-trust", "labels": [ "Hardware Root of Trust" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "hardware-security-key", "title": "Hardware Security Key", "domain": "security", "domain_name": "Security", "definition": "A hardware security key is a dedicated physical authenticator \u2014 typically a USB, NFC, or Bluetooth device \u2014 that stores cryptographic keys and performs origin-bound public-key challenge responses to prove possession of a second factor. By keeping private keys in tamper-resistant hardware and signing only challenges scoped to the legitimate web origin, it provides strong resistance to phishing, credential replay, and man-in-the-middle attacks that defeat one-time-password methods. Hardware keys implement open standards such as FIDO2 and U2F and are a core enabler of passwordless and passkey-based authentication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hardware-security-key", "labels": [ "Hardware Security Key" ], "is_subclass_of": [ "Multi-Factor Authentication" ], "wikilinks": [] }, { "id": "hardware-security-module", "title": "hardware security module", "domain": "security", "domain_name": "Security", "definition": "A Hardware Security Module (HSM) is a dedicated, tamper-evident and tamper-resistant hardware appliance that generates, stores, and manages cryptographic keys in a physically protected environment, performing sensitive cryptographic operations\u2014such as digital signing, bulk encryption, key derivation, and random-number generation\u2014entirely within its secure boundary so that plaintext key material is never exposed to the host system. HSMs are validated against formal security standards including FIPS 140-2/140-3 (Levels 1\u20134) and Common Criteria EAL4+, and are mandated by payment card schemes (PCI-DSS, PCI-P2PE), certificate authority trust frameworks, and government PKI and national-security infrastructures. They are available as PCIe cards, rack-mounted network appliances, USB tokens, and cloud-hosted dedicated services, all exposing a standardised PKCS#11 (Cryptoki) API. Unlike software keystores or Trusted Execution Environments, HSMs respond to physical tamper events by irreversibly zeroing all stored key material, making them the highest-assurance key-custody mechanism in mainstream deployment.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:hardware-security-module", "labels": [ "Hardware Security Module", "Hardware Security Module (optional)" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "hardware-security", "title": "Hardware Security", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Hardware security protects systems by building trust and isolation into physical components, using features such as secure enclaves, key storage and tamper resistance to defend against attacks software alone cannot stop.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hardware-security", "labels": [ "Hardware Security" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [ "Computer Hardware", "Trusted Execution Environment", "Hardware Security Module", "Cryptography" ] }, { "id": "hardware-wallet", "title": "Hardware Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A hardware wallet is a dedicated physical device that stores cryptocurrency private keys in a secure element and signs transactions internally, so the keys never leave the device or touch an internet-connected computer. It protects against malware and remote key theft by isolating signing operations behind on-device confirmation. Hardware wallets are a foundational tool for self-custody of digital assets.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:hardware-wallet", "labels": [ "Hardware Wallet" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "hardware-and-edge", "title": "Hardware and Edge", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hardware and Edge refers to the integrated class of specialised silicon architectures, embedded compute platforms, and distributed inference runtimes designed to execute artificial intelligence workloads at or near the point of data generation, without mandatory dependence on centralised cloud da...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hardware-and-edge", "labels": [ "Hardware and Edge" ], "is_subclass_of": [ "Computing and Cloud", "Distributed Computing", "Edge Computing", "AI Accelerator", "Embedded Systems", "Neural Processing Unit" ], "wikilinks": [ "AI Accelerator", "AIDeploymentDomain", "Anomaly Detection", "Apache TVM", "Apple Neural Engine", "ARM Architecture", "ARM Ethos NPU", "Autonomous Systems", "Centralised Inference", "ComputeHardwareDomain", "CPU-Only Inference", "CUDA", "Data Centre GPU", "Embedded Systems", "ETSI MEC", "ExecuTorch", "FirmwareLayer", "Google Coral Edge TPU", "Hailo Accelerator", "Hardware Abstraction Layer" ] }, { "id": "hardware-in-the-loop-testing", "title": "Hardware-in-the-Loop Testing", "domain": "robotics", "domain_name": "Robotics", "definition": "Hardware-in-the-loop (HIL) testing is a validation technique in which real physical hardware, such as a controller or actuator, is connected to a real-time simulation of the rest of the system. It lets engineers exercise embedded control software against realistic, repeatable plant dynamics without risking expensive or dangerous full-system runs. HIL is standard practice in robotics, automotive, and aerospace development.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hardware-in-the-loop-testing", "labels": [ "Hardware-in-the-Loop Testing", "Hardware-in-the-Loop" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "hardware", "title": "Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Hardware denotes the physical, tangible components of a computing system \u2014 processors, memory modules, storage devices, input/output peripherals, power subsystems, and interconnects \u2014 that collectively provide the substrate on which software executes. In the spatial-computing and infrastructure context it encompasses both general-purpose devices (CPUs, GPUs, FPGAs, ASICs) and specialised sensors and actuators (depth cameras, inertial measurement units, haptic controllers, head-mounted displays) that enable immersive and intelligent workloads. Hardware capability sets the fundamental performance, latency, power, and thermal envelope that software stacks must respect, making hardware design decisions inseparable from system architecture, operating-system abstractions, and application-level trade-offs. The discipline spans semiconductor fabrication, PCB design, embedded firmware, hardware security primitives, and the standards that allow heterogeneous devices to interoperate.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:hardware", "labels": [ "Hardware", "HardwareInventory" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "harm-taxonomy", "title": "Harm Taxonomy", "domain": "governance", "domain_name": "Governance", "definition": "A harm taxonomy is a structured classification of the potential negative impacts arising from a technology, used to organise risk assessment and mitigation. In AI it categorises harms such as misinformation, discrimination, privacy violation, manipulation, and physical or economic damage. A clear taxonomy enables systematic red-teaming, policy mapping, and accountability for deployed systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:harm-taxonomy", "labels": [ "Harm Taxonomy" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "harmful-bias", "title": "Harmful Bias", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Systematic and unjust discrimination in AI system outcomes that disadvantages individuals or groups based on protected characteristics (race, gender, age, disability, religion, etc.) or other sensitive attributes, resulting in material harm, dignity violations, or perpetuation of societal inequal...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:harmful-bias", "labels": [ "Harmful Bias" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "harmlessness", "title": "Harmlessness", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An alignment objective ensuring AI systems avoid generating outputs that could cause harm, including toxic, dangerous, misleading, or unethical content. Harmlessness is one of the three core alignment dimensions alongside helpfulness and honesty, implemented through techniques such as Constitutional AI and RLHF to constrain model behaviour without sacrificing utility.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:harmlessness", "labels": [ "Harmlessness" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "\ud83e\udd16", "MetaverseDomain", "Research Tools", "Sam Hammond" ] }, { "id": "harmonic-drive", "title": "Harmonic Drive", "domain": "robotics", "domain_name": "Robotics", "definition": "A harmonic drive is a strain-wave gearing mechanism that achieves very high reduction ratios in a compact, lightweight package with near-zero backlash. It uses a flexible spline deformed by an elliptical wave generator to mesh with a rigid outer gear, transmitting motion with high precision. Harmonic drives are widely used in robot joints, where accuracy and torque density are critical.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:harmonic-drive", "labels": [ "Harmonic Drive" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "harmonised-standard", "title": "Harmonised Standard", "domain": "robotics", "domain_name": "Robotics", "definition": "A harmonised standard is a technical specification developed by a recognised European standards organisation and published in the Official Journal of the EU, compliance with which grants a presumption of conformity with the essential requirements of an EU directive or regulation. For robotics and machinery, harmonised standards translate broad legal requirements, such as those in the Machinery Regulation, into concrete, testable engineering criteria used to prepare a declaration of conformity. Manufacturers are not legally obliged to follow a harmonised standard, but doing so is the practical route to CE marking since deviation requires demonstrating equivalent safety by other means. They are maintained by bodies such as CEN and CENELEC and revised periodically.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:harmonised-standard", "labels": [ "Harmonised Standard" ], "is_subclass_of": [ "Robot Standard" ], "wikilinks": [] }, { "id": "harness-configuration-packs", "title": "Harness Configuration Packs", "domain": "ai", "domain_name": "Ai", "definition": "Skill packs, slash-command libraries, and meta-prompting frameworks that enhance existing coding agents with structured workflows, multi-role capabilities, and progressive disclosure \u2014 includes Superpowers, Everything Claude Code, GStack, Anthropic Skills, Get-Shit-Done (GSD), wshobson/agents, and pmstack.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:harness-configuration-packs", "labels": [ "Harness Configuration Packs" ], "is_subclass_of": [ "Agent Harness", "Agent Frameworks", "Agentic AI", "Prompt Engineering" ], "wikilinks": [ "Agent Harness", "IDE Coding Agents", "Terminal Coding Agents", "Internal AI Harness", "External AI Harness", "Model Context Protocol", "Large Language Model", "Prompt Engineering", "Agent Frameworks", "Agentic AI", "Autonomous Coding", "Hook System", "Tool Use", "Multi-Agent Orchestration Frameworks", "Plan-and-Execute Pattern", "Context Window", "Workflow Automation", "Multi-Agent Collaboration", "Agent Evaluation Benchmarks", "Prompt Template" ] }, { "id": "hash-collision", "title": "Hash Collision", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A hash collision occurs when two distinct inputs produce the same output from a cryptographic hash function, violating the collision-resistance property that is essential to blockchain data integrity. In blockchain systems, collision resistance ensures that no adversary can craft two different transactions or blocks yielding the same hash digest, making Merkle tree roots and block headers tamper-evident. While collisions are computationally infeasible for production-grade functions such as SHA-256, their theoretical possibility drives ongoing cryptographic research and post-quantum security planning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hash-collision", "labels": [ "Hash Collision" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "hash-function", "title": "Hash Function", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Hash Function is a deterministic computational mapping H: {0,1}* \u2192 {0,1}^n from arbitrary-length input strings (preimages, messages) to fixed-length output strings (digests, hashes, fingerprints) of n bits (typically n \u2208 {128, 160, 224, 256, 384, 512}), whose security and utility derive fro...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hash-function", "labels": [ "Hash Function", "Hash-Function", "HashFunction" ], "is_subclass_of": [ "Cryptographic Primitive", "Compression Function", "Cryptographic Algorithm", "One-Way Function", "Deterministic Function", "Digital Fingerprint" ], "wikilinks": [ "Albrecht Grassi Rechberger Roy Tiessen 2016 MiMC ASIACRYPT", "Aumasson et al. 2013 BLAKE2 ACNS", "Avalanche Effect", "Bellare & Rogaway 1993 Random Oracle Model", "Bellare Canetti Krawczyk 1996 HMAC CRYPTO", "Bernstein et al. 2019 SPHINCS+ ACM CCS", "Bertoni Daemen Peeters Van Assche 2007 Sponge Functions", "Bertoni et al. 2013 Keccak EUROCRYPT", "Bertoni et al. 2016 KangarooTwelve", "Birthday Attack", "Biryukov Dinu Khovratovich 2016 Argon2 EuroS&P", "Bitcoin Whitepaper", "Bitwise Operations", "Blockchain Immutability", "Bloom Filter", "Boolean Algebra", "Certificate Transparency", "Checksum", "Code Signing", "Commitment Scheme" ] }, { "id": "hash-functions", "title": "Hash Functions", "domain": "security", "domain_name": "Security", "definition": "Hash functions are algorithms that map data of arbitrary size to a fixed-size output, with cryptographic hash functions designed to be one-way and collision resistant. They are central to security and blockchains.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hash-functions", "labels": [ "Hash Functions" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Merkle Tree", "Transaction Validation", "Hash Function", "Cryptography", "https://csrc.nist.gov/projects/hash-functions", "https://en.wikipedia.org/wiki/Cryptographic_hash_function" ] }, { "id": "hash-rate", "title": "Hash Rate", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hash rate is the aggregate computational throughput of a proof-of-work blockchain network, measured as the number of hash function evaluations performed per unit of time across all participating mining nodes. It serves as the primary quantitative indicator of a network's security: a higher hash rate means an attacker must control and operate more hardware to execute a 51% attack, making double-spend attacks proportionally more expensive. Hash rate is typically expressed in hashes per second (H/s) with SI prefixes (kH/s, MH/s, GH/s, TH/s, PH/s, EH/s) and fluctuates with the entry or exit of miners, hardware efficiency improvements, and changes in mining profitability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hash-rate", "labels": [ "Hash Rate", "HashRate" ], "is_subclass_of": [ "Proof Of Work" ], "wikilinks": [] }, { "id": "hash-time-locked-contract", "title": "Hash Time-Locked Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Hash Time-Locked Contract (HTLC) is a type of smart contract that conditionally releases funds to a recipient only if they present a valid cryptographic preimage satisfying a specified hash condition within a defined time window; if the condition remains unmet before the timeout expires, the funds automatically revert to the sender. HTLCs combine two complementary mechanisms\u2014a hashlock, which binds settlement to knowledge of a secret, and a timelock, which enforces a bounded settlement window\u2014to achieve trustless atomicity across one or more blockchain ledgers. They are foundational to payment channel networks such as the Lightning Network and to cross-chain atomic swap protocols, enabling conditional payment routing without custodial intermediaries or mutual trust. As a composable on-chain primitive, HTLCs underpin a wide range of decentralised finance and interoperability constructs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hash-time-locked-contract", "labels": [ "Hash Time-Locked Contract", "Hash Time Locked Contract", "Hash Timelock Contract", "Hashed Time-Lock Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "hash-time-locked-contracts", "title": "Hash Time-Locked Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A type of smart contract that conditions a payment on the recipient revealing a cryptographic preimage before a deadline, enabling trustless conditional and cross-chain transfers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hash-time-locked-contracts", "labels": [ "Hash Time-Locked Contracts" ], "is_subclass_of": [ "Smart Contracts" ], "wikilinks": [ "Cryptographic Hash", "Timelock", "Atomic Swap", "Lightning Network", "Smart Contracts" ] }, { "id": "hashcash", "title": "Hashcash", "domain": "security", "domain_name": "Security", "definition": "Hashcash is a proof-of-work system invented by Adam Back in 1997 that requires a sender to compute a partial SHA-1 (later SHA-256) hash collision by finding a nonce such that the resulting digest has a specified number of leading zero bits. The asymmetry between expensive computation and cheap verification makes it suitable as an anti-abuse token: the sender bears a measurable cost while the recipient verifies in microseconds. Originally designed to combat email spam and denial-of-service attacks, the mechanism was directly adopted by Bitcoin as the basis of its mining consensus algorithm, making Hashcash one of the foundational primitives of the blockchain era.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:hashcash", "labels": [ "Hashcash" ], "is_subclass_of": [ "Proof Of Work" ], "wikilinks": [ "Hash Function", "Bitcoin Protocol", "Mining", "Cryptographic Hash Function" ] }, { "id": "hashed-timelock-contract", "title": "Hashed Timelock Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Hashed Timelock Contract (HTLC) is a conditional payment construct that locks funds until either a recipient reveals a preimage matching a published hash, or a timeout elapses and the funds revert to the sender. By combining a hashlock with a timelock, it enables trustless, atomic transfers without a central intermediary. HTLCs are the foundational primitive behind Lightning Network payment routing and cross-chain atomic swaps.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hashed-timelock-contract", "labels": [ "Hashed Timelock Contract" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [] }, { "id": "hateoas", "title": "Hateoas", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "HATEOAS, hypermedia as the engine of application state, is the REST constraint requiring that a client interact with an application entirely through hypermedia links and controls supplied dynamically by the server in its responses. Rather than hard-coding endpoint structures, the client discovers available actions and transitions at runtime by following links the server provides, mirroring how a browser navigates the web. This constraint decouples clients from fixed URI schemes and is the distinguishing feature of a fully RESTful, self-describing API.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hateoas", "labels": [ "Hateoas", "HATEOAS" ], "is_subclass_of": [ "REST" ], "wikilinks": [] }, { "id": "hazard-analysis", "title": "Hazard Analysis", "domain": "governance", "domain_name": "Governance", "definition": "Hazard analysis is the systematic identification, characterisation, and prioritisation of conditions or events with the potential to cause harm to people, systems, or environments, forming the foundational step in safety engineering processes that design controls to prevent or mitigate those harms. Methods range from qualitative techniques such as HAZOP and FMEA to quantitative probabilistic risk assessment, applied across aviation, automotive, nuclear, medical devices, and increasingly AI systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:hazard-analysis", "labels": [ "Hazard Analysis", "Hazard Analysis and Critical Control Points" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "hazard-identification", "title": "Hazard Identification", "domain": "robotics", "domain_name": "Robotics", "definition": "Hazard identification is the systematic process of recognising and cataloguing sources of potential harm within a system, task, or environment before they cause injury, damage, or loss. It is the first and enabling stage of risk assessment, producing the list of hazards that downstream analysis, evaluation, and mitigation depend on. In robotics it underpins functional-safety compliance by exposing mechanical, electrical, control, and human-interaction hazards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hazard-identification", "labels": [ "Hazard Identification" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "head-related-transfer-function", "title": "Head Related Transfer Function", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A head-related transfer function (HRTF) describes how sound from a point in space is filtered by the listener's head, torso and outer ears before reaching each eardrum. It captures the frequency-dependent level, time and spectral cues that the auditory system uses to localise sound. Convolving a mono source with the appropriate left and right HRTFs synthesises a convincing three-dimensional position over headphones, making the HRTF the mathematical core of binaural and spatial audio rendering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:head-related-transfer-function", "labels": [ "Head Related Transfer Function", "Head-Related Transfer Function" ], "is_subclass_of": [ "Spatial Audio" ], "wikilinks": [] }, { "id": "head-mounted-display", "title": "Head-Mounted Display", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A head-mounted display (HMD) is a wearable optoelectronic device worn on the head that positions one or more display panels or optical projectors in front of the user's eyes, delivering immersive visual content for virtual reality, augmented reality, or mixed reality applications. HMDs incorporate dedicated optics \u2014 ranging from Fresnel lenses and pancake optics to diffractive waveguides \u2014 to focus near-focal-plane displays at a perceptually comfortable vergence distance, alongside inertial and visual tracking systems that measure head orientation and six-degrees-of-freedom position in real time. The design space spans sealed VR systems that fully occlude the real world, optical see-through AR waveguide systems that superimpose digital graphics onto the physical environment, and video-passthrough mixed reality headsets that combine high-fidelity camera streams with real-time scene reconstruction. HMDs serve as the primary platform through which users perceive and interact with spatial computing environments.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:head-mounted-display", "labels": [ "Head-Mounted Display", "Head Mounted Display" ], "is_subclass_of": [ "XR Headset" ], "wikilinks": [] }, { "id": "health-check", "title": "Health Check", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A health check is a periodic probe that determines whether a service instance is functioning correctly and able to handle requests. Health checks distinguish between liveness, whether a process is running and should be restarted if not, and readiness, whether it is prepared to receive traffic. Load balancers, orchestrators, and service meshes use health checks to route around failed instances, trigger restarts, and gate traffic, making them a foundational primitive for high availability and self-healing distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:health-check", "labels": [ "Health Check" ], "is_subclass_of": [ "High Availability" ], "wikilinks": [] }, { "id": "health-metaverse-application", "title": "Health Metaverse Application", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A specialized virtual platform integrating healthcare delivery, medical training, therapeutic interventions, and patient engagement through immersive environments that comply with health data regulations and clinical standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:health-metaverse-application", "labels": [ "Health Metaverse Application" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "3D Medical Imaging", "Biometric Sensor Integration", "Clinical AI", "Diagnostic Interface", "DICOM", "End-to-End Encryption", "FDA Digital Health", "Haptic Feedback System", "Health Record System", "HL7 FHIR", "Medical Education", "Medical Simulation", "Mental Health Therapy", "OpenXR Healthcare", "Patient Portal", "Rehabilitation Program", "Surgical Training", "Telemedicine", "Therapy Environment", "ApplicationLayer" ] }, { "id": "health-monitoring", "title": "Health Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Health monitoring is the continuous observation of a system's components, services, and dependencies to determine whether they are operating correctly and are able to serve requests. It uses signals such as heartbeats, readiness and liveness probes, resource metrics, and synthetic checks to produce a real-time view of system health. Health monitoring underpins high availability by enabling automated detection of failures and triggering recovery, failover, or load redistribution.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:health-monitoring", "labels": [ "Health Monitoring" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "healthcare-ai", "title": "healthcare ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Healthcare AI is the systematic application of machine learning, natural language processing, computer vision, and large language models to clinical and operational problems in medicine\u2014encompassing diagnostic imaging analysis, clinical decision support, drug discovery, genomic interpretation, patient outcome prediction, and administrative automation. The field operates under stringent regulatory oversight (FDA, MHRA, EU MDR/IVDR) requiring prospective clinical validation, post-market surveillance, and clearly defined human-AI workflow integration to ensure patient safety. Fairness, explainability, and bias auditing are central technical and ethical concerns given the high-stakes nature of clinical decisions, where model underperformance across demographic subgroups can directly harm patients. Federated learning, differential privacy, and synthetic data generation are increasingly adopted to enable model training without centralising sensitive patient records.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:healthcare-ai", "labels": [ "Healthcare AI" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "healthcare-analytics", "title": "Healthcare Analytics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence and data science to healthcare data for population health management, operational efficiency optimisation, clinical outcome prediction, resource allocation, and healthcare policy decision-making. Healthcare analytics systems analyse electronic health records, claims data, public health datasets, and operational metrics to derive actionable insights for clinical and administrative improvement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:healthcare-analytics", "labels": [ "Healthcare Analytics" ], "is_subclass_of": [ "AI Application", "Machine Learning Discipline" ], "wikilinks": [ "Population Health", "Data Analytics", "Medical AI", "MetaverseDomain", "Update Cycle" ] }, { "id": "healthcare-application-classification", "title": "Healthcare Application Classification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A taxonomy framework categorising metaverse healthcare applications by technology type (AR, VR, lifelogging, mirror world), use case (telemedicine, training, therapy), and end user (patient, clinician), enabling systematic eand deployment of immersive medical technologies.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:healthcare-application-classification", "labels": [ "Healthcare Application Classification" ], "is_subclass_of": [ "AI Application", "Healthcare Technology" ], "wikilinks": [ "Structured Healthcare Innovation", "Healthcare Technology", "metaverse" ] }, { "id": "healthcare-records", "title": "Healthcare Records", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based electronic health record systems that employ immutable distributed ledgers, smart contracts for consent management, and cryptographic security to enable secure patient data sharing across healthcare providers. These systems give patients controlled access to their own records whilst maintaining HIPAA and GDPR compliance through hybrid on-chain/off-chain architectures integrating HL7 FHIR standards.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:healthcare-records", "labels": [ "Healthcare Records" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BC-0426-hyperledger-fabric", "BC-0456-self-sovereign-identity", "BC-0457-decentralized-identifiers", "BC-0458-verifiable-credentials", "BC-0459-digital-identity-wallet", "BC-0476-aml-kyc-compliance", "BC-0492-clinical-trials", "BlockchainDomain" ] }, { "id": "healthcare-technology", "title": "Healthcare Technology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Healthcare Technology encompasses the digital tools, software systems, and AI-driven applications deployed in clinical and health management contexts, including medical imaging, electronic health records, clinical decision support, and telemedicine platforms. It bridges spatial computing, data analytics, and AI to improve diagnostic accuracy, treatment outcomes, and health system efficiency.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:healthcare-technology", "labels": [ "Healthcare Technology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "healthcare", "title": "Healthcare", "domain": "ai", "domain_name": "Ai", "definition": "Healthcare is the organised provision of medical services, preventive care, diagnostics, treatment, and rehabilitation to individuals and populations. It encompasses clinical practice, health informatics, medical devices, pharmaceutical supply chains, and public health systems. As an application domain for AI and spatial computing, healthcare is distinguished by stringent regulatory requirements, sensitivity of patient data, and direct impact on human wellbeing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:healthcare", "labels": [ "Healthcare" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "heartbeat-mechanism", "title": "Heartbeat Mechanism", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A heartbeat mechanism is a distributed-systems technique in which a node periodically sends a lightweight signal to its peers or a coordinator to indicate that it remains alive and responsive, with the absence of expected heartbeats within a timeout interval treated as evidence of failure. It underlies failure detection in health checks, cluster membership protocols and consensus algorithms such as Raft, where a leader's heartbeats suppress follower election timeouts and its absence triggers a new leader election. The heartbeat interval and timeout must be tuned to balance fast failure detection against false positives caused by transient network delay.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:heartbeat-mechanism", "labels": [ "Heartbeat Mechanism" ], "is_subclass_of": [ "Distributed System Protocol" ], "wikilinks": [] }, { "id": "heat-pipe", "title": "Heat Pipe", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:heat-pipe", "labels": [ "Heat Pipe" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "heatmap-regression", "title": "Heatmap Regression (Keypoint Localisation)", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A dense prediction technique for keypoint localisation in which a neural network outputs a 2D likelihood map per landmark instead of regressing coordinates directly. Ground-truth targets are rendered as Gaussian peaks centred on each keypoint, giving a spatially smooth supervision signal that preserves the convolutional structure of the feature maps. The final location is decoded from the argmax (or a sub-pixel refinement) of each predicted heatmap. It is the dominant formulation for human pose estimation, facial landmark detection, and anatomical landmark localisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:heatmap-regression", "labels": [ "Heatmap Regression (Keypoint Localisation)", "Heatmap Regression" ], "is_subclass_of": [ "Regression" ], "wikilinks": [ "Regression", "Keypoint Detection", "Pose Estimation", "Bounding Box Regression", "Non Maximum Suppression" ] }, { "id": "risk-intensity-heatmap-regression", "title": "Heatmap Regression", "domain": "ai", "domain_name": "Ai", "definition": "Heatmap regression is a computer-vision technique for keypoint localisation in which a network predicts a 2D probability map per landmark rather than directly regressing coordinates. The peak of each predicted heatmap indicates the most likely location, and Gaussian-blurred ground-truth targets make training spatially smooth and robust. It is the dominant approach for human pose estimation and facial landmark detection.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-intensity-heatmap-regression", "labels": [ "Heatmap Regression" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "hedera-hashgraph", "title": "Hedera Hashgraph", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hedera is a public distributed ledger that uses the hashgraph consensus algorithm based on gossip about gossip and virtual voting. It is governed by a council of organisations and uses the HBAR token.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hedera-hashgraph", "labels": [ "Hedera Hashgraph", "Hashgraph" ], "is_subclass_of": [ "Distributed Ledger Technology" ], "wikilinks": [ "Consensus Algorithm", "Byzantine Fault Tolerance", "Cryptocurrency", "Smart Contract", "Distributed Ledger Technology" ] }, { "id": "hedging", "title": "Hedging", "domain": "finance", "domain_name": "Finance", "definition": "A risk management practice in which an exposure to adverse price, rate, or credit movements is deliberately offset by taking a counterbalancing position \u2014 typically in derivatives such as futures, options, and swaps, or in correlated instruments \u2014 so that losses on the primary holding are compensated by gains on the hedge; the aim is not profit but the reduction of variance, exchanging upside potential for predictability of outcomes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:hedging", "labels": [ "Hedging" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [ "Risk Management", "Market Making", "Synthetic Asset", "Exchange Rate" ] }, { "id": "heliopause", "title": "Heliopause", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:heliopause", "labels": [ "Heliopause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "heliophysics", "title": "Heliophysics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:heliophysics", "labels": [ "Heliophysics" ], "is_subclass_of": [ "Space Science" ], "wikilinks": [] }, { "id": "heliosphere", "title": "Heliosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:heliosphere", "labels": [ "Heliosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "helpfulness", "title": "Helpfulness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An alignment objective ensuring AI systems provide useful, relevant, and informative responses to user queries. Helpfulness represents a key dimension of AI utility that must be balanced against harmlessness and honesty.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:helpfulness", "labels": [ "Helpfulness" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Alignment" ], "wikilinks": [ "MUST", "MetaverseDomain" ] }, { "id": "heuristic-clustering", "title": "Heuristic Clustering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Heuristic clustering, in blockchain forensics, is the practice of grouping addresses that are inferred to be controlled by the same entity using behavioural heuristics such as common-input-ownership or change-address detection, rather than direct cryptographic proof of common control. It is the core analytic technique used by chain-analysis firms to de-anonymise transaction graphs and attribute activity to real-world actors. Its outputs are probabilistic, and heuristics can misfire against wallets that deliberately break the assumed patterns.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:heuristic-clustering", "labels": [ "Heuristic Clustering" ], "is_subclass_of": [ "Clustering" ], "wikilinks": [] }, { "id": "heuristic-evaluation", "title": "Heuristic Evaluation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Heuristic evaluation is a usability inspection method in which a small number of expert evaluators judge an interface against a set of recognised usability principles, or heuristics, to identify usability problems. It is a discount technique that requires no test participants, producing a ranked list of issues with severity estimates. Because it depends on evaluator expertise rather than observed user behaviour, it complements, rather than replaces, empirical usability testing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:heuristic-evaluation", "labels": [ "Heuristic Evaluation" ], "is_subclass_of": [ "Usability" ], "wikilinks": [] }, { "id": "heuristic-function", "title": "Heuristic Function", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A heuristic function is an estimating function used in search and optimisation that approximates the cost or distance from a given state to a goal state. It guides informed search algorithms by prioritising the exploration of states that appear most promising, trading guaranteed optimality for improved efficiency. A heuristic is admissible when it never overestimates the true cost, and consistent when it satisfies the triangle inequality, properties that determine the optimality guarantees of algorithms that use it.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:heuristic-function", "labels": [ "Heuristic Function" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [] }, { "id": "heuristic-methods", "title": "Heuristic Methods", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Heuristic methods in AI are problem-solving approaches that employ practical, experience-based techniques to find satisfactory solutions when optimal solutions are computationally infeasible. They include search heuristics such as A* and hill climbing, rule-of-thumb strategies, and metaheuristics such as genetic algorithms and simulated annealing, trading completeness for efficiency in combinatorial optimisation and planning tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:heuristic-methods", "labels": [ "Heuristic Methods", "Heuristic", "Heuristic Functions" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Metaheuristics", "Optimization", "Planning", "Search Algorithms" ] }, { "id": "heuristic-search", "title": "Heuristic Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Heuristic search is a family of search algorithms that use a problem-specific evaluation function to estimate the cost or promise of candidate states, focusing exploration on the most promising regions of a search space. By trading exhaustive coverage for informed guidance, heuristic methods such as A* and best-first search solve large combinatorial problems that are intractable for uninformed search. The quality of the heuristic determines both efficiency and, in admissible cases, optimality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:heuristic-search", "labels": [ "Heuristic Search" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [] }, { "id": "heuristics", "title": "Heuristics", "domain": "ai", "domain_name": "Ai", "definition": "Mental shortcuts or simple decision rules that trade completeness and guaranteed optimality for speed and low cognitive or computational cost. In psychology, heuristics such as availability, representativeness, and anchoring explain how humans judge under uncertainty; in computer science, heuristic functions guide search and optimisation algorithms towards good solutions when exhaustive evaluation is intractable. Heuristics are adaptive in the environments they evolved for but produce systematic biases outside them.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:heuristics", "labels": [ "Heuristics" ], "is_subclass_of": [ "Decision Making" ], "wikilinks": [ "Decision Making", "Bounded Rationality", "Decision Theory", "Cognitive Psychology" ] }, { "id": "hidden-dimension", "title": "Hidden Dimension", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The dimensionality of the internal representations in a neural network, determining the capacity of each layer to encode information, typically denoted as d_model in transformers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:hidden-dimension", "labels": [ "Hidden Dimension" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "hidden-hand", "title": "Hidden Hand", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hidden Hand is a governance incentive marketplace operated by Redacted Cartel that lets protocols offer rewards to direct gauge votes across multiple DeFi systems. It generalises the bribery market model beyond a single protocol.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hidden-hand", "labels": [ "Hidden Hand" ], "is_subclass_of": [ "Gauge Voting" ], "wikilinks": [ "Gauge Voting", "Governance Token", "Tokenomics", "Votium" ] }, { "id": "hidden-layer", "title": "Hidden Layer", "domain": "ai", "domain_name": "Ai", "definition": "A hidden layer is an intermediate layer of neurons in a neural network situated between the input and output layers, whose activations are not directly observed. Each hidden layer applies a learned linear transformation followed by a non-linear activation, building increasingly abstract feature representations. Stacking multiple hidden layers is what gives deep networks their representational power.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hidden-layer", "labels": [ "Hidden Layer" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "hidden-markov-model", "title": "Hidden Markov Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A hidden Markov model (HMM) is a probabilistic model for sequences in which an unobserved Markov chain of discrete states generates observable outputs, one per state, according to state-dependent emission distributions. The model is defined by transition probabilities between hidden states and emission probabilities for observations, plus an initial state distribution. HMMs support efficient inference for filtering, decoding, and learning over sequential data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hidden-markov-model", "labels": [ "Hidden Markov Model" ], "is_subclass_of": [ "Graphical Model", "AI Technique" ], "wikilinks": [] }, { "id": "hidden-state", "title": "Hidden State", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The vector representation of a token or sequence at any layer in a neural network, encoding contextualised information learned by the model. Hidden states are progressively refined through self-attention and feed-forward transformations, with deeper layers capturing increasingly abstract semantic features used in downstream tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:hidden-state", "labels": [ "Hidden State" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "hierarchical-deterministic-wallet", "title": "Hierarchical Deterministic Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A hierarchical deterministic (HD) wallet is a cryptocurrency wallet that derives a tree of key pairs from a single master seed using a deterministic algorithm, as standardised in BIP-32 and related proposals. From one human-readable mnemonic phrase the wallet can regenerate an effectively unlimited hierarchy of addresses, allowing backup of an entire wallet from a single seed. The structure improves privacy by using fresh addresses while keeping recovery and organisation manageable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hierarchical-deterministic-wallet", "labels": [ "Hierarchical Deterministic Wallet" ], "is_subclass_of": [ "Blockchain Wallet" ], "wikilinks": [] }, { "id": "hierarchical-organisation", "title": "Hierarchical Organisation", "domain": "governance", "domain_name": "Governance", "definition": "A governance and coordination structure in which authority, decision-making, and information flow are arranged in ranked tiers, with each level reporting upward to a narrower level above it, culminating in a single apex of control; the dominant organisational pattern of firms, bureaucracies, and traditional software architectures, offering clear accountability and command efficiency at the cost of communication bottlenecks, single points of failure, and reduced local autonomy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:hierarchical-organisation", "labels": [ "Hierarchical Organisation" ], "is_subclass_of": [ "Governance Structure" ], "wikilinks": [ "Governance Structure", "Centralised Control", "Decentralised Coordination" ] }, { "id": "hierarchical-task-network", "title": "Hierarchical Task Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A hierarchical task network (HTN) is an automated planning formalism in which planning proceeds by recursively decomposing high-level compound tasks into networks of smaller subtasks until only primitive, directly executable actions remain. Decomposition is guided by domain-specific methods that encode expert knowledge about how tasks may be accomplished, together with ordering constraints between subtasks. HTN planning contrasts with classical state-space planning by searching over task decompositions rather than over world states alone, which often yields stronger guidance and greater efficiency in well-structured domains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hierarchical-task-network", "labels": [ "Hierarchical Task Network" ], "is_subclass_of": [ "Automated Planning" ], "wikilinks": [] }, { "id": "high-availability", "title": "high availability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "High Availability (HA) is an infrastructure design property that specifies and enforces the conditions under which a system continues delivering its intended service despite component failures, maintenance windows, or demand spikes. HA is quantified by availability targets expressed as 'nines' \u2014 for example 99.9% permits roughly 8.7 hours of downtime per year, while 99.999% ('five nines') permits only 5.26 minutes \u2014 achieved through redundant components, active-active or active-passive failover, continuous health monitoring, and automated recovery orchestration. It is operationalised through the complementary metrics of Mean Time Between Failures (MTBF) and Mean Time To Recovery (MTTR), with recovery time objectives (RTO) and recovery point objectives (RPO) anchoring HA targets to business continuity requirements. HA is a foundational non-functional requirement in cloud-native platforms, telecommunications networks, financial trading systems, and safety-critical infrastructure where service interruption carries regulatory, commercial, or life-safety consequences.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:high-availability", "labels": [ "High Availability" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "high-bandwidth-interconnect", "title": "High Bandwidth Interconnect", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A high-bandwidth interconnect is a communication link or fabric engineered to move very large volumes of data between processors, accelerators, or nodes with minimal latency, enabling tightly-coupled parallel computation. In machine learning it is the substrate over which gradients, activations, and parameters are exchanged during distributed training, directly bounding how efficiently models scale across many devices. Technologies such as NVLink, InfiniBand, and RDMA fabrics provide the throughput that large-scale training demands.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:high-bandwidth-interconnect", "labels": [ "High Bandwidth Interconnect", "High-Bandwidth Interconnect" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [] }, { "id": "high-bandwidth-memory", "title": "High Bandwidth Memory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "High Bandwidth Memory (HBM) is a 3D-stacked DRAM technology that places multiple memory dies on a silicon interposer adjacent to a processor, delivering very wide buses and high memory bandwidth at low power per bit. It is the standard memory for AI accelerators and GPUs where feeding compute units with data is the dominant bottleneck. HBM enables training and inference of large models that would otherwise be memory-bandwidth bound.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:high-bandwidth-memory", "labels": [ "High Bandwidth Memory", "HBM Memory", "High-Bandwidth Memory" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "high-earth-orbit", "title": "High Earth Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:high-earth-orbit", "labels": [ "High Earth Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "high-energy-consumption", "title": "High Energy Consumption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "High Energy Consumption characterises Proof-of-Work blockchain networks that secure the ledger through computationally intensive mining, requiring significant and ongoing electricity expenditure proportional to network hash rate. This property creates environmental concerns, drives carbon footprint assessments, and motivates regulatory scrutiny of PoW chains, contrasting sharply with the energy profile of stake-based consensus alternatives.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:high-energy-consumption", "labels": [ "High Energy Consumption" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Energy Consumption" ], "wikilinks": [ "Blockchain" ] }, { "id": "high-risk-ai-system", "title": "High Risk AI System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An AI system classified as presenting significant risk to health, safety, fundamental rights, or other critical interests based on its intended purpose, deployment context, and potential for substantial adverse impact, subject to stringent regulatory requirements under the EU AI Act and similar f...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:high-risk-ai-system", "labels": [ "High Risk AI System" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Conformity Assessment (AI-0103)", "Human Oversight (AI-0041)", "user experience", "Introduction to me", "MetaverseDomain" ] }, { "id": "high-frequency-trading", "title": "High-Frequency Trading", "domain": "finance", "domain_name": "Finance", "definition": "High-frequency trading (HFT) is a form of automated electronic trading characterised by very high order submission rates, extremely short holding periods and a reliance on minimising latency to gain advantage. HFT firms use co-located servers, optimised networking and algorithmic strategies to react to market signals in microseconds, often acting as market makers or capturing fleeting price discrepancies. It is a dominant participant in modern equity, futures and foreign-exchange markets and a central subject of market microstructure research.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:high-frequency-trading", "labels": [ "High-Frequency Trading", "High Frequency Trading" ], "is_subclass_of": [ "Market Microstructure" ], "wikilinks": [] }, { "id": "high-performance-computing", "title": "High-Performance Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "High-Performance Computing (HPC) refers to the use of aggregated computational resources \u2014 clusters, supercomputers, and massively parallel systems \u2014 to solve problems requiring sustained compute throughput, memory bandwidth, or I/O rates far beyond those achievable on commodity servers. HPC systems combine specialised processors (CPUs, GPUs, FPGAs, TPUs), high-speed interconnects (InfiniBand, NVLink), parallel file systems, and workload-management software to execute scientific simulations, large-scale machine learning training, genomics, climate modelling, and computational fluid dynamics at peta- or exascale performance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:high-performance-computing", "labels": [ "High-Performance Computing", "High Performance Computing" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "high-speed-networking", "title": "High-Speed Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "High-speed networking refers to interconnect technologies and protocols engineered to sustain very high data-transfer rates, typically tens to hundreds of gigabits per second, with minimal latency, spanning link types such as fibre optics, InfiniBand and high-bandwidth Ethernet. It is a prerequisite wherever large data volumes must move between compute nodes, sensors or accelerators without becoming a system bottleneck. Applications range from ASIC-accelerated pipelines to synchronised multi-camera motion capture rigs.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:high-speed-networking", "labels": [ "High-Speed Networking" ], "is_subclass_of": [ "Network Layer" ], "wikilinks": [] }, { "id": "highly-elliptical-orbit", "title": "Highly Elliptical Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:highly-elliptical-orbit", "labels": [ "Highly Elliptical Orbit" ], "is_subclass_of": [ "Elliptical Orbit" ], "wikilinks": [] }, { "id": "hiroshima-ai-process", "title": "Hiroshima AI Process", "domain": "governance", "domain_name": "Governance", "definition": "The Hiroshima AI Process is a G7-led international initiative, launched in 2023, to develop shared principles and a voluntary code of conduct for organisations developing advanced AI systems. It promotes safe, secure, and trustworthy AI through international cooperation on guardrails for foundation and generative models. The process represents a multilateral, soft-law approach to AI governance complementing national regulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hiroshima-ai-process", "labels": [ "Hiroshima AI Process", "Hiroshima AI Code of Conduct" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "historical-research", "title": "Historical Research", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Historical research is the systematic investigation of past events, peoples, and material culture through the examination of primary and secondary sources. In digital contexts it increasingly draws on digitised archives, 3D reconstructions, and computational analysis to interpret and present the past. It provides the evidentiary and interpretive basis for cultural heritage applications such as site reconstruction and virtual museums.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:historical-research", "labels": [ "Historical Research" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [] }, { "id": "history-and-path-to-agi", "title": "History and Path to AGI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "History and Path to AGI is the intellectual and institutional chronicle of artificial intelligence research from its philosophical origins through contemporary frontier AI development, tracing the succession of paradigms, breakthroughs, and failures that collectively constitute the discipline's t...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:history-and-path-to-agi", "labels": [ "History and Path to AGI" ], "is_subclass_of": [ "AI Research Area", "Artificial General Intelligence", "AI Research", "Technology History", "Cognitive Science", "Scientific Paradigm" ], "wikilinks": [ "AAAI", "ACL", "AI Policy", "AI Research", "AI Safety Research", "AI Winter", "Autonomous Systems", "Cognitive Psychology", "Cognitive Science", "Computational Substrate", "Computer Science", "Connectionism", "Drug Discovery", "Frontier AI Evaluation", "GPU Compute", "ICLR", "ICML", "Institutional Funding", "MachineLearningDomain", "Mathematical Foundations" ] }, { "id": "hohmann-transfer", "title": "Hohmann Transfer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hohmann-transfer", "labels": [ "Hohmann Transfer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "holder-binding", "title": "Holder Binding", "domain": "security", "domain_name": "Security", "definition": "Holder binding is the cryptographic mechanism that ties a verifiable credential to the legitimate holder's controlled key material, ensuring that only the entity to whom a credential was issued can present it. It prevents credential theft and replay by requiring the presenter to prove possession of a private key bound to the credential at presentation time. Common realisations include key binding in SD-JWT and proof-of-possession challenges during presentation exchange.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:holder-binding", "labels": [ "Holder Binding" ], "is_subclass_of": [ "Decentralized Identifier" ], "wikilinks": [] }, { "id": "holographic-consensus", "title": "Holographic Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Holographic consensus is a DAO governance mechanism that uses a prediction market of staked tokens to surface which proposals reflect the collective will, allowing a small attentive subset to make decisions that represent the whole. Predictors stake on whether a proposal will pass, boosting promising proposals to a faster majority-vote track while filtering spam. It addresses scalability of decentralised governance without requiring every member to vote on everything.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:holographic-consensus", "labels": [ "Holographic Consensus" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "holographic-display", "title": "Holographic Display", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A holographic display is a display system that reconstructs the full optical wavefront of a three-dimensional scene, enabling viewers to perceive genuine depth cues \u2014 including motion parallax, focus accommodation, and binocular disparity \u2014 without wearing specialised eyewear. It achieves this by encoding scenes as holograms (interference fringe patterns) and illuminating them with coherent or structured light to recreate the original light field. Unlike stereoscopic or autostereoscopic displays, holographic displays avoid the vergence-accommodation conflict because the eye can naturally refocus at different depths within the reconstructed scene. Practical implementations typically use spatial light modulators, diffractive optical elements, or photopolymer recording media to shape wavefronts, placing extreme demands on computational throughput and optical bandwidth.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:holographic-display", "labels": [ "Holographic Display" ], "is_subclass_of": [ "Display Technology" ], "wikilinks": [ "Holography", "Spatial Computing", "Computer Graphics", "Display Technology" ] }, { "id": "holographic-rendering", "title": "Holographic Rendering", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Holographic rendering is the process of computing and displaying imagery that reconstructs a light field so that virtual content appears with genuine depth, parallax, and view-dependent shading. Unlike conventional flat-screen rendering, it produces multiple correct perspectives across a viewing volume, allowing observers to perceive three-dimensional structure without stereoscopic eyewear or with depth cues that match natural accommodation. It combines light-field computation, display technology, and real-time graphics pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:holographic-rendering", "labels": [ "Holographic Rendering" ], "is_subclass_of": [ "Holography" ], "wikilinks": [] }, { "id": "holographic-telepresence", "title": "Holographic Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Holographic Telepresence is a form of remote presence technology that projects three-dimensional, life-size representations of remote participants into a shared physical or virtual space. It combines light field capture, display, and rendering techniques to create the illusion that a remote person is physically co-located with local participants. This approach enables natural interaction cues such as eye contact, spatial positioning, and gestural communication that are absent in conventional video conferencing.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:holographic-telepresence", "labels": [ "Holographic Telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "holography", "title": "Holography", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A technique for recording and reconstructing the full wavefront of light, including both amplitude and phase, to reproduce three-dimensional images. It records the interference pattern between a reference beam and light scattered from an object.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:holography", "labels": [ "Holography", "Computational Holography" ], "is_subclass_of": [ "Display and Rendering", "owl:Thing" ], "wikilinks": [ "Camera", "Holographic Display", "Image Processing", "owl:Thing" ] }, { "id": "home-assistant", "title": "Home Assistant", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Home Assistant is a free and open-source home automation platform written in Python (backend) and TypeScript (frontend), enabling local-first integration and control of heterogeneous smart-home devices and services across protocols including Zigbee Protocol, Z-Wave Protocol,", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:home-assistant", "labels": [ "Home Assistant" ], "is_subclass_of": [ "Computing and Cloud", "Smart Home", "IoT Platform", "Edge Computing", "Home Automation", "Open Source Software" ], "wikilinks": [ "AI Task Integration", "Amazon Alexa", "AppDaemon", "Apple HomeKit", "ARM64 Linux", "Assist Voice Pipeline", "Automation Engine", "Bluetooth LE", "Building Automation", "Buildroot", "CoAP", "CSA Matter Specification", "Demand Side Response", "Device Interoperability", "Domoticz", "Edge AI", "EdgeComputingLayer", "Energy Dashboard", "Energy Management", "ESPHome" ] }, { "id": "homeostasis", "title": "Homeostasis", "domain": "ai", "domain_name": "Ai", "definition": "Homeostasis is the property of a biological or engineered system by which it maintains internal state variables within a defined physiological or operational range despite external perturbations, achieved through negative feedback control loops that sense deviations from a setpoint and activate corrective effectors. It is a foundational principle of physiology, cybernetics, and adaptive control engineering.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:homeostasis", "labels": [ "Homeostasis" ], "is_subclass_of": [ "Cybernetics" ], "wikilinks": [] }, { "id": "homogeneous-transformation", "title": "Homogeneous Transformation", "domain": "robotics", "domain_name": "Robotics", "definition": "A homogeneous transformation is a 4x4 matrix that compactly represents both rotation and translation of a rigid body in three-dimensional space using homogeneous coordinates. By embedding rotation and translation into a single matrix, transformations can be composed through matrix multiplication, making them the standard tool for relating coordinate frames in robotics and graphics. Homogeneous transformations underpin forward and inverse kinematics, pose representation and frame chaining along kinematic links.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:homogeneous-transformation", "labels": [ "Homogeneous Transformation" ], "is_subclass_of": [ "Coordinate Transformation" ], "wikilinks": [] }, { "id": "homography", "title": "Homography", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A homography is a projective transformation, represented by a 3x3 matrix, that maps points from one plane to another in homogeneous coordinates, preserving straight lines but not parallelism or angles. In computer vision it relates two images of the same planar surface or two views taken from the same camera centre, enabling tasks such as image rectification, perspective correction, and mosaicking. Homographies are estimated from corresponding feature points, often robustly via methods that reject outlier matches.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:homography", "labels": [ "Homography" ], "is_subclass_of": [ "Projective Geometry" ], "wikilinks": [] }, { "id": "homomorphic-encryption-for-machine-learning", "title": "Homomorphic Encryption for Machine Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Homomorphic Encryption for Machine Learning is a cryptographic paradigm that enables arithmetic computations to be performed directly on ciphertext, so that AI model training and inference can proceed on encrypted data without any decryption step, guaranteeing that neither cloud servers nor third parties ever observe plaintext inputs, intermediate activations, or model weights. The approach relies on algebraic homomorphisms\u2014addition and multiplication over encrypted values\u2014combined with bootstrapping techniques to manage noise accumulation, with schemes such as CKKS targeting approximate real-number arithmetic well-suited to neural-network workloads. Applications span privacy-preserving inference, encrypted federated learning aggregation, and collaborative multi-party model training on sensitive data spanning healthcare, finance, and government.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:homomorphic-encryption-for-machine-learning", "labels": [ "Homomorphic Encryption for Machine Learning", "Encrypted Machine Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "CKKS Scheme", "IBM HELib", "Microsoft SEAL", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "homomorphic-encryption", "title": "homomorphic encryption", "domain": "security", "domain_name": "Security", "definition": "Homomorphic Encryption (HE) is a cryptographic paradigm that permits arbitrary arithmetic and logical operations to be performed directly on ciphertext, yielding an encrypted result that, upon decryption, matches the outcome of the same operations applied to the original plaintext. Fully Homomorphic Encryption (FHE), first constructed by Craig Gentry in 2009 using ideal-lattice hard problems, supports an unbounded depth of operations and enables third parties \u2014 such as cloud compute providers \u2014 to process sensitive data without ever gaining access to it in plaintext form. Practical FHE schemes include BGV and BFV for exact integer arithmetic, CKKS for approximate real-number arithmetic (widely used in machine-learning inference), and TFHE for fast gate-by-gate bootstrapping over Boolean circuits.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:homomorphic-encryption", "labels": [ "Homomorphic Encryption", "Fully Homomorphic Encryption", "Paillier Encryption" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "honesty", "title": "Honesty", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An alignment objective ensuring AI systems provide truthful and accurate information, avoiding false claims and acknowledging uncertainty when appropriate. Honesty (also called truthfulness) represents a critical dimension of trustworthy AI alongside helpfulness and harmlessness.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:honesty", "labels": [ "Honesty" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Alignment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "hong-kong", "title": "Hong Kong", "domain": "finance", "domain_name": "Finance", "definition": "Hong Kong is a Special Administrative Region (SAR) of the People's Republic of China, operating under the 'one country, two systems' framework that preserves a distinct legal, financial, and regulatory order. It functions as one of the world's leading international financial centres, hosting a deep capital market, extensive banking sector, and a highly active financial technology ecosystem. The Hong Kong Monetary Authority (HKMA) has conducted prominent pilots in central bank digital currency, cross-border wholesale settlement, and tokenised asset infrastructure, including through the BIS Innovation Hub centre established in the territory. Hong Kong's regulatory regime, common-law courts, and open capital account make it a critical interface between global finance and the broader Chinese economy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hong-kong", "labels": [ "Hong Kong" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Financial Technology", "Central Bank Digital Currency", "BIS Innovation Hub", "Entity" ] }, { "id": "hookes-law", "title": "Hooke's Law", "domain": "robotics", "domain_name": "Robotics", "definition": "Hooke's law states that the restoring force exerted by an ideal spring is directly proportional to its displacement from equilibrium, expressed as F = -kx, where k is the spring's stiffness constant. It is the foundational physical relationship behind compliant and series elastic actuation in robotics, where a spring element is deliberately placed in the drivetrain so that measuring its deflection yields an accurate estimate of applied force or torque. The law holds only within a spring's elastic limit, beyond which deformation becomes non-linear or permanent.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:hookes-law", "labels": [ "Hooke's Law", "Hookes Law" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "hop-protocol", "title": "Hop Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Hop Protocol is a cross-chain bridge that enables the transfer of tokens between Ethereum and its layer-two rollup networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hop-protocol", "labels": [ "Hop Protocol" ], "is_subclass_of": [ "Blockchain", "Cross-Chain Bridge" ], "wikilinks": [ "Token", "Interoperability", "Web3 Infrastructure", "Blockchain", "https://hop.exchange/", "https://docs.hop.exchange/" ] }, { "id": "horizon-sensor", "title": "Horizon Sensor", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:horizon-sensor", "labels": [ "Horizon Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "horizon-workrooms", "title": "Horizon Workrooms", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"Meta's enterprise virtual reality telepresence application for Quest headsets, providing immersive meeting rooms where distributed teams collaborate as avatars around virtual conference tables with spatial audio, shared whiteboards, screen sharing, and keyboard passthrough, representing mainstre...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:horizon-workrooms", "labels": [ "Horizon Workrooms", "TELE-028-horizon-workrooms" ], "is_subclass_of": [ "Telepresence", "Virtual Reality Telepresence", "TELE-020-virtual-reality-telepresence" ], "wikilinks": [ "TELE-020-virtual-reality-telepresence", "TELE-026-microsoft-mesh\\", "TELE-026-microsoft-mesh", "TELE-027-spatial-platform\\", "TELE-100-ai-avatars", "TELE-105-real-time-language-translation", "TELE-107-ai-meeting-assistants", "TELE-203-haptic-feedback-telepresence", "TELE-302-shared-whiteboards", "VirtualWhiteboarding", "Metaverse-Telepresence Bridge", "TELE-001-telepresence" ] }, { "id": "horizontal-coordinate-system", "title": "Horizontal Coordinate System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:horizontal-coordinate-system", "labels": [ "Horizontal Coordinate System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "horizontal-coordinate", "title": "Horizontal Coordinate", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:horizontal-coordinate", "labels": [ "Horizontal Coordinate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "horizontal-datum", "title": "Horizontal Datum", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:horizontal-datum", "labels": [ "Horizontal Datum" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "horizontal-dilution-of-precision", "title": "Horizontal Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:horizontal-dilution-of-precision", "labels": [ "Horizontal Dilution of Precision" ], "is_subclass_of": [ "Dilution of Precision" ], "wikilinks": [] }, { "id": "horizontal-scalability", "title": "Horizontal Scalability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Horizontal scalability is the capacity of a system to increase throughput by adding more independent nodes rather than upgrading a single machine (vertical scaling). It relies on partitioning work and state so that load can be distributed across commodity servers behind a balancer. Horizontal scaling underpins cloud-native and distributed architectures because it offers near-linear, fault-tolerant growth.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:horizontal-scalability", "labels": [ "Horizontal Scalability", "Horizontal Scaling" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "hot-wallet", "title": "Hot Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A hot wallet is a cryptocurrency wallet whose private keys are held on an internet-connected device or service, enabling rapid signing and broadcasting of transactions. Its constant connectivity makes it convenient for frequent transfers, exchange operations, and decentralised-application interaction, but also exposes it to remote compromise. It contrasts with cold storage, where keys are kept offline to minimise attack surface. Operators typically hold only operationally necessary balances in hot wallets and sweep surplus funds to cold storage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hot-wallet", "labels": [ "Hot Wallet" ], "is_subclass_of": [ "Crypto Wallet" ], "wikilinks": [] }, { "id": "hot-stuff-consensus", "title": "HotStuff Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Byzantine fault-tolerant state machine replication protocol that achieves the first simultaneous combination of linear message complexity, optimistic responsiveness, and a simple three-phase voting pipeline. HotStuff replaces PBFT's quadratic message complexity with a leader-based threshold signature aggregation scheme, enabling safe view changes with O(n) messages and forming the foundation for the DiemBFT/LibraBFT and Aptos consensus protocols.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hot-stuff-consensus", "labels": [ "HotStuff Consensus", "HotStuff BFT" ], "is_subclass_of": [ "Protocol and Consensus", "Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Byzantine Fault Tolerance" ] }, { "id": "hot-stuff", "title": "HotStuff", "domain": "blockchain", "domain_name": "Blockchain", "definition": "HotStuff is a leader-based Byzantine fault-tolerant consensus protocol that achieves linear communication complexity and responsiveness through a pipelined, three-phase voting structure with threshold signatures. Its linear view-change cost and rotating leadership make it well suited to large validator sets. HotStuff influenced modern BFT systems and underpins protocols such as Libra/Diem's consensus.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hot-stuff", "labels": [ "HotStuff" ], "is_subclass_of": [ "Byzantine Fault Tolerance" ], "wikilinks": [] }, { "id": "howey-test", "title": "Howey Test", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Howey Test is a four-part legal framework established by the U.S. Supreme Court in SEC v. W.J. Howey Co. (1946) to determine whether a transaction qualifies as an 'investment contract' and therefore constitutes a security subject to federal securities regulation. An instrument is a security if it involves (1) an investment of money, (2) in a common enterprise, (3) with an expectation of profits, (4) derived from the efforts of others.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:howey-test", "labels": [ "Howey Test" ], "is_subclass_of": [ "Investment Contract Analysis" ], "wikilinks": [] }, { "id": "http2", "title": "Http2", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "HTTP/2 is a major revision of the Hypertext Transfer Protocol that introduces a binary framing layer, multiplexed streams over a single TCP connection, header compression and server push. It reduces latency and head-of-line blocking at the application layer compared with HTTP/1.1 while preserving the protocol's semantics. It is standardised in RFC 7540 (later RFC 9113) and is widely deployed across the modern web.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:http2", "labels": [ "Http2", "HTTP/2", "HTTP2" ], "is_subclass_of": [ "HTTP" ], "wikilinks": [] }, { "id": "huawei-chips", "title": "Huawei Chips", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A family of AI accelerators and processors developed by Huawei, serving as a critical domestic alternative to US-based hardware for training and inference in the Chinese AI ecosystem.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:huawei-chips", "labels": [ "Huawei Chips" ], "is_subclass_of": [ "GPU Computing" ], "wikilinks": [] }, { "id": "hugging-face-accelerate", "title": "Hugging Face Accelerate", "domain": "ai", "domain_name": "Ai", "definition": "Hugging Face Accelerate is an open-source Python library that lets PyTorch training code run unchanged across CPUs, single or multiple GPUs, and TPUs by abstracting device placement and distributed launch. It handles mixed precision, gradient accumulation, and sharded data/model parallelism with minimal boilerplate. Accelerate lowers the barrier to scaling deep-learning training and inference across hardware configurations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hugging-face-accelerate", "labels": [ "Hugging Face Accelerate", "Accelerate", "Accelerate Training Framework" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "hugging-face-diffusers", "title": "hugging face diffusers", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hugging Face Diffusers is an open-source Python library that provides a modular, composable toolkit for training, fine-tuning, and running inference with state-of-the-art diffusion models for image, audio, and video generation. The library abstracts the scheduling, noise prediction, and decoder stages of diffusion pipelines behind a consistent API, exposing interchangeable components \u2014 noise schedulers (DDPM, DDIM, DPM-Solver), denoising U-Net or Diffusion Transformer (DiT) backbones, and variational autoencoders \u2014 that researchers can recombine freely. It integrates natively with the Hugging Face Hub for model discovery, versioning, and model-card governance, and supports PyTorch and JAX as compute backends with optional xFormers memory-efficient attention and Accelerate-based distributed training.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hugging-face-diffusers", "labels": [ "Hugging Face Diffusers", "Diffusers Library" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "hugging-face-model-hub", "title": "Hugging Face Model Hub", "domain": "ai", "domain_name": "Ai", "definition": "The Hugging Face Model Hub is a public repository and platform for sharing, discovering, and versioning machine-learning models, datasets, and demo spaces. Built on Git and Git-LFS, it hosts hundreds of thousands of pretrained models with standardised model cards, metadata, and direct integration into the Transformers ecosystem. It functions as the central distribution layer for open machine-learning artefacts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hugging-face-model-hub", "labels": [ "Hugging Face Model Hub", "Model Hub" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "hugging-face", "title": "hugging face", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hugging Face is an AI company and open-source platform that operates the Transformers, Diffusers, Datasets, PEFT, and TRL libraries alongside the Hugging Face Hub \u2014 a centralised model and dataset repository hosting hundreds of thousands of community-contributed checkpoints spanning natural language processing, computer vision, audio, multimodal, and reinforcement learning domains. The Hub standardises model cards, dataset cards, and Spaces (interactive Gradio or Streamlit demos), and has become the de facto distribution platform for open-weight large language models and their fine-tuned derivatives. Through its inference API, AutoTrain service, and parameter-efficient fine-tuning tooling, Hugging Face significantly lowers the barrier to deploying and adapting state-of-the-art machine learning models for research and production use.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hugging-face", "labels": [ "Hugging Face", "HuggingFace" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "hugging-face-hub", "title": "huggingface hub", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hugging Face Hub is a centralised, version-controlled platform for hosting, discovering, and sharing machine learning models, datasets, and interactive Spaces under a collaborative open-source ecosystem. It provides Git-LFS-backed repositories for large artefact storage, model cards for documentation and responsible AI metadata, and a Python SDK (huggingface_hub) for programmatic access. The Hub functions as the de-facto registry for open-weight foundation models, transformer checkpoints, and diffusion model pipelines, serving millions of downloads daily across research and production workflows.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:hugging-face-hub", "labels": [ "HuggingFace Hub", "Hugging Face Hub" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "human-agency-and-oversight", "title": "Human Agency and Oversight", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Human Agency and Oversight is a core trustworthiness dimension of responsible AI that encompasses two coupled principles: human agency\u2014protecting individuals' freedom to make informed, uncoerced decisions when interacting with or affected by AI systems\u2014and human oversight\u2014establishing technical and organisational mechanisms that allow authorised humans to monitor, intervene in, correct, or deactivate AI operations at appropriate granularity. The EU AI Act Article 14 mandates these mechanisms for high-risk AI systems, requiring that oversight be achievable by qualified natural persons who understand system outputs and can exercise meaningful authority over them.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:human-agency-and-oversight", "labels": [ "Human Agency and Oversight" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU AI Act Article 14", "EU HLEG AI", "IEEE P7000", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "human-annotation", "title": "Human Annotation Data", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Human annotation data is the output of people labelling, ranking, or judging raw content, such as text, images, audio, or model responses, to create ground-truth signals for training and evaluating machine learning systems. It is produced through structured annotation workflows involving guidelines, multiple annotators, and inter-annotator agreement checks such as Cohen's kappa. Human annotation data underpins supervised learning, evaluation metrics such as COMET, and reinforcement learning from human feedback, where annotators' preference judgements directly shape model behaviour.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-annotation", "labels": [ "Human Annotation Data", "Human Annotation" ], "is_subclass_of": [ "Data Annotation" ], "wikilinks": [] }, { "id": "human-annotator", "title": "Human Annotator", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A human annotator is a person who labels, categorises or reviews raw data samples \u2014 text, images, audio or model outputs \u2014 to produce the ground-truth or preference signals used to train and evaluate machine learning models. Annotators follow labelling guidelines and, for subjective tasks, their agreement is measured via inter-annotator agreement to assess label quality. Human annotators are central to supervised data annotation pipelines and to collecting human feedback for reinforcement learning from human feedback.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:human-annotator", "labels": [ "Human Annotator" ], "is_subclass_of": [ "Data Annotation" ], "wikilinks": [] }, { "id": "human-capital", "title": "Human Capital", "domain": "ai", "domain_name": "Ai", "definition": "Human capital is the classical economic concept treating the skills, knowledge, experience, and health embodied in individuals as a stock of productive capital that yields economic returns through employment and innovation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-capital", "labels": [ "Human Capital" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "human-capture-and-recognition", "title": "Human Capture & Recognition", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Techniques for digitally acquiring and interpreting human appearance, motion, and biometric data for use in virtual and augmented environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:human-capture-and-recognition", "labels": [ "Human Capture & Recognition" ], "is_subclass_of": [ "Interaction Technology", "Reality Capture" ], "wikilinks": [ "3D Scanning", "3D Visualization", "Biometric Analysis", "Depth Cameras", "ETSI ARF 010", "Facial Recognition", "Image Processing", "Optical Sensors", "Reality Capture", "Reality Modeling", "3D Reconstruction", "Avatar Creation", "ComputeLayer", "Computer Vision", "CreativeMediaDomain", "Digital Twin Generation", "Machine Learning Models", "Motion Tracking", "NetworkLayer", "Pattern Recognition" ] }, { "id": "human-centred-design", "title": "Human Centred Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Human centred design is a problem-solving approach that places the needs, capabilities and behaviours of people at the centre of the design process, iteratively shaping products and systems around real user contexts. It draws on empirical understanding of users and tasks, involves stakeholders throughout development, and refines solutions through repeated evaluation. The practice underpins usable, accessible and inclusive interactive systems across digital and physical domains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:human-centred-design", "labels": [ "Human Centred Design", "Human-Centred Design" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "human-centred-values", "title": "Human Centred Values", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AI systems should be designed and operated in ways that respect the rule of law, human rights, democratic values and diversity, with appropriate safeguards to ensure human determination and control over consequential decisions (human agency), whilst incorporating mechanisms to protect fairness an...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-centred-values", "labels": [ "Human Centred Values" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Metaverse", "MetaverseDomain" ] }, { "id": "human-computer-interaction", "title": "Human Computer Interaction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Human-Computer Interaction (HCI) in the AI context examines the design, evaluation, and implementation of interactive systems that incorporate artificial intelligence capabilities. This interdisciplinary field addresses usability, accessibility, user experience, and cognitive aspects of AI-powered interfaces, emphasising transparency, trust calibration, and ethical implications of algorithmic decision-making on human users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:human-computer-interaction", "labels": [ "Human Computer Interaction", "Human-Computer Interaction", "Human-Computer Interaction Framework", "HumanComputerInteraction" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Human-AI Collaboration", "User Experience Design", "Conversational AI", "Explainable AI" ] }, { "id": "human-computer-interface", "title": "Human Computer Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The set of hardware and software components through which a human user perceives, commands, and receives feedback from a computing system. In spatial computing, human computer interfaces extend beyond screens and keyboards to include gesture input, voice commands, gaze control, haptic feedback, and brain-computer interfaces, fundamentally shaping how users navigate and manipulate virtual environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-computer-interface", "labels": [ "Human Computer Interface", "Human-AI Interface", "Human-Computer Interface", "Human-Machine Interface" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "human-development-index", "title": "Human Development Index", "domain": "governance", "domain_name": "Governance", "definition": "The Human Development Index (HDI) is a composite statistic published by the United Nations Development Programme that ranks countries by aggregating life expectancy, education (mean and expected years of schooling), and gross national income per capita. It measures human wellbeing beyond purely economic output and is used to compare development levels and to track progress over time. The HDI underpins many analyses of inequality and policy effectiveness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:human-development-index", "labels": [ "Human Development Index", "UNDP Human Development Index" ], "is_subclass_of": [ "Evaluation Metric" ], "wikilinks": [] }, { "id": "human-evaluation", "title": "Human Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "Human evaluation is the assessment of machine-learning system outputs by human judges against quality criteria such as relevance, fluency, helpfulness, factuality, or preference between alternatives. It complements automatic metrics by capturing nuanced, subjective, and context-dependent judgements that proxy measures miss, and is central to evaluating generative and conversational models. It contrasts with automatic evaluation in cost, latency, and the need to manage rater agreement and bias.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:human-evaluation", "labels": [ "Human Evaluation", "Human Evaluation Correlation" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation" ], "wikilinks": [] }, { "id": "human-factors", "title": "Human Factors", "domain": "robotics", "domain_name": "Robotics", "definition": "Human factors encompasses the interdisciplinary study of how humans interact with, perceive, and respond to robotic systems, integrating ergonomics, psychology, cognitive science, and design principles to ensure robots enhance rather than impede human performance and safety.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:human-factors", "labels": [ "Human Factors", "Human Factors Engineering" ], "is_subclass_of": [ "Human-Robot Interaction", "Interaction Design", "Systems Engineering" ], "wikilinks": [ "Biomechanics", "Cognitive Load Analysis", "Cognitive Science", "Collaborative Robotics", "Efficient Collaboration", "Ergonomics", "Human-Robot Teamwork", "Interaction Design", "Interface Design", "Intuitive Control", "Moral Agency", "Operator Safety", "Psychology", "Safety Assessment", "Systems Engineering", "Task Analysis", "Trust Calibration", "Usability Testing", "User Acceptance", "Gaze Control" ] }, { "id": "human-feedback", "title": "Human Feedback", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Information provided by human evaluators about model outputs, typically in the form of rankings, ratings, demonstrations, or corrections. Human feedback serves as the training signal for aligning AI systems with human preferences and values, enabling learning of complex objectives that are difficult to specify formally.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:human-feedback", "labels": [ "Human Feedback", "Human Feedback Data" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "RLHF", "MetaverseDomain" ] }, { "id": "human-interface-device", "title": "Human Interface Device", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical hardware component enabling user input or feedback in immersive systems through controllers, sensors, and actuators.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:human-interface-device", "labels": [ "Human Interface Device" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Bluetooth", "Calibration", "Communication Interface", "Device Drivers", "ETSI GR ARF 010", "Input Sensors", "Interaction System", "Output Actuators", "Power Management", "Tracking Components", "USB Protocol", "User Input", "Wireless Communication", "Haptic Feedback", "Haptics", "InteractionDomain", "Motion Tracking", "PhysicalLayer", "Spatial Interaction" ] }, { "id": "human-interface-layer-hil", "title": "Human Interface Layer (HIL)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software and hardware layer encompassing devices and modalities that connect users physically and sensorily to immersive environments, managing interaction design and user experience.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:human-interface-layer-hil", "labels": [ "Human Interface Layer (HIL)" ], "is_subclass_of": [ "Interaction Technology", "Interaction Technology" ], "wikilinks": [ "Haptic Systems", "Input Devices", "Interaction Models", "MSF Taxonomy 2025", "Multimodal Feedback", "Natural Interaction", "Output Devices", "Tracking System", "Tracking Systems", "User Immersion", "Hardware Abstraction Layer (HAL)", "InteractionDomain", "Interaction Domain", "Network Layer", "Presence", "Rendering Engine" ] }, { "id": "human-oversight", "title": "Human Oversight", "domain": "governance", "domain_name": "Governance", "definition": "The continuous or periodic involvement of competent human actors in the governance, development, deployment, and operation of artificial intelligence systems, exercising meaningful control, judgment, and intervention capabilities to ensure AI system decisions and actions remain aligned with human values, ethical principles, legal requirements, and intended purposes, with particular emphasis on preventing, detecting, and correcting harmful or inappropriate AI behaviours through informed human decision-making authority.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:human-oversight", "labels": [ "Human Oversight", "Human Oversight (AI-0041)" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [ "Automation Bias", "continuous improvement", "error correction", "Accountability", "AI Audit", "AI Governance", "AI Operator", "Explainability", "Human in the Loop", "MetaverseDomain", "Responsible AI", "Risk Management", "Transparency" ] }, { "id": "human-pose-slam-capture-system", "title": "Human Pose SLAM Capture System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Human Pose SLAM Capture System is an integrated sensing and computation pipeline that simultaneously localises a device within an unknown environment (SLAM) while continuously tracking the full-body skeletal pose of one or more human occupants in real time. It fuses data from depth cameras, inertial measurement units, and RGB imagery through probabilistic state estimation \u2014 typically particle filters or factor-graph optimisers \u2014 to produce a joint world model of both the static scene geometry and dynamic human kinematics. The output drives applications including markerless motion capture, avatar animation in extended reality, safety-aware robot navigation around people, and persistent spatial AI anchoring. The discipline sits at the intersection of computer vision, human-computer interaction, and spatial computing, with maturing industrial deployments in XR headsets, telepresence rigs, and autonomous vehicle pedestrian tracking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:human-pose-slam-capture-system", "labels": [ "Human Pose SLAM Capture System", "Human tracking and SLAM capture" ], "is_subclass_of": [ "Simultaneous Localisation and Mapping" ], "wikilinks": [] }, { "id": "human-preference", "title": "Human Preference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Human Preference refers to the explicit or implicit judgements made by human evaluators that express which of two or more AI outputs is more aligned with human values, intentions, or quality criteria; these preference signals are collected through comparative annotation tasks and used as training signal to guide reinforcement learning from human feedback (RLHF) and preference optimisation methods. The aggregate of human preferences encodes desired model behaviour on dimensions such as helpfulness, harmlessness, and honesty, providing an empirical proxy for human values that is more tractable to collect than hand-crafted reward functions. Preference data quality, evaluator diversity, and annotator agreement rates critically determine the fidelity with which the resulting reward model captures genuine human intent.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:human-preference", "labels": [ "Human Preference", "Human Preference Data" ], "is_subclass_of": [ "AI Governance and Ethics", "Neural Network Component" ], "wikilinks": [ "Artificial Intelligence", "Neural Network Component" ] }, { "id": "human-rights-law", "title": "Human Rights Law", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Human rights law is the body of international and domestic legal instruments that define and protect the fundamental freedoms and entitlements owed to all individuals. It encompasses treaties, conventions, and constitutional provisions covering rights such as privacy, non-discrimination, freedom of expression, and due process. In the context of artificial intelligence, human rights law increasingly shapes obligations around algorithmic fairness, surveillance, and the protection of personal data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:human-rights-law", "labels": [ "Human Rights Law" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "human-rights", "title": "Human Rights", "domain": "governance", "domain_name": "Governance", "definition": "Fundamental rights and freedoms inherent to all human beings as recognised in international instruments including the Universal Declaration of Human Rights, which AI systems must respect and protect throughout their lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:human-rights", "labels": [ "Human Rights", "Human Rights and Technology" ], "is_subclass_of": [ "Governance and Safety", "Human Centred Values" ], "wikilinks": [ "AIGroundedDomain", "Blockchain Technology", "Council of Europe AI Treaty", "Fundamental Rights Impact Assessment", "OECD AI Principles 2024", "Privacy Rights", "Universal Declaration of Human Rights", "AI Ethics", "Algorithmic Accountability", "Democratic Values", "Digital Rights", "EU AI Act", "Fairness (OECD)", "Human Centred Values" ] }, { "id": "human-robot-interaction", "title": "Human Robot Interaction", "domain": "robotics", "domain_name": "Robotics", "definition": "Interdisciplinary field studying how humans and robots communicate, collaborate, and interact safely and effectively in shared physical and virtual spaces. HRI combines robotics engineering, AI, human factors engineering, and cognitive science to design systems with natural interaction modalities including gesture recognition, natural language processing, and haptic feedback.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:human-robot-interaction", "labels": [ "Human Robot Interaction", "Human-Robot Interaction", "HumanRobotInteraction", "Physical Human-Robot Interaction" ], "is_subclass_of": [ "Robotics", "RoboticsDomain" ], "wikilinks": [ "Authentication", "dt:enhancedBy", "dt:presentedIn", "dt:securedBy", "dt:trainedVia", "dt:uses", "enablesCollaboration", "ensuresSafety", "GestureRecognition", "IntuitiveInterface", "providesInterface", "RoboticsEngineering", "SafetyMetrics", "SafetyMetrics", "SafetyProtocol", "SharedWorkspace", "TeachPendant", "usesModality", "ArtificialIntelligence", "ArtificialIntelligence" ] }, { "id": "human-in-the-loop", "title": "Human in the Loop", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A design pattern and operational approach for artificial intelligence systems in which human judgment, decision-making, or validation is integrated as an essential component of the AI system's decision cycle, requiring active human participation at critical points before AI-generated outputs are finalised or actions are executed, thereby ensuring meaningful human control, accountability, and the application of human values and contextual understanding to consequential AI-assisted decisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:human-in-the-loop", "labels": [ "Human in the Loop", "Human In The Loop", "Human-in-the-Loop" ], "is_subclass_of": [ "Human Oversight" ], "wikilinks": [ "Automation Bias", "Decision Support", "error correction", "ethical alignment", "Human-on-the-Loop", "Accountability", "AI Governance", "AI Operator", "Convergence", "Disruption", "Explainability", "Human Oversight", "MetaverseDomain", "Risk Management", "Social contract and jobs" ] }, { "id": "human-on-the-loop", "title": "Human on the Loop", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A supervisory model of human oversight in which an automated or AI system selects and executes actions autonomously while a human monitors its operation and retains the authority to intervene, veto, or shut it down; distinct from human-in-the-loop control, where each consequential action requires affirmative human approval before execution rather than after-the-fact supervision.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:human-on-the-loop", "labels": [ "Human on the Loop", "Human-on-the-Loop" ], "is_subclass_of": [ "Human Oversight" ], "wikilinks": [ "Human Oversight", "Human in the Loop", "Meaningful Human Control" ] }, { "id": "human-ai-capability-complementarity", "title": "Human-AI Capability Complementarity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The contested domain examining comparative capabilities, limitations, and complementarities between human cognition and artificial intelligence systems, encompassing debates about autonomy, agency, creativity, and the appropriate division of decision-making authority between people and machines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-ai-capability-complementarity", "labels": [ "Human-AI Capability Complementarity", "Human vs AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "human-ai-collaboration", "title": "Human-AI Collaboration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Human-AI collaboration is the design of systems and workflows in which people and artificial-intelligence agents work jointly, combining human judgement and oversight with machine speed and scale. It encompasses interaction patterns, division of labour, and trust mechanisms that keep humans meaningfully in or on the loop. Effective collaboration improves decision quality and accountability while harnessing AI as an augmenting rather than replacing force.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:human-ai-collaboration", "labels": [ "Human-AI Collaboration" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [] }, { "id": "human-agent-interaction-surfaces", "title": "Human-Agent Interaction Surfaces", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The set of software boundaries, interaction surfaces, and protocol contracts through which humans, AI agents, and system components communicate. In the context of LLMs and spatial computing, interfaces include node-based visual editors, chat frontends, API gateways, and multimodal input layers that mediate access to underlying AI or infrastructure capabilities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-agent-interaction-surfaces", "labels": [ "Human-Agent Interaction Surfaces", "Interfaces" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Langflow", "Agents", "Hardware and Edge", "Infrastructure", "Interfaces", "Large Language Models", "Node based visual interfaces", "Stable Diffusion" ] }, { "id": "human-robot-collaboration", "title": "Human-Robot Collaboration", "domain": "robotics", "domain_name": "Robotics", "definition": "Human-robot collaboration is the design and operation of robots that work alongside people in a shared workspace, coordinating tasks safely through sensing, control and interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:human-robot-collaboration", "labels": [ "Human-Robot Collaboration", "Human-Robot Teaming", "Safe Human-Robot Collaboration" ], "is_subclass_of": [ "Robotics", "Human-Robot Interaction", "Robotics Domain" ], "wikilinks": [ "Human Robot Interaction", "Sensor Fusion", "Collaborative Robot", "Assistive Robotics", "Exoskeleton", "Robotics Domain" ] }, { "id": "human-in-the-loop-learning", "title": "Human-in-the- Loop Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A machine learning paradigm that integrates human expertise into the training process through iterative feedback, active learning queries, and collaborative validation. HITL learning combines automated model updates with human judgement for data labelling, error correction, and safety-critical decision review, and is essential in domains where ground truth is subjective or expert-dependent.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:human-in-the-loop-learning", "labels": [ "Human-in-the- Loop Learning", "Human-in-the-Loop Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Interactive Machine Learning", "Active Learning", "Data Annotation", "Reinforcement Learning from Human Feedback" ] }, { "id": "humaninthe-loop-learning", "title": "Humaninthe Loop Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A machine learning paradigm in which human annotators or domain experts are incorporated into the training loop to provide labels, corrections, or preference signals at points where automated methods are insufficient. Human-in-the-loop learning improves model quality on ambiguous or high-stakes tasks and is foundational to techniques such as active learning and RLHF.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:humaninthe-loop-learning", "labels": [ "Humaninthe Loop Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Computer Science", "Artificial Intelligence", "Machine Learning", "owl:Thing" ] }, { "id": "humanities-last-exam", "title": "Humanities Last Exam", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A benchmark designed to evaluate the capabilities of AI models in understanding and reasoning about humanities subjects.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:humanities-last-exam", "labels": [ "Humanities Last Exam" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [] }, { "id": "humanity-attestation", "title": "Humanity Attestation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Verification process that confirms a digital identity represents a human rather than an automated agent, bot, or AI system.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:humanity-attestation", "labels": [ "Humanity Attestation" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Account Security", "Authentication System", "Behavioral Analysis", "Biometric Verification", "Bot Prevention", "CAPTCHA", "Challenge Protocol", "Challenge-Response Protocol", "Cryptographic Proof", "ETSI GR ARF 010", "Fraud Prevention", "MSF Use Cases", "Trust Establishment", "Verification Mechanism", "Digital Identity", "Identity Verification", "Machine Learning", "MiddlewareLayer", "Pattern Recognition", "TrustAndGovernanceDomain" ] }, { "id": "humanoid-robot", "title": "Humanoid Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A humanoid robot is an autonomous or semi-autonomous mechanical system whose overall morphology, kinematic chain, and sensorimotor organisation deliberately mirrors the human body plan: a vertical torso supported on two bipedal legs, bilateral upper limbs terminating in multi-fingered end-effecto...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:humanoid-robot", "labels": [ "Humanoid Robot" ], "is_subclass_of": [ "Robot Type", "Robot", "Mobile Robot", "Embodied Agent", "Autonomous System", "Bipedal Robot" ], "wikilinks": [ "Autonomous System", "Bipedal Balance", "Bipedal Locomotion", "Bipedal Robot", "Black et al. 2024 pi-zero", "Brohan et al. 2023 RT-2", "Cognition Layer", "Contact Dynamics", "Control Layer", "Control Systems", "Dexterous Manipulation", "Domain Randomisation", "Drone", "Elder Care Robotics", "Embodied Agent", "Embodied AI", "Exoskeleton", "General-Purpose Manipulation", "Hazardous Environment Inspection", "Human-Robot Collaboration" ] }, { "id": "humanoid-robotics", "title": "Humanoid Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Humanoid robotics is the subfield of robotics concerned with the design, construction, and control of robots that exhibit a human-like morphology, typically including a bipedal lower body and articulated upper limbs. Such robots are engineered to operate in environments built for humans, using the same tools, furniture, and physical interfaces. Key technical challenges include stable bipedal locomotion across uneven terrain, dexterous manipulation of varied objects, real-time perception and planning under uncertainty, and safe physical interaction with humans. Advances in machine learning and actuator technology have accelerated commercial deployments in manufacturing, logistics, and care.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:humanoid-robotics", "labels": [ "Humanoid Robotics" ], "is_subclass_of": [ "Humanoid Robot" ], "wikilinks": [] }, { "id": "hybrid-cloud", "title": "Hybrid Cloud", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hybrid cloud is a computing architecture that combines on-premises or private-cloud infrastructure with one or more public clouds, orchestrated so that workloads and data can move between them. It lets organisations keep sensitive systems in controlled environments while bursting to public capacity for scale or specialised services. Hybrid cloud emphasises interoperability, unified management and consistent identity and networking across environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hybrid-cloud", "labels": [ "Hybrid Cloud" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "hybrid-consensus", "title": "Hybrid Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A consensus mechanism that combines two or more distinct consensus approaches\u2014typically pairing a proof-based method such as Proof of Work or Proof of Stake with a Byzantine Fault Tolerant finality layer\u2014to balance security, throughput, and finality properties. Hybrid designs aim to capture the best characteristics of each constituent mechanism while mitigating their individual weaknesses.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hybrid-consensus", "labels": [ "Hybrid Consensus" ], "is_subclass_of": [ "Protocol and Consensus", "Consensus Mechanism" ], "wikilinks": [ "Blockchain", "Consensus Mechanism" ] }, { "id": "hybrid-retrieval", "title": "Hybrid Retrieval", "domain": "ai", "domain_name": "Ai", "definition": "Hybrid retrieval is an information-retrieval technique that combines lexical (sparse, keyword-based) scoring such as BM25 with semantic (dense, embedding-based) vector search to rank documents. By fusing the complementary strengths of exact-term matching and meaning-based similarity, it improves recall and precision over either method alone. Fusion is typically performed with reciprocal rank fusion or weighted score combination.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hybrid-retrieval", "labels": [ "Hybrid Retrieval", "First-Stage Retrieval" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "hybrid-robot", "title": "Hybrid Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotic system that combines two or more distinct locomotion or manipulation modalities\u2014such as wheeled and legged movement, or fixed-base and mobile operation\u2014to extend operational range and adaptability across heterogeneous environments. Hybrid robots exploit the efficiency of specialised subsystems whilst maintaining versatility that no single modality can achieve alone.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:hybrid-robot", "labels": [ "Hybrid Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "owl:Thing" ] }, { "id": "hybrid-search", "title": "Hybrid Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hybrid search is an information retrieval approach that combines sparse lexical retrieval \u2014 typically BM25 or TF-IDF \u2014 with dense vector search over neural embeddings, fusing their complementary strengths: precise keyword matching and semantic understanding respectively. Score fusion via Reciprocal Rank Fusion or learned weighting combines ranked lists from both systems. The combination consistently outperforms either method alone across diverse query types, and has become the dominant retrieval pattern underpinning Retrieval-Augmented Generation pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:hybrid-search", "labels": [ "Hybrid Search" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "hybrid-work", "title": "Hybrid Work", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Hybrid work is a working model that combines on-site presence with remote work, allowing employees to split their time between a central workplace and distributed locations. It blends synchronous in-person collaboration with asynchronous and video-mediated coordination, and depends on digital collaboration tooling to keep co-located and remote participants on equal footing. Hybrid work has become a default operating model for many knowledge organisations seeking flexibility without abandoning physical workspace.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:hybrid-work", "labels": [ "Hybrid Work" ], "is_subclass_of": [ "Remote Work" ], "wikilinks": [] }, { "id": "hydraulic-actuator", "title": "Hydraulic Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "HydraulicActuator is a mechanical transduction device that converts the energy stored in pressurised hydraulic fluid into controlled mechanical work\u2014linear force and stroke via hydraulic cylinders, continuous rotational torque and speed via hydraulic motors, or limited angular displacement via ro...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hydraulic-actuator", "labels": [ "Hydraulic Actuator" ], "is_subclass_of": [ "Actuation and Control", "Actuator", "Fluid Power System", "Mechanical Transducer", "Motion Control Device", "Force Control System" ], "wikilinks": [ "Accumulator", "Actuator", "Bernoulli Equation", "Boston Dynamics Atlas", "Boston Dynamics BigDog", "Bulk Modulus", "Closed-Loop Position Control", "Compliant Force Control", "Construction Robotics", "ControlArchitectureLayer", "ControlSystemsDomain", "DARPA Robotics Challenge", "Deep-Sea Robotics", "Differential Pressure Sensing", "DIN 24346 Hydraulic Components", "Electric BLDC Actuator", "Electrohydraulic Servo Valve", "Electrohydrostatic Drive", "Exoskeleton", "Exoskeleton Actuation" ] }, { "id": "hydraulic-cylinder", "title": "Hydraulic Cylinder", "domain": "robotics", "domain_name": "Robotics", "definition": "A hydraulic cylinder is a mechanical actuator that converts hydraulic pressure and fluid flow into unidirectional linear force and motion. It consists of a cylindrical barrel, piston, piston rod, end caps, and seals; pressurised fluid acts on the piston face to extend or retract the rod, generating forces from tens of newtons to several meganewtons depending on bore diameter and system pressure. Hydraulic cylinders are foundational components in industrial robots, heavy machinery, manufacturing automation, and construction equipment where high force density and precise position control are required.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hydraulic-cylinder", "labels": [ "Hydraulic Cylinder" ], "is_subclass_of": [ "Actuation and Control", "Hydraulic Actuator" ], "wikilinks": [ "Hydraulic Actuator", "Robotics" ] }, { "id": "hydraulic-motor", "title": "Hydraulic Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "A Hydraulic Motor is an actuation device that converts pressurised hydraulic fluid flow into continuous rotational mechanical torque, functioning as the rotary counterpart to the hydraulic cylinder (which produces linear force). It is distinguished by extremely high power-to-weight ratio and the ability to sustain high torques at low rotational speeds without gearbox inefficiencies, making it the preferred actuator for heavy robotic joints, industrial manipulators, and mobile machinery operating in harsh environments. Common design types include gear motors, vane motors, and axial-piston motors, each offering different torque-speed-efficiency trade-offs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:hydraulic-motor", "labels": [ "Hydraulic Motor" ], "is_subclass_of": [ "Actuation and Control", "Hydraulic Actuator" ], "wikilinks": [ "Hydraulic Actuator", "Robotics" ] }, { "id": "hydrographic-bank", "title": "Hydrographic Bank", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hydrographic-bank", "labels": [ "Hydrographic Bank" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hydrosphere-feature", "title": "Hydrosphere Feature", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hydrosphere-feature", "labels": [ "Hydrosphere Feature" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hyper-personalisation", "title": "Hyper personalisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hyper personalisation is the application of real-time behavioural signals, rich first-party identity data, and AI/ML inference \u2014 spanning collaborative filtering, content-based filtering, hybrid ensemble models, transformer-based sequential recommendation, reinforcement-learning-from-feedback loo...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hyper-personalisation", "labels": [ "Hyper personalisation", "Hyper Personalisation" ], "is_subclass_of": [ "AI Application", "AI Agent System", "Machine Learning Discipline", "Recommender Systems", "Predictive Personalization", "Behavioural AI", "Customer Experience Management" ], "wikilinks": [ "A/B Testing Framework", "Acquisition Function", "Batch Recommendation", "Behavioural AI", "Collaborative Filtering", "Consent Management Platform", "Content-Based Filtering", "Contextual Bandits", "Customer Data Platform", "Customer Experience Management", "Customer Lifetime Value Optimisation", "DataIntelligenceDomain", "Dynamic Pricing", "Dynamic Pricing Engine", "Embedding Model", "Embeddings", "Financial Services Personalisation", "GDPR", "GDPR Compliance", "Generic Content Delivery" ] }, { "id": "hyperautomation", "title": "Hyperautomation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Hyperautomation is a disciplined approach to automating as many business and IT processes as possible by orchestrating a coordinated set of technologies including robotic process automation, machine learning, process mining and low-code platforms. Rather than automating tasks in isolation, it discovers, designs and integrates end-to-end automated workflows with intelligence layered in. The goal is scalable, adaptive automation that augments human work across the enterprise.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:hyperautomation", "labels": [ "Hyperautomation" ], "is_subclass_of": [ "Intelligent Automation" ], "wikilinks": [] }, { "id": "hyperbitcoinization", "title": "Hyperbitcoinization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hyperbitcoinization is a theoretical macroeconomic scenario, first articulated by Daniel Krawisz in 2014, in which Bitcoin undergoes a self-reinforcing demonetisation of fiat currencies and becomes the dominant global monetary unit. The process is modelled as a currency substitution accelerated by the Cantillon effect: as fiat monetary expansion erodes purchasing power, rational economic actors progressively shift savings and transactional balances into Bitcoin, increasing its adoption and liquidity, which in turn makes it more attractive as a medium of exchange. Unlike hyperinflation, which is an involuntary collapse of a fiat currency, hyperbitcoinization is theorised as a voluntary migration driven by Bitcoin's superior monetary properties \u2014 fixed supply, self-custody, and censorship resistance \u2014 ultimately displacing state-issued money.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:hyperbitcoinization", "labels": [ "Hyperbitcoinization" ], "is_subclass_of": [ "Bitcoin Value Proposition" ], "wikilinks": [] }, { "id": "hyperbolic-trajectory", "title": "Hyperbolic Trajectory", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hyperbolic-trajectory", "labels": [ "Hyperbolic Trajectory" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hypergravity", "title": "Hypergravity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hypergravity", "labels": [ "Hypergravity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hyperlane", "title": "Hyperlane", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hyperlane is an interoperability protocol that allows smart contracts on different blockchains to send messages and transfer assets between chains. It supports permissionless deployment to new chains.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:hyperlane", "labels": [ "Hyperlane" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [ "Smart Contract", "Blockchain Interoperability", "Bridge", "Cross-Chain Bridge", "https://hyperlane.xyz", "https://docs.hyperlane.xyz" ] }, { "id": "hyperledger-aries", "title": "Hyperledger Aries", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hyperledger Aries is an open-source toolkit providing the infrastructure for peer-to-peer interactions, secure messaging, and the exchange of verifiable credentials between decentralised identity agents. It implements DIDComm messaging and credential protocols, sitting above the ledger layer so that agents can issue, hold, and verify credentials independently of any specific blockchain. Aries is commonly paired with Hyperledger Indy as its verifiable-data registry.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:hyperledger-aries", "labels": [ "Hyperledger Aries" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "hyperledger-besu", "title": "Hyperledger Besu", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Enterprise-grade Ethereum client developed by the Hyperledger Foundation supporting both public and permissioned private blockchain deployments with pluggable consensus mechanisms including Proof of Work, Proof of Authority (Clique/IBFT), and Practical Byzantine Fault Tolerance (PBFT). Besu provides full EVM compatibility, privacy extensions via private transaction groups, permissioning, and a JSON-RPC/WebSocket API surface, making it suitable for enterprise consortia, regulated financial networks, and cross-chain interoperability scenarios.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:hyperledger-besu", "labels": [ "Hyperledger Besu", "BC-0427-hyperledger-besu" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BC-0001-blockchain", "BC-0142-smart-contract", "BC-0315-zero-knowledge-proof", "BC-0426-hyperledger-fabric", "BC-0428-enterprise-blockchain-architecture", "BC-0429-permissioned-blockchain", "BC-0431-privacy-preserving-blockchain", "Hyperledger Foundation", "HyperledgerFoundation", "InterledgerProtocol", "BlockchainDomain", "ConsensusProtocol", "Hyperledger Fabric", "Practical Byzantine Fault Tolerance", "PrivateChannels", "ProofOfAuthority", "ProofOfWork" ] }, { "id": "hyperledger-fabric", "title": "Hyperledger Fabric", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissioned blockchain framework enabling enterprise consortia to build modular, confidential distributed-ledger systems through private channels, chaincode (smart contracts), pluggable consensus mechanisms, and membership service providers enforcing identity and access control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:hyperledger-fabric", "labels": [ "Hyperledger Fabric", "BC-0067-hyperledger-fabric", "BC-0426-hyperledger-fabric", "HyperledgerFabric" ], "is_subclass_of": [ "Permissioned Blockchain" ], "wikilinks": [ "BC-0120-consensus-mechanism", "BC-0142-smart-contract", "BC-0315-zero-knowledge-proof", "BC-0427-hyperledger-besu", "BC-0428-enterprise-blockchain-architecture", "BC-0429-permissioned-blockchain", "BC-0430-private-channels", "BC-0446-supply-chain-traceability", "HyperledgerFoundation", "AccessControl", "BlockchainDomain", "PracticalByzantineFaultTolerance", "PrivateChannels", "PublicBlockchain", "SmartContract" ] }, { "id": "hyperledger-foundation", "title": "hyperledger foundation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Hyperledger Foundation is an open-source collaborative initiative hosted by the Linux Foundation, founded in December 2015, that stewards a portfolio of enterprise-grade distributed ledger frameworks, identity toolkits, and interoperability libraries. Its flagship projects include Hyperledger Fabric (a modular permissioned DLT with pluggable consensus and chaincode-based smart contracts), Hyperledger Besu (an Ethereum-compatible client for public and private enterprise networks), Hyperledger Aries (a decentralised identity and verifiable credentials stack), and Hyperledger Cacti (a cross-ledger interoperability framework). The Foundation operates under an open-governance model with a Technical Steering Committee, project lifecycle governance (Incubation, Graduated, Dormant), and membership tiers spanning major technology vendors, financial institutions, and government bodies worldwide.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hyperledger-foundation", "labels": [ "Hyperledger Foundation", "Hyperledger Project", "HyperledgerFoundation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "hyperledger-indy", "title": "Hyperledger Indy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hyperledger Indy is the purpose-built, permissioned, Byzantine-fault-tolerant distributed-ledger project hosted by the Linux Foundation-resident Hyperledger Foundation (now consolidated under the LF Decentralized Trust umbrella since June 2024) that provides the canonical reference implem...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hyperledger-indy", "labels": [ "Hyperledger Indy" ], "is_subclass_of": [ "Network Component", "Blockchain", "Verifiable Data Registry", "Permissioned Distributed Ledger", "Decentralized Identifier Method", "Byzantine Fault Tolerant System", "Hyperledger Project" ], "wikilinks": [ "Aardvark BFT", "Allen 2016 The Path to Self-Sovereign Identity", "AnonCreds", "AnonCreds Specification v1.0", "Aries Askar", "Aries Cloud Agent Python", "Aries Cloud Agent Python", "Atala PRISM", "Aublin Mokhtar Qu\u00e9ma 2013 ICDCS RBFT", "BC Wallet", "Bifold Wallet", "BLS12-381", "Boneh Boyen Shacham 2004 CRYPTO Short Group Signatures", "Byzantine Fault Tolerant System", "Camenisch Drijvers Lehmann 2016 TRUST Anonymous Attestation", "Camenisch Kohlweiss Soriente 2009 PKC Accumulator Anonymous Credentials", "Camenisch Lysyanskaya 2002 SCN Signature Scheme", "Camenisch Lysyanskaya 2004 CRYPTO Bilinear Maps", "Camenisch-Lysyanskaya Signature", "Cameron 2005 The Laws of Identity" ] }, { "id": "hyperledger-iroha", "title": "Hyperledger Iroha", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Hyperledger Iroha is a general-purpose permissioned Byzantine-fault-tolerant Distributed Ledger framework originally designed by Soramitsu (Tokyo, Japan) in collaboration with Hitachi, NTT Data and Colu, contributed to the Hyperledger Foundation (now Linux Foundation Decentralised Trust) ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:hyperledger-iroha", "labels": [ "Hyperledger Iroha" ], "is_subclass_of": [ "Network Component", "Blockchain", "Permissioned Blockchain", "Distributed Ledger", "Byzantine Fault Tolerant System", "Hyperledger Project", "Enterprise Blockchain Framework" ], "wikilinks": [ "Aleo", "Androulaki et al 2018 Hyperledger Fabric EuroSys", "Asset Tokenisation", "Atlantic Council CBDC Tracker Cambodia", "Atomic Batch Transactions", "Atomic Multi-Asset Transfer", "Aumasson 2013 BLAKE2 ACNS", "Aztec Protocol", "Bakong", "Bernstein et al 2011 Ed25519 High-speed Signatures", "BIS 2021-2024 CBDC Country Briefs Cambodia Bakong", "BIS 2024 Project Agora", "BIS Innovation Hub 2022-2024 Project mBridge", "BLAKE2b", "BLAKE2b Hashing", "BoE BIS Innovation Hub 2023 Project Rosalind", "Bokolo Cash", "Brown 2018 Corda Platform R3 White Paper", "BSI PAS 19668", "Buchman 2016 Tendermint MSc Thesis" ] }, { "id": "hyperparameter-optimisation", "title": "Hyperparameter Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hyperparameter optimisation is the automated search process for the configuration values \u2014 such as learning rate, regularisation strength, architecture depth, and batch size \u2014 that govern a machine learning model's training dynamics but are not learned directly from data, with the aim of maximising held-out validation performance. It encompasses grid search, random search, Bayesian optimisation, and gradient-based meta-learning, operating over an outer loop that wraps the inner model training procedure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:hyperparameter-optimisation", "labels": [ "Hyperparameter Optimisation" ], "is_subclass_of": [ "Hyperparameter", "Bayesian Optimisation" ], "wikilinks": [] }, { "id": "hyperparameter-tuning", "title": "Hyperparameter Tuning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hyperparameter tuning is the systematic process of searching over the configuration space of parameters that govern the training process of a machine learning model \u2014 distinct from the learnable parameters updated during training itself. Common hyperparameters include learning rate, batch size, network depth and width, regularisation coefficients, dropout rate, optimiser choice, and architectural decisions such as kernel size or number of attention heads. The tuning process employs search strategies \u2014 including grid search, random search, Bayesian optimisation, population-based methods, and evolutionary algorithms \u2014 to identify configurations that maximise model performance on a held-out validation set while controlling for overfitting. Efficient hyperparameter optimisation is critical to practical machine learning deployment because model generalisation is often highly sensitive to these configuration choices, and naive exhaustive search is computationally intractable in high-dimensional spaces.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:hyperparameter-tuning", "labels": [ "Hyperparameter Tuning", "Parameter Tuning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "hyperparameter", "title": "Hyperparameter", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A configuration variable set before training that controls the learning process but is not learned from data. Examples include learning rate, batch size, number of layers, dropout rate, and regularisation coefficients. Hyperparameter selection directly determines model capacity, convergence speed, and generalisation, making their tuning a critical step in building effective machine learning systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:hyperparameter", "labels": [ "Hyperparameter", "Model Hyperparameter" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique" ], "wikilinks": [ "MachineLearningDomain" ] }, { "id": "hyperscale-cloud", "title": "Hyperscale Cloud", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The tier of cloud computing infrastructure operated by major providers like AWS, Azure, and Google Cloud, characterized by massive global capacity and enterprise-grade services.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:hyperscale-cloud", "labels": [ "Hyperscale Cloud" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "hyperspectral-imaging", "title": "Hyperspectral Imaging", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:hyperspectral-imaging", "labels": [ "Hyperspectral Imaging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "hypertext-transfer-protocol", "title": "Hypertext Transfer Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Hypertext Transfer Protocol (HTTP) is an application-layer, request-response protocol for exchanging hypermedia documents and data across the web. Clients issue requests with methods and headers, and servers return responses with status codes and content. HTTP is the foundational protocol of the World Wide Web and of most modern web APIs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hypertext-transfer-protocol", "labels": [ "Hypertext Transfer Protocol" ], "is_subclass_of": [ "Application Layer" ], "wikilinks": [] }, { "id": "hypervisor", "title": "Hypervisor", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A hypervisor, or virtual machine monitor, is system software that creates and runs virtual machines by abstracting and partitioning a host's physical CPU, memory, and I/O resources. Type 1 (bare-metal) hypervisors run directly on hardware, while Type 2 (hosted) hypervisors run atop a conventional operating system. By isolating multiple guest operating systems on shared hardware, the hypervisor is the foundational technology of server virtualisation and cloud computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:hypervisor", "labels": [ "Hypervisor" ], "is_subclass_of": [ "Virtualisation", "Computing and Cloud" ], "wikilinks": [] }, { "id": "hypothesis-testing", "title": "Hypothesis Testing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hypothesis testing is a statistical inference procedure for deciding, on the basis of sample data, whether to reject a null hypothesis in favour of an alternative, using a test statistic and a pre-specified significance level. It underpins experimental design by providing the formal framework for determining whether an observed effect is unlikely to have arisen by chance. Bayesian decision theory offers an alternative, probability-based framework for the same class of decisions.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:hypothesis-testing", "labels": [ "Hypothesis Testing" ], "is_subclass_of": [ "Statistical Inference" ], "wikilinks": [] }, { "id": "iata-temperature-control-regulations", "title": "IATA Temperature Control Regulations", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The IATA Temperature Control Regulations (TCR) are the International Air Transport Association's standards governing the handling, packaging, labelling, and documentation of temperature-sensitive air cargo such as pharmaceuticals, vaccines, and perishables. They define acceptable temperature ranges, qualified packaging, and chain-of-custody requirements to preserve product integrity throughout air transport. Compliance is mandatory for carriers and shippers operating in regulated cold-chain logistics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iata-temperature-control-regulations", "labels": [ "IATA Temperature Control Regulations" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ibc-specification", "title": "IBC Specification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Inter-Blockchain Communication (IBC) specification is an open protocol standard for authenticated, ordered, and reliable message passing between independent distributed ledgers. Originating in the Cosmos ecosystem, it uses light-client verification and Merkle proofs so that two chains can trustlessly relay packets such as token transfers without a trusted intermediary. IBC underpins much of the modular and cross-chain interoperability landscape.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ibc-specification", "labels": [ "IBC Specification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ibc", "title": "IBC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Inter-Blockchain Communication is a protocol for relaying authenticated data and tokens between independent blockchains. It is the native interoperability standard of the Cosmos ecosystem.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ibc", "labels": [ "IBC", "IBC Protocol" ], "is_subclass_of": [ "Blockchain Interoperability" ], "wikilinks": [ "Cosmos SDK", "Merkle Tree", "Cross-Chain Bridge", "Tendermint", "Blockchain Interoperability", "https://www.ibcprotocol.dev", "https://github.com/cosmos/ibc" ] }, { "id": "ibm-food-trust", "title": "IBM Food Trust", "domain": "blockchain", "domain_name": "Blockchain", "definition": "IBM Food Trust is a permissioned blockchain network, built on Hyperledger Fabric, that enables food industry participants \u2014 growers, processors, distributors, retailers, and regulators \u2014 to digitally record, share, and verify supply chain events from farm to shelf. By anchoring provenance records in an immutable distributed ledger, it supports rapid traceability during food safety incidents, reduces waste through better freshness visibility, and builds consumer confidence in product origin claims. The platform operates as a multi-party ecosystem in which each participant controls the data they contribute while selectively sharing access with downstream trading partners. Walmart, Carrefour, Nestl\u00e9, and other major retailers have deployed it to meet regulatory traceability mandates such as the US FDA Food Safety Modernization Act.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ibm-food-trust", "labels": [ "IBM Food Trust" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [ "Hyperledger Fabric", "Food Safety", "Distributed Ledger", "Supply Chain" ] }, { "id": "ibm-true-north", "title": "IBM TrueNorth", "domain": "robotics", "domain_name": "Robotics", "definition": "IBM TrueNorth is a neuromorphic processor developed by IBM that implements spiking neural network architecture in hardware.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ibm-true-north", "labels": [ "IBM TrueNorth" ], "is_subclass_of": [ "Neuromorphic Computing" ], "wikilinks": [ "Hardware", "Neuromorphic Computing", "Neural Network" ] }, { "id": "ibm", "title": "IBM", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "IBM is a technology company providing enterprise hardware, software, cloud services and consulting, with a long history in mainframe and research computing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ibm", "labels": [ "IBM" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [ "Database Systems", "Web3 Infrastructure", "Cloud Computing", "https://www.ibm.com/", "https://www.ibm.com/cloud" ] }, { "id": "iccv", "title": "ICCV", "domain": "ai", "domain_name": "Ai", "definition": "ICCV (International Conference on Computer Vision) is a biennial top-tier academic conference sponsored by IEEE that focuses on the full spectrum of computer vision research, from low-level image processing to high-level scene understanding and embodied intelligence. Held in odd-numbered years and alternating with CVPR in global prestige, ICCV features highly selective peer review and is associated with the Marr Prize for outstanding contributions to the field.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iccv", "labels": [ "ICCV", "IEEE ICCV" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "ice-protocol", "title": "ICE Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Interactive Connectivity Establishment (ICE) is an IETF framework that enables two peers behind network address translators (NATs) or firewalls to discover and negotiate the best path for direct media and data connections. It gathers candidate transport addresses via STUN and TURN servers, then performs connectivity checks to select a working candidate pair. ICE is a foundational component of real-time peer-to-peer communication.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ice-protocol", "labels": [ "ICE Protocol", "ICE Framework" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "iclr", "title": "ICLR", "domain": "ai", "domain_name": "Ai", "definition": "The International Conference on Learning Representations (ICLR) is a premier academic venue dedicated to research in deep learning and representation learning, co-founded in 2013 by Yoshua Bengio and Yann LeCun to provide a focused home for the emerging field. ICLR is distinguished by its fully open, internet-based double-open peer review conducted on the OpenReview platform, making submitted manuscripts, reviewer critiques, and author rebuttals publicly visible throughout the review cycle. The conference covers neural network architectures, optimisation methods, generalisation theory, self-supervised learning, reinforcement learning, and the intersection of deep learning with natural language processing, computer vision, and scientific domains. ICLR ranks among the three most competitive machine-learning venues globally and has been the publication forum for many landmark advances including attention-based sequence models, large-language-model scaling investigations, and representation-learning theory.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:iclr", "labels": [ "ICLR" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "icml", "title": "ICML", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The International Conference on Machine Learning (ICML) is the premier annual academic conference for machine learning research, organised by the International Machine Learning Society (IMLS) since 1980. ICML serves as the primary venue for disseminating foundational and applied advances across supervised learning, unsupervised learning, reinforcement learning, deep learning, optimisation, and the theory of machine learning. Proceedings are published through the open-access Proceedings of Machine Learning Research (PMLR), making ICML papers freely available and widely cited. Together with NeurIPS and ICLR, ICML forms the trifecta of elite machine learning publication venues that shape research directions and career trajectories across academia and industry globally.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:icml", "labels": [ "ICML" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "ico-ai-guidance", "title": "ICO AI Guidance", "domain": "governance", "domain_name": "Governance", "definition": "The ICO AI Guidance is the body of guidance issued by the UK Information Commissioner's Office on applying data-protection law to artificial-intelligence systems. It covers lawful basis, fairness, transparency, automated decision-making, and the trade-offs between accuracy and privacy when processing personal data in AI. The guidance helps organisations demonstrate accountability and conduct data-protection impact assessments for AI deployments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ico-ai-guidance", "labels": [ "ICO AI Guidance", "ICO Guidance on AI", "UK ICO Guidance" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "ico", "title": "ICO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Initial Coin Offering (ICO) is a blockchain-based fundraising mechanism in which a project or organisation sells newly issued cryptographic tokens to investors in exchange for established cryptocurrencies (typically ETH or BTC) or fiat currency, with the tokens granting access to a future service, governance rights, or speculative value appreciation. ICOs gained widespread use in 2017\u20132018 as an alternative to traditional venture capital and IPO processes, raising billions of dollars before facing significant regulatory scrutiny from securities authorities worldwide. They differ from traditional securities offerings in being permissionless, globally accessible, and typically not subject to investor accreditation requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ico", "labels": [ "ICO" ], "is_subclass_of": [ "Crypto Token" ], "wikilinks": [] }, { "id": "icra", "title": "ICRA", "domain": "robotics", "domain_name": "Robotics", "definition": "ICRA, the International Conference on Robotics and Automation, is an annual academic conference organised by the IEEE Robotics and Automation Society.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:icra", "labels": [ "ICRA" ], "is_subclass_of": [ "Robotics", "Academic Conference" ], "wikilinks": [ "IROS", "Mobile Manipulation", "Robotics" ] }, { "id": "icvcm-core-carbon-principles", "title": "ICVCM Core Carbon Principles", "domain": "governance", "domain_name": "Governance", "definition": "The ICVCM Core Carbon Principles (CCPs) are a set of ten threshold standards established by the Integrity Council for the Voluntary Carbon Market to define what constitutes a high-quality carbon credit. They cover governance, emissions impact, and sustainable development dimensions, requiring that credits be real, additional, quantified, permanent, independently verified, and contribute positively to sustainable development goals without causing harm to local communities or ecosystems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:icvcm-core-carbon-principles", "labels": [ "ICVCM Core Carbon Principles", "Core Carbon Principles" ], "is_subclass_of": [ "ICVCM Framework" ], "wikilinks": [] }, { "id": "icvcm-framework", "title": "ICVCM Framework", "domain": "governance", "domain_name": "Governance", "definition": "The ICVCM Framework is the governance framework established by the Integrity Council for the Voluntary Carbon Market to set quality benchmarks for carbon credits. Centred on the Core Carbon Principles and an associated Assessment Framework, it defines criteria for additionality, permanence, robust quantification, and transparent governance. The framework aims to restore trust and comparability across voluntary carbon-credit issuers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:icvcm-framework", "labels": [ "ICVCM Framework" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "ide-coding-agents", "title": "IDE Coding Agents", "domain": "ai", "domain_name": "Ai", "definition": "AI coding agents embedded within integrated development environments as extensions or sidebars, implementing plan-then-act loops with approval gates and cost transparency, enabling autonomous multi-step software development from within the developer's primary workspace \u2014 includes Cline, Kilo Code (successor to Roo Code), OpenHands, and claw-code-agent.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:ide-coding-agents", "labels": [ "IDE Coding Agents" ], "is_subclass_of": [ "Agent Harness", "Agentic AI", "Agent Frameworks", "Autonomous Coding" ], "wikilinks": [ "Agent Harness", "Terminal Coding Agents", "Harness Configuration Packs", "Internal AI Harness", "External AI Harness", "Model Context Protocol", "Tool Call Loop", "Large Language Model", "Agent Frameworks", "Agentic AI", "Autonomous Coding", "Browser Automation", "Computer Use", "Agent Evaluation Benchmarks", "Multi-Agent Orchestration Frameworks", "Plan-and-Execute Pattern", "ReAct Pattern", "Tool Use", "Function Calling", "Context Window" ] }, { "id": "idnow", "title": "IDnow", "domain": "security", "domain_name": "Security", "definition": "A German identity verification company that provides electronic identification, know-your-customer, and digital signing services. It offers automated and agent-assisted verification of identity documents and biometrics.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:idnow", "labels": [ "IDnow" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [ "Identity Verification", "Biometric Authentication", "Regulatory Compliance" ] }, { "id": "iec-61131-3", "title": "IEC 61131-3", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IEC 61131-3 is the third part of the IEC 61131 international standard for programmable logic controllers, defining the syntax and semantics of five programming languages used in industrial automation: Ladder Diagram (LD), Function Block Diagram (FBD), Structured Text (ST), Instruction List (IL), and Sequential Function Chart (SFC). The standard provides a common programming model across PLC vendors, enabling portability of control logic and reducing the effort required to maintain industrial systems. It is the primary reference for software engineering practice in the industrial automation domain and forms the basis of IEC 61131-3-compliant toolchains and runtime environments used in manufacturing, process control, and infrastructure systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iec-61131-3", "labels": [ "IEC 61131-3", "IEC 61131" ], "is_subclass_of": [ "ISO/IEC" ], "wikilinks": [] }, { "id": "iec-61508", "title": "iec 61508", "domain": "standards", "domain_name": "Standards", "definition": "IEC 61508 is the foundational international functional safety standard published by the International Electrotechnical Commission (IEC), specifying systematic requirements for the full safety lifecycle of electrical, electronic, and programmable electronic (E/E/PE) safety-related systems across all industry sectors. It introduces the Safety Integrity Level (SIL 1\u20134) framework as a quantitative measure of risk reduction, mandating specific probabilistic failure-rate targets and qualitative development-process constraints for each level. The standard governs hardware, software, and system-level requirements from hazard identification and risk assessment through design, validation, operation, and decommissioning. Sector-specific derivative standards\u2014including ISO 26262 (automotive), EN 50128 (rail), IEC 62061 (machinery), and IEC 62443 (industrial cybersecurity)\u2014inherit its SIL concepts and lifecycle structure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:iec-61508", "labels": [ "IEC 61508", "IEC 61508 Functional Safety" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [] }, { "id": "iec-61800-adjustable-speed-electrical-power-drive-systems", "title": "IEC 61800 Adjustable Speed Electrical Power Drive Systems", "domain": "standards", "domain_name": "Standards", "definition": "An IEC standard series covering adjustable speed electrical power drive systems, addressing ratings, electromagnetic compatibility, functional safety and energy efficiency. It applies to power drive systems used to control motor speed.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iec-61800-adjustable-speed-electrical-power-drive-systems", "labels": [ "IEC 61800 Adjustable Speed Electrical Power Drive Systems", "IEC 61800", "IEC 61800 Adjustable Speed Drives", "IEC 61800 Drives Standard", "IEC 61800-7" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iec-62061", "title": "IEC 62061", "domain": "standards", "domain_name": "Standards", "definition": "IEC 62061 is an international standard published by the International Electrotechnical Commission that specifies requirements for the design, integration, and validation of safety-related electrical, electronic, and programmable electronic control systems (SRECS) for machinery. It defines a risk-based framework for achieving Safety Integrity Levels (SIL 1\u20133) in machinery safety applications, covering the entire safety lifecycle from hazard identification through to decommissioning. The standard is sector-specific machinery application of the generic IEC 61508 functional safety framework, harmonised under the EU Machinery Directive and its successor the EU Machinery Regulation. It provides quantitative methods for calculating Probability of Dangerous Failure per Hour (PFH) and Diagnostic Coverage (DC) for hardware and software subsystems.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:iec-62061", "labels": [ "IEC 62061" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iec-62443", "title": "IEC 62443", "domain": "security", "domain_name": "Security", "definition": "IEC 62443 is a multi-part international standards series developed by IEC Technical Committee 65 (in collaboration with ISA99) that defines requirements, policies, and lifecycle processes for securing Industrial Automation and Control Systems (IACS) against cyber threats. It establishes a risk-based framework covering security management systems, security policies and procedures, system and component requirements, and security levels (SL 1\u20134) that grade protection against progressively sophisticated threat actors. The standard addresses all stakeholders in the IACS lifecycle \u2014 asset owners, system integrators, and product suppliers \u2014 assigning distinct roles, responsibilities, and conformance obligations to each. IEC 62443 is widely adopted in sectors such as energy, water, chemicals, manufacturing, and critical national infrastructure as the authoritative basis for operational technology (OT) cybersecurity governance.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:iec-62443", "labels": [ "IEC 62443" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iec", "title": "IEC", "domain": "robotics", "domain_name": "Robotics", "definition": "The International Electrotechnical Commission (IEC) is the leading global standards organisation that prepares and publishes international standards for all electrical, electronic and related technologies, collectively known as electrotechnology. Its standards cover areas such as industrial automation, machinery safety, programmable controllers and functional safety, and are frequently adopted jointly with ISO. IEC standards are widely referenced in robotics and industrial systems to ensure interoperability, safety and regulatory compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:iec", "labels": [ "IEC" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "ieee-1451", "title": "IEEE 1451", "domain": "standards", "domain_name": "Standards", "definition": "An IEEE standard family defining interfaces for smart transducers, including transducer electronic data sheets (TEDS). It supports interoperability between sensors, actuators and networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-1451", "labels": [ "IEEE 1451", "IEEE 1451 Sensor Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-1588-ptp", "title": "IEEE 1588 PTP", "domain": "standards", "domain_name": "Standards", "definition": "An IEEE standard defining the Precision Time Protocol (PTP) for synchronising clocks across a packet network. It enables sub-microsecond clock synchronisation in measurement and control systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-1588-ptp", "labels": [ "IEEE 1588 PTP", "IEEE 1588" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-1872", "title": "IEEE 1872", "domain": "standards", "domain_name": "Standards", "definition": "IEEE 1872 defines an ontology for robotics and automation, providing a standard vocabulary and structure for representing knowledge in that domain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-1872", "labels": [ "IEEE 1872", "IEEE 1872 Ontology for Robotics", "IEEE 1872-2015" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-2418-1", "title": "IEEE 2418.1", "domain": "blockchain", "domain_name": "Blockchain", "definition": "IEEE 2418.1 is the IEEE Standard for the Framework of Blockchain Use in Internet of Things (IoT), establishing a common vocabulary, reference architecture, and conceptual building blocks for integrating distributed ledger technology with IoT systems. It specifies the relationship between blockchain components and IoT devices, defines trust and security requirements for IoT data provenance, and provides a foundation for interoperability across heterogeneous IoT deployments. The standard addresses data integrity, device identity, and transactional automation through smart contracts within constrained IoT environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ieee-2418-1", "labels": [ "IEEE 2418.1", "IEEE P2418.1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-7000-model-process", "title": "IEEE 7000 Model Process", "domain": "standards", "domain_name": "Standards", "definition": "An IEEE standard defining a model process for addressing ethical concerns during system design. It provides a process to translate values into system requirements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-7000-model-process", "labels": [ "IEEE 7000 Model Process" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-7000", "title": "IEEE 7000", "domain": "governance", "domain_name": "Governance", "definition": "IEEE 7000 (formally IEEE Std 7000-2021) is an international standard that defines a process model for embedding ethical considerations into the design and engineering of autonomous and software-intensive systems. It provides a structured methodology \u2014 spanning concept exploration, value elicitation, ethical risk analysis, and requirements specification \u2014 enabling engineers and organisations to systematically address value conflicts during the systems development life cycle. Grounded in value-sensitive design theory, the standard bridges engineering practice with stakeholder ethics, offering normative guidance on transparency, accountability, and harm avoidance in technology development.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:ieee-7000", "labels": [ "IEEE 7000", "IEEE 7000 Series", "IEEE 7000 Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-802-x", "title": "IEEE 802-X", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IEEE 802 is the family of IEEE standards governing local area, metropolitan area, and personal area networks, defining the physical and data-link layers for technologies such as Ethernet (802.3), Wi-Fi (802.11), and low-rate wireless PANs (802.15.4). The suite standardises framing, medium access control, and addressing so that heterogeneous equipment can interoperate. It forms the backbone of most wired and wireless networking in use today.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ieee-802-x", "labels": [ "IEEE 802-X", "IEEE 802", "IEEE 802 Committee", "IEEE 802 Standards", "IEEE 802.1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ieee-802-11", "title": "IEEE 802.11", "domain": "standards", "domain_name": "Standards", "definition": "IEEE 802.11 specifies the medium access control and physical layer protocols for wireless local area networks, commonly known as Wi-Fi.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-802-11", "labels": [ "IEEE 802.11", "IEEE 802.11 Wireless LAN", "IEEE 802.11ad" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-802-15-4", "title": "IEEE 802.15.4", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IEEE 802.15.4 is a low-rate wireless personal area network (LR-WPAN) standard that specifies the physical layer (PHY) and medium access control (MAC) sublayer for short-range, low-power, low-data-rate wireless communication, operating primarily in the 2.4 GHz ISM band at up to 250 kbps. It serves as the foundational radio layer for higher-level IoT protocols including Zigbee, Thread, WirelessHART, and 6LoWPAN, enabling battery-powered sensor and actuator networks with multi-year device lifetimes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:ieee-802-15-4", "labels": [ "IEEE 802.15.4" ], "is_subclass_of": [ "IEEE 802-X" ], "wikilinks": [] }, { "id": "ieee-802-3", "title": "IEEE 802.3", "domain": "standards", "domain_name": "Standards", "definition": "An IEEE standard defining Ethernet, including the physical layer and media access control for wired local area networks. It covers a range of speeds and media types.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-802-3", "labels": [ "IEEE 802.3" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-control-systems-society", "title": "IEEE Control Systems Society", "domain": "robotics", "domain_name": "Robotics", "definition": "The IEEE Control Systems Society (CSS) is a professional technical organisation within the Institute of Electrical and Electronics Engineers dedicated to advancing the theory, design, and application of control systems across engineering disciplines. It publishes flagship journals including IEEE Transactions on Automatic Control and the IEEE Control Systems Magazine, and organises premier conferences such as the Conference on Decision and Control (CDC). The Society bridges classical control theory with modern machine learning and autonomous systems research, promoting standards, educational resources, and an international community of practitioners. Its technical committees address domains ranging from robotics and aerospace to networked systems and intelligent transportation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:ieee-control-systems-society", "labels": [ "IEEE Control Systems Society" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "ieee-p2048-1", "title": "IEEE P2048-1", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P2048-1 is the part of the IEEE 2048 virtual and augmented reality standards series that establishes device taxonomy and definitions: a common vocabulary for classifying VR, AR, and mixed-reality hardware \u2014 head-mounted displays, handheld and projection systems, tracking and input devices \u2014 together with the defining attributes of each category. By fixing shared terminology and classification criteria, it gives the sibling parts of the series, regulators, and procurement specifications a consistent language for describing immersive devices and their capabilities.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ieee-p2048-1", "labels": [ "IEEE P2048-1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE P2048", "Technical Standard", "XR Device", "IEEE P2048-3" ] }, { "id": "ieee-p2048-2", "title": "IEEE P2048-2", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P2048-2 is the part of the IEEE 2048 virtual and augmented reality standards series addressing immersive video taxonomy and quality metrics. It classifies immersive video types \u2014 180-degree and 360-degree monoscopic and stereoscopic video, volumetric capture, and view-dependent formats \u2014 and defines measurable quality attributes such as resolution per eye, frame rate, motion-to-photon latency, and perceptual quality scores, giving producers, platforms, and device makers a common basis for specifying and comparing immersive video experiences.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ieee-p2048-2", "labels": [ "IEEE P2048-2" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE P2048", "Technical Standard", "Immersive Media", "IEEE P2048-3" ] }, { "id": "ieee-p-2048-3", "title": "IEEE P2048-3", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "IEEE P2048-3 is a standard in the IEEE 2048 series for immersive reality (XR) environments, specifically addressing person identification within virtual, augmented, and mixed reality systems. It defines terminology, reference architectures, and interoperability requirements for identifying, representing, and authenticating users and their avatars across XR platforms. The standard underpins identity continuity, access control, and social trust in multi-user immersive environments, forming a foundational component of the broader IEEE 2048 XR standardisation ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ieee-p-2048-3", "labels": [ "IEEE P2048-3" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-p2048-4", "title": "IEEE P2048-4", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P2048-4 is the part of the IEEE 2048 virtual and augmented reality standards series that specifies requirements and methods for verifying a person's identity in immersive (virtual and augmented reality) environments \u2014 Person Identity. It lets a user identify themselves to a site or service through one or more identity authorities, authenticate individual pieces of information without disclosing additional data, and present a certified visual appearance (avatar). Its sibling IEEE P2048-3 addresses the complementary media question of how immersive video files and streams are containerised, signalled and delivered, so that verified identity can govern entitlement to the immersive content those formats carry.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ieee-p2048-4", "labels": [ "IEEE P2048-4" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE P2048", "Technical Standard", "IEEE P2048-3", "Video Compression" ] }, { "id": "ieee-p-2048-9", "title": "IEEE P2048-9", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P2048-9 is a draft standard within the IEEE 2048 series on virtual reality and augmented reality, addressing a specific part of that family.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-p-2048-9", "labels": [ "IEEE P2048-9" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-p-2048", "title": "IEEE P2048", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P2048 is the draft project covering the IEEE 2048 series of standards on virtual reality and augmented reality.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-p-2048", "labels": [ "IEEE P2048", "IEEE 2048" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-p-7000", "title": "IEEE P7000", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P7000 is the draft project that produced the IEEE 7000 process model for addressing ethical concerns during system design.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-p-7000", "labels": [ "IEEE P7000" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-p-7003-2021", "title": "IEEE P7003-2021", "domain": "governance", "domain_name": "Governance", "definition": "IEEE P7003-2021 is a technical standard developed under the IEEE P7000 series on ethically aligned AI, specifically addressing algorithmic bias considerations in autonomous and intelligent systems. It provides a structured methodology for identifying, characterising, and mitigating unintended bias in algorithmic decision-making systems, covering the full development lifecycle from requirements elicitation through deployment and monitoring. The standard establishes processes for bias impact assessment, stakeholder engagement, and documentation requirements so that developers and organisations can systematically evaluate whether their AI systems produce discriminatory or inequitable outcomes across protected characteristics. It is complementary to IEEE P7001 (Transparency), IEEE P7002 (Data Privacy), and other standards in the IEEE 7000 series that collectively operationalise responsible AI development.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:ieee-p-7003-2021", "labels": [ "IEEE P7003-2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-p-7009", "title": "IEEE P7009", "domain": "standards", "domain_name": "Standards", "definition": "IEEE P7009 addresses fail-safe design for autonomous and semi-autonomous systems, defining methodologies for safe shutdown behaviour.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ieee-p-7009", "labels": [ "IEEE P7009" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IEEE", "Technical Standard" ] }, { "id": "ieee-ras", "title": "IEEE RAS", "domain": "robotics", "domain_name": "Robotics", "definition": "The IEEE Robotics and Automation Society (RAS) is the professional society within IEEE dedicated to advancing robotics and automation through publications, conferences, and the development of technical standards. It sponsors standards such as IEEE 1872 (ontologies for robotics and automation) and maintains working groups covering terminology, safety, and interoperability. RAS is a primary standards-developing body for the robotics field.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ieee-ras", "labels": [ "IEEE RAS" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ieee-robotics-and-automation-society", "title": "IEEE Robotics And Automation Society", "domain": "robotics", "domain_name": "Robotics", "definition": "The IEEE Robotics and Automation Society (IEEE RAS) is the preeminent international professional and technical society dedicated to advancing the theory, design, practice, and application of robotics, automation, and related technologies. Founded in 1984 as a successor to the IEEE Robotics and Automation Council, it serves a global membership of engineers, researchers, and practitioners through flagship conferences including ICRA and IROS, peer-reviewed publications such as the IEEE Transactions on Robotics, and technical committee activities that coordinate standards development and community building across the full breadth of robotics disciplines.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ieee-robotics-and-automation-society", "labels": [ "IEEE Robotics And Automation Society" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ieee-robotics-standards", "title": "IEEE Robotics Standards", "domain": "robotics", "domain_name": "Robotics", "definition": "IEEE Robotics Standards are the technical standards developed under IEEE for the robotics and automation domain, covering robot ontologies, ethical design, terminology, and component interoperability. Notable examples include IEEE 1872 for core ontologies and the IEEE 7000 series for ethically aligned design of autonomous systems. They provide a normative foundation for safe, interoperable, and accountable robotic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ieee-robotics-standards", "labels": [ "IEEE Robotics Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ieee-standards-association", "title": "IEEE Standards Association", "domain": "standards", "domain_name": "Standards", "definition": "The IEEE Standards Association (IEEE SA) is the standards development arm of the Institute of Electrical and Electronics Engineers, operating a consensus-based process through which industry, government, and academic stakeholders collaboratively develop and maintain technical standards covering electrical, electronic, computing, and communications technologies, with over 1,400 active standards and more than 900 standards under active development at any time.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:ieee-standards-association", "labels": [ "IEEE Standards Association" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ieee-standards", "title": "IEEE Standards", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IEEE Standards are technical standards developed and published by the Institute of Electrical and Electronics Engineers and its standards association, spanning electrical, electronic, computing and communications technologies. Well-known examples include the IEEE 802 family for local and metropolitan networks (Ethernet and Wi-Fi) and IEEE 754 for floating-point arithmetic. These standards promote interoperability, safety and consistent engineering practice across hardware and networking systems worldwide.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ieee-standards", "labels": [ "IEEE Standards" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "ieee", "title": "IEEE", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Institute of Electrical and Electronics Engineers (IEEE) is the world's largest technical professional organisation, comprising over 400,000 members across 160 countries, whose Standards Association (IEEE SA) develops and maintains more than 1,300 active standards governing electrical engineering, electronics, telecommunications, computer science, and emerging disciplines including artificial intelligence and autonomous systems. IEEE standards define foundational networking protocols such as IEEE 802.11 (Wi-Fi), IEEE 802.3 (Ethernet), and IEEE 802.15.4 (low-power wireless), as well as critical infrastructure specifications including IEEE 754 (floating-point arithmetic) and IEEE 1588 (Precision Time Protocol). The organisation conducts standards development through consensus-based volunteer working groups spanning industry, academia, and government, and publishes its technical library via IEEE Xplore. IEEE has expanded into AI ethics (IEEE 7000 series), autonomous vehicles, quantum computing, and neural interfaces, making it a primary governance body across both traditional and emerging technology domains.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:ieee", "labels": [ "IEEE", "IEEE (Institute of Electrical and Electronics Engineers)", "IEEE 2030", "IEEE Audio Standards", "IEEE Blockchain Standards", "IEEE Computer Society", "IEEE Conference on Games", "IEEE Ethically Aligned Design", "IEEE Learning Technology Standards Committee", "IEEE P2957", "IEEE P7001-2021", "IEEE Transactions on Pattern Analysis and Machine Intelligence" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ietf-rfc", "title": "IETF RFC", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An IETF RFC (Request for Comments) is a numbered document published by the Internet Engineering Task Force and the RFC Editor that records Internet specifications, protocols, procedures, and informational notes. Standards-track RFCs progress through community review to define the protocols that constitute the Internet, while other categories capture experimental, informational, or historic material. The RFC series is the canonical record through which Internet technology is standardised and disseminated.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ietf-rfc", "labels": [ "IETF RFC" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ietf", "title": "IETF", "domain": "standards", "domain_name": "Standards", "definition": "The Internet Engineering Task Force (IETF) is the principal international open-standards organisation responsible for developing and promoting voluntary Internet standards, particularly those comprising the Internet protocol suite (TCP/IP). It operates through volunteer working groups organised into technical areas, publishing specifications as Requests for Comments (RFCs) \u2014 the canonical technical definitions of protocols such as HTTP, TLS, QUIC, SMTP, and DNS. The IETF is a bottom-up, consensus-driven body with no formal membership; participation is open to any technically engaged individual. Its decision ethos of 'rough consensus and running code' has made it the dominant model for pragmatic, implementation-first standards development.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:ietf", "labels": [ "IETF", "IETF (Internet Engineering Task Force)", "IETF BCP 47", "IETF OAuth Working Group", "IETF PRIO Protocol", "IETF RFC 4120", "IETF RFC 7574 BitTorrent-like Gossip", "IETF Standard", "IETF WebRTC Standards" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ifac", "title": "IFAC", "domain": "robotics", "domain_name": "Robotics", "definition": "The International Federation of Automatic Control (IFAC) is the world's primary international organisation for the science and engineering of automatic control, founded in 1957 and comprising national member organisations from over fifty countries. IFAC coordinates research, education, and standards activity across all aspects of control theory and its applications, including process control, robotics, mechatronics, aerospace, and intelligent transportation. It convenes the triennial IFAC World Congress and sponsors over forty technical committees covering specialised control domains. IFAC's publications and conference proceedings constitute a core reference body for the global control engineering community.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ifac", "labels": [ "IFAC" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "ifrs-s1", "title": "IFRS S1", "domain": "governance", "domain_name": "Governance", "definition": "IFRS S1 is the inaugural sustainability-disclosure standard issued by the International Sustainability Standards Board, setting general requirements for disclosing sustainability-related financial information. It requires entities to report material risks and opportunities across governance, strategy, risk management, and metrics that could affect their prospects. IFRS S1 establishes the common baseline on which topic-specific standards such as IFRS S2 build.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ifrs-s1", "labels": [ "IFRS S1", "IFRS S1 S2" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ifrs-s2", "title": "IFRS S2", "domain": "governance", "domain_name": "Governance", "definition": "IFRS S2 is the climate-related disclosure standard issued by the International Sustainability Standards Board, specifying how entities report climate-related risks and opportunities. Building on the TCFD recommendations, it requires disclosure of physical and transition risks, greenhouse-gas emissions across Scopes 1 to 3, and climate-resilience analysis. IFRS S2 operates alongside the general IFRS S1 baseline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ifrs-s2", "labels": [ "IFRS S2", "ISSB IFRS S2" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ilo-core-labour-standards", "title": "ILO Core Labour Standards", "domain": "governance", "domain_name": "Governance", "definition": "The ILO Core Labour Standards are the fundamental principles and rights at work codified by the International Labour Organization, covering freedom of association and collective bargaining, the elimination of forced and child labour, non-discrimination, and a safe and healthy working environment. They are expressed through ratifiable conventions that member states commit to uphold. The standards form the baseline against which ethical sourcing and labour due-diligence are assessed.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ilo-core-labour-standards", "labels": [ "ILO Core Labour Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "imf-cbdc-framework", "title": "IMF CBDC Framework", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The IMF CBDC Framework is the International Monetary Fund's analytical and policy guidance for the design, adoption, and macro-financial implications of central bank digital currencies. It addresses monetary-policy transmission, financial stability, cross-border interoperability, and capacity development for member-state central banks. The framework helps jurisdictions evaluate whether and how to issue a CBDC.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:imf-cbdc-framework", "labels": [ "IMF CBDC Framework" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "imf-crypto-asset-classification-framework", "title": "IMF Crypto Asset Classification Framework", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The IMF Crypto Asset Classification Framework is the International Monetary Fund's taxonomy and policy guidance for categorising crypto assets by their economic function and legal characteristics, such as whether they serve as means of payment, investment instruments, or stablecoins. It informs consistent statistical treatment, regulation, and macro-financial risk monitoring across jurisdictions. The framework supports coherent global oversight of the crypto-asset sector.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:imf-crypto-asset-classification-framework", "labels": [ "IMF Crypto Asset Classification Framework" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "imf", "title": "IMF", "domain": "governance", "domain_name": "Governance", "definition": "The International Monetary Fund (IMF) is an intergovernmental organisation of 190 member countries, established in 1944 under the Bretton Woods Agreement, whose core mandate is to foster global monetary cooperation, secure exchange rate stability, facilitate balanced international trade, and provide financial assistance and policy advice to members experiencing balance-of-payments difficulties. The Fund operates as the world's primary multilateral lender of last resort for sovereign balance-of-payments crises, deploying conditional lending programmes backed by Special Drawing Rights (SDR) quotas contributed by members. Beyond crisis finance, the IMF conducts macroeconomic surveillance through bilateral Article IV consultations, publishes the World Economic Outlook and Global Financial Stability Report, and delivers technical assistance in fiscal, monetary, and financial-sector policy domains. Since 2017 the IMF has expanded into digital asset governance, publishing classification frameworks for crypto assets, CBDC design handbooks, and a Finternet unified-ledger vision that positions it as a de-facto norm-setter for sovereign digital money.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:imf", "labels": [ "IMF", "International Monetary Fund" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "imu-sensors", "title": "IMU Sensors", "domain": "robotics", "domain_name": "Robotics", "definition": "An inertial measurement unit (IMU) is a sensor package that measures a body's specific force and angular rate, typically combining a three-axis accelerometer and a three-axis gyroscope, often with a magnetometer. By integrating these measurements it estimates orientation, velocity, and relative motion without external references. IMUs are central to motion tracking, navigation, and stabilisation in robotics, XR, and aerospace.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:imu-sensors", "labels": [ "IMU Sensors", "Inertial Sensors" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "imu", "title": "imu", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Inertial Measurement Unit (IMU) is a self-contained electronic sensor module that integrates tri-axial accelerometers, gyroscopes, and optionally magnetometers to measure a rigid body's specific force, angular rate, and magnetic heading relative to an inertial reference frame without dependence on external infrastructure. MEMS-fabricated IMUs fuse their outputs through Kalman or complementary filter algorithms to yield real-time pose and orientation estimates at high sample rates, feeding inertial navigation, SLAM pipelines, and 6-DoF tracking systems. IMUs span performance grades from low-cost consumer MEMS units (bias instability >1\u00b0/hr) to navigation-grade fibre-optic and ring-laser gyro systems used in aerospace and submarine applications. They are integral to XR headsets, autonomous vehicles, UAVs, legged robots, wearables, and surgical instruments wherever low-latency, infrastructure-independent motion awareness is required.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:imu", "labels": [ "IMU", "IMU Preintegration", "IMU Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "iosco", "title": "iosco", "domain": "governance", "domain_name": "Governance", "definition": "The International Organisation of Securities Commissions (IOSCO) is the principal global standard-setting body for securities and capital markets regulation, founded in 1983 and headquartered in Madrid, with membership spanning national securities regulators and self-regulatory organisations from over 130 jurisdictions. IOSCO develops and publishes Objectives and Principles of Securities Regulation, Recommendations, and targeted policy reports adopted by member regulators to promote fair, efficient, and transparent capital markets, protect investors, and reduce systemic risk. Its remit has expanded to cover digital asset markets, crypto-asset intermediaries, decentralised finance, and AI-driven algorithmic trading, making it a central node in the global financial regulatory architecture alongside the Financial Stability Board and Basel Committee.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:iosco", "labels": [ "IOSCO", "IOSCO Recommendations" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "ip-addressing", "title": "IP Addressing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IP addressing is the scheme by which every host and interface on an Internet Protocol network is assigned a numeric identifier used to locate and route datagrams to it. It encompasses the structure of IPv4 and IPv6 address spaces, the partition of addresses into network and host portions via subnet masks and prefixes, and the assignment mechanisms that allocate addresses to devices. Correct addressing is the precondition for routing decisions and end-to-end delivery across interconnected networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ip-addressing", "labels": [ "IP Addressing" ], "is_subclass_of": [ "Internet Protocol", "Network Infrastructure" ], "wikilinks": [] }, { "id": "ip-adapter", "title": "IP-Adapter", "domain": "ai", "domain_name": "Ai", "definition": "IP-Adapter (Image Prompt Adapter) is a lightweight adapter module for pre-trained text-to-image diffusion models that enables image-conditioned generation by injecting reference image features via a decoupled cross-attention mechanism. Introduced by Tencent AI Lab in 2023, it allows users to supply a reference image alongside a text prompt to control style, subject identity, or composition without fine-tuning the base diffusion model. The adapter architecture inserts parallel cross-attention layers that process image embeddings from a pre-trained image encoder such as CLIP, keeping base model weights frozen.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ip-adapter", "labels": [ "IP-Adapter", "IP Adapter", "IP-Adapter Image Conditioning", "Image Prompt IP-Adapter Module" ], "is_subclass_of": [ "Adapter Modules" ], "wikilinks": [] }, { "id": "ipfs-content-addressing", "title": "IPFS Content Addressing", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The addressing scheme used by the InterPlanetary File System, in which content is identified by a cryptographic hash of its data rather than by location. The resulting content identifier changes if the content changes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ipfs-content-addressing", "labels": [ "IPFS Content Addressing" ], "is_subclass_of": [ "IPFS" ], "wikilinks": [ "Cryptographic Hash", "Distributed Storage", "IPFS" ] }, { "id": "ipfs", "title": "ipfs", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The InterPlanetary File System (IPFS) is a peer-to-peer, content-addressed hypermedia protocol and distributed file system in which each piece of content is identified by a Content Identifier (CID) \u2014 a self-describing cryptographic hash derived from the content itself \u2014 rather than by its network location. Nodes exchange data blocks via Bitswap and route lookups through a Kademlia-based Distributed Hash Table implemented in libp2p, while the underlying data model (IPLD) structures blocks as a Merkle DAG enabling deduplication and efficient versioning. IPFS operates as the primary off-chain storage layer for Web3 applications, decentralised websites, NFT metadata, and distributed knowledge repositories, complemented by Filecoin's incentive layer for long-term data persistence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ipfs", "labels": [ "IPFS", "Ipfs" ], "is_subclass_of": [ "Peer-to-Peer Network" ], "wikilinks": [] }, { "id": "ipld", "title": "IPLD", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IPLD (InterPlanetary Linked Data) is a data model and set of specifications for building content-addressed, hash-linked data structures that interoperate across distributed systems. It defines how to represent linked data as directed acyclic graphs whose edges are content identifiers (CIDs), allowing any hash-linked structure \u2014 Git commits, blockchain blocks, IPFS files \u2014 to be traversed through a common addressing scheme. IPLD provides the data layer beneath IPFS and Filecoin, decoupling the logical structure of data from the protocol used to store or transport it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ipld", "labels": [ "IPLD", "IPLD Specifications", "W3C IPLD" ], "is_subclass_of": [ "Content Addressing" ], "wikilinks": [] }, { "id": "iri", "title": "IRI", "domain": "data", "domain_name": "Data", "definition": "An Internationalised Resource Identifier (IRI) is a compact string that uniquely identifies an abstract or physical resource, generalising the Uniform Resource Identifier (URI) to permit characters from the full Unicode repertoire rather than only ASCII. IRIs are the foundational naming mechanism of the Semantic Web: every node and predicate in an RDF graph is named by an IRI, allowing data published by independent parties to refer unambiguously to the same entity. An IRI may be mapped to an equivalent URI through percent-encoding, preserving compatibility with legacy web infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iri", "labels": [ "IRI" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [] }, { "id": "iros", "title": "IROS", "domain": "robotics", "domain_name": "Robotics", "definition": "IROS, the International Conference on Intelligent Robots and Systems, is an annual academic robotics conference co-sponsored by the IEEE and the Robotics Society of Japan.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iros", "labels": [ "IROS" ], "is_subclass_of": [ "Robotics", "Academic Conference" ], "wikilinks": [ "ICRA", "Mobile Manipulation", "Robotics" ] }, { "id": "isa-95", "title": "ISA-95", "domain": "robotics", "domain_name": "Robotics", "definition": "ISA-95 is an international standard for integrating enterprise and control systems in manufacturing, defining models for the interface between business and production operations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:isa-95", "labels": [ "ISA-95" ], "is_subclass_of": [ "OPC UA" ], "wikilinks": [ "OPC UA" ] }, { "id": "isda-cdm", "title": "ISDA CDM", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ISDA CDM (Common Domain Model) is a standardised, machine-readable data and process model for derivatives and other financial transactions, published by the International Swaps and Derivatives Association. It provides a single, consistent representation of trade events, lifecycle processes, and legal terms so that disparate systems and smart contracts interpret them identically. It is foundational to interoperable enterprise tokenisation and automated post-trade processing on distributed ledgers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:isda-cdm", "labels": [ "ISDA CDM", "FINOS CDM" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "isda-common-domain-model", "title": "ISDA Common Domain Model", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The ISDA Common Domain Model is the full name of the standardised, machine-readable representation of financial products, trade events, and lifecycle processes maintained by the International Swaps and Derivatives Association. It establishes a single canonical model so that distributed ledgers, smart contracts, and institutional systems process derivatives transactions identically and without reconciliation. It is a cornerstone standard for interoperable, automated post-trade infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:isda-common-domain-model", "labels": [ "ISDA Common Domain Model", "FINOS Common Domain Model" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "isda", "title": "ISDA", "domain": "finance", "domain_name": "Finance", "definition": "ISDA, the International Swaps and Derivatives Association, is a trade body that produces standard documentation and definitions for over-the-counter derivatives markets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:isda", "labels": [ "ISDA" ], "is_subclass_of": [ "Traditional Finance" ], "wikilinks": [ "Financial Regulation", "Traditional Finance" ] }, { "id": "iso-10218-robot-safety", "title": "ISO 10218 Robot Safety", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard, in multiple parts, specifying safety requirements for industrial robots and robot systems. It addresses hazards and protective measures for robot design and integration.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-10218-robot-safety", "labels": [ "ISO 10218 Robot Safety" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-10218-1", "title": "ISO 10218-1", "domain": "standards", "domain_name": "Standards", "definition": "ISO 10218-1 specifies safety requirements for the design and construction of industrial robots.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-10218-1", "labels": [ "ISO 10218-1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-10218", "title": "iso 10218", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO 10218 is a two-part international safety standard published by the International Organisation for Standardisation specifying mandatory requirements for the design, construction, and safeguarding of industrial robots and robotic systems. Part 1 (ISO 10218-1) governs the robot unit itself, requiring manufacturers to implement safety-rated monitored stop, speed and force limitation, workspace monitoring, and control-reliable safety stop circuits before market placement. Part 2 (ISO 10218-2) governs robot system integration into workcells, requiring formal risk assessments per ISO 12100, safety control architecture documentation per ISO 13849 performance levels, and comprehensive declarations of conformity by system integrators. Together the two parts form the normative safety foundation for industrial robotics globally and are the prerequisite for CE marking of robot workcells in European markets under the EU Machinery Directive.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:iso-10218", "labels": [ "ISO 10218", "ISO 10218 Industrial Robot Safety", "ISO 10218 Safety Standard" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "iso-10816", "title": "ISO 10816", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ISO 10816 is an international standard specifying the measurement and evaluation of mechanical vibration of rotating machinery by readings taken on non-rotating parts. It defines vibration severity zones that classify equipment condition from acceptable to unacceptable, guiding when intervention is required. It is widely used as the reference basis for condition monitoring and predictive maintenance of pumps, motors, fans, and turbines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-10816", "labels": [ "ISO 10816" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "iso-11179", "title": "ISO 11179", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard, in multiple parts, for metadata registries that specifies how to represent and manage data element definitions and metadata. It supports consistent description and sharing of data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-11179", "labels": [ "ISO 11179", "ISO/IEC 11179" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-12100", "title": "ISO 12100", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO 12100 is the foundational international standard for the safety of machinery, setting out general principles for design, risk assessment and risk reduction. It defines a structured methodology in which designers identify hazards, estimate and evaluate risk, then apply inherently safe design, safeguarding and information-for-use in a defined order of priority. As a Type-A standard it provides the overarching framework that more specific Type-B and Type-C machinery standards build upon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iso-12100", "labels": [ "ISO 12100" ], "is_subclass_of": [ "Safety Standard" ], "wikilinks": [] }, { "id": "iso-13482", "title": "ISO 13482", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO 13482 is the international standard specifying safety requirements for personal care robots\u2014autonomous or semi-autonomous robots designed to perform tasks that directly assist humans in daily life, including mobility assistance, physical assistance, and person-carrier applications. Published in 2014, the standard defines risk assessment methodologies, protective measure requirements, and verification procedures tailored to the unique hazards of robots operating in close physical proximity to vulnerable populations such as the elderly and disabled. It complements the industrial robot safety standard ISO 10218 and is increasingly referenced in regulatory frameworks for assistive and service robotics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-13482", "labels": [ "ISO 13482", "ISO 13482 Robot Safety" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [] }, { "id": "iso-13849", "title": "ISO 13849", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard on safety of machinery, specifying safety-related parts of control systems. It provides requirements and guidance for the design of such parts to achieve required performance levels.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-13849", "labels": [ "ISO 13849" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-14001", "title": "ISO 14001", "domain": "governance", "domain_name": "Governance", "definition": "ISO 14001 is an international standard, published by the International Organization for Standardization, that specifies the requirements for an environmental management system (EMS). It provides a framework for an organisation to identify, manage, monitor, and continually improve its environmental performance through a plan-do-check-act cycle. Certification against ISO 14001 demonstrates conformance to a recognised set of environmental governance practices and supports regulatory compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iso-14001", "labels": [ "ISO 14001", "ISO 14001 Environmental Management" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "iso-14040", "title": "ISO 14040", "domain": "standards", "domain_name": "Standards", "definition": "ISO 14040 sets out the principles and framework for life cycle assessment as part of environmental management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-14040", "labels": [ "ISO 14040", "ISO 14040 LCA", "ISO 14040 LCA Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-14064", "title": "iso 14064", "domain": "governance", "domain_name": "Governance", "definition": "ISO 14064 is a three-part international standard series published by the International Organisation for Standardisation that specifies principles and requirements for the quantification, monitoring, reporting, and independent verification of greenhouse gas (GHG) emissions and removals. Part 1 addresses organisational-level GHG inventories including Scope 1, 2, and 3 emissions; Part 2 covers project-level GHG mitigation quantification underpinning carbon credit issuance; and Part 3 prescribes competence and process requirements for validation and verification bodies. Together the three parts form an auditable, globally recognised framework that underpins corporate sustainability reporting, voluntary and compliance carbon markets, and emerging blockchain-based carbon credit tokenisation schemes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-14064", "labels": [ "ISO 14064", "ISO 14064 GHG Protocol" ], "is_subclass_of": [ "Environmental Standards" ], "wikilinks": [] }, { "id": "iso-14067", "title": "ISO 14067", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "ISO 14067 is an international standard that specifies principles, requirements, and guidelines for quantifying and reporting the carbon footprint of a product based on life-cycle assessment. It defines how greenhouse-gas emissions associated with a product's life cycle are calculated and communicated, expressed in CO2-equivalent. It provides the methodological backbone for credible product-level carbon-footprint measurement and disclosure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-14067", "labels": [ "ISO 14067" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "iso-20022", "title": "ISO 20022", "domain": "finance", "domain_name": "Finance", "definition": "ISO 20022 is the international standard for financial services messaging, providing a universal methodology that decouples business concepts from syntax through a central repository of business components, message definitions, and multi-syntax encoding rules covering XML, ASN.1, and JSON. Published by the International Organisation for Standardisation and maintained by ISO Technical Committee 68, the standard spans payments (credit transfers, direct debits), securities settlement and clearing, trade finance, and foreign exchange, replacing legacy formats such as SWIFT MT and CHIPS. Its rich, structured data model carries full remittance information, Legal Entity Identifiers, purpose codes, and beneficiary details that legacy formats cannot accommodate, enabling straight-through processing, automated reconciliation, and enhanced regulatory reporting. Major financial market infrastructures worldwide \u2014 including TARGET2, CHAPS, Fedwire, and the SWIFT network \u2014 have migrated or are actively migrating to ISO 20022, making it the convergence standard for the global payments landscape.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:iso-20022", "labels": [ "ISO 20022", "ISO 20022 Payment Messaging" ], "is_subclass_of": [ "Interoperability Standard" ], "wikilinks": [] }, { "id": "iso-20400", "title": "ISO 20400", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "ISO 20400 is an international standard providing guidance on integrating sustainability into procurement processes and decisions across an organisation and its supply chains. It helps buyers consider environmental, social, and economic impacts when sourcing goods and services, aligning purchasing with responsible and ethical principles. As a guidance standard rather than a certifiable requirement, it shapes policy, supplier engagement, and accountability for sustainable sourcing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-20400", "labels": [ "ISO 20400" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "iso-21448", "title": "ISO 21448", "domain": "standards", "domain_name": "Standards", "definition": "ISO 21448 addresses the safety of the intended functionality for road vehicles, covering hazards arising from functional insufficiencies rather than system faults.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-21448", "labels": [ "ISO 21448", "ISO 21448 SOTIF" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-22000", "title": "ISO 22000", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "ISO 22000 is an international standard specifying the requirements for a food safety management system, enabling any organisation in the food chain to demonstrate control of food-safety hazards. It integrates HACCP principles with management-system structure, prerequisite programmes, and interactive communication across the supply chain. It is widely certified by producers, processors, and logistics providers, including those operating cold-chain monitoring.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-22000", "labels": [ "ISO 22000" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "iso-22144", "title": "ISO 22144", "domain": "standards", "domain_name": "Standards", "definition": "A standard published by ISO. The specific subject area is not determined from the identifier alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-22144", "labels": [ "ISO 22144" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-22166", "title": "ISO 22166", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard, in the robotics area, concerning modularity for service robots. It addresses requirements and information for modular robotic components.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-22166", "labels": [ "ISO 22166" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-22739", "title": "ISO 22739", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ISO 22739:2020 is the foundational International Organisation for Standardisation standard that establishes a comprehensive vocabulary of terms and definitions for distributed ledger technology (DLT) and blockchain systems, providing a normative reference for terminology used across all subsequent blockchain-related ISO standards, regulatory documents, and technical specifications. The standard was developed by ISO/TC 307 (Blockchain and distributed ledger technologies) and covers over 60 core concepts including blockchain, distributed ledger, smart contract, consensus mechanism, node, fork, digital asset, cryptographic key, and wallet.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-22739", "labels": [ "ISO 22739" ], "is_subclass_of": [ "Iso Standards" ], "wikilinks": [] }, { "id": "iso-23247", "title": "ISO 23247", "domain": "standards", "domain_name": "Standards", "definition": "ISO 23247 defines a framework for digital twins in manufacturing, covering reference architecture and the representation of manufacturing elements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-23247", "labels": [ "ISO 23247" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-23257", "title": "ISO 23257", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ISO/IEC 23257 is an international standard published jointly by ISO and IEC that defines a reference architecture for blockchain and distributed ledger technology (DLT) systems, providing a common vocabulary, conceptual model, and set of architectural views covering nodes, networks, consensus mechanisms, and system roles. It establishes a vendor-neutral framework that enables interoperability assessments, system design, and procurement across heterogeneous DLT ecosystems. The standard complements the broader ISO/TC 307 series on blockchain and DLT, serving as the architectural foundation upon which specific technology standards, conformance frameworks, and security profiles are built.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-23257", "labels": [ "ISO 23257" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-24165", "title": "ISO 24165", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ISO 24165 is an international standard specifying the Digital Token Identifier (DTI), a universal system for uniquely identifying digital tokens regardless of the distributed ledger or blockchain platform on which they reside. It defines the structure, syntax, and assignment procedures of the DTI, enabling interoperability across exchanges, regulators, financial institutions, and infrastructure providers. The standard is maintained by the Digital Token Identifier Foundation (DTIF) as the Registration Authority under ISO, and addresses persistent identification needs for cryptocurrencies, stablecoins, tokenised assets, and other on-chain instruments. ISO 24165 complements financial instrument identification standards such as ISIN and FIGI by providing blockchain-native unambiguous references usable in regulatory reporting, settlement, and data aggregation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:iso-24165", "labels": [ "ISO 24165", "ISO 24165 Digital Token Identifier" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-26000", "title": "ISO 26000", "domain": "standards", "domain_name": "Standards", "definition": "ISO 26000 provides guidance on social responsibility for organisations, covering principles and core subjects without specifying certifiable requirements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-26000", "labels": [ "ISO 26000" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-26262", "title": "ISO 26262", "domain": "standards", "domain_name": "Standards", "definition": "ISO 26262 is the international functional safety standard for electrical and electronic (E/E) systems in road vehicles, first published in 2011 by ISO Technical Committee 22 and revised in a second edition in 2018 to extend coverage to motorcycles (Part 12) and semiconductor components (Part 11). The standard introduces the Automotive Safety Integrity Level (ASIL) classification system, ranging from ASIL A (lowest rigour) to ASIL D (highest rigour), derived from a Hazard Analysis and Risk Assessment (HARA) process that combines Severity, Exposure, and Controllability parameters for each hazardous event. Compliance requires evidence of a structured safety lifecycle encompassing concept phase, system design, hardware, software, production, and operation, providing automotive OEMs, Tier 1 suppliers, and semiconductor manufacturers with a shared framework for demonstrating that residual risk of hazardous E/E failure is acceptably low. The standard is widely mandated for ADAS, powertrain control, braking, and steering systems globally, and is increasingly complemented by SOTIF (ISO 21448) and the cybersecurity standard ISO/SAE 21434 for AI-enabled autonomous driving systems.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:iso-26262", "labels": [ "ISO 26262" ], "is_subclass_of": [ "Safety Standard" ], "wikilinks": [] }, { "id": "iso-27001", "title": "ISO 27001", "domain": "security", "domain_name": "Security", "definition": "ISO/IEC 27001 is the internationally recognised standard specifying requirements for establishing, implementing, maintaining, and continually improving an Information Security Management System (ISMS). Published jointly by the International Organisation for Standardisation (ISO) and the International Electrotechnical Commission (IEC), the standard takes a risk-based approach to information security governance, requiring organisations to systematically identify information assets, assess threats and vulnerabilities, select appropriate controls from the accompanying Annex A catalogue, and demonstrate conformance through a formal audit and certification process. The 2022 edition restructured the Annex A catalogue to 93 controls across four themes \u2014 Organisational, People, Physical, and Technological \u2014 and introduced eleven new controls addressing cloud security, threat intelligence, and secure coding. Certification is granted by accredited third-party certification bodies and has become a de facto procurement prerequisite in technology supply chains, financial services, healthcare, and government contracting worldwide.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:iso-27001", "labels": [ "ISO 27001", "ISO 27001 Certification", "ISO 27001 Controls" ], "is_subclass_of": [ "Cybersecurity Standard" ], "wikilinks": [] }, { "id": "iso-27701", "title": "ISO 27701", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard extending ISO/IEC 27001 and 27002 with requirements and guidance for a privacy information management system. It supports management of personally identifiable information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-27701", "labels": [ "ISO 27701" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-28000", "title": "ISO 28000", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard specifying requirements for a security management system, including aspects relevant to supply chain security. It can be used as a basis for assessing and certifying an organisation's security management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-28000", "labels": [ "ISO 28000" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-29100", "title": "ISO 29100", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 29100 provides a privacy framework defining terminology, actors, roles and principles for the protection of personally identifiable information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-29100", "labels": [ "ISO 29100" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-29134", "title": "ISO 29134", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard providing guidelines for privacy impact assessment, including process and structure. It supports assessing the impact of processing on the privacy of individuals.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-29134", "labels": [ "ISO 29134" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-29184", "title": "ISO 29184", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard providing requirements and recommendations for the content and structure of online privacy notices and for obtaining consent. It addresses how personally identifiable information processing is communicated.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-29184", "labels": [ "ISO 29184" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-31000-risk-management", "title": "ISO 31000 Risk Management", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard providing principles and guidelines for risk management applicable to any organisation. It describes a framework and process for managing risk and is not intended for certification.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-31000-risk-management", "labels": [ "ISO 31000 Risk Management" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-31000", "title": "ISO 31000", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard providing principles and guidelines for risk management applicable to organisations of any type or size. It describes a framework and process for managing risk rather than certifiable requirements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-31000", "labels": [ "ISO 31000" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-3691-4", "title": "ISO 3691-4", "domain": "standards", "domain_name": "Standards", "definition": "ISO 3691-4 specifies safety requirements for driverless industrial trucks and their systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-3691-4", "labels": [ "ISO 3691-4" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-37301-2021", "title": "ISO 37301 2021", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard, published in 2021, specifying requirements and guidance for a compliance management system. It provides a basis for establishing, developing and maintaining organisational compliance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-37301-2021", "labels": [ "ISO 37301 2021", "ISO 37301:2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-42001", "title": "ISO 42001", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 42001 specifies requirements for establishing, implementing, maintaining and continually improving an artificial intelligence management system within an organisation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-42001", "labels": [ "ISO 42001" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-4217", "title": "ISO 4217", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard defining three-letter alphabetic and numeric codes for the representation of currencies and funds. It is widely used in banking and finance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-4217", "labels": [ "ISO 4217", "ISO 4217 XAU Currency Code" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-639-language-codes", "title": "ISO 639 Language Codes", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard, in multiple parts, defining codes for the representation of names of languages. It is widely used to identify languages in data and software.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-639-language-codes", "labels": [ "ISO 639 Language Codes" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-8373-2021", "title": "ISO 8373 2021", "domain": "standards", "domain_name": "Standards", "definition": "ISO 8373:2021 defines terms used in relation to robots and robotic devices operating in industrial and non-industrial environments.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-8373-2021", "labels": [ "ISO 8373 2021", "ISO 8373:2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-8373-robotics-vocabulary", "title": "ISO 8373 Robotics Vocabulary", "domain": "standards", "domain_name": "Standards", "definition": "ISO 8373 defines terms used in relation to robots and robotic devices operating in industrial and non-industrial environments.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-8373-robotics-vocabulary", "labels": [ "ISO 8373 Robotics Vocabulary", "Robotics Vocabulary" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-8373", "title": "ISO 8373", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard defining terms and vocabulary for robots and robotic devices. It establishes standardised terminology used across robotics standards, including industrial robots, service robots, manipulators, and related components.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-8373", "labels": [ "ISO 8373" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-9241-112", "title": "ISO 9241-112", "domain": "standards", "domain_name": "Standards", "definition": "ISO 9241-112 is part of the ISO 9241 series on ergonomics of human-system interaction, addressing principles for the presentation of information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-9241-112", "labels": [ "ISO 9241-112" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-9241-940", "title": "ISO 9241-940", "domain": "standards", "domain_name": "Standards", "definition": "ISO 9241-940 is part of the ISO 9241 series on ergonomics of human-system interaction, addressing the evaluation of tactile and haptic interactions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-9241-940", "labels": [ "ISO 9241-940" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-9241-960", "title": "ISO 9241-960", "domain": "standards", "domain_name": "Standards", "definition": "A part of the ISO 9241 series on ergonomics of human-system interaction, concerning gesture-based interactions. It provides guidance within the broader human-system interaction framework.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-9241-960", "labels": [ "ISO 9241-960" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-9283-manipulating-industrial-robots-performance-criteria", "title": "ISO 9283 Manipulating Industrial Robots Performance Criteria", "domain": "standards", "domain_name": "Standards", "definition": "An ISO standard specifying performance criteria and related test methods for manipulating industrial robots. It defines how to measure characteristics such as pose accuracy and repeatability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-9283-manipulating-industrial-robots-performance-criteria", "labels": [ "ISO 9283 Manipulating Industrial Robots Performance Criteria", "ISO 9283 Robot Performance Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-9283", "title": "ISO 9283", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO 9283 is an international standard that defines performance criteria and test methods for manipulating industrial robots, specifying how to measure and report pose accuracy, pose repeatability, path accuracy, and path velocity characteristics under controlled conditions. The standard provides a common vocabulary and methodology enabling manufacturers, integrators, and end users to make meaningful, reproducible comparisons of robot performance across different systems and vendors. It covers both Cartesian-space and joint-space metrics and prescribes the environmental and loading conditions under which measurements are to be taken. ISO 9283 is maintained by ISO/TC 299 and is complemented by related standards covering safety, collaborative operation, and application-specific performance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-9283", "labels": [ "ISO 9283" ], "is_subclass_of": [ "Robot Standard" ], "wikilinks": [] }, { "id": "iso-iec-42001-ai-management-system", "title": "ISO IEC 42001 AI Management System", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard specifying requirements for an artificial intelligence management system. It provides a basis for establishing, implementing and improving the governance of AI within an organisation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-42001-ai-management-system", "labels": [ "ISO IEC 42001 AI Management System", "ISO/IEC 42001 AI Management" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-42001", "title": "iso iec 42001", "domain": "governance", "domain_name": "Governance", "definition": "ISO/IEC 42001:2023 is the first international management system standard specifically designed for artificial intelligence, published by ISO/IEC Joint Technical Committee 1, Subcommittee 42, establishing requirements for an AI Management System (AIMS) that organisations use to responsibly develop, provide, or use AI-based products and services. It follows the Annex SL high-level structure shared with ISO 9001 and ISO 27001, enabling seamless integration with existing management frameworks through a Plan-Do-Check-Act cycle covering context, leadership, planning, support, operation, performance evaluation, and improvement. The normative Annex A provides a reference control set addressing data governance, transparency, human oversight, algorithmic accountability, and AI supply-chain management, from which organisations select controls according to their risk profile. Certification by accredited third-party audit bodies offers independent assurance of AI governance maturity and serves as a documented compliance aid for regulations such as the EU AI Act.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:iso-iec-42001", "labels": [ "ISO IEC 42001", "ISO/IEC 42001" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "iso-iec-jtc-1-sc-17", "title": "ISO/IEC JTC 1/SC 17", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC JTC 1/SC 17 is the joint ISO and IEC subcommittee responsible for standards covering cards and security devices for personal identification, including integrated circuit cards, machine-readable travel documents, driving licences, and mobile identity documents. Its work programme includes ISO/IEC 7816 for smart cards, ISO/IEC 14443 for contactless proximity cards, and ISO/IEC 18013-5 for mobile driving licences, making it the principal international venue for interoperable identity credential formats consumed by digital identity wallets.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:iso-iec-jtc-1-sc-17", "labels": [ "ISO/IEC JTC 1/SC 17" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [ "Standards Organization", "ISO/IEC JTC 1", "Digital Identity Wallet", "Digital Identity" ] }, { "id": "iso-iec-jtc1-sc29", "title": "ISO/IEC JTC 1/SC 29", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC JTC 1/SC 29 is the joint ISO and IEC subcommittee for coding of audio, picture, multimedia, and hypermedia information \u2014 the home of the JPEG and MPEG families of standards. Its working groups have produced the compression and carriage formats that underpin digital imaging and video worldwide, from JPEG and JPEG 2000 for still images through MPEG-2, AVC, HEVC, and VVC for video to immersive and point-cloud coding, making it the primary international venue for standardising how visual media are represented, compressed, and transported.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:iso-iec-jtc1-sc29", "labels": [ "ISO/IEC JTC 1/SC 29", "ISO/IEC JTC1 SC29" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [ "Standards Organization", "ISO/IEC JTC 1", "MPEG", "Image Processing" ] }, { "id": "iso-iec-jtc-1", "title": "ISO/IEC JTC 1", "domain": "governance", "domain_name": "Governance", "definition": "ISO/IEC JTC 1 is the Joint Technical Committee established by the International Organization for Standardization and the International Electrotechnical Commission to develop standards in information technology. Its subcommittees produce standards across areas such as coding, data management, AI, cloud computing, IoT, and multimedia. It is the principal international body coordinating IT standardisation that enables cross-domain convergence and interoperability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-iec-jtc-1", "labels": [ "ISO/IEC JTC 1", "ISO/IEC JTC1", "ISO/IEC JTC1 SC27" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "iso-iec-jtc1-sc42", "title": "ISO IEC JTC1 SC42", "domain": "governance", "domain_name": "Governance", "definition": "ISO/IEC JTC 1/SC 42 is the joint ISO and IEC subcommittee responsible for international standardisation in the field of artificial intelligence. It develops foundational standards covering AI concepts and terminology, risk management, trustworthiness, data quality for machine learning, governance, and the AI system lifecycle. Its outputs, including ISO/IEC 22989 and ISO/IEC 23894, provide a common vocabulary and management framework that regulators and industry reference when assuring AI systems. SC 42 coordinates with national standards bodies and aligns with emerging AI regulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iso-iec-jtc1-sc42", "labels": [ "ISO IEC JTC1 SC42", "ISO/IEC JTC 1/SC 42", "ISO/IEC JTC1 SC42", "ISO/IEC JTC1/SC42" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "iso-tc-299", "title": "ISO/TC 299", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO/TC 299 is the ISO technical committee responsible for standardization in the field of robotics, excluding toys and military applications. It develops international standards covering robot terminology, safety requirements, performance criteria and modularity for industrial, service and collaborative robots. Its work underpins the regulatory and interoperability framework for robot deployment worldwide.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-tc-299", "labels": [ "ISO/TC 299", "ISO TC 299 Robotics" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "iso-tc-307-blockchain-standards", "title": "ISO TC 307 Blockchain Standards", "domain": "standards", "domain_name": "Standards", "definition": "ISO Technical Committee 307, responsible for standardisation of blockchain and distributed ledger technologies. It is a standards-development committee that produces multiple standards rather than a single document.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-tc-307-blockchain-standards", "labels": [ "ISO TC 307 Blockchain Standards", "ISO TC 307 Blockchain" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "ISO", "Technical Standard" ] }, { "id": "iso-tc-307", "title": "iso tc 307", "domain": "blockchain", "domain_name": "Blockchain", "definition": "ISO/TC 307 is the ISO Technical Committee established in 2016 with Australia as secretariat, mandated to develop international standards for blockchain and distributed ledger technologies (DLT). Its normative outputs span terminology (ISO 22739), reference architecture, security and privacy frameworks, smart contract interaction models, digital identity standards, and governance structures for decentralised systems. The committee coordinates with ISO/TC 68 (Financial Services), ISO/IEC JTC 1/SC 27 (Cryptographic Security), W3C, ITU-T SG17, and IEEE to ensure interoperability standards and shared vocabulary underpin regulatory, procurement, and legislative language across jurisdictions. It operates through specialised working groups covering foundations, security, smart contracts, data governance, DLT governance, and interoperability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:iso-tc-307", "labels": [ "ISO TC 307", "ISO TC307", "ISO/TC 307" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "iso-ts-15066", "title": "iso ts 15066", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO/TS 15066:2016 is an ISO Technical Specification that supplements ISO 10218-1 and ISO 10218-2 by providing detailed guidance and biomechanical data for the safety of collaborative robot applications in which industrial robots and human workers share a common workspace without a physical separating guard. It defines four collaboration modes \u2014 safety-rated monitored stop, hand guiding, speed and separation monitoring, and power and force limiting \u2014 along with body-part-specific pain thresholds that govern permissible contact forces and pressures for 29 anatomical regions. The specification is a foundational reference for cobot risk assessment, CE marking under the EU Machinery Directive, and the basis for evolving ISO 10218-3 work within ISO TC 299.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:iso-ts-15066", "labels": [ "ISO TS 15066", "ISO 15066 Collaborative Robots", "ISO/TS 15066", "ISO/TS 15066 Compliance" ], "is_subclass_of": [ "Robot Standard" ], "wikilinks": [] }, { "id": "iso-iec-14888", "title": "ISO-IEC 14888", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard, in multiple parts, specifying digital signature schemes giving message recovery and signatures with appendix. It defines cryptographic mechanisms for digital signatures.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-14888", "labels": [ "ISO-IEC 14888", "ISO/IEC 14888", "ISO/IEC 14888-3" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-17820", "title": "ISO-IEC 17820", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 17820 is a standard designation attributed to the joint ISO/IEC technical committee. Within this corpus it is cited as a normative reference for motion capture rigs and reality capture systems, but no published standard under this number appears in the public ISO catalogue as of 2026, so the designation should be treated as provisional pending verification.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-17820", "labels": [ "ISO-IEC 17820", "ISO/IEC 17820" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-18013-5-m-dl", "title": "ISO-IEC 18013-5 mDL", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 18013-5 specifies the mobile driving licence application, defining the interface and data model for a driving licence held on a mobile device.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-18013-5-m-dl", "labels": [ "ISO-IEC 18013-5 mDL", "ISO 18013-5", "ISO 18013-5 mDL", "ISO 18013-5 mDL Standard", "ISO/IEC 18013-5 mDL", "mDL ISO 18013-5", "mDoc ISO 18013-5" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-18033", "title": "ISO-IEC 18033", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 18033 is a multi-part standard specifying encryption algorithms for the protection of data confidentiality.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-18033", "labels": [ "ISO-IEC 18033", "ISO/IEC 18033" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-18039", "title": "ISO-IEC 18039", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 18039 addresses mixed and augmented reality, defining a reference model for systems in that field.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-18039", "labels": [ "ISO-IEC 18039", "ISO/IEC 18039" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-22989-2022", "title": "ISO-IEC 22989 2022", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 22989:2022 establishes terminology for artificial intelligence and describes concepts in the field of AI.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-22989-2022", "labels": [ "ISO-IEC 22989 2022", "ISO/IEC 22989", "ISO/IEC 22989:2022" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23053-2022", "title": "ISO-IEC 23053 2022", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 23053:2022 establishes a framework for describing artificial intelligence systems that use machine learning, including their components and terminology.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23053-2022", "labels": [ "ISO-IEC 23053 2022", "ISO/IEC 23053:2022" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23053", "title": "ISO-IEC 23053", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 23053 establishes a framework for describing artificial intelligence systems that use machine learning, including their components and terminology.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23053", "labels": [ "ISO-IEC 23053", "ISO/IEC 23053" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23090-3", "title": "ISO-IEC 23090-3", "domain": "standards", "domain_name": "Standards", "definition": "Part 3 of the ISO/IEC 23090 series on coded representation of immersive media (MPEG-I), covering versatile video coding (VVC). It defines a video compression standard.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23090-3", "labels": [ "ISO-IEC 23090-3", "ISO/IEC 23090-3" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23247", "title": "ISO-IEC 23247", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard, in multiple parts, defining a digital twin framework for manufacturing. It specifies a reference architecture and concepts for digital twins of manufacturing assets and processes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23247", "labels": [ "ISO-IEC 23247", "ISO/IEC 23247" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23257-2021", "title": "ISO-IEC 23257 2021", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 23257:2021 specifies a reference architecture for blockchain and distributed ledger technology systems, covering concepts, components, roles and the relationships between them.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23257-2021", "labels": [ "ISO-IEC 23257 2021", "ISO/IEC 23257", "ISO/IEC 23257:2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23894-2023", "title": "ISO-IEC 23894 2023", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 23894:2023 provides guidance on managing risk related to the development and use of artificial intelligence, applying risk management principles to AI activities.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23894-2023", "labels": [ "ISO-IEC 23894 2023", "ISO/IEC 23894:2023" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-23894", "title": "ISO-IEC 23894", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard providing guidance on managing risk related to artificial intelligence for organisations that develop or use AI. It applies risk management principles to the AI context.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-23894", "labels": [ "ISO-IEC 23894", "ISO 23894", "ISO IEC 23894", "ISO/IEC 23894", "ISO/IEC 23894:2023" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-24760", "title": "ISO-IEC 24760", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard, in multiple parts, providing a framework for identity management, including terminology and concepts. It defines core concepts for managing identity information.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-24760", "labels": [ "ISO-IEC 24760", "ISO/IEC 24760" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-25010", "title": "ISO-IEC 25010", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard defining a quality model for systems and software product quality, part of the SQuaRE series. It specifies characteristics such as functional suitability, reliability and usability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-25010", "labels": [ "ISO-IEC 25010", "ISO/IEC 25010" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-25024", "title": "ISO-IEC 25024", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 25024 is part of the ISO/IEC 25000 SQuaRE (Systems and software Quality Requirements and Evaluation) series. It defines quantitative measures for the data quality characteristics established in ISO/IEC 25012, enabling the quality of data retained in structured formats within an information system to be measured, evaluated and compared across its life cycle.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-25024", "labels": [ "ISO-IEC 25024", "ISO/IEC 25024" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-25059", "title": "ISO-IEC 25059", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 25059 extends the SQuaRE series with a quality model for artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-25059", "labels": [ "ISO-IEC 25059", "ISO/IEC 25059" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-27001-2022", "title": "ISO-IEC 27001 2022", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard, in its 2022 revision, specifying requirements for an information security management system (ISMS). Organisations can be certified against its requirements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-27001-2022", "labels": [ "ISO-IEC 27001 2022", "ISO/IEC 27001:2022" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-27001", "title": "iso/iec 27001", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ISO/IEC 27001 is an internationally recognised standard that specifies the requirements for establishing, implementing, maintaining, and continually improving an Information Security Management System (ISMS) within the context of an organisation's overall business risks. It adopts a risk-based approach, requiring organisations to systematically identify information security risks and apply appropriate controls drawn from Annex A. Certification against ISO/IEC 27001 provides third-party assurance of an organisation's commitment to protecting the confidentiality, integrity, and availability of information assets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:iso-iec-27001", "labels": [ "ISO/IEC 27001", "ISO/IEC 27000 Family" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "iso-iec-27701", "title": "ISO-IEC 27701", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 27701 specifies requirements and guidance for a privacy information management system as an extension to ISO/IEC 27001 and ISO/IEC 27002.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-27701", "labels": [ "ISO-IEC 27701", "ISO/IEC 27701" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-30170", "title": "ISO-IEC 30170", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 30170 is the international standard specifying the syntax and semantics of the Ruby programming language. Published in 2012 and derived from the Japanese standard JIS X 3017, it defines Ruby's object model, execution behaviour and core language constructs so that independent implementations can conform to a common specification.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-30170", "labels": [ "ISO-IEC 30170", "ISO/IEC 30170" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-38500", "title": "ISO-IEC 38500", "domain": "standards", "domain_name": "Standards", "definition": "An ISO/IEC standard providing principles and guidance for the corporate governance of information technology. It addresses the responsibilities of governing bodies for the use of IT.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-38500", "labels": [ "ISO-IEC 38500", "ISO/IEC 38500", "ISO/IEC 38500:2024" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-42001-2023", "title": "ISO-IEC 42001 2023", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 42001:2023 specifies requirements for establishing, implementing, maintaining and continually improving an artificial intelligence management system within an organisation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-42001-2023", "labels": [ "ISO-IEC 42001 2023", "ISO/IEC 42001:2023" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-5338-2023", "title": "ISO-IEC 5338 2023", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC 5338:2023 defines processes for the life cycle of artificial intelligence systems, building on established system and software life cycle process standards.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-5338-2023", "labels": [ "ISO-IEC 5338 2023", "ISO/IEC 5338:2023" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-9075", "title": "ISO/IEC 9075 SQL Standard", "domain": "data", "domain_name": "Data", "definition": "ISO/IEC 9075 is the international standard that defines SQL, the Structured Query Language used to define, query and manipulate data in relational database systems. Published jointly by ISO and IEC and revised across multiple editions, it specifies the language's data model, syntax, semantics and conformance levels through a series of parts covering the core framework, foundation, call-level interface and extensions such as JSON and property-graph queries. It provides the common reference that lets relational databases interoperate and that vendors extend with proprietary dialects.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:iso-iec-9075", "labels": [ "ISO/IEC 9075 SQL Standard", "ISO/IEC 9075", "ISO/IEC 9075 (SQL Standard)" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "iso-iec-jtc-1-sc-27", "title": "ISO/IEC JTC 1/SC 27", "domain": "security", "domain_name": "Security", "definition": "ISO/IEC JTC 1/SC 27 is the joint ISO and IEC subcommittee responsible for developing international standards on information security, cybersecurity and privacy protection. Its remit includes the ISO/IEC 27000 family of information security management standards, cryptographic techniques, security evaluation criteria and identity management. It is the principal global body coordinating consensus standards in the security and privacy domain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:iso-iec-jtc-1-sc-27", "labels": [ "ISO/IEC JTC 1/SC 27" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "iso-iec-jtc-1-sc-24", "title": "ISO-IEC JTC 1-SC 24", "domain": "standards", "domain_name": "Standards", "definition": "A subcommittee of ISO/IEC Joint Technical Committee 1 responsible for standardisation of computer graphics, image processing and environmental data representation. It is a standards-development body rather than a single document.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-jtc-1-sc-24", "labels": [ "ISO-IEC JTC 1-SC 24", "ISO/IEC JTC 1/SC 24" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-jtc-1-sc-42", "title": "ISO/IEC JTC 1/SC 42", "domain": "governance", "domain_name": "Governance", "definition": "ISO/IEC JTC 1/SC 42 (Artificial Intelligence) is the international standardisation subcommittee within the ISO/IEC Joint Technical Committee 1 responsible for developing and maintaining global standards for artificial intelligence. Established in 2017, it coordinates with national bodies across more than 40 participating countries to produce standards covering AI terminology, foundational concepts, trustworthiness, bias, risk management, data lifecycle, governance frameworks, and AI system evaluation. Its publications inform regulatory frameworks worldwide, including the EU AI Act, and provide the shared vocabulary necessary for interoperability of AI governance instruments across jurisdictions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-iec-jtc-1-sc-42", "labels": [ "ISO/IEC JTC 1/SC 42", "ISO IEC JTC 1 SC 42", "ISO/IEC JTC1 SC42" ], "is_subclass_of": [ "AI Regulation" ], "wikilinks": [] }, { "id": "iso-iec-tr-24027", "title": "ISO-IEC TR 24027", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC TR 24027 is a technical report addressing bias in artificial intelligence systems and in AI-aided decision making.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-tr-24027", "labels": [ "ISO-IEC TR 24027", "ISO/IEC TR 24027" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec-tr-24028", "title": "ISO/IEC TR 24028", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC TR 24028 is a technical report, published by ISO/IEC JTC 1/SC 42, that provides an overview of trustworthiness in artificial intelligence. It surveys the properties that make AI systems trustworthy \u2014 reliability, robustness, resilience, safety, transparency, explainability, controllability, and privacy \u2014 and catalogues the threats to trustworthiness and the mitigation approaches available across the AI lifecycle. As a foundational overview it frames the vocabulary and concerns that later, more specific standards build upon. It is a distinct deliverable from ISO/IEC TR 24027, which addresses bias in AI systems and AI-aided decision making; the two carry different numbers and cover different subjects within the same SC 42 programme.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-iec-tr-24028", "labels": [ "ISO/IEC TR 24028" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard", "ISO-IEC TR 24027", "AI Trustworthiness" ] }, { "id": "iso-iec-tr-24029-1-2021", "title": "ISO-IEC TR 24029-1 2021", "domain": "standards", "domain_name": "Standards", "definition": "ISO/IEC TR 24029-1:2021 is a technical report providing an overview of methods for assessing the robustness of neural networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:iso-iec-tr-24029-1-2021", "labels": [ "ISO-IEC TR 24029-1 2021", "ISO IEC 24029", "ISO/IEC TR 24029-1:2021" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard" ] }, { "id": "iso-iec", "title": "ISO/IEC", "domain": "governance", "domain_name": "Governance", "definition": "ISO/IEC refers to the joint technical committee structure formed by the International Organisation for Standardisation (ISO) and the International Electrotechnical Commission (IEC) to develop international standards in information technology and related fields. The primary body is ISO/IEC JTC 1, which oversees standards in areas including software engineering, cybersecurity, artificial intelligence, and biometrics. ISO/IEC standards are developed through consensus among national standardisation bodies and carry significant weight in regulatory compliance, procurement, and international trade agreements. Landmark outputs include ISO/IEC 27001 for information security management and ISO/IEC 42001 for AI management systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:iso-iec", "labels": [ "ISO/IEC", "ISO IEC", "ISO IEC Standard", "ISO IEC Standards", "ISO-IEC", "ISO/IEC 29101", "ISO/IEC 30107", "ISO/IEC 39075:2024 GQL", "ISO/IEC Standard", "ISO/IEC Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "iso", "title": "ISO", "domain": "governance", "domain_name": "Governance", "definition": "The International Organisation for Standardisation (ISO) is an independent, non-governmental international body comprising the national standards institutes of 167 countries, responsible for developing and publishing voluntary international standards across virtually every sector of industry, technology, and commerce. ISO standards are produced by technical committees composed of experts nominated by member bodies and are adopted through a consensus-based process requiring defined levels of national body approval. ISO standards span quality management, environmental management, information security, robotics, artificial intelligence, financial services, and hundreds of other domains. Published standards carry normative authority in many regulatory and procurement frameworks worldwide.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:iso", "labels": [ "ISO", "ISO (International Organization for Standardization)", "ISO 14021", "ISO 15552", "ISO 15552 Standard", "ISO 19100 Series", "ISO 19115", "ISO 29115", "ISO 30401", "ISO TC172 Optics Standardisation", "MPEG DASH ISO 23009" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "itu-t", "title": "ITU-T", "domain": "standards", "domain_name": "Standards", "definition": "ITU-T is the Telecommunication Standardization Sector of the International Telecommunication Union, a United Nations agency that develops international standards (Recommendations) for telecommunications and information technology. Its work spans video and audio coding, network protocols, public-key infrastructure and digital signatures. ITU-T Recommendations such as the H-series video codecs and X.509 are foundational to global communications interoperability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:itu-t", "labels": [ "ITU-T", "ITU Standards", "ITU-T G.711 Audio Codec", "ITU-T H.264 H.265", "ITU-T Standards", "ITU-T X.1255" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "itu", "title": "ITU", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The International Telecommunication Union (ITU) is the United Nations specialised agency responsible for information and communication technologies, coordinating global use of the radio spectrum and satellite orbits, developing technical standards and working to expand connectivity worldwide. Through its sectors for radiocommunication, standardization and development it produces recommendations that underpin interoperable telecommunications infrastructure. It is one of the principal international bodies governing the global telecommunications system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:itu", "labels": [ "ITU" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "ivms-101", "title": "IVMS 101", "domain": "blockchain", "domain_name": "Blockchain", "definition": "IVMS 101 (interVASP Messaging Standard 101) is an open data model and schema specification developed by the Joint Working Group on interVASP Messaging Standards (JWG) to enable Virtual Asset Service Providers (VASPs) to exchange originator and beneficiary identity information in compliance with the FATF Travel Rule. Published in 2020 and subsequently adopted by the FATF as the preferred global standard for Travel Rule data transmission, IVMS 101 defines canonical field names, value types, and encoding rules for natural persons, legal persons, and address objects in a technology-neutral JSON schema. It is designed to be embedded within any Travel Rule messaging protocol \u2014 including TRP, TRUST, VerifyVASP, and Sygna Bridge \u2014 ensuring semantic interoperability across different technical implementations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ivms-101", "labels": [ "IVMS 101" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "ice-base", "title": "Ice Base", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-base", "labels": [ "Ice Base" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-calf", "title": "Ice Calf", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-calf", "labels": [ "Ice Calf" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-cap", "title": "Ice Cap", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-cap", "labels": [ "Ice Cap" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-core", "title": "Ice Core", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-core", "labels": [ "Ice Core" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-field", "title": "Ice Field", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-field", "labels": [ "Ice Field" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-giant", "title": "Ice Giant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-giant", "labels": [ "Ice Giant" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-sheet-monitoring", "title": "Ice Sheet Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-sheet-monitoring", "labels": [ "Ice Sheet Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-sheet", "title": "Ice Sheet", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-sheet", "labels": [ "Ice Sheet" ], "is_subclass_of": [ "Glacier" ], "wikilinks": [] }, { "id": "ice-shelf", "title": "Ice Shelf", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-shelf", "labels": [ "Ice Shelf" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-snow-interface", "title": "Ice Snow Interface", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-snow-interface", "labels": [ "Ice Snow Interface" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-stream", "title": "Ice Stream", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-stream", "labels": [ "Ice Stream" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ice-surface", "title": "Ice Surface", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ice-surface", "labels": [ "Ice Surface" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "iceberg", "title": "Iceberg", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:iceberg", "labels": [ "Iceberg" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "idempotency", "title": "Idempotency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Idempotency is the property of an operation whereby applying it multiple times produces the same result as applying it once. In distributed systems and APIs it is essential for safe retries, ensuring that duplicated requests caused by network failures do not corrupt state or cause unintended side effects. Idempotency is commonly enforced via idempotency keys, deterministic merge functions, or naturally commutative state updates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:idempotency", "labels": [ "Idempotency" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "identification-system", "title": "Identification System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An Identification System is a system that determines or confirms the identity of an entity\u2014person, device, organisation, or digital asset\u2014by collecting, processing, and evaluating identity evidence against registered credentials or biometric templates. In AI contexts, identification systems increasingly use machine learning models for biometric recognition, document verification, and behavioural profiling, raising significant questions about accuracy, fairness, and privacy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:identification-system", "labels": [ "Identification System", "Person Identification" ], "is_subclass_of": [ "System" ], "wikilinks": [ "Artificial Intelligence", "System" ] }, { "id": "identification", "title": "Identification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identification is the process of establishing the claimed identity of an entity \u2014 person, device, or service \u2014 by associating it with a unique identifier and supporting evidence such as credentials, biometrics, or cryptographic proofs. It is a prerequisite step to authentication and authorisation in any access-control or trust framework. Identification schemes range from simple username\u2013password pairs to government-issued digital identity documents and decentralised self-sovereign identity systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:identification", "labels": [ "Identification" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "identifier-layer", "title": "Identifier Layer", "domain": "security", "domain_name": "Security", "definition": "The Identifier Layer is the stratum that defines and manages the names and references used to denote entities unambiguously. It sits above the data and cryptographic primitives that make identifiers durable and verifiable and below the Identity Layer that binds meaning to them. It contains naming schemes, resolution mechanisms, and uniqueness guarantees.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:identifier-layer", "labels": [ "Identifier Layer" ], "is_subclass_of": [ "Security and Identity", "owl:Thing" ], "wikilinks": [ "Data Layer", "Cryptographic Layer", "Identity Layer", "Custody Layer", "Uniform Resource Identifier", "Namespace", "owl:Thing", "W3C (World Wide Web Consortium)" ] }, { "id": "identifier", "title": "Identifier", "domain": "standards", "domain_name": "Standards", "definition": "A symbol, string, or token that unambiguously denotes an entity \u2014 a resource, person, organisation, device, dataset, or concept \u2014 within a defined scope, enabling that entity to be referenced, located, compared, and linked independently of its descriptive attributes. Identifiers are the foundational primitive of naming systems: their guarantees of uniqueness, persistence, and resolvability within a namespace determine what URIs, UUIDs, DIDs, and registry codes can be trusted to do.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:identifier", "labels": [ "Identifier" ], "is_subclass_of": [ "Identification" ], "wikilinks": [ "Identification", "URI", "Namespace", "Decentralised Identifier" ] }, { "id": "identity-attestation", "title": "Identity Attestation", "domain": "security", "domain_name": "Security", "definition": "Identity attestation is a cryptographically signed claim, issued by a trusted party, that vouches for an attribute or fact about a digital identity, such as personhood, credential possession or eligibility. It is verified through digital-signature verification against the issuer's public key, allowing a relying party to trust the claim without contacting the issuer directly. Attestation schemes are used in decentralised identity systems and in platforms such as Gitcoin to establish sybil-resistant, verifiable identity without revealing underlying personal data.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:identity-attestation", "labels": [ "Identity Attestation" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "identity-federation", "title": "Identity Federation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A distributed authentication workflow that enables cross-domain identity linking through trust relationships, allowing users to access resources across multiple organizations using a single set of credentials.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:identity-federation", "labels": [ "Identity Federation" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Attribute Exchange", "Attribute Schema", "B2B Collaboration", "Credential Mapping", "Cross-Domain SSO", "Decentralized Identity", "Federation Agreement", "Federation Protocol", "Metadata Exchange", "Multi-Organization Access", "NIST SP 800-63C", "OASIS SAML", "OpenID Foundation", "PKI Infrastructure", "Policy Negotiation", "Session Propagation", "Token Translation", "Trust Establishment", "Distributed Authentication Architecture", "Identity Management System" ] }, { "id": "identity-graph", "title": "Identity Graph", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network-based data structure representing entities, their digital identities, and relationships in a graph database optimized for identity resolution, trust scoring, and social network analysis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:identity-graph", "labels": [ "Identity Graph" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "DID Resolution", "DIF Universal Resolver", "Fraud Detection", "Identity Node", "Identity Resolution", "Relationship Edge", "Social Network Analysis", "Trust Attribute", "Trust Metrics", "W3C Verifiable Credentials", "Cryptographic Keys", "Graph Database", "Identity Management System", "Identity Provider", "MiddlewareLayer", "Reputation System", "TrustAndGovernanceDomain", "Trust Framework", "Verifiable Credentials" ] }, { "id": "identity-layer", "title": "Identity Layer", "domain": "security", "domain_name": "Security", "definition": "The Identity Layer is the cross-cutting stratum that establishes, represents, and verifies the identities of participants and resources. It sits above the Cryptographic Layer, on which its proofs rest, and below access, policy, and governance strata that reason about who is acting. It contains identifiers, credentials, authentication mechanisms, and the bindings between them.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:identity-layer", "labels": [ "Identity Layer" ], "is_subclass_of": [ "Security and Identity", "owl:Thing" ], "wikilinks": [ "Cryptographic Layer", "Identifier Layer", "Policy Layer", "Custody Layer", "Public Key Infrastructure", "Decentralised Identifiers", "owl:Thing", "W3C (World Wide Web Consortium)" ] }, { "id": "identity-management-system", "title": "Identity Management System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Identity Management System (IdM/IAM \u2014 Identity and Access Management) is a comprehensive technology framework that governs the full lifecycle of digital identities\u2014creation, provisioning, authentication, authorisation, federation, delegation, and decommissioning\u2014across enterprise, cloud, c...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:identity-management-system", "labels": [ "Identity Management System" ], "is_subclass_of": [ "Security and Identity", "Zero Trust Architecture", "Access Control System", "Governance Framework", "Security Architecture", "Digital Infrastructure" ], "wikilinks": [ "Authorisation Engine", "Decentralised Identity", "DevSecOps", "Directory Service", "Entra ID", "FAPI 2.0", "Federated Identity", "FIDO Alliance", "FIDO2", "GovernanceDomain", "Hardware Security Module", "IETF", "Keycloak", "LDAP", "MFA Service", "NIST", "OASIS", "OAuth 2.0", "OpenID Connect", "OpenID Foundation" ] }, { "id": "identity-management", "title": "Identity Management", "domain": "security", "domain_name": "Security", "definition": "Identity management refers to the systems, protocols, and policies enabling individuals and organisations to create, control, and verify digital identities. In decentralised contexts this encompasses DID architectures, verifiable credentials, and self-sovereign identity models that eliminate reliance on centralised authorities through cryptographic proofs and distributed ledger infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:identity-management", "labels": [ "Identity Management", "Device Identity Management", "Enterprise Identity Management", "Identity Management Integration", "IdentityManagement" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain", "Self-Sovereign Identity" ] }, { "id": "identity-portability", "title": "Identity Portability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identity portability is the capacity for a user to move their digital identity, credentials, and associated reputation across platforms, services, or virtual worlds without re-establishing them from scratch. It relies on interoperable identity standards and user-controlled wallets so that authentication and verified attributes remain valid across boundaries. Portability is a cornerstone of self-sovereign identity and an open metaverse.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:identity-portability", "labels": [ "Identity Portability" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "identity-proofing", "title": "Identity Proofing", "domain": "security", "domain_name": "Security", "definition": "Identity proofing is the process of collecting and validating evidence to establish that a claimed identity corresponds to a real, unique person before issuing credentials or granting access. It typically combines document verification, biometric checks, and authoritative-source confirmation, graded by assurance levels such as those defined in NIST SP 800-63A. Identity proofing precedes authentication and is foundational to trustworthy onboarding.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:identity-proofing", "labels": [ "Identity Proofing", "Remote Identity Proofing" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "identity-provider-id-p", "title": "Identity Provider (IdP)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An authentication service system that creates, maintains, and manages identity information for principals while providing authentication services to relying party applications within a federation or distributed network.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:identity-provider-id-p", "labels": [ "Identity Provider (IdP)" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Authentication Protocol", "Authentication Server", "Credential Schema", "Credential Store", "Cryptographic Key Store", "Directory Service", "Federation Infrastructure", "IETF OAuth", "Multi-Factor Authentication", "OASIS SAML", "OpenID Foundation", "PKI Infrastructure", "Session Manager", "Single Sign-On (SSO)", "Token Issuer", "User Database", "User Directory", "User Provisioning", "Access Control", "Identity Federation" ] }, { "id": "identity-provider", "title": "Identity Provider", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Identity Provider (IdP) is a specialised security system that authenticates principals \u2014 humans, service accounts, devices, and workloads \u2014 and issues cryptographically signed tokens or assertions that downstream service providers accept as proof of identity and authorised attributes, oper...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:identity-provider", "labels": [ "Identity Provider" ], "is_subclass_of": [ "Security and Identity", "Security Infrastructure", "Identity Management System", "Access Control System", "Authentication Service", "Trust Service" ], "wikilinks": [ "Audit Log", "Authentication Engine", "B2C Identity", "B2E Identity", "Biometric Authentication", "Continuous Authentication", "Credential Store", "Cryptographic Hash Function", "Delegated Authorization", "Directory Service", "Federated Identity", "Federation Gateway", "FIDO Alliance", "FIDO2", "Hardware Security Module", "IdentityAndAccessManagementDomain", "Identity Governance", "Identity Governance", "IETF", "JSON Web Signature" ] }, { "id": "identity-resolution", "title": "Identity Resolution", "domain": "data", "domain_name": "Data", "definition": "Identity resolution is the process of determining that multiple data records, identifiers, or signals across different systems refer to the same real-world entity \u2014 whether a person, organisation, or device \u2014 and consolidating them into a unified, persistent representation. It combines probabilistic and deterministic matching algorithms, data enrichment, and graph-linking to resolve fragmented identities across first-party, second-party, and third-party data sources.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:identity-resolution", "labels": [ "Identity Resolution", "Identity Resolver" ], "is_subclass_of": [ "Entity Resolution" ], "wikilinks": [] }, { "id": "identity-service", "title": "Identity Service", "domain": "security", "domain_name": "Security", "definition": "An identity service is a system component that manages the lifecycle of digital identities, providing authentication, authorisation, and identity data to other services within a platform architecture. It centralises functions such as user registration, credential issuance, session management, and federation so that applications can delegate trust decisions. Identity services are typically exposed via standard protocols such as OAuth 2.0, OpenID Connect, and SAML.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:identity-service", "labels": [ "Identity Service" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "identity-standards", "title": "Identity Standards", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identity Standards are agreed specifications and protocols governing how digital identities are created, expressed, verified and exchanged across systems. They include standards for decentralised identifiers, verifiable credentials and federated authentication, enabling secure, interoperable identity management across platforms and virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:identity-standards", "labels": [ "Identity Standards" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "owl:Thing" ] }, { "id": "identity-system", "title": "Identity System", "domain": "security", "domain_name": "Security", "definition": "An organised set of processes and technologies used to establish, manage and verify the identities of subjects within a defined context.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:identity-system", "labels": [ "Identity System", "Cryptographic Identity System", "Identity Store" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [ "Authentication", "Identity Verification", "Access Control", "Digital Identity", "Identity Provider", "Identity Management" ] }, { "id": "identity-systems", "title": "Identity Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identity Systems are the technical and organisational frameworks that establish, manage, verify, and revoke digital identities for users, devices, and services. They encompass authentication mechanisms, credential issuance, public-key infrastructure, and decentralised identity models (self-sovereign identity, DIDs) that enable secure, interoperable identification across platforms and jurisdictions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:identity-systems", "labels": [ "Identity Systems" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "owl:Thing" ] }, { "id": "identity-verification-system", "title": "Identity Verification System", "domain": "security", "domain_name": "Security", "definition": "An Identity Verification System (IVS) is an integrated set of processes, technologies, and controls that establish with high assurance that a person presenting a claimed identity is genuinely that individual. It typically orchestrates document authentication, biometric matching with liveness detection, and cross-referential data validation against authoritative databases to produce a risk-scored trust signal. IVSs operate across regulated sectors \u2014 financial services, healthcare, government, and telecommunications \u2014 where Know Your Customer (KYC) and Anti-Money Laundering (AML) obligations mandate identity proofing at onboarding and at trust-elevation events. Modern IVSs increasingly integrate machine-learning fraud-detection models, cryptographic document signatures, and decentralised identity standards to resist synthetic-identity fraud and deepfake attacks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:identity-verification-system", "labels": [ "Identity Verification System" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [ "Biometric Authentication", "Onfido", "iProov", "Identity Management" ] }, { "id": "identity-verification", "title": "Identity Verification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identity Verification (IDV, often used interchangeably with identity proofing for the one-time onboarding event) is the trust-establishment process by which a relying party tests an identity claim \u2014 that a specific natural person or legal entity is who they purport to be \u2014 by collecting evide...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:identity-verification", "labels": [ "Identity Verification", "AI Identity Verification", "Customer Identity Verification", "Electronic Identity Verification", "Identity Document Verification", "NIP-05 Identity Verification", "Reusable Identity Verification" ], "is_subclass_of": [ "Security and Identity", "Identity Proofing", "Trust Establishment", "Authentication Method", "Security Mechanism", "Compliance Control" ], "wikilinks": [ "Account Opening", "Age Verification", "Alloy", "Anti-Money Laundering", "ArcFace", "Au10tix", "Authentication Method", "Authorisation", "Authoritative Data Source", "Behavioural Biometrics", "Biometric Matching Algorithm", "Biometric Template", "Biometric Verification", "C2PA", "Capture Device", "Civic Pass", "ComplianceLayer", "Consent Mechanism", "Credit Bureau", "Credit Bureau" ] }, { "id": "identity-and-access-management", "title": "identity and access management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Identity and Access Management (IAM) is a security discipline comprising the frameworks, policies, technologies, and processes that govern how digital identities are created, authenticated, authorised, and managed throughout their lifecycle within and across organisational boundaries. IAM systems enforce the principle of least privilege by ensuring that subjects\u2014users, applications, devices, and service accounts\u2014can access only the resources required for their legitimate purpose at the appropriate time. The discipline spans directory services, multi-factor authentication, role-based and attribute-based access control, privileged access management, identity governance and administration, and federated identity protocols that extend trust across cloud and partner environments. IAM is a foundational control domain within information security frameworks such as ISO/IEC 27001, NIST SP 800-53, and Zero Trust architecture models.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:identity-and-access-management", "labels": [ "Identity and Access Management", "Access Management", "Cloud Identity and Access Management", "Customer Identity and Access Management", "Identity Governance", "Permission Management" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "identity", "title": "Identity", "domain": "security", "domain_name": "Security", "definition": "Identity is the complete, verifiable representation of a principal \u2014 person, organisation, device, or agent \u2014 within a digital system, constituted by a set of attributes, credentials, and cryptographically anchored claims that distinguish that principal from all others. Digital identity underpins authentication, authorisation, and accountability across networked systems, and increasingly extends to self-sovereign, decentralised forms in which control of the identity record rests with the subject rather than a centralised authority. In spatial computing and metaverse contexts, identity further encompasses persistent avatars, reputational history, and cross-platform portability anchored to decentralised identifiers (DIDs). The coherent management of identity is the foundation upon which access control, trust hierarchies, and privacy-preserving interactions are built.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:identity", "labels": [ "Identity" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "owl:Thing" ] }, { "id": "iec-61511", "title": "Iec 61511", "domain": "standards", "domain_name": "Standards", "definition": "IEC 61511 is an international standard for the functional safety of safety instrumented systems in the process industry sector. It applies the generic principles of IEC 61508 to end users and integrators, defining a safety lifecycle that spans hazard and risk assessment, allocation of safety functions to protection layers, and verification against target safety integrity levels. It is the dominant reference for emergency shutdown and high-integrity protection systems on process plant.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iec-61511", "labels": [ "Iec 61511", "IEC 61511" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [] }, { "id": "ieee-vr", "title": "Ieee Vr", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "IEEE VR is the IEEE Conference on Virtual Reality and 3D User Interfaces, the leading international academic venue for research in virtual, augmented and mixed reality. Sponsored by the IEEE, it publishes peer-reviewed work on immersive display, tracking, interaction techniques, perception and applications, and shapes the technical agenda of the field. It functions as both a standardising community and a dissemination channel for spatial-computing research.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ieee-vr", "labels": [ "Ieee Vr", "IEEE VR" ], "is_subclass_of": [ "Academic Conference" ], "wikilinks": [] }, { "id": "image-captioning", "title": "Image Captioning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Image captioning is the artificial-intelligence task of generating a natural-language description of the content of an image. It sits at the intersection of computer vision and natural-language generation, typically pairing a visual encoder that extracts image features with a language decoder that produces a fluent sentence. Modern systems use attention mechanisms and large vision-language models to ground the generated text in salient regions of the image.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:image-captioning", "labels": [ "Image Captioning" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "image-classification", "title": "Image Classification", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Image Classification is the computer vision task of assigning a categorical label to an entire image from a predefined set of classes, determining what is depicted in the image as a whole. Modern image classification employs deep convolutional neural networks (ResNet, EfficientNet, Vision Transformers) trained on large-scale datasets such as ImageNet to achieve human-level or super-human performance on diverse visual recognition tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:image-classification", "labels": [ "Image Classification" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "automation", "documentation", "optimization", "presentation", "taleb2012antifragile", "Transfer Learning", "computer vision", "Computer Vision", "Convolutional Neural Network", "Landscape", "machine learning", "MetaverseDomain", "Proprietary Video" ] }, { "id": "image-co-registration", "title": "Image Co-registration", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:image-co-registration", "labels": [ "Image Co-registration" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "image-compositing", "title": "Image Compositing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:image-compositing", "labels": [ "Image Compositing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "image-compression", "title": "Image Compression", "domain": "data", "domain_name": "Data", "definition": "The process of encoding digital images with fewer bits by exploiting spatial redundancy, statistical structure, and the limits of human visual perception. Lossless methods (PNG, lossless WebP) permit exact reconstruction, while lossy methods (JPEG, HEIC, AVIF, JPEG XL) discard perceptually insignificant detail through transform coding, quantisation, and entropy coding to achieve far higher ratios, trading fidelity against file size for storage and transmission.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:image-compression", "labels": [ "Image Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [ "Data Compression", "Entropy Coding", "Image Processing" ] }, { "id": "image-editing", "title": "Image Editing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Image editing is the process of altering or enhancing digital images using software tools, encompassing operations such as colour correction, compositing, retouching, masking, and applying filters or effects. Modern image editing spans a spectrum from manual pixel-level manipulation to AI-driven automated transformations that interpret semantic content to intelligently modify or generate imagery.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:image-editing", "labels": [ "Image Editing", "Photo Editing Software" ], "is_subclass_of": [ "Content Creation" ], "wikilinks": [] }, { "id": "image-generation", "title": "Image Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Image Generation is the synthesis of realistic or stylised images using generative AI models including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and Diffusion Models. Modern systems such as DALL-E, Stable Diffusion, Midjourney, and Flux produce high-fidelity images from text descriptions, sketches, or latent representations, enabling creative applications, data augmentation, virtual production, and content creation at scale.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:image-generation", "labels": [ "Image Generation", "AI Image Generation", "Custom Image Generation", "Diffusion Image Generation", "Pose-Guided Image Generation" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "automation", "BibTeX", "Code", "community", "Diffusion Model", "documentation", "Llama", "MotionDirector", "neural networks", "optimization", "Paper", "research", "Stable Video Diffusion", "troubleshooting", "user experience", "visionflow", "workflow management", "Adoption of Convergent Technologies", "Agents", "AI Video" ] }, { "id": "image-mosaicking", "title": "Image Mosaicking", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:image-mosaicking", "labels": [ "Image Mosaicking" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "image-preprocessing", "title": "Image Preprocessing", "domain": "ai", "domain_name": "Ai", "definition": "Image preprocessing is the set of operations applied to raw image data before it is fed into a computer vision or machine learning model. It standardises and conditions images through resizing, normalisation, colour-space conversion, denoising and contrast adjustment to improve downstream model accuracy and robustness. Preprocessing also includes augmentation transforms that synthetically expand training data and feature-oriented steps that emphasise salient structures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:image-preprocessing", "labels": [ "Image Preprocessing" ], "is_subclass_of": [ "Image Processing" ], "wikilinks": [] }, { "id": "image-processing-software", "title": "Image Processing Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software systems that acquire, process, analyse, and visualise digital images for metaverse applications, employing computer vision algorithms for user tracking, environment creation, gesture recognition, 3D reconstruction, and real-time visual enhancement in virtual and augmented reality environ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:image-processing-software", "labels": [ "Image Processing Software" ], "is_subclass_of": [ "Display and Rendering", "Computer Vision System" ], "wikilinks": [ "Visual Metaverse Interaction", "Computer Vision System", "metaverse" ] }, { "id": "image-processing", "title": "Image Processing", "domain": "ai", "domain_name": "Ai", "definition": "Image Processing is the computational manipulation of digital images using mathematical operations\u2014including spatial filtering, morphological transforms, frequency-domain analysis, and learned convolutional operations\u2014to enhance visual quality, extract structured information, or transform image representations for downstream tasks. It encompasses both classical signal processing techniques (Fourier and wavelet transforms, histogram equalisation, edge detection via Sobel or Canny operators, morphological erosion and dilation) and modern deep-learning approaches implemented through convolutional neural networks, vision transformers, and diffusion models. Image processing forms the foundational preprocessing and analysis layer for computer vision pipelines, medical imaging workflows, remote sensing, autonomous navigation, and industrial quality inspection, operating on discrete pixel grids to produce processed images or structured semantic outputs. The field bridges raw sensor data acquisition and higher-level scene understanding, with applications spanning from embedded real-time systems to large-scale cloud inference infrastructure.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:image-processing", "labels": [ "Image Processing", "Image Pyramid" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "image-recognition", "title": "Image Recognition", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Image recognition is the computer-vision task of identifying and categorising the objects, scenes, people or attributes present in a digital image. It maps raw pixel data to semantic labels, ranging from whole-image classification through to localisation of multiple distinct entities within a single frame. Modern image recognition is dominated by deep convolutional and transformer-based neural networks trained on large labelled datasets, which learn hierarchical visual features rather than relying on hand-engineered descriptors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:image-recognition", "labels": [ "Image Recognition" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "image-resampling", "title": "Image Resampling", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:image-resampling", "labels": [ "Image Resampling" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "image-segmentation", "title": "Image Segmentation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Image Segmentation is a foundational computer vision task that partitions a digital image into multiple discrete, semantically meaningful regions or segments, assigning each pixel (or group of pixels) a label that represents its category, instance identity, or both, enabling downstream systems to...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:image-segmentation", "labels": [ "Image Segmentation", "Promptable Segmentation", "Segmentation" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "Deep Learning", "Structured Prediction", "Scene Understanding", "Dense Prediction" ], "wikilinks": [ "ADE20K", "AlgorithmLayer", "Atrous Convolution", "Autonomous Driving", "Autonomous Vehicles", "Average Precision", "Background Removal", "Cancer Detection", "Cityscapes Dataset", "COCO Dataset", "Convolutional Neural Networks", "Cross Entropy Loss", "Dense Prediction", "Dice Coefficient", "Evaluation Metrics", "Feature Pyramid Network", "Fully Convolutional Networks", "GPU Compute", "ImageNet", "Interactive Segmentation" ] }, { "id": "image-sensor", "title": "Image Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An image sensor is a solid-state device that converts incident light into electrical signals to capture a digital image, typically implemented as a CMOS or CCD array of photodetector pixels. Each pixel accumulates charge proportional to received photons, which is read out, digitised, and assembled into a frame. Image sensors are the core capture element in cameras and many computer-vision and tracking systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:image-sensor", "labels": [ "Image Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "image-synchronisation", "title": "Image Synchronisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Image synchronisation is the precise temporal alignment of frame capture across two or more cameras so that corresponding images represent the same instant in the scene. It is essential for stereo vision, where depth is computed by triangulating matched features between left and right images, since even small timing offsets introduce triangulation error for moving scenes or moving cameras. Hardware synchronisation, typically a shared trigger signal, achieves tighter alignment than software timestamp matching, which is limited by operating system scheduling jitter. It is a prerequisite for reliable multi-camera sensor fusion in robotic perception systems operating on dynamic scenes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:image-synchronisation", "labels": [ "Image Synchronisation" ], "is_subclass_of": [ "Sensor Fusion" ], "wikilinks": [] }, { "id": "image-synthesis", "title": "Image Synthesis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The generation of images from models or descriptions rather than direct capture, including rendering from scene data and machine learning models that produce images from learned distributions. Encompasses classical computer graphics rendering, generative adversarial networks, diffusion models, and other learned generative approaches conditioned on text, semantic maps, or latent codes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:image-synthesis", "labels": [ "Image Synthesis", "Conditional Image Synthesis" ], "is_subclass_of": [ "Image Generation" ], "wikilinks": [ "Generative Model", "Computer Graphics", "Diffusion Model", "Generative Adversarial Network", "Image Generation" ] }, { "id": "image-and-video-restoration", "title": "Image and Video Restoration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Image and Video Restoration is a computational imaging discipline within the artificial-intelligence and computer-vision domain concerned with recovering high-quality, perceptually faithful visual content from degraded observations, where degradation encompasses noise corruption (Gaussian/Poisson...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:image-and-video-restoration", "labels": [ "Image and Video Restoration", "Photo Restoration" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "Inverse Problems", "Computational Photography", "Image Processing", "Signal Recovery" ], "wikilinks": [ "AlgorithmLayer", "CodeFormer", "Computational Imaging", "Computational Photography", "ComputationalPhotographyDomain", "ComputerVisionDomain", "Convolutional Neural Networks", "Cultural Heritage Preservation", "CVPR", "Deblurring Network", "Deep Learning Framework", "Degradation Model", "Degradation Prior", "Denoising Network", "DifFace", "DnCNN", "ECCV", "Face Restoration Module", "Film and Television Production", "Forensic Image Enhancement" ] }, { "id": "image-to-image-translation", "title": "Image to Image Translation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Image-to-Image Translation transforms images from one visual domain to another whilst preserving content structure, converting between image modalities such as sketch-to-photo, day-to-night, satellite-to-map, or style transfer between artistic styles.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:image-to-image-translation", "labels": [ "Image to Image Translation" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "Generative Adversarial Network", "Image Generation", "MetaverseDomain", "Style Transfer" ] }, { "id": "image-to-image", "title": "Image-to-Image", "domain": "ai", "domain_name": "Ai", "definition": "Image-to-image is a class of generative tasks where a model transforms an input image into an output image, conditioned on the input and often a text prompt. Examples include style transfer, editing and translation between domains.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:image-to-image", "labels": [ "Image-to-Image", "Image-to-Image Generation" ], "is_subclass_of": [ "Image Generation" ], "wikilinks": [ "Diffusion Model", "Inpainting", "Stable Diffusion", "Image Generation" ] }, { "id": "imaging-parameters", "title": "Imaging Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Imaging Parameters are the configurable settings\u2014such as exposure time, aperture, ISO, focal length, white balance, and sensor gain\u2014that govern how a camera or depth sensor captures light and produces a digital image. In spatial computing pipelines, correct imaging parameter calibration is essential for photogrammetric reconstruction, volumetric capture, and computer-vision model accuracy across varying lighting conditions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:imaging-parameters", "labels": [ "Imaging Parameters" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "imaging-spectrometer", "title": "Imaging Spectrometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:imaging-spectrometer", "labels": [ "Imaging Spectrometer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "imitation-learning", "title": "Imitation Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Imitation Learning (IL), also termed Learning from Demonstration (LfD) or Programming by Demonstration (PbD), is the sequential-decision-making paradigm in which an autonomous agent acquires a policy \u03c0(a|s) mapping observations to actions by mimicking expert demonstrations rather than by maximisi...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:imitation-learning", "labels": [ "Imitation Learning", "ImitationLearning" ], "is_subclass_of": [ "AI Technique", "AI Agent System", "Machine Learning Discipline", "Sequential Decision Making", "Policy Learning", "Supervised Learning", "Learning from Demonstration" ], "wikilinks": [ "Abbeel Ng 2004 Apprenticeship Learning", "Action Chunking Transformer", "Action Decoder", "Action Space", "AIRL", "AlgorithmLayer", "Apprenticeship Learning", "Argall Chernova Veloso Browning 2009 LfD Survey", "Autonomous Driving", "Bain Sammut 1995 Behavioural Cloning", "Behavioural Cloning", "Brohan et al 2022 RT-1", "Brohan et al 2023 RT-2", "CALVIN", "Chi et al 2023 Diffusion Policy", "Classical Control", "ControlLayer", "Cross-Embodiment Transfer", "Demonstration Dataset", "Demonstrator" ] }, { "id": "immersion-cooling", "title": "Immersion Cooling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Immersion cooling is a thermal-management technique in which electronic hardware is submerged directly in a thermally conductive but electrically insulating dielectric fluid to dissipate heat. It offers far higher heat-transfer efficiency than air cooling, enabling denser deployments and lower cooling energy overhead. The technique is widely adopted in high-density data centres and cryptocurrency-mining facilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersion-cooling", "labels": [ "Immersion Cooling" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "immersion", "title": "Immersion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Subjective experience of psychological engagement and sense of presence within a virtual environment, characterized by reduced awareness of physical surroundings and absorption in virtual context.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:immersion", "labels": [ "Immersion", "User Immersion" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ACM", "Cognitive Immersion", "Content Quality", "Display Technology", "Flow State", "Interaction Mechanism", "PresentationLayer", "Sensory Immersion", "User Engagement", "Visual Fidelity", "ApplicationLayer", "Audio Spatialization", "Emotional Immersion", "Haptic Feedback", "InteractionDomain", "Presence" ] }, { "id": "immersive-audio-system", "title": "Immersive Audio System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Audio technologies enabling three-dimensional soundscapes in virtual environments through spatial audio processing, binaural rendering, and head-related transfer functions (HRTF), creating realistic acoustic experiences that respond to user position and movement within VR, AR, and metaverse appli...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:immersive-audio-system", "labels": [ "Immersive Audio System" ], "is_subclass_of": [ "Display and Rendering", "Audio Technology" ], "wikilinks": [ "Audio Technology", "metaverse", "Spatial Presence" ] }, { "id": "immersive-audio-technology", "title": "Immersive Audio Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technologies and techniques for rendering spatially accurate, three-dimensional soundscapes within virtual and augmented reality environments. Encompasses binaural rendering, ambisonics, head-related transfer function (HRTF) personalisation, and real-time positional audio to create convincing auditory presence that reinforces visual immersion.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:immersive-audio-technology", "labels": [ "Immersive Audio Technology" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "immersive-audio", "title": "Immersive Audio", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive audio refers to audio reproduction and processing technologies that create a convincing three-dimensional sound field, enveloping the listener in a spatially accurate sonic environment. It encompasses object-based audio formats, binaural rendering, ambisonics, and head-related transfer function (HRTF) personalisation, enabling the perception of sounds positioned above, below, and around the listener rather than confined to a flat stereo plane.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:immersive-audio", "labels": [ "Immersive Audio" ], "is_subclass_of": [ "Audio Technology" ], "wikilinks": [] }, { "id": "immersive-collaboration", "title": "Immersive Collaboration", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Immersive collaboration is the practice of working together within shared spatial virtual or mixed-reality environments where participants, represented by avatars, interact with 3D content and one another in real time. It extends remote teamwork beyond flat video calls by providing spatial audio, embodied presence, and manipulable shared objects. Immersive collaboration is a key application of metaverse and XR technologies for the enterprise.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersive-collaboration", "labels": [ "Immersive Collaboration", "Immersive Enterprise Collaboration" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "immersive-communication", "title": "Immersive Communication", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive Communication encompasses communication systems and experiences that leverage extended reality (XR) technologies \u2014 including virtual reality (VR) meetings, augmented reality (AR) collaboration tools, spatial audio conferencing, and mixed reality platforms \u2014 to create spatially-aware, embodied interaction environments that transcend conventional 2D video interfaces. These systems convey presence, embodiment, and spatial context by rendering participants as avatars or volumetric representations within shared virtual or hybrid spaces, enabling natural non-verbal cues, gaze, gesture, and proxemics that flat-screen media cannot replicate. The field draws on human-computer interaction research, telepresence theory, and spatial computing infrastructure to reduce cognitive load and social distance in remote work, education, and social engagement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersive-communication", "labels": [ "Immersive Communication" ], "is_subclass_of": [ "Telecollaboration" ], "wikilinks": [ "Telecollaboration" ] }, { "id": "immersive-computing", "title": "Immersive Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive computing is the broad discipline of designing and operating computing systems that surround users with synthetic or blended sensory environments \u2014 encompassing virtual reality, augmented reality, and mixed reality \u2014 to create a subjective sense of presence within a digitally mediated space. It integrates real-time 3D rendering, spatial tracking, haptic feedback, and multi-modal interaction to replace or augment the user's natural perceptual field.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersive-computing", "labels": [ "Immersive Computing" ], "is_subclass_of": [ "Extended Reality", "Platform and Environment" ], "wikilinks": [] }, { "id": "immersive-display-infrastructure", "title": "Immersive Display Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The hardware and software infrastructure used to install, calibrate, and drive immersive visual displays in XR, digital signage, and spatial computing contexts. This encompasses projection mapping, light-field displays, LED walls, head-mounted display optics, and the rendering pipelines that feed them.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:immersive-display-infrastructure", "labels": [ "Immersive Display Infrastructure", "Installation and display tech" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Light field" ] }, { "id": "immersive-education", "title": "Immersive Education", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Immersive education is the use of virtual, augmented, and mixed-reality environments to deliver learning experiences in which students are spatially present within simulated or augmented contexts. It supports experiential and situated learning, letting learners practise skills, explore inaccessible places, and visualise abstract concepts in 3D. Immersive education is an applied intersection of educational technology and metaverse platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersive-education", "labels": [ "Immersive Education" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [] }, { "id": "immersive-entertainment", "title": "Immersive Entertainment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Entertainment experiences leveraging VR, AR, and mixed reality technologies to create interactive, participatory content that transforms audiences from passive spectators into active participants within gaming, theme parks, live events, and digital experiences.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:immersive-entertainment", "labels": [ "Immersive Entertainment" ], "is_subclass_of": [ "Platform and Environment", "Digital Entertainment" ], "wikilinks": [ "Digital Entertainment", "Interactive Storytelling", "metaverse", "Telecollaboration" ] }, { "id": "immersive-experience-pipeline", "title": "Immersive Experience Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The end-to-end workflow for creating, processing, and delivering XR content, encompassing concept design, 3D asset creation, game engine integration, optimisation, platform deployment, and cloud streaming to produce immersive virtual experiences for metaverse applications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:immersive-experience-pipeline", "labels": [ "Immersive Experience Pipeline" ], "is_subclass_of": [ "Platform and Environment", "Content Production Workflow" ], "wikilinks": [ "XR Content Delivery", "Content Production Workflow", "metaverse" ] }, { "id": "immersive-experience", "title": "Immersive Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Immersive Experience is a deeply engaging interaction or environment that induces a compelling sense of presence by combining high-fidelity multi-sensory stimulation, responsive interaction, and narrative coherence across visual, auditory, and haptic channels. Underpinned by virtual reality, augmented reality, and spatial audio technologies, immersive experiences span VR training simulations, therapeutic applications, collaborative virtual workspaces, and entertainment, with quality measured through presence questionnaires, physiological indicators, and behavioural engagement metrics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:immersive-experience", "labels": [ "Immersive Experience", "Immersive User Experience", "Immersive XR Experience", "ImmersiveExperience" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "immersive-experiences", "title": "Immersive Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital interactions utilising VR, AR, and mixed reality technologies to create engaging, interactive environments that generate a sense of presence and participation, spanning gaming, education, enterprise collaboration, and consumer applications within metaverse platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:immersive-experiences", "labels": [ "Immersive Experiences" ], "is_subclass_of": [ "Digital Experience" ], "wikilinks": [ "Digital Experience", "metaverse", "Virtual Presence" ] }, { "id": "immersive-gaming", "title": "Immersive Gaming", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Video gaming experiences enhanced through VR, AR, and metaverse technologies that place players within interactive 3D environments, enabling physical interaction, spatial awareness, and presence-based gameplay through advanced headsets, motion tracking, and haptic feedback systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:immersive-gaming", "labels": [ "Immersive Gaming", "Immersive Gameplay", "Immersive Gaming Experiences" ], "is_subclass_of": [ "Platform and Environment", "Digital Gaming" ], "wikilinks": [ "VR Gaming Experience", "Digital Gaming", "metaverse" ] }, { "id": "immersive-interface", "title": "Immersive Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User interaction systems for VR, AR, and metaverse applications encompassing spatial user interfaces, haptic feedback devices, gesture recognition, voice control, and multimodal input mods that enable natural, intuitive engagement with virtual environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:immersive-interface", "labels": [ "Immersive Interface", "ImmersiveInterface" ], "is_subclass_of": [ "Interaction Technology", "Human Computer Interface" ], "wikilinks": [ "Natural VR Interaction", "Human-Computer Interface", "metaverse" ] }, { "id": "immersive-learning", "title": "Immersive Learning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Educational approaches utilising VR, AR, and metaverse technologies to create engaging, interactive learning environments that enable experiential skill development, realistic scenario simulation, and enhanced knowledge retention through presence-based educational experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:immersive-learning", "labels": [ "Immersive Learning" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [ "Experiential Education", "Educational Technology", "metaverse" ] }, { "id": "immersive-media", "title": "Immersive Media", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive media is content designed to surround and respond to the viewer so they perceive themselves as present within the experience rather than observing it from outside. It spans virtual, augmented and mixed reality, 360-degree and volumetric video, and spatial audio, typically consumed through head-mounted displays or spatial devices. Effective immersive media combines stereoscopic visuals, spatialised sound and low-latency tracking to sustain the sense of presence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:immersive-media", "labels": [ "Immersive Media" ], "is_subclass_of": [ "Interactive Media" ], "wikilinks": [] }, { "id": "immersive-presence", "title": "Immersive Presence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive presence is the subjective sense of physically being within a virtual or mediated environment, rather than merely observing it on a screen. It is produced by consistent, low-latency sensory feedback across vision, audio and touch, so that a user's perception and reactions align with the simulated space as though it were real. High presence is a key quality metric for VR and AR systems, driven by technologies such as spatial audio and haptic feedback that reinforce the illusion of embodiment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:immersive-presence", "labels": [ "Immersive Presence" ], "is_subclass_of": [ "Immersive Experience" ], "wikilinks": [] }, { "id": "immersive-storytelling", "title": "Immersive Storytelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive storytelling is a narrative form that uses extended reality technologies \u2014 virtual, augmented and mixed reality \u2014 to place an audience within a story environment, allowing them to experience events from a first-person perspective and, in interactive formats, to influence how the narrative unfolds. It draws on techniques from film, game design and spatial audio to create a sense of presence, replacing the fixed frame of traditional media with a fully surrounding, often responsive, environment. Immersive storytelling is used in film production, journalism, museum exhibits and personalised virtual experiences to deepen audience engagement beyond passive viewing.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:immersive-storytelling", "labels": [ "Immersive Storytelling" ], "is_subclass_of": [ "Extended Reality" ], "wikilinks": [] }, { "id": "immersive-technology-exhibition-event", "title": "Immersive Technology Exhibition Event", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An exhibition is a curated public or professional presentation of artefacts, technologies, artworks, or innovations staged in a physical venue, virtual environment, or hybrid setting. In the context of immersive technology and AI, exhibitions serve as deployment contexts for spatial computing experiences, interactive demonstrations, and knowledge communication \u2014 including trade fairs, museum installations, and dedicated industry events such as AIX. They sit at the intersection of event management, immersive experience design, and audience engagement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:immersive-technology-exhibition-event", "labels": [ "Immersive Technology Exhibition Event", "exhibition" ], "is_subclass_of": [ "Event" ], "wikilinks": [ "AIX" ] }, { "id": "immersive-technology", "title": "Immersive Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The collective technologies comprising virtual reality (VR), augmented reality (AR), mixed reality (MR), and extended reality (XR) that create digital experiences blending physical and virtual environments, enabling user presence and interaction within computer-generated or enhanced spaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:immersive-technology", "labels": [ "Immersive Technology", "ImmersiveTechnology" ], "is_subclass_of": [ "Digital Technology" ], "wikilinks": [ "Digital Technology", "metaverse", "Metaverse Platform" ] }, { "id": "immersive-workspaces", "title": "Immersive Workspaces", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive workspaces are physical or virtual environments enhanced with spatial and extended-reality technologies\u2014including AR, VR, and mixed reality\u2014that allow users to interact intuitively with digital content and collaborators in three-dimensional space, fostering a strong sense of presence and engagement. They integrate spatial computing principles to support real-time manipulation of data, collaborative visualisation, immersive training, and seamless transitions between physical and digital environments. Immersive workspaces are increasingly deployed across corporate, industrial, and educational settings to support hybrid and remote collaboration at scale.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:immersive-workspaces", "labels": [ "Immersive Workspaces" ], "is_subclass_of": [ "Platform and Environment", "Metaverse and Telecollaboration" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "immutability", "title": "Immutability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The tamper-resistant property of blockchain ledgers whereby confirmed records cannot be altered without invalidating the cryptographic chain of hashes, providing verifiable finality, audit-trail integrity, and resistance to retroactive manipulation. Achieved through chained Merkle roots, accumulated proof-of-work, or BFT-based finality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:immutability", "labels": [ "Immutability" ], "is_subclass_of": [ "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "immutable-infrastructure", "title": "Immutable Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Immutable infrastructure is an operational model in which servers and components are never modified after deployment; instead, any change is delivered by building a new versioned artefact (image or container) and replacing the running instance. This eliminates configuration drift, makes deployments reproducible and rollbacks trivial, and pairs naturally with infrastructure-as-code and automated pipelines. It contrasts with mutable, in-place patching of long-lived servers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:immutable-infrastructure", "labels": [ "Immutable Infrastructure" ], "is_subclass_of": [ "Infrastructure", "Digital Infrastructure" ], "wikilinks": [] }, { "id": "immutable-record", "title": "Immutable Record", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An immutable record is a data entry that, once written, cannot be altered or deleted without cryptographic or consensus-based detection of tampering. Immutability is enforced through hash chaining, Merkle trees, distributed ledger consensus mechanisms, or append-only data structures, making immutable records foundational to audit trails, provenance tracking, and trustworthy data archiving.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:immutable-record", "labels": [ "Immutable Record", "Immutable Data Model" ], "is_subclass_of": [ "Distributed Ledger" ], "wikilinks": [] }, { "id": "immutable-storage", "title": "Immutable Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Immutable storage is a data persistence model in which written records cannot be modified or deleted for a defined period, preserving their integrity against tampering and accidental loss. It is realised through write-once-read-many policies, append-only logs, content addressing, and cryptographic hashing, and underpins audit trails, regulatory retention, and ransomware resilience. Immutability provides verifiable evidence that data has not changed since it was committed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:immutable-storage", "labels": [ "Immutable Storage" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "impact-assessment", "title": "Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Impact assessment is a structured process for evaluating the likely effects of a policy, project, technology or decision before it is adopted, and for monitoring effects once implemented. It identifies affected stakeholders, weighs benefits against harms and risks, and informs decisions on whether and how to proceed, with mitigations where necessary. Domain-specific variants include environmental, privacy, data-protection and algorithmic impact assessments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:impact-assessment", "labels": [ "Impact Assessment" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "impact-crater", "title": "Impact Crater", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:impact-crater", "labels": [ "Impact Crater" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "impact-investing", "title": "Impact Investing", "domain": "finance", "domain_name": "Finance", "definition": "Impact investing is an investment approach that intentionally seeks to generate measurable positive social or environmental outcomes alongside a financial return. It is distinguished from conventional investing by the explicit intent to create impact and by the commitment to measure and report that impact against defined indicators. Impact investing spans asset classes and is a core practice within sustainable finance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:impact-investing", "labels": [ "Impact Investing" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "impact-metrics", "title": "Impact Metrics", "domain": "governance", "domain_name": "Governance", "definition": "Impact metrics are quantitative and qualitative indicators used to measure the social, environmental, or economic outcomes produced by an intervention, project, or organisation. They translate a theory of change into measurable outputs, outcomes, and longer-term impacts, enabling comparison, accountability, and decision-making. Standardised metric catalogues such as IRIS+ promote comparability across programmes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:impact-metrics", "labels": [ "Impact Metrics" ], "is_subclass_of": [ "Evaluation Metric" ], "wikilinks": [] }, { "id": "impedance-control", "title": "Impedance Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot control strategy that dynamically modulates mechanical compliance (stiffness, damping, and inertia) to regulate the dynamic relationship between force and motion at the robot end-effector, enabling compliant and safe interaction with objects, surfaces, and humans without requiring explicit force feedback in all configurations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:impedance-control", "labels": [ "Impedance Control", "Cartesian Impedance Control", "Force Impedance Control" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Control Systems" ], "wikilinks": [ "Adaptive Manipulation", "Compliant Motion Control", "Control Systems", "Damping Control", "Force Regulation", "Hybrid Control", "Safe Human Interaction", "Soft Contact Tasks", "Stiffness Modulation", "Virtual Dynamics Model", "Robotics", "RoboticsDomain" ] }, { "id": "imperial-college-london", "title": "Imperial College London", "domain": "ai", "domain_name": "Ai", "definition": "Imperial College London is a public research university in South Kensington, London, specialising in science, engineering, medicine, and business, founded in 1907 through the merger of the Royal College of Science, the Royal School of Mines, and the City and Guilds College. It gained full independence from the University of London on its centenary in 2007, and is consistently ranked among the world's top universities for STEM disciplines. The college is internationally recognised for research across artificial intelligence, machine learning, bioengineering, robotics, quantum computing, climate science, and computational biology. As a founding member of the Russell Group and a partner in numerous global research consortia, Imperial serves as a key node in the United Kingdom's knowledge and innovation infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:imperial-college-london", "labels": [ "Imperial College London" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "University of Cambridge", "University of Edinburgh", "owl:Thing" ] }, { "id": "impermanent-loss", "title": "Impermanent Loss", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Impermanent loss is the opportunity cost incurred by a liquidity provider in an automated market maker (AMM) when the price ratio of deposited assets diverges from the ratio at deposit time, causing the provider's pool share to be worth less than simply holding the assets would have been. The loss is 'impermanent' because it reverses if prices return to the original ratio, but becomes realised upon withdrawal.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:impermanent-loss", "labels": [ "Impermanent Loss" ], "is_subclass_of": [ "Liquidity Provision" ], "wikilinks": [] }, { "id": "implicit-neural-representation", "title": "Implicit Neural Representation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An Implicit Neural Representation (INR) is a method of encoding continuous signals\u2014such as 3D shapes, scenes, or images\u2014as the weights of a neural network rather than as discrete grids or meshes. The network acts as a function that maps spatial or temporal coordinates to signal values, enabling theoretically infinite resolution. INRs are widely used in novel-view synthesis, shape reconstruction, and physics simulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:implicit-neural-representation", "labels": [ "Implicit Neural Representation" ], "is_subclass_of": [ "Machine Learning", "AI Model Architecture" ], "wikilinks": [] }, { "id": "importance-sampling", "title": "Importance Sampling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Importance sampling is a variance-reduction technique in Monte Carlo estimation that draws samples from a proposal distribution that concentrates probability mass in regions contributing most to the quantity being estimated, then corrects for the distributional mismatch using importance weights. It is foundational to Bayesian inference, reinforcement learning, path-tracing renderers, and off-policy evaluation. Importance sampling allows tractable estimation of expectations under distributions that are difficult or impossible to sample from directly. Poor choice of proposal distribution can however lead to high-variance or even infinite-variance estimators, necessitating careful design.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:importance-sampling", "labels": [ "Importance Sampling", "Importance Weight", "Multiple Importance Sampling", "Sampling Importance Resampling" ], "is_subclass_of": [ "Monte Carlo Methods" ], "wikilinks": [] }, { "id": "importer", "title": "Importer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A natural or legal person located or established in the Union that places on the market an AI system that bears the name or trademark of a natural or legal person established in a third country.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:importer", "labels": [ "Importer" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "in-hand-manipulation", "title": "In Hand Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "In-hand manipulation is the robotic skill of repositioning or reorienting a grasped object using the fingers and palm of a hand, without placing it down or relying on the arm to reposition it. It demands coordinated finger control, tactile sensing and contact modelling to maintain stable grasps while controlled slip and finger gaiting move the object. It is a core capability of dexterous manipulation in multi-fingered robotic hands.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:in-hand-manipulation", "labels": [ "In Hand Manipulation", "In-Hand Manipulation" ], "is_subclass_of": [ "Dexterous Manipulation" ], "wikilinks": [] }, { "id": "in-situ-observation", "title": "In Situ Observation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:in-situ-observation", "labels": [ "In Situ Observation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "in-situ-resource-utilisation", "title": "In Situ Resource Utilisation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:in-situ-resource-utilisation", "labels": [ "In Situ Resource Utilisation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "in-camera-vfx", "title": "In-Camera VFX", "domain": "metaverse", "domain_name": "Metaverse", "definition": "In-camera visual effects (ICVFX) is a virtual-production technique in which final-pixel visual effects are captured live during photography, typically by displaying real-time rendered 3D environments on large LED volumes behind the actors. Camera tracking drives the rendered background's perspective so that parallax and lighting match the physical camera, producing realistic in-camera composites without later green-screen keying. ICVFX collapses much of post-production into the shoot itself.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:in-camera-vfx", "labels": [ "In-Camera VFX" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "in-context-learning", "title": "In-Context Learning", "domain": "ai", "domain_name": "Ai", "definition": "In-Context Learning (ICL) is the capability of large language models to adapt their output distribution to a novel task at inference time by conditioning on a small number of labelled input-output demonstrations embedded directly within the prompt, without any gradient-based parameter updates. The model implicitly extracts task structure and decision boundaries from the provided examples and applies that inferred structure to unseen queries, enabling rapid generalisation across domains. ICL is an emergent phenomenon that scales with model size and training-data diversity, and is theoretically framed as implicit Bayesian inference over a latent space of task hypotheses. It underpins the practical utility of modern large language models by eliminating the need for task-specific fine-tuning pipelines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:in-context-learning", "labels": [ "In-Context Learning" ], "is_subclass_of": [ "Meta-Learning" ], "wikilinks": [] }, { "id": "in-house-ai-infrastructure", "title": "In-House AI Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The internal computational and software resources an organization builds and maintains to train, fine-tune, and deploy AI models independently of external providers.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:in-house-ai-infrastructure", "labels": [ "In-House AI Infrastructure" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "in-memory-computing", "title": "In-Memory Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "In-memory computing is an architectural approach that holds working data sets in a system's main memory (RAM) rather than on disk, eliminating storage-layer I/O from the critical path of data access and processing. By keeping data resident in fast volatile memory, it delivers order-of-magnitude reductions in latency and supports high-throughput analytics, transaction processing, and real-time decisioning. It typically pairs with techniques such as columnar layouts, distributed caching, and durability mechanisms (logging, replication, persistence) to combine speed with resilience.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:in-memory-computing", "labels": [ "In-Memory Computing" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "in-page-executable-notebook-pattern", "title": "In-Page Executable Notebook Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A knowledge-management pattern that embeds executable code cells with shared global state within a page or document, mimicking the interactive notebook paradigm of Jupyter. Code blocks execute in sequence against a common variable scope, enabling data loading, transformation, and display without leaving the note-taking environment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:in-page-executable-notebook-pattern", "labels": [ "In-Page Executable Notebook Pattern", "Jupyter like behaviour within a page" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "in-process-communication", "title": "In-Process Communication", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "In-process communication is the exchange of data and control between software components running inside a single operating-system process: direct function and method calls, shared objects on the heap, and in-memory events. Because callers and callees share one address space, invocation costs nanoseconds, arguments pass by reference without serialisation, and failures are shared \u2014 the defining communication style of monolithic architectures, and the deliberate opposite of inter-process communication, which crosses process boundaries via sockets, pipes, or messaging at the price of serialisation, latency, and partial-failure semantics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:in-process-communication", "labels": [ "In-Process Communication" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [ "Software Architecture", "Monolithic Architecture", "Inter Process Communication", "Shared Memory" ] }, { "id": "inaccessible-design", "title": "Inaccessible Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Design of products, interfaces, environments, or experiences that erects barriers for people with disabilities or situational limitations \u2014 relying on a single sensory channel, demanding fine motor precision, ignoring assistive technologies, or assuming one body, language, or cognitive style. The antonym of accessible design, it excludes users not through intent but through unexamined defaults, and in immersive XR contexts can render entire experiences unusable.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:inaccessible-design", "labels": [ "Inaccessible Design" ], "is_subclass_of": [ "User Interface Design" ], "wikilinks": [ "User Interface Design", "Accessible Design", "Universal Design", "Accessible Experience" ] }, { "id": "incentive-alignment", "title": "Incentive Alignment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Incentive Alignment is the design principle by which blockchain protocols structure economic rewards and penalties such that individual rational behaviour converges with the collective goals of the network. When correctly engineered, participants who act in their own self-interest\u2014validators confirming blocks, miners extending chains, token holders participating in governance\u2014simultaneously reinforce system security, liveness, and integrity. Misaligned incentives produce attack vectors such as selfish mining, validator collusion, or governance capture.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:incentive-alignment", "labels": [ "Incentive Alignment", "Contributor Incentive Alignment" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "incentive-compatibility", "title": "Incentive Compatibility", "domain": "finance", "domain_name": "Finance", "definition": "A property of a mechanism or protocol in which every participant's dominant strategy is to act truthfully, such that honest reporting of private information is individually rational. Incentive compatibility ensures that the designed rules align agents' self-interest with socially desirable outcomes, eliminating the benefit of strategic misrepresentation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:incentive-compatibility", "labels": [ "Incentive Compatibility" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "incentive-mechanism", "title": "Incentive Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An incentive mechanism is a designed system of rewards and penalties that aligns the self-interested behaviour of participants with a desired collective outcome. In distributed systems and blockchain networks it uses tokens, fees and slashing to motivate honest participation, resource provision and protocol-conformant behaviour. Drawing on game theory and mechanism design, it makes cooperation the rational strategy even among mutually distrustful actors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:incentive-mechanism", "labels": [ "Incentive Mechanism" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "incentive-structures", "title": "Incentive Structures", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Incentive structures are the systems of rewards and penalties that shape the behaviour of agents within an economic or protocol system, aligning individual self-interest with desired collective outcomes. In decentralised systems they encode rewards such as block subsidies, fees, and staking yields, alongside slashing or penalties, to make honest participation the rational choice. Well-designed incentive structures are central to mechanism design and the security of token economies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:incentive-structures", "labels": [ "Incentive Structures", "Incentive Structure" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "inception-v3", "title": "Inception v3", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Inception v3 is a convolutional neural network architecture from Google that refined the earlier Inception design with factorised convolutions, auxiliary classifiers, and label smoothing to improve accuracy and reduce computational cost on image classification tasks. It replaces large convolutional filters with sequences of smaller asymmetric convolutions, cutting parameter count while preserving representational capacity. Pretrained on ImageNet, its penultimate-layer activations are the standard feature extractor used to compute the Fr\u00e9chet Inception Distance and Inception Score for evaluating generative image models. It remains a common baseline and building block in computer vision pipelines despite the emergence of newer architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:inception-v3", "labels": [ "Inception v3" ], "is_subclass_of": [ "Convolutional Neural Network" ], "wikilinks": [] }, { "id": "incidence-angle", "title": "Incidence Angle", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:incidence-angle", "labels": [ "Incidence Angle" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "incident-investigation", "title": "Incident Investigation", "domain": "security", "domain_name": "Security", "definition": "Incident investigation is the structured process of determining the root cause, scope, and impact of a security or operational incident by collecting and analysing evidence. In a cybersecurity context it follows the breach lifecycle\u2014identification, containment, evidence preservation, forensic analysis, and lessons learned\u2014to understand how an incident occurred and to prevent recurrence. It relies heavily on audit trails and digital-forensics methods to reconstruct events.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:incident-investigation", "labels": [ "Incident Investigation" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "incident-management", "title": "Incident Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Incident management is the operational discipline of detecting, responding to, resolving, and learning from unplanned disruptions to a service. It coordinates people and tooling through detection, triage, escalation, mitigation, and recovery, then conducts blameless post-mortems to prevent recurrence. Closely associated with site reliability engineering and ITIL practice, it aims to minimise mean time to recovery and protect service-level objectives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:incident-management", "labels": [ "Incident Management" ], "is_subclass_of": [ "Site Reliability Engineering" ], "wikilinks": [] }, { "id": "incident-reporting", "title": "Incident Reporting", "domain": "governance", "domain_name": "Governance", "definition": "Incident Reporting is a systematic process by which organisations detect, document, classify, and communicate adverse events, near-misses, or anomalous system behaviours to relevant stakeholders, regulatory bodies, and affected parties. In AI and digital systems contexts, it covers cybersecurity breaches, AI system failures, algorithmic harms, and data protection violations. Effective incident reporting is foundational to organisational learning, regulatory compliance, and public accountability, enabling causal analysis that drives systemic improvement. Mandatory disclosure obligations are increasingly codified in sector-specific legislation including NIS2, the EU AI Act, and financial services regulations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:incident-reporting", "labels": [ "Incident Reporting", "AI Incident Reporting", "Critical Safety Incident Reporting", "Serious Incident Reporting" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "incident-response", "title": "Incident Response", "domain": "security", "domain_name": "Security", "definition": "Incident Response is the structured organisational process for detecting, containing, eradicating, and recovering from cybersecurity incidents, followed by post-incident analysis to prevent recurrence and strengthen defensive posture. It is operationalised through lifecycle models such as NIST SP 800-61 and the SANS PICERL framework (Preparation, Identification, Containment, Eradication, Recovery, Lessons Learned), enacted via security operations centre (SOC) playbooks, SIEM-driven detection pipelines, and SOAR-automated response actions. Effective incident response minimises dwell time\u2014the interval between initial compromise and detection\u2014limits lateral movement and data exfiltration, and satisfies regulatory notification obligations under frameworks such as GDPR and NIS2. AI-assisted triage, automated containment orchestration, and threat-intelligence enrichment have become defining characteristics of mature programmes.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:incident-response", "labels": [ "Incident Response", "AI Incident Response", "Incident Response Protocol", "Incident Triage", "IncidentResponse" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "inclusive-design", "title": "Inclusive Design", "domain": "governance", "domain_name": "Governance", "definition": "Inclusive design is a design methodology and philosophy that creates products, services, environments, and digital experiences to be usable by the widest possible range of people across diverse abilities, ages, languages, cultures, and contexts, without requiring specialised adaptations or separate versions for different user groups. It treats human diversity as a design resource rather than a problem to be accommodated, recognising that designing for people with disabilities, situational impairments, or non-standard use contexts typically produces solutions that benefit all users. The approach is distinct from universal design in emphasising process \u2014 including people with diverse characteristics as active co-designers throughout \u2014 rather than solely optimising for a single configuration serving all. Across digital, spatial, and AI-mediated systems, inclusive design establishes both participatory methods and measurable technical criteria that reduce systematic exclusion.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:inclusive-design", "labels": [ "Inclusive Design", "Inclusive Design Principles", "InclusiveDesign" ], "is_subclass_of": [ "Design Thinking" ], "wikilinks": [] }, { "id": "inclusive-experience", "title": "Inclusive Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An immersive or interactive experience deliberately shaped so that people of differing abilities, bodies, languages, cultures, and levels of technical confidence can all participate fully and feel that the experience was made for them. In extended reality contexts an inclusive experience goes beyond meeting accessibility conformance: it offers multiple interaction modalities, adaptable comfort and representation options, and social norms that make diverse participants welcome.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:inclusive-experience", "labels": [ "Inclusive Experience" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [ "User Experience", "Accessible Experience", "Inclusive Xr Design", "Universal Design" ] }, { "id": "inclusive-growth", "title": "Inclusive Growth", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "AI should benefit people and planet by augmenting human capabilities, enhancing creativity, advancing inclusion of underrepresented populations, reducing economic, social and geographical inequalities, and protecting natural environments, thereby invigorating inclusive growth, sustainable develop...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:inclusive-growth", "labels": [ "Inclusive Growth" ], "is_subclass_of": [ "Legal and Regulatory", "Policy Enforcement" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "inclusive-participation", "title": "Inclusive Participation", "domain": "governance", "domain_name": "Governance", "definition": "Inclusive participation is the design principle and operational practice of structuring technologies, platforms, and governance processes to enable full and meaningful engagement by individuals of diverse abilities, economic circumstances, cultural backgrounds, and technical literacy levels, removing barriers that would otherwise exclude underrepresented or marginalised populations. In technology contexts, inclusive participation encompasses accessible user interface design, multilingual support, low-bandwidth alternatives, assistive technology compatibility, and governance mechanisms that distribute decision-making power beyond technically or economically privileged groups. It is a cross-cutting concern in XR platform design, DAO governance, metaverse standards development, and public goods infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:inclusive-participation", "labels": [ "Inclusive Participation" ], "is_subclass_of": [ "Participation Framework" ], "wikilinks": [] }, { "id": "inclusive-xr-design", "title": "Inclusive Xr Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Design modologies for creating accessible VR, AR, and metaverse experiences that accommodate diverse user abilities, including those with visual, auditory, motor, and cognitive disabilities, ensuring equitable access through alternative input mods, adaptive interfaces, and assistive technology in...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:inclusive-xr-design", "labels": [ "Inclusive Xr Design", "Inclusive XR Design" ], "is_subclass_of": [ "Governance and Safety", "Accessible Design" ], "wikilinks": [ "Universal XR Access", "Accessible Design", "metaverse" ] }, { "id": "inclusive-xr-experience", "title": "Inclusive Xr Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Extended reality applications and environments designed to provide equitable, accessible experiences for users of all abilities, incorporating assistive technologies, alternative interaction modalities, and adaptive features that ensure meaningful participation in VR, AR, and metaverse spaces.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:inclusive-xr-experience", "labels": [ "Inclusive Xr Experience", "Inclusive Virtual Experiences" ], "is_subclass_of": [ "Governance and Safety", "Accessible Experience" ], "wikilinks": [ "Equitable Metaverse Access", "Accessible Experience", "metaverse", "Telecollaboration" ] }, { "id": "independent-living", "title": "Independent Living", "domain": "robotics", "domain_name": "Robotics", "definition": "Independent living is the outcome of enabling people with disabilities, reduced mobility or age-related impairment to carry out daily activities and remain in their own homes with minimal reliance on carers. It is supported by accessibility technology and assistive robotics that compensate for physical or cognitive limitations, such as mobility aids, environmental sensors and robotic assistance with domestic tasks. As a design goal it shapes requirements for assistive systems, prioritising safety, dignity and user autonomy over full automation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:independent-living", "labels": [ "Independent Living" ], "is_subclass_of": [ "Assistive Robotics" ], "wikilinks": [] }, { "id": "independent-verification", "title": "Independent Verification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Independent Verification is the ability of any participant to check the validity of transactions or state transitions themselves, using publicly available data and consensus rules, rather than trusting a third party's assertion. In blockchain systems it is what a full node performs when it re-executes and validates every block, giving the network its trust-minimised guarantees. It is a precondition for meaningful decentralisation, since a system that cannot be independently verified reduces to trusting whoever operates it.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:independent-verification", "labels": [ "Independent Verification" ], "is_subclass_of": [ "Verification Process" ], "wikilinks": [] }, { "id": "indexing", "title": "Indexing", "domain": "data", "domain_name": "Data", "definition": "Indexing is the technique of building auxiliary data structures that allow a system to locate records satisfying a query without scanning the entire dataset. By maintaining ordered or hashed mappings from key values to record locations, indexes turn linear searches into logarithmic or constant-time lookups, dramatically improving query performance at the cost of additional storage and update overhead. It is fundamental to databases, search engines and information retrieval.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:indexing", "labels": [ "Indexing" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "indoor-navigation", "title": "Indoor Navigation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Indoor Navigation encompasses the technologies and systems that provide wayfinding, positioning, and routing guidance within enclosed spaces where satellite-based positioning (GPS) is unavailable or unreliable. It relies on alternative positioning signals such as Wi-Fi fingerprinting, Bluetooth beacons, Ultra-Wideband (UWB) ranging, visual markers, inertial sensors, and lidar-based spatial maps. Indoor navigation is critical for large venues such as airports, hospitals, shopping centres, factories, and warehouses where occupant or asset routing improves operational efficiency and user experience.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:indoor-navigation", "labels": [ "Indoor Navigation" ], "is_subclass_of": [ "Spatial Computing", "Environmental Mapping" ], "wikilinks": [] }, { "id": "inductive-bias", "title": "Inductive Bias", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Inductive bias is the set of assumptions a learning algorithm uses to generalise from finite training data to unseen inputs. Because infinitely many functions fit any finite sample, a learner must prefer some hypotheses over others, and that preference \u2014 encoded in model architecture, regularisation, priors, or the choice of hypothesis space \u2014 is its inductive bias. Appropriate inductive bias is what lets a model extrapolate sensibly rather than merely memorise.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inductive-bias", "labels": [ "Inductive Bias" ], "is_subclass_of": [ "Machine Learning", "AI Research Area" ], "wikilinks": [] }, { "id": "industrial-ethernet", "title": "Industrial Ethernet", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Industrial Ethernet is the adaptation of standard IEEE 802.3 Ethernet for factory-floor and process control: ruggedised hardware combined with real-time protocol extensions \u2014 PROFINET, EtherNet/IP, EtherCAT, POWERLINK, Modbus TCP \u2014 that add the determinism, cyclic exchange, and device profiles plain Ethernet lacks. It delivers 100 Mbit/s and gigabit speeds, cycle times down to tens of microseconds in hard-real-time variants, and seamless connectivity between automation devices and IT systems, and has overtaken classic fieldbus as the dominant networking technology in new industrial installations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:industrial-ethernet", "labels": [ "Industrial Ethernet" ], "is_subclass_of": [ "Ethernet" ], "wikilinks": [ "Ethernet", "IndustrialAutomation", "Profinet", "Fieldbus", "Time-Sensitive Networking" ] }, { "id": "industrial-inspection", "title": "Industrial Inspection", "domain": "robotics", "domain_name": "Robotics", "definition": "Industrial inspection is the systematic examination of manufactured parts, assemblies, infrastructure, and production processes to verify conformance with quality specifications, detect defects, and identify safety or structural hazards before product deployment or during operational service. It encompasses both non-destructive testing methods (ultrasonic, eddy-current, X-ray, thermographic) and machine vision approaches that use computer vision algorithms to automate defect classification at production speeds. AI-driven industrial inspection replaces or augments human visual inspectors with deep learning models trained on labelled defect images, enabling consistent sub-millimetre defect detection at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:industrial-inspection", "labels": [ "Industrial Inspection" ], "is_subclass_of": [ "Quality Assurance" ], "wikilinks": [] }, { "id": "industrial-io-t", "title": "industrial iot", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Industrial IoT (IIoT) is the application of Internet of Things technologies to industrial operational environments, connecting sensors, actuators, programmable logic controllers, and edge gateways to analytics platforms and supervisory control systems to enable real-time operational visibility, process optimisation, predictive maintenance, and remote asset management. IIoT bridges historically isolated operational technology (OT) networks\u2014including SCADA systems, distributed control systems (DCS), and fieldbus networks\u2014with IT infrastructure, creating a converged cyber-physical stack requiring rigorous security governance under frameworks such as IEC 62443. Key application-layer standards include OPC UA (IEC 62541), MQTT (ISO/IEC 20922), AMQP, and time-sensitive networking extensions (IEEE 802.1 TSN) for deterministic Ethernet.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:industrial-io-t", "labels": [ "Industrial IoT", "Industrial Internet of Things" ], "is_subclass_of": [ "Internet of Things" ], "wikilinks": [] }, { "id": "industrial-manipulation", "title": "Industrial Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Industrial Manipulation refers to the robotic discipline and practice of using articulated robot arms and end-effectors to physically interact with objects in manufacturing, assembly, logistics, and processing environments, performing tasks such as pick-and-place, assembly, welding, painting, and material handling at production scale with repeatability and precision exceeding human capability. It encompasses the full kinematic and dynamic modelling of robot mechanisms, trajectory planning, force and torque control, and the integration of sensing for adaptive behaviour in structured industrial settings. Industrial manipulation robots are among the most economically significant deployed robotics systems, forming the backbone of automotive, electronics, and consumer goods production worldwide.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:industrial-manipulation", "labels": [ "Industrial Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "industrial-metaverse", "title": "Industrial Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A virtual platform integrating digital twin technology, simulation environments, and collaborative workspaces for manufacturing operations, supply chain management, remote equipment control, and industrial training across geographically distributed facilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:industrial-metaverse", "labels": [ "Industrial Metaverse" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Application Platform" ], "wikilinks": [ "3D CAD Integration", "Collaborative Design", "Industrial AI", "Industrial Protocol Gateway", "ISO 23247 Digital Twin Framework", "OPC UA", "Physics Simulation Engine", "Real-Time Data Synchronization", "Remote Control Interface", "Remote Operations", "Siemens Xcelerator", "Smart Manufacturing", "Supply Chain Visualization", "Training Simulation", "ApplicationLayer", "ComputationAndIntelligenceDomain", "Digital Twin", "Edge Computing", "InfrastructureDomain", "IoT Sensor Network" ] }, { "id": "industrial-network", "title": "Industrial Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An industrial network is a communications network designed to connect sensors, controllers, actuators and machines on a factory floor or industrial site, prioritising deterministic timing, reliability and safety over the general-purpose flexibility of enterprise IT networks. It typically uses fieldbus or industrial Ethernet protocols such as Profinet or EtherCAT to meet the real-time control requirements of robots and process equipment, and forms a core part of operational technology infrastructure. Industrial networks are increasingly connected to enterprise IT systems as part of Industry 4.0 initiatives, raising distinct security and segmentation requirements.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:industrial-network", "labels": [ "Industrial Network" ], "is_subclass_of": [ "Operational Technology" ], "wikilinks": [] }, { "id": "industrial-robot", "title": "Industrial Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Industrial Robot is a reprogrammable, automatically controlled manipulator programmable in three or more axes, fixed in place or mobile, for use in industrial automation applications as defined by ISO 8373:2012.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:industrial-robot", "labels": [ "Industrial Robot", "Industrial Manipulators" ], "is_subclass_of": [ "Robot Type", "Actuation and Control", "Robotics Systems", "Manufacturing Automation", "Electromechanical Systems", "Cyber Physical Systems", "Programmable Logic Controller" ], "wikilinks": [ "Automotive Manufacturing", "Autonomous Mobile Robot", "Calibration System", "Collaborative Automation", "Computer Numerical Control", "ControlLayer", "Cyber Physical Systems", "Electromechanical Systems", "Electronics Assembly", "Embedded Systems", "EN 60204", "EtherCAT", "Fixed Automation", "Flexible Manufacturing", "Food Processing Automation", "High Speed Assembly", "IEC 62061", "IndustrialAutomationDomain", "Industrial Network", "ISO 10218" ] }, { "id": "industrial-robotics", "title": "Industrial Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Industrial robotics is the application of programmable, reprogrammable robots to manufacturing tasks including assembly, welding, material handling, painting and inspection, operating within structured production environments to improve consistency, throughput and safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:industrial-robotics", "labels": [ "Industrial Robotics", "Traditional Industrial Robotics" ], "is_subclass_of": [ "Robotics", "Robotics Application" ], "wikilinks": [ "Actuators", "Collaborative Robots", "Control Theory", "Robotics", "https://ifr.org/", "https://en.wikipedia.org/wiki/Industrial_robot" ] }, { "id": "industrial-strategy", "title": "Industrial Strategy", "domain": "governance", "domain_name": "Governance", "definition": "The UK government's long-term policy framework for shaping the structure of the national economy through targeted support for priority sectors, exemplified by the 2017 Industrial Strategy white paper and the 2025 Modern Industrial Strategy, which designate growth-driving sectors \u2014 including digital technologies and artificial intelligence \u2014 for coordinated investment, skills development, regulatory reform, and public procurement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:industrial-strategy", "labels": [ "Industrial Strategy" ], "is_subclass_of": [ "Public Policy" ], "wikilinks": [ "Public Policy", "AI Investment", "National Ai Strategy", "Economic Growth" ] }, { "id": "industrial-symbiosis", "title": "Industrial Symbiosis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Industrial symbiosis is a collaborative approach within industrial ecology in which geographically proximate companies exchange materials, energy, water, and by-products to achieve collective environmental and economic benefits that are unattainable individually. By treating the waste or surplus of one facility as a resource input for another, industrial symbiosis operationalises circular economy principles at an industrial park or regional scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:industrial-symbiosis", "labels": [ "Industrial Symbiosis" ], "is_subclass_of": [ "Circular Economy" ], "wikilinks": [] }, { "id": "industrial-automation", "title": "industrialautomation", "domain": "robotics", "domain_name": "Robotics", "definition": "Industrial automation is the systematic application of control technologies\u2014including programmable logic controllers (PLCs), distributed control systems (DCS), supervisory control and data acquisition (SCADA), and robotic manipulators\u2014together with information technologies to operate industrial processes and machinery with minimal or no human intervention. It encompasses both discrete manufacturing (assembly lines, packaging, welding) and continuous process industries (chemical plants, oil refineries, power generation), delivering gains in throughput, consistency, quality, and safety by removing workers from hazardous or repetitive environments. The discipline integrates mechanical, electrical, software, and network engineering and is increasingly augmented by machine vision, collaborative robots (cobots), edge computing, and AI-driven process optimisation. Modern industrial automation converges with the Industrial Internet of Things (IIoT), digital twin modelling, and cyber-physical systems to enable adaptive, closed-loop manufacturing at scale.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:industrial-automation", "labels": [ "IndustrialAutomation", "High-Force Industrial Automation", "Industrial Automation" ], "is_subclass_of": [ "Cyber Physical Systems" ], "wikilinks": [] }, { "id": "industry-4-0", "title": "Industry 4.0", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Industry 4.0 denotes the fourth industrial revolution, characterised by the integration of cyber-physical systems, industrial Internet of Things, cloud computing, advanced robotics, and artificial intelligence into manufacturing and supply chain operations to create smart, autonomous, and highly flexible production environments. Originating from a German government initiative, it represents a strategic framework for digitalising industrial production through real-time data exchange, machine intelligence, and networked automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:industry-4-0", "labels": [ "Industry 4.0" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "industry-cloud-platforms", "title": "Industry Cloud Platforms", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Industry Cloud Platforms (ICPs) are collections of integrated cloud IT assets \u2014 including applications, data services, AI pipelines, and compliance tooling \u2014 tailored to the specific workflows, regulatory requirements, and data models of a vertical industry such as healthcare, financial services, manufacturing, or retail. They combine horizontal cloud infrastructure with deep domain knowledge, enabling organisations to run both vertical and cross-functional business applications on a unified platform while maintaining appropriate security controls. Gartner identified ICPs as a top strategic technology trend for 2024, driven by the need to reduce IT sprawl and accelerate industry-specific digital transformation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:industry-cloud-platforms", "labels": [ "Industry Cloud Platforms" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "industry-consortium", "title": "Industry Consortium", "domain": "ai", "domain_name": "Ai", "definition": "An Industry Consortium is a collaborative alliance of multiple independent organisations within the same sector that pool resources, expertise, and influence to pursue shared objectives such as standards development, research, or market promotion. Consortia typically operate under a formal governance charter and may produce specifications, reference implementations, or advocacy positions. They differ from regulatory bodies in that membership is voluntary and outputs are generally non-binding unless adopted by standards organisations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:industry-consortium", "labels": [ "Industry Consortium" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "inertia-tensor", "title": "Inertia Tensor", "domain": "robotics", "domain_name": "Robotics", "definition": "The inertia tensor is a 3x3 symmetric matrix that characterises how a rigid body's mass is distributed about a reference point, relating the body's angular velocity to its angular momentum. Its diagonal entries are the moments of inertia about the coordinate axes and its off-diagonal entries are the products of inertia. It is a foundational quantity in rigid-body dynamics, enabling computation of rotational acceleration under applied torques.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inertia-tensor", "labels": [ "Inertia Tensor" ], "is_subclass_of": [ "Robot Dynamics" ], "wikilinks": [] }, { "id": "inertia", "title": "Inertia", "domain": "robotics", "domain_name": "Robotics", "definition": "The physical property of a body that resists changes in its state of motion, whether translational or rotational. In robotics and actuation systems, inertia determines how much force or torque is required to accelerate or decelerate a component, and must be modelled accurately for stable, precise control.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:inertia", "labels": [ "Inertia" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "inertial-measurement-unit", "title": "inertial measurement unit", "domain": "robotics", "domain_name": "Robotics", "definition": "An Inertial Measurement Unit (IMU) is an electronic device that integrates tri-axial accelerometers, tri-axial gyroscopes, and optionally tri-axial magnetometers to measure a rigid body's specific force, angular rate, and magnetic field, enabling computation of linear acceleration, orientation, and heading relative to an inertial reference frame. MEMS-based IMUs dominate consumer and robotics applications due to low cost, small form factor, and adequate noise performance, while tactical, navigation, and strategic grades using ring-laser or fibre-optic gyroscopes serve aerospace, defence, and surveying applications demanding higher accuracy and lower drift. The raw sensor data are fused through algorithms such as extended Kalman filters, complementary filters, or pre-integration on manifolds to produce robust pose and velocity estimates for autonomous systems. IMUs are foundational to visual-inertial odometry, SLAM pipelines, flight control, pedestrian dead reckoning, and augmented reality tracking.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:inertial-measurement-unit", "labels": [ "Inertial Measurement Unit" ], "is_subclass_of": [ "Robot Sensor" ], "wikilinks": [] }, { "id": "inference-algorithm", "title": "Inference Algorithm", "domain": "ai", "domain_name": "Ai", "definition": "An inference algorithm is a computational procedure for deriving conclusions about unobserved quantities from a model and observed data, typically by computing or approximating posterior distributions in probabilistic models. Common families include exact methods (variable elimination, belief propagation), sampling methods (Markov chain Monte Carlo, importance sampling), and variational approximations. It is the engine that turns a model specification into actionable estimates or predictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inference-algorithm", "labels": [ "Inference Algorithm" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "inference-compute", "title": "Inference Compute", "domain": "ai", "domain_name": "Ai", "definition": "Inference compute is the computational capacity consumed when a trained machine-learning model generates outputs from inputs, as distinct from the compute used during training. For large language models it scales with model size, context length, and the number of generated tokens, and increasingly with test-time reasoning techniques that spend more compute per query to improve answers. It is a primary cost and latency driver for deployed AI systems and agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inference-compute", "labels": [ "Inference Compute", "Inference-Time Compute" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "inference-cost-efficiency", "title": "Inference Cost Efficiency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The optimization of computational resources and financial expenditure required to execute AI model predictions, often measured by cost per token or latency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:inference-cost-efficiency", "labels": [ "Inference Cost Efficiency" ], "is_subclass_of": [ "Inference" ], "wikilinks": [] }, { "id": "inference-economics", "title": "Inference Economics", "domain": "economics", "domain_name": "Economics", "definition": "The study of costs, margins, and pricing strategies associated with running AI model inference, including token pricing and provider profitability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:inference-economics", "labels": [ "Inference Economics" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "inference-engine", "title": "Inference Engine", "domain": "ai", "domain_name": "Ai", "definition": "An Inference Engine is a specialised software runtime optimised for executing trained machine learning models in production environments, transforming input data into predictions, classifications, embeddings, or generated content with primary objectives of minimising latency, maximising throughput, and efficiently utilising hardware accelerators such as GPUs, TPUs, and NPUs. Inference engines apply techniques including operator fusion, kernel auto-tuning, mixed-precision quantisation, and memory layout optimisation to close the performance gap between a training-time model representation and optimal hardware utilisation. They typically accept models in portable interchange formats such as ONNX or TensorRT engine plans, decoupling model architecture from the serving runtime, and are deployed within MLOps pipelines to serve AI applications at production scale. Modern inference engines address both traditional deep learning models (CNNs, transformers for classification and detection) and large language model serving, where techniques such as continuous batching, paged KV-cache management, and speculative decoding are critical to economic viability.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:inference-engine", "labels": [ "Inference Engine" ], "is_subclass_of": [ "Inference" ], "wikilinks": [] }, { "id": "inference-hardware", "title": "Inference Hardware", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Specialized processors, accelerators, and system-on-chip designs optimized for running trained machine learning models in production, prioritizing low latency, energy efficiency, high throughput, and cost-effectiveness across data centers, edge devices, and embedded systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:inference-hardware", "labels": [ "Inference Hardware", "GPU Inference Hardware" ], "is_subclass_of": [ "AI Hardware" ], "wikilinks": [ "Real-Time AI", "AI Hardware" ] }, { "id": "inference-infrastructure", "title": "Inference Infrastructure", "domain": "ai", "domain_name": "Ai", "definition": "Inference infrastructure is the stack of hardware, serving software, and orchestration used to deploy machine-learning models for low-latency, high-throughput prediction in production. It encompasses accelerator fleets, model servers, autoscaling, load balancing, batching engines, and caching layers that route requests and manage GPU memory. It is what makes real-time AI services such as search and chat economically and operationally viable at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inference-infrastructure", "labels": [ "Inference Infrastructure", "AI Inference Infrastructure", "Cloud Inference Infrastructure" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "inference-layer", "title": "Inference Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Inference Layer is the stratum that executes trained models to produce predictions from new inputs. In the canonical stack it sits above the Model Layer and below the Middleware Layer, turning static artefacts into a live serving capability. It contains serving runtimes, batching and caching logic, and the request handling that delivers predictions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:inference-layer", "labels": [ "Inference Layer" ], "is_subclass_of": [ "AI Model Architecture", "owl:Thing" ], "wikilinks": [ "Model Layer", "Middleware Layer", "Application Layer", "Model Serving", "Quantisation", "owl:Thing" ] }, { "id": "inference-optimisation", "title": "Inference Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Inference Optimisation encompasses techniques and processes for reducing the computational cost, latency, and memory footprint of deploying trained machine learning models at runtime. Methods include quantisation, pruning, knowledge distillation, and hardware-specific kernel fusion. The goal is to make model inference faster and more efficient without significantly degrading predictive accuracy.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:inference-optimisation", "labels": [ "Inference Optimisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "inference-runtime", "title": "Inference Runtime", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An inference runtime is the software layer that loads a trained machine-learning model and executes its forward pass to produce predictions on new inputs. It schedules computation across CPUs, GPUs or accelerators, applies graph optimisations such as operator fusion and quantisation, and manages memory, batching and concurrency for low-latency serving. Inference runtimes are the execution engine beneath model-serving infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inference-runtime", "labels": [ "Inference Runtime" ], "is_subclass_of": [ "Artificial Intelligence", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "inference-serving", "title": "Inference Serving", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Inference serving is the end-to-end runtime discipline of deploying trained machine learning models as production services that respond to prediction requests at scale, with controlled latency and high throughput. It encompasses the inference engine, request routing, dynamic batching, accelerator scheduling, autoscaling, and observability infrastructure needed to meet service-level objectives. Unlike model training, inference serving must optimise for request-level latency, cost-per-query, and concurrent user load simultaneously. Modern inference serving platforms add specialised techniques such as continuous batching, quantisation, and KV-cache management to sustain high GPU utilisation across large language models and other deep neural network workloads.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:inference-serving", "labels": [ "Inference Serving", "AI Inference Serving" ], "is_subclass_of": [ "Model Serving", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [ "AI Model Inference Engine", "Model Serving", "Latency", "GPU" ] }, { "id": "inference-speed", "title": "Inference Speed", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The rate at which an AI model generates output tokens, typically measured in tokens per second, serving as a critical performance metric for user experience and latency-sensitive applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:inference-speed", "labels": [ "Inference Speed" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [] }, { "id": "inference", "title": "Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Inference is the process of applying a trained AI model to new, unseen data to produce predictions, classifications, or generated outputs. It is distinct from training in that model parameters are fixed; the computational objective is throughput, latency, and memory efficiency. Inference is the primary execution path in production deployments and is governed by ISO/IEC 22989:2022 clause 3.3.4.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:inference", "labels": [ "Inference", "Activity Inference", "Deep Learning Inference", "Hardware-Aware Inference", "Inference at Scale", "Knowledge Inference", "LLM Inference", "Local Inference", "Machine Learning Inference", "Single-Step Inference", "Streaming Inference" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "infiniband", "title": "Infiniband", "domain": "ai", "domain_name": "Ai", "definition": "InfiniBand is a high-throughput, low-latency switched-fabric interconnect standard used to connect servers, storage, and accelerators in high-performance computing and large-scale AI training clusters. It provides remote direct memory access (RDMA) that bypasses the operating system kernel, enabling near-wire-speed data movement between nodes. InfiniBand is widely deployed as the backbone fabric for GPU clusters where collective communication bandwidth determines training scalability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:infiniband", "labels": [ "Infiniband", "InfiniBand" ], "is_subclass_of": [ "Artificial Intelligence", "Interconnect" ], "wikilinks": [] }, { "id": "inflation-control", "title": "Inflation Control", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Inflation Control encompasses the mechanisms, policies, and economic design patterns used within virtual and metaverse economies to regulate the creation, circulation, and destruction of virtual currency and digital assets. These controls prevent hyperinflation and value collapse by balancing supply-side minting with demand-side sinks. Effective inflation control underpins the long-term viability of play-to-earn ecosystems, NFT markets, and decentralised finance platforms built on virtual worlds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:inflation-control", "labels": [ "Inflation Control" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Game Economics", "Virtual Economics", "Metaverse", "MetaverseDomain" ] }, { "id": "inflation-hedge", "title": "Inflation Hedge", "domain": "finance", "domain_name": "Finance", "definition": "An inflation hedge is an asset expected to retain or increase its value as the general price level rises, protecting purchasing power. Commonly cited examples include gold, real estate and index-linked bonds.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:inflation-hedge", "labels": [ "Inflation Hedge" ], "is_subclass_of": [ "Store of Value" ], "wikilinks": [ "Inflation", "Gold", "Store of Value" ] }, { "id": "inflation-hedging", "title": "Inflation Hedging", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Inflation hedging is an investment strategy that protects the real value of capital against the erosion of purchasing power caused by rising consumer prices. It allocates wealth to assets whose nominal value tends to rise with or outpace inflation, such as commodities, real estate, inflation-linked bonds, and scarce stores of value. Both gold and Bitcoin are frequently cited as inflation hedges because their supply is constrained relative to fiat currency issuance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inflation-hedging", "labels": [ "Inflation Hedging" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "inflation-targeting", "title": "Inflation Targeting", "domain": "finance", "domain_name": "Finance", "definition": "Inflation targeting is a monetary-policy framework in which a central bank publicly commits to achieving a specified rate of inflation over the medium term and adjusts its policy instruments to meet that target. By anchoring expectations to a clear, numerical goal, it aims to deliver price stability while preserving transparency and accountability. The framework links interest-rate decisions to forecasts of how inflation will deviate from the announced target.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inflation-targeting", "labels": [ "Inflation Targeting" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "inflation", "title": "Inflation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The scheduled or dynamic increase in token supply within a blockchain network, used to fund validator rewards, incentivise participation, and manage monetary policy. Inflation rate parameters are typically encoded in the protocol and may be adjusted via governance, directly affecting token holder purchasing power and long-term economic security.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:inflation", "labels": [ "Inflation" ], "is_subclass_of": [ "Economic Mechanism", "Blockchain Entity", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "inflationary-token", "title": "Inflationary Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Inflationary Token is a blockchain token whose total circulating supply increases over time through a programmatic emission schedule, typically as a mechanism to reward network participants\u2014validators, miners, stakers, or liquidity providers\u2014and incentivise ongoing network participation. Unlike fixed-supply or deflationary tokens, inflationary tokens accept dilution of existing holders as the cost of sustaining economic participation incentives. Inflation rates can be fixed (constant annual issuance), variable (decreasing block rewards as in Bitcoin's halving), or algorithmically adjusted in response to network conditions such as staking participation rates. Design of the emission schedule is a critical tokenomics decision balancing security, participation, and value preservation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:inflationary-token", "labels": [ "Inflationary Token" ], "is_subclass_of": [ "Token and Asset", "Blockchain Entity", "Economic Mechanism", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "influence-maximisation", "title": "Influence Maximisation", "domain": "data", "domain_name": "Data", "definition": "Influence maximisation is the combinatorial optimisation problem of selecting a small set of seed nodes in a network so as to maximise the expected spread of information, adoption or behaviour under a diffusion model such as independent cascade or linear threshold. It is NP-hard in general, so practical algorithms rely on submodularity-based greedy approximation or scalable heuristics. It is applied in social network analysis for viral marketing, epidemic containment planning and identifying key influencers.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:influence-maximisation", "labels": [ "Influence Maximisation" ], "is_subclass_of": [ "Network Analysis" ], "wikilinks": [] }, { "id": "information-architecture", "title": "Information Architecture", "domain": "data", "domain_name": "Data", "definition": "Information Architecture (IA) is the discipline of structuring, organising, labelling, and navigating shared information environments so that users and automated systems can find, understand, and act on content efficiently. It encompasses the design of taxonomies, controlled vocabularies, navigation schemas, metadata frameworks, and search systems that govern how information is classified and retrieved across digital products, knowledge bases, and interconnected platforms. IA bridges human cognitive models and machine-interpretable representations, drawing on ontologies, thesauri, and faceted classification to ensure content remains discoverable, reusable, and interoperable. In distributed and spatial contexts it extends to the semantic namespacing of assets, entities, and services that underpin cross-platform data exchange.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:information-architecture", "labels": [ "Information Architecture" ], "is_subclass_of": [ "Knowledge Organization System" ], "wikilinks": [ "owl:Thing" ] }, { "id": "information-asymmetry", "title": "Information Asymmetry", "domain": "finance", "domain_name": "Finance", "definition": "Information asymmetry is a condition in which one party to a transaction or interaction possesses more or better information than another, distorting decisions and market outcomes. It gives rise to adverse selection and moral hazard, and motivates institutions such as signalling, screening, disclosure rules, and reputation systems. It is a foundational concept in microeconomics and is increasingly relevant to AI systems whose training data and outputs can encode hidden biases.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:information-asymmetry", "labels": [ "Information Asymmetry" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "information-extraction", "title": "Information Extraction", "domain": "ai", "domain_name": "Ai", "definition": "Information Extraction (IE) is the automated process of identifying and structuring specific pieces of information from unstructured or semi-structured natural language text, producing machine-readable records such as typed entities, relational tuples, and event frames. Core subtasks include Named Entity Recognition, relation extraction, event extraction, coreference resolution, and slot filling. IE pipelines underpin knowledge graph population, question answering, and downstream analytics by converting free-form prose into structured representations consumable by databases and reasoning systems. Modern IE systems leverage pretrained transformer language models fine-tuned with supervised or few-shot learning, substantially improving generalisation across domains compared to earlier rule-based and feature-engineered approaches.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:information-extraction", "labels": [ "Information Extraction", "Data Extraction" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Natural Language Processing", "Computer Science" ] }, { "id": "information-governance", "title": "Information Governance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Information Governance is the framework of policies, processes, roles, and standards that organisations use to manage the availability, usability, integrity, and security of their information assets throughout the data lifecycle. It encompasses data stewardship, regulatory compliance, records management, and accountability structures. Effective information governance ensures that data is trustworthy, properly classified, and handled in accordance with legal and business requirements.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:information-governance", "labels": [ "Information Governance" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "information-integrity", "title": "Information Integrity", "domain": "governance", "domain_name": "Governance", "definition": "Information integrity is the property and governance objective of ensuring that information remains accurate, complete, authentic and resistant to unauthorised or deceptive alteration across its lifecycle. In the digital-content context it spans provenance, authenticity verification and resilience against misinformation and disinformation. It draws on technical measures such as digital watermarking and cryptographic provenance alongside institutional trust-and-safety practices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:information-integrity", "labels": [ "Information Integrity" ], "is_subclass_of": [ "Content Provenance" ], "wikilinks": [] }, { "id": "information-management", "title": "Information Management", "domain": "data", "domain_name": "Data", "definition": "The organisational discipline of acquiring, organising, storing, curating, securing, distributing, and disposing of information across its lifecycle so that the right information reaches the right people, in the right form, at the right time. It spans structured data and unstructured content alike, sitting between data management (which stewards raw data assets) and knowledge management (which cultivates human understanding), and is governed by policies for quality, retention, access, and compliance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:information-management", "labels": [ "Information Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Data Management", "Knowledge Management", "Records Management", "Information Governance" ] }, { "id": "information-protection", "title": "Information Protection", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The set of policies, cryptographic mechanisms, and technical controls applied in blockchain and distributed systems to ensure the confidentiality, integrity, and availability of on-chain and off-chain data. Information protection encompasses encryption, access control, privacy-preserving computation, and secure key management.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:information-protection", "labels": [ "Information Protection" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "owl:Thing" ] }, { "id": "information-retrieval", "title": "Information Retrieval", "domain": "ai", "domain_name": "Ai", "definition": "Information Retrieval (IR) is the science and engineering discipline concerned with representing, storing, organising, and providing access to items of information \u2014 typically documents, passages, or structured records \u2014 so that a user's information need, expressed as a query, can be satisfied efficiently and accurately. Classical IR models such as the Boolean model, the vector space model (TF-IDF), and probabilistic models (BM25) underpin search engines and document ranking systems; neural IR extends these with dense vector representations derived from transformer language models, enabling semantic matching that generalises beyond exact term overlap. Modern IR encompasses sparse retrieval, dense retrieval, re-ranking, and hybrid architectures, and is the core subsystem enabling Retrieval-Augmented Generation (RAG), question answering, and knowledge-graph-grounded AI systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:information-retrieval", "labels": [ "Information Retrieval", "Accurate Retrieval", "Biomedical Information Retrieval", "Information Retrieval Metric" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "owl:Thing" ] }, { "id": "information-security", "title": "Information Security", "domain": "security", "domain_name": "Security", "definition": "Information Security addresses the protection of data, systems, models, and AI infrastructure from unauthorised access, adversarial attacks, privacy breaches, and malicious exploitation. Security measures encompass differential privacy, federated learning, robust training methods, secure multi-party computation, and encryption to ensure confidentiality, integrity, and availability of systems and data as critical infrastructure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:information-security", "labels": [ "Information Security" ], "is_subclass_of": [ "Security and Identity", "AI Governance and Ethics" ], "wikilinks": [ "Adversarial Machine Learning", "Secure Computation", "Differential Privacy", "Federated Learning" ] }, { "id": "information-sharing", "title": "Information Sharing", "domain": "governance", "domain_name": "Governance", "definition": "Information sharing is the coordinated exchange of data, findings or risk signals between organisations or stakeholders, used in AI governance to spread safety-relevant knowledge across otherwise competing or independent actors. Bodies such as the Frontier Model Forum and regulatory coordination groups rely on structured information-sharing arrangements to track emerging risks and align practice. It is a form of data sharing applied specifically to governance and oversight contexts.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:information-sharing", "labels": [ "Information Sharing" ], "is_subclass_of": [ "Data Sharing" ], "wikilinks": [] }, { "id": "information-theoretic-security", "title": "Information Theoretic Security", "domain": "security", "domain_name": "Security", "definition": "Information theoretic security is a class of cryptographic guarantee in which a scheme is secure against an adversary with unlimited computational power, because the ciphertext or protocol transcript carries no statistical information about the secret. Unlike computational security, which rests on the assumed hardness of mathematical problems, these guarantees follow from the structure of information itself and remain valid even against future advances such as quantum computers. Canonical examples include the one-time pad and threshold secret sharing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:information-theoretic-security", "labels": [ "Information Theoretic Security", "Information-Theoretic Security" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "information-theory", "title": "Information Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Information Theory is the mathematical study of the quantification, storage and communication of information, founded by Claude Shannon in 1948. It introduces entropy as a measure of uncertainty in a random source and defines channel capacity, the maximum rate at which information can be transmitted reliably over a noisy channel. The theory underpins data compression, error-correcting codes and modern digital communication. It also connects to statistics, cryptography and machine learning through measures such as mutual information and Kullback-Leibler divergence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:information-theory", "labels": [ "Information Theory", "Shannon Information Theory" ], "is_subclass_of": [ "Applied Mathematics", "owl:Thing" ], "wikilinks": [ "Entropy", "Mutual Information", "Channel Capacity", "Probability Theory", "Data Compression", "Error Correcting Code", "Cryptography", "Machine Learning", "Statistics", "owl:Thing", "Shannon 1948, A Mathematical Theory of Communication" ] }, { "id": "informed-consent", "title": "Informed Consent", "domain": "governance", "domain_name": "Governance", "definition": "Informed consent is a legal and ethical doctrine requiring that an individual voluntarily agrees to a procedure, treatment, data use, or research participation after receiving and comprehending all material information about its nature, risks, benefits, and alternatives. It rests on four elements: disclosure of relevant information, comprehension by the consenting party, voluntariness free from coercion, and capacity to make the decision. Codified in post-Nuremberg biomedical ethics (Belmont Report, Declaration of Helsinki), GDPR data protection law, and clinical trials regulations, informed consent is the foundational mechanism protecting individual autonomy across medical, research, and digital data contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:informed-consent", "labels": [ "Informed Consent", "Consent", "Informed Consent Mechanism" ], "is_subclass_of": [ "Consent Management" ], "wikilinks": [] }, { "id": "informed-search", "title": "Informed Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A class of search algorithms in artificial intelligence that use domain-specific heuristic knowledge to guide exploration of the search space, reducing computational cost compared to uninformed search. Canonical examples include A* and greedy best-first search, which evaluate states using an estimate of the remaining cost to a goal.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:informed-search", "labels": [ "Informed Search" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "owl:Thing" ] }, { "id": "informed-decision-making", "title": "Informed decision-making", "domain": "ai", "domain_name": "Ai", "definition": "Informed decision-making is a cognitive and organisational process in which choices are grounded in accurate, relevant, and timely evidence rather than intuition, habit, or incomplete information. It encompasses the collection, analysis, and interpretation of data, the surfacing of relevant context through knowledge management systems, and the mitigation of cognitive biases that distort judgement. In digital and AI-augmented environments, informed decision-making is increasingly supported by decision support tools, predictive analytics, and transparent reasoning systems that make the basis of recommendations auditable.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:informed-decision-making", "labels": [ "Informed decision-making", "Informed Decision Making" ], "is_subclass_of": [ "Decision Support", "Decision Making" ], "wikilinks": [] }, { "id": "infrared-astronomy", "title": "Infrared Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:infrared-astronomy", "labels": [ "Infrared Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "infrared-camera", "title": "Infrared Camera", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Imaging devices that capture infrared light wavelengths for VR/AR tracking applications, enabling position tracking, eye tracking, hand gesture recognition, and movement detection in immersive environments through dedicated IR-sensitive sensors and algorithms.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:infrared-camera", "labels": [ "Infrared Camera" ], "is_subclass_of": [ "Interaction Technology", "Tracking Hardware" ], "wikilinks": [ "metaverse", "Motion Tracking", "Tracking Hardware" ] }, { "id": "infrared-illuminator", "title": "Infrared Illuminator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Light-emitting devices that project infrared wavelengths to enable VR/AR tracking sensors to function in low-light or dark environments, enhancing hand tracking accuracy, controller detection, and spatial awareness for immersive experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:infrared-illuminator", "labels": [ "Infrared Illuminator", "Infrared Illumination" ], "is_subclass_of": [ "Interaction Technology", "Tracking Hardware" ], "wikilinks": [ "Dark Environment VR", "metaverse", "Tracking Hardware" ] }, { "id": "infrared-led-illuminator", "title": "Infrared Led Illuminator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "LED-based infrared light sources integrated into or used alongside VR/AR headsets to provide illumination for tracking systems, featuring compact form factors, low power consumption, and wavelengths optimised for camera sensor detection.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:infrared-led-illuminator", "labels": [ "Infrared Led Illuminator" ], "is_subclass_of": [ "Interaction Technology", "Infrared Illuminator" ], "wikilinks": [ "Hand Tracking Enhancement", "Infrared Illuminator", "metaverse" ] }, { "id": "infrared-light-source", "title": "Infrared Light Source", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Emission devices providing infrared wavelength light for VR/AR tracking applications, encompassing LEDs, illuminators, and integrated headset components that enable eye tracking, position detection, and hand gesture recognition in extended reality systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:infrared-light-source", "labels": [ "Infrared Light Source", "Infrared Emitter", "Infrared Laser", "Infrared Lighting" ], "is_subclass_of": [ "Interaction Technology", "Tracking Hardware" ], "wikilinks": [ "Foveated Rendering", "metaverse", "Tracking Hardware" ] }, { "id": "infrared-sensor", "title": "InfraredSensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An optoelectronic sensor device that detects electromagnetic radiation in the infrared spectrum (wavelengths approximately 700 nanometers to 1 millimeter) to measure heat emission, enable proximity detection, support autonomous navigation, or facilitate object recognition in robotic systems, empl...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:infrared-sensor", "labels": [ "InfraredSensor", "Infrared Sensor", "Thermal Infrared Sensing" ], "is_subclass_of": [ "Perception and Sensing", "ExteroceptiveSensor", "OptoelectronicDevice", "ProximitySensor", "ThermalImager" ], "wikilinks": [ "AnalogDigitalConverter", "AutonomousNavigationDomain", "Calibration", "GestureRecognition", "IEC 60825 Safety of Laser Products", "IEEE Std 1855 Fuzzy Markup Language", "International Society for Optics and Photonics (SPIE)", "IREmitter", "IRReceiver", "ISO 8373 Robotics Vocabulary", "LensAssembly", "LineFollowing", "NavigationControl", "OpticalAlignment", "OpticalFilter", "OptoelectronicDevice", "Robot Operating System (ROS) Sensor Standards", "SensorTechnologyDomain", "SignalAmplifier", "TemperatureCompensation" ] }, { "id": "infrastructure-architecture", "title": "Infrastructure Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The foundational technical framework supporting metaverse platforms, comprising cloud computing, edge networks, distributed systems, and computing power networks that enable scalable, low-latency delivery of immersive virtual experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:technology-infrastructure-domain-architecture", "labels": [ "Infrastructure Architecture", "InfrastructureArchitecture" ], "is_subclass_of": [ "Computing and Cloud", "Technical Architecture" ], "wikilinks": [ "metaverse", "Metaverse Platform", "Technical Architecture" ] }, { "id": "infrastructure-as-a-service", "title": "Infrastructure As A Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Infrastructure as a Service (IaaS) is a cloud computing service model in which a provider delivers virtualised computing resources \u2014 including virtual machines, storage, networking, and bare-metal servers \u2014 over the internet on a pay-per-use basis, while the customer manages the operating system, middleware, and applications. IaaS abstracts away physical hardware procurement and data centre operations, enabling organisations to provision and de-provision compute capacity elastically. Major IaaS providers include Amazon Web Services (EC2/S3), Microsoft Azure (Virtual Machines), and Google Cloud Platform (Compute Engine).", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:infrastructure-as-a-service", "labels": [ "Infrastructure As A Service", "Infrastructure as a Service", "Infrastructure-as-a-Service" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "infrastructure-component", "title": "Infrastructure Component", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A discrete, independently deployable unit of technical infrastructure that fulfils a specific function within a larger system, such as a compute node, network layer, storage service, or security module, and that can be composed with other components to deliver end-to-end platform capabilities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-infrastructure-domain-component", "labels": [ "Infrastructure Component", "InfrastructureComponent" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "owl:Thing" ] }, { "id": "infrastructure-inspection", "title": "Infrastructure Inspection", "domain": "robotics", "domain_name": "Robotics", "definition": "Infrastructure inspection is the robotic application of surveying and assessing physical assets such as power lines, pipelines, bridges, wind turbines, and buildings to detect defects, corrosion, and structural risk. Robots equipped with cameras, thermal and LiDAR sensors capture data in environments that are hazardous, remote, or costly for human crews. It improves safety and inspection frequency while generating quantitative condition records for predictive maintenance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:technology-infrastructure-domain-inspection", "labels": [ "Infrastructure Inspection" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "infrastructure-layer", "title": "Infrastructure Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Foundational base layer providing computing, storage, and network capabilities that enable metaverse applications and services to operate at scale.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:technology-infrastructure-domain-layer", "labels": [ "Infrastructure Layer", "InfrastructureLayer" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "5G Network", "CDN", "Data Centers", "High Availability", "Low Latency", "MSF Taxonomy 2025", "Power Systems", "Scalability", "Cloud Computing", "Distributed Computing", "Edge Computing", "Hardware Abstraction Layer (HAL)", "InfrastructureDomain", "Network Infrastructure", "Physical Hardware", "Physical Layer", "Spatial Computing Layer", "Storage Layer" ] }, { "id": "infrastructure-as-code", "title": "Infrastructure as Code", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Infrastructure as Code (IaC) is the practice of defining and provisioning computing infrastructure through machine-readable definition files rather than manual configuration. Declarative or imperative specifications describe the desired state of servers, networks, and services, which tooling then realises idempotently and reproducibly. Treating infrastructure like software allows version control, peer review, automated testing, and consistent deployment across environments, eliminating configuration drift.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:infrastructure-as-code", "labels": [ "Infrastructure as Code", "Infrastructure-as-Code" ], "is_subclass_of": [ "Automation" ], "wikilinks": [] }, { "id": "infrastructure", "title": "Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Infrastructure domain encompasses the foundational computing, networking, storage, data-management, security, and software-engineering systems that underpin every other technology domain. It is the substrate on which artificial-intelligence, spatial-computing, blockchain, robotics, and distributed-collaboration applications operate, spanning cloud and compute platforms, communications networks, data and identity management, legal/regulatory frameworks, and software engineering practice.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:infrastructure", "labels": [ "Infrastructure", "Development Infrastructure", "Infrastructure Layer", "Large-Scale Data Infrastructure", "Server Infrastructure" ], "is_subclass_of": [ "Thing" ], "wikilinks": [] }, { "id": "initial-coin-offering", "title": "Initial Coin Offering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An initial coin offering (ICO) is a blockchain-based fundraising mechanism in which a project issues and sells newly created cryptographic tokens to early backers, typically in exchange for established cryptocurrencies such as Bitcoin or Ether. Tokens are usually distributed via smart contracts and may grant utility within a planned protocol, governance rights, or speculative value. Pioneered around 2013\u20132017, ICOs enabled permissionless global capital formation but attracted intense regulatory scrutiny over investor protection and securities classification, prompting more structured successors such as security token offerings and initial exchange offerings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:initial-coin-offering", "labels": [ "Initial Coin Offering" ], "is_subclass_of": [ "Cryptocurrency Token" ], "wikilinks": [] }, { "id": "initial-dex-offering", "title": "Initial Dex Offering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An initial DEX offering (IDO) is a token fundraising and distribution model in which a new token is launched directly on a decentralised exchange, with liquidity provided into an automated market maker pool so trading begins immediately. Unlike centralised or regulated offerings, an IDO is permissionless and settled on-chain through smart contracts, often using launchpads that manage allocation and anti-bot measures. It contrasts with the initial coin offering and the regulated initial public offering by emphasising immediate liquidity and decentralised access.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:initial-dex-offering", "labels": [ "Initial Dex Offering", "Initial DEX Offering" ], "is_subclass_of": [ "Token Sale" ], "wikilinks": [] }, { "id": "initial-public-offering", "title": "Initial Public Offering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An initial public offering (IPO) is the process by which a privately held company first sells its shares to the public on a regulated stock exchange, transitioning into a publicly traded entity. It is intermediated by investment banks that underwrite the issue, set a price range, and allocate shares to institutional and retail investors under securities regulation. In blockchain discourse the IPO serves as the regulated, equity-based reference point against which token-based fundraising mechanisms are contrasted.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:initial-public-offering", "labels": [ "Initial Public Offering" ], "is_subclass_of": [ "Capital Markets" ], "wikilinks": [] }, { "id": "initialization-vector", "title": "Initialization Vector", "domain": "security", "domain_name": "Security", "definition": "An initialization vector (IV) is a fixed-size input to a symmetric encryption mode of operation that randomises the encryption process so that identical plaintexts produce different ciphertexts under the same key. By introducing fresh, unpredictable variation for each message, it prevents an attacker from detecting repetition and defeats certain chosen-plaintext attacks. Depending on the mode, an IV must be unpredictable, unique, or both, and is typically transmitted in the clear alongside the ciphertext because its secrecy is not required, only its non-repetition. Misusing an IV, such as reusing it, can catastrophically undermine the security of an otherwise sound cipher.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:initialization-vector", "labels": [ "Initialization Vector" ], "is_subclass_of": [ "Symmetric Encryption" ], "wikilinks": [] }, { "id": "inner-product", "title": "Inner Product", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An inner product is a bilinear operation on a vector space that combines two vectors to produce a scalar, generalising the geometric notion of a dot product to abstract vector spaces. It defines notions of length and angle, and its normalised form underlies cosine similarity, a standard measure of semantic closeness between embeddings. Inner products are computed extensively in semantic search systems to rank candidate vectors against a query vector.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:inner-product", "labels": [ "Inner Product" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "inner-source", "title": "Inner Source", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Inner source is the adoption of open-source development practices, tools and culture within the boundaries of a single organisation, allowing teams to discover, use and contribute to each other's code as if it were open source internally. It promotes shared repositories, transparent collaboration, peer review and meritocratic contribution while keeping the code private to the company. The approach aims to reduce duplication, improve quality and spread knowledge across organisational silos.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:inner-source", "labels": [ "Inner Source" ], "is_subclass_of": [ "Distributed Collaboration", "Open Source" ], "wikilinks": [] }, { "id": "innovation-diffusion", "title": "Innovation Diffusion", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Innovation diffusion is the process by which a new technology, practice or idea spreads through a population over time. It is studied with adoption curves that describe how successive groups take up an innovation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:innovation-diffusion", "labels": [ "Innovation Diffusion", "Technological Diffusion" ], "is_subclass_of": [ "Technology Adoption" ], "wikilinks": [ "Technology Adoption", "Agent-Based Modelling", "Economics", "Simulation" ] }, { "id": "innovation-ecosystems", "title": "Innovation Ecosystems", "domain": "governance", "domain_name": "Governance", "definition": "Interdependent networks of firms, universities, research institutions, investors, and governments that jointly create, diffuse, and commercialise new technologies, where co-located talent pools, venture funding, shared infrastructure, and knowledge spillovers reinforce one another so that the productive capacity of the whole system exceeds the sum of its individual actors.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:innovation-ecosystems", "labels": [ "Innovation Ecosystems" ], "is_subclass_of": [ "Innovation" ], "wikilinks": [ "Innovation", "Venture Capital", "AI Talent", "Technology Transfer" ] }, { "id": "innovation", "title": "Innovation", "domain": "ai", "domain_name": "Ai", "definition": "Innovation is the process of creating and successfully introducing new or significantly improved products, services, processes or business models that deliver value. It spans the journey from invention and creative ideation through development, adoption and diffusion across markets and organisations. Innovation is a primary driver of economic growth, competitiveness and societal change, and is closely linked to research, knowledge transfer and entrepreneurship.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:innovation", "labels": [ "Innovation" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [] }, { "id": "inpainting", "title": "Inpainting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Inpainting is the computational task of plausibly reconstructing missing, occluded, masked or unwanted regions of an image or video so the completed output appears coherent with the surrounding (known) context, originally formalised in the digital domain by Bertalmio, Sapiro, Caselles and Ballest...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:inpainting", "labels": [ "Inpainting", "Image Inpainting", "Inpainting Model" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "Image Editing", "Conditional Image Synthesis", "Image Restoration", "Computer Vision Task", "Generative AI" ], "wikilinks": [ "Adobe Firefly", "AlgorithmLayer", "Apple Intelligence", "Barnes Shechtman Finkelstein Goldman 2009 PatchMatch", "Bertalmio Bertozzi Sapiro 2001 Navier-Stokes Inpainting", "Bertalmio Sapiro Caselles Ballester 2000 Image Inpainting SIGGRAPH", "Black Forest Labs 2024 Flux.1 Fill Tools", "C2PA", "Chan Shen 2001 Mathematical Models for Local Non-Texture Inpaintings", "Classifier-Free Guidance", "Clone Stamp Tool", "ComputerVisionDomain", "Conditional Generation", "Conditional Image Synthesis", "Conditioning Encoder", "Context Region", "ControlNet", "Convolutional Neural Networks", "Criminisi Perez Toyama 2004 Exemplar-Based Inpainting", "Cultural Heritage Restoration" ] }, { "id": "input", "title": "Input", "domain": "blockchain", "domain_name": "Blockchain", "definition": "In the UTXO blockchain model, an Input is a reference to a previous unspent transaction output that is being consumed to fund a new transaction. Each input includes a pointer to the referenced output, a cryptographic signature proving ownership, and an unlocking script that satisfies the output's locking conditions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:input", "labels": [ "Input" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "inscription", "title": "Inscription", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An inscription is arbitrary data, such as an image, text or document, written directly onto an individual satoshi on the Bitcoin blockchain so that the content is stored entirely on-chain. Enabled by the Taproot upgrade and the Ordinals numbering scheme, inscriptions embed their payload in the witness portion of a transaction, making each inscribed satoshi a verifiable, transferable digital artefact. They underpin Bitcoin-native non-fungible assets and digital collectibles without relying on external metadata storage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:inscription", "labels": [ "Inscription" ], "is_subclass_of": [ "Ordinals" ], "wikilinks": [] }, { "id": "inside-out-tracking", "title": "Inside Out Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Inside-out tracking is a positional-tracking method in which cameras and sensors mounted on a head-mounted display observe the surrounding environment to determine the device's own pose in space. It requires no external base stations, computing six-degrees-of-freedom position and orientation from features detected in the scene. The approach underpins standalone virtual and mixed reality headsets by enabling self-contained spatial tracking.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inside-out-tracking", "labels": [ "Inside Out Tracking", "Inside-Out Tracking" ], "is_subclass_of": [ "Positional Tracking" ], "wikilinks": [] }, { "id": "insider-trading", "title": "Insider Trading", "domain": "finance", "domain_name": "Finance", "definition": "The buying or selling of a publicly traded company's securities by individuals with access to material, non-public information about that company, in breach of a fiduciary duty or other relationship of trust. It undermines market integrity and investor confidence by giving informed parties an unfair advantage over ordinary participants, and is prohibited under securities law in most jurisdictions, with enforcement led by regulators such as the SEC and the FCA.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:insider-trading", "labels": [ "Insider Trading" ], "is_subclass_of": [ "Market Manipulation" ], "wikilinks": [ "Market Manipulation", "Market Integrity", "Investor Protection", "Securities Law" ] }, { "id": "inspection-robot", "title": "Inspection Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A specialised robotic system designed to autonomously examine physical structures, components, or manufactured products for defects, faults, or deviations from specification. Inspection robots combine computer vision, sensor fusion, and AI-driven anomaly detection to replace or augment manual quality-assurance processes in hazardous or high-throughput environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:inspection-robot", "labels": [ "Inspection Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "instance-segmentation", "title": "Instance Segmentation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A computer vision task that extends object detection by predicting precise pixel-level masks for each individual object instance, jointly performing detection and segmentation to delineate the exact boundaries of distinct objects. Architectures such as Mask R-CNN, YOLACT, and SOLOv2 enable fine-grained object localisation essential for robotics manipulation, autonomous driving, and medical image analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:instance-segmentation", "labels": [ "Instance Segmentation" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "MetaverseDomain", "Object Detection", "Panoptic Segmentation", "Semantic Segmentation" ] }, { "id": "instant-messaging", "title": "Instant Messaging", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Instant messaging is a form of real-time, text-based direct communication between two or more participants over a network. Messages are delivered with minimal latency, enabling synchronous or near-synchronous conversation without the formality of email. It serves as a foundational channel for informal coordination and rapid decision-making in distributed teams.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:instant-messaging", "labels": [ "Instant Messaging", "Chat Messaging" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "instant-payment", "title": "Instant Payment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Instant payment refers to a funds transfer that is settled and made available to the recipient within seconds, rather than the minutes, hours or days typical of traditional clearing systems. On blockchain networks this is commonly achieved through payment channel networks such as the Lightning Network, which route transactions off-chain to avoid base-layer confirmation latency, or through purpose-built low-latency settlement layers. Instant payment is a key usability requirement for point-of-sale and micropayment use cases, where users expect confirmation comparable to card payments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:instant-payment", "labels": [ "Instant Payment" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [] }, { "id": "instant-settlement", "title": "Instant Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Instant settlement is the near-immediate, final transfer of value between parties, eliminating the multi-day clearing and settlement delays of traditional banking and card networks. In blockchain and payment-channel systems it is achieved through cryptographic finality or off-chain channel updates that make funds usable within seconds. It reduces counterparty and credit risk, frees up working capital, and is a defining property of digital money.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:instant-settlement", "labels": [ "Instant Settlement", "Instant Bitcoin Payment", "Instant Payments", "Instant Settlements" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "instant-value-settlement", "title": "Instant value settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Instant value settlement is the process by which the transfer of economic value between counterparties \u2014 wher central-bank fiat currency, commercial bank money, tokenised deposits, stablecoins, securities, or programmable digital money \u2014 is completed in real time (sub-second to at most a few seco...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:instant-value-settlement", "labels": [ "Instant value settlement" ], "is_subclass_of": [ "DeFi and Economics", "Cryptocurrency", "Blockchain Network", "Distributed Ledger", "Smart Contract", "Financial Infrastructure" ], "wikilinks": [ "CryptographyDomain", "FinancialInfrastructureDomain", "PaymentsDomain", "RegulatoryDomain", "RegulatoryLayer", "AML KYC Compliance", "ApplicationLayer", "Atomic Swap", "Bitcoin As Money", "Bitcoin Mining", "Bitcoin Technical Overview", "BlockchainDomain", "Blockchain Interoperability", "Blockchain Network", "BTC Layer 3", "Cashu", "CBDC Frameworks", "CBDCs", "Consortium Blockchain", "Cross Border Compliance" ] }, { "id": "instant-id", "title": "InstantID", "domain": "ai", "domain_name": "Ai", "definition": "InstantID is a diffusion-model technique for identity-preserving image generation that synthesises new images of a specific person from a single reference photograph, without per-subject fine-tuning. It combines a face encoder with an IdentityNet adapter that injects facial identity and spatial landmarks into a text-to-image diffusion backbone. While powerful for personalised avatars and stylisation, its zero-shot fidelity also raises misuse concerns around impersonation and synthetic media.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:instant-id", "labels": [ "InstantID" ], "is_subclass_of": [ "Generative Model", "Diffusion Models" ], "wikilinks": [] }, { "id": "institutional-adoption", "title": "Institutional Adoption", "domain": "finance", "domain_name": "Finance", "definition": "Institutional adoption is the process by which regulated, large-scale organisations \u2014 including banks, asset managers, pension funds, sovereign wealth funds, and corporations \u2014 integrate a novel technology, protocol, or asset class into their core operations, risk frameworks, and balance sheets. It is characterised by formal governance procedures, compliance with applicable regulatory regimes, and deployment of significant capital under fiduciary obligations. Unlike retail or speculative uptake, institutional adoption requires custodial infrastructure, legal certainty, and integration with existing settlement, reporting, and audit systems. It typically follows a maturation arc: proof-of-concept pilots by innovation teams, limited production deployment, and finally full strategic integration with dedicated product lines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:institutional-adoption", "labels": [ "Institutional Adoption", "Institutional Crypto Adoption" ], "is_subclass_of": [ "Technology Adoption" ], "wikilinks": [ "Financial Regulation", "Traditional Finance", "Asset Tokenisation", "owl:Thing" ] }, { "id": "institutional-custody", "title": "Institutional Custody", "domain": "finance", "domain_name": "Finance", "definition": "Institutional custody is the regulated safekeeping of digital assets on behalf of organisations such as funds, exchanges, and corporations, combining cryptographic key-management infrastructure with legal, operational, and insurance controls that meet fiduciary standards. Providers use cold storage, multi-signature and multi-party-computation schemes, hardware security modules, and segregated accounts to protect client assets against theft, loss, and insider risk while supporting auditability and regulatory reporting. It is a precondition for large-scale institutional participation in crypto markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:institutional-custody", "labels": [ "Institutional Custody" ], "is_subclass_of": [ "Custody" ], "wikilinks": [] }, { "id": "institutional-design", "title": "Institutional Design", "domain": "governance", "domain_name": "Governance", "definition": "Institutional design is the deliberate construction of the rules, roles, and decision-making procedures that structure how a collective coordinates, allocates authority, and resolves disputes. Drawing on economics, political science, and mechanism design, it shapes incentives so that self-interested actors produce outcomes aligned with shared goals. It applies to states, firms, standards bodies, and on-chain organisations alike, where formal rules and enforcement determine legitimacy and durability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:institutional-design", "labels": [ "Institutional Design" ], "is_subclass_of": [ "Institutional Economics" ], "wikilinks": [] }, { "id": "institutional-digital-asset-custody", "title": "Institutional Digital Asset Custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Institutional digital asset custody is the regulated safekeeping of cryptocurrencies and tokenised assets on behalf of funds, banks, and corporations, with controls meeting fiduciary and compliance standards. It combines hardware security modules, multi-party computation or multi-signature key management, segregation of duties, insurance, and audited operational procedures. It is the trust layer that enables large pools of capital to hold crypto assets without bearing single-key loss or theft risk.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:institutional-digital-asset-custody", "labels": [ "Institutional Digital Asset Custody" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "institutional-economics", "title": "Institutional Economics", "domain": "governance", "domain_name": "Governance", "definition": "A school of economic thought that analyses how formal institutions (laws, regulations, contracts) and informal institutions (norms, customs, culture) shape economic behaviour, incentive structures, and aggregate outcomes. It departs from neoclassical assumptions of frictionless exchange by foregrounding transaction costs, bounded rationality, and the evolutionary path-dependence of rules. The field spans the original institutionalism of Veblen and Commons and the New Institutional Economics (NIE) of Coase, North, and Williamson, which provided micro-foundations for why institutions exist and how they change. Institutional economics informs applied work in development economics, regulatory design, organisational theory, and governance of digital platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:institutional-economics", "labels": [ "Institutional Economics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [ "Economics", "Property Rights", "Behavioural Economics", "Game Theory" ] }, { "id": "institutional-framework", "title": "Institutional Framework", "domain": "governance", "domain_name": "Governance", "definition": "An institutional framework is the set of formal rules, such as laws, regulations, and property rights, and informal norms and enforcement mechanisms that structure economic and social interaction within a jurisdiction or organisation. Institutional economics treats it as a primary determinant of economic outcomes, since well-defined property rights and enforceable contracts reduce transaction costs and enable investment that would otherwise be too risky to undertake. A robust institutional framework is widely identified as a precondition for sustained economic growth, because it allows actors to make credible long-term commitments. Within organisations, an institutional framework similarly comprises the governance structures, reporting lines, and internal controls that constrain and enable corporate decision-making.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:institutional-framework", "labels": [ "Institutional Framework" ], "is_subclass_of": [ "Institutional Economics" ], "wikilinks": [] }, { "id": "institutional-investment", "title": "Institutional Investment", "domain": "finance", "domain_name": "Finance", "definition": "The investment of pooled funds by organisations such as pension funds, insurers, endowments and asset managers on behalf of their beneficiaries or clients.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:institutional-investment", "labels": [ "Institutional Investment", "Institutional Capital Allocation" ], "is_subclass_of": [ "Asset Management" ], "wikilinks": [ "Asset Management", "Cryptocurrency", "Institutional Economics" ] }, { "id": "institutional-layer", "title": "Institutional Layer", "domain": "finance", "domain_name": "Finance", "definition": "The Institutional Layer is the cross-cutting stratum that represents the organisations, roles, and durable arrangements through which a system operates in the wider world. It sits above the Governance Layer, embodying its decisions in standing bodies, and depends on regulatory and compliance structures. It contains organisational entities, mandates, and the relationships between them.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:institutional-layer", "labels": [ "Institutional Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Governance Layer", "Regulatory Layer", "Organisational Layer", "Social Layer", "Institutional Economics", "Principal-Agent Problem", "owl:Thing" ] }, { "id": "institutional-trust", "title": "Institutional Trust", "domain": "finance", "domain_name": "Finance", "definition": "Institutional trust is the confidence that individuals and organisations place in formal institutions, such as central banks, regulators, courts, and clearing infrastructures, to act competently, fairly, and predictably in accordance with their stated mandates. Unlike interpersonal trust, it is impersonal and systemic: it rests on transparency, accountability, the rule of law, and a credible track record rather than personal acquaintance. Institutional trust underpins the functioning of fiat currency, financial markets, and governance, and its erosion is a primary driver of demand for trust-minimising alternatives such as blockchain systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:institutional-trust", "labels": [ "Institutional Trust" ], "is_subclass_of": [ "Trust" ], "wikilinks": [] }, { "id": "instruction-following", "title": "instruction following", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Instruction following is a language model capability enabling accurate parsing and faithful execution of explicit user or system directives specified in natural language, encompassing multi-step tasks, output format constraints, persona assignments, conditional branching, and constraint satisfaction. It is principally acquired through supervised instruction tuning on curated (instruction, response) datasets and further refined via reinforcement learning from human feedback (RLHF) or direct preference optimisation (DPO). Instruction following is evaluated by the degree to which a model correctly fulfils all stated requirements simultaneously without omitting or violating any constraint, and is a prerequisite for reliable agentic behaviour where a model must decompose and execute multi-step plans expressed as natural language specifications. Failure modes include instruction forgetting in long contexts, sycophantic overriding, and specification gaming.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:instruction-following", "labels": [ "Instruction Following" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "instruction-set-architecture", "title": "Instruction Set Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Instruction Set Architecture (ISA) is the abstract contract between hardware and software that defines the instructions a processor can execute, its registers, data types, addressing modes and memory model. It is the stable interface that allows compilers and operating systems to target a processor family without knowing its microarchitectural implementation. Examples include x86, ARM and the open RISC-V ISA, each balancing complexity, power efficiency and ecosystem support.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:instruction-set-architecture", "labels": [ "Instruction Set Architecture" ], "is_subclass_of": [ "CPU" ], "wikilinks": [] }, { "id": "instruction-tuning", "title": "Instruction Tuning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A fine-tuning technique that trains language models to follow natural language instructions by learning from diverse instruction-response pairs. Instruction tuning enables models to generalise to new tasks described through instructions without task-specific training data, bridging raw language modelling and practical assistive behaviour.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:instruction-tuning", "labels": [ "Instruction Tuning" ], "is_subclass_of": [ "Fine Tuning", "AI Technique" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "instruction-following-conversational-ai-system", "title": "Instruction-Following Conversational AI System", "domain": "ai", "domain_name": "Ai", "definition": "An instruction-following conversational AI system is a natural language processing architecture trained to interpret, decompose, and execute open-ended user directives within a multi-turn dialogue context, producing contextually coherent and task-appropriate responses. Such systems combine large-scale pre-training on diverse corpora with alignment techniques \u2014 notably reinforcement learning from human feedback (RLHF) and instruction fine-tuning \u2014 to bridge the gap between raw language modelling capability and safe, helpful behaviour. They are distinguished from earlier rule-based chatbots by their generalisation across task types (summarisation, coding, question-answering, reasoning) without requiring task-specific engineering. Representative instances include OpenAI's ChatGPT, Google's Gemini, and Anthropic's Claude.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:instruction-following-conversational-ai-system", "labels": [ "Instruction-Following Conversational AI System", "ChatGPT Integration" ], "is_subclass_of": [ "Conversational AI" ], "wikilinks": [] }, { "id": "instrumental-convergence", "title": "Instrumental Convergence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The hypothesis in AI safety research that sufficiently capable goal-directed agents will pursue similar instrumental subgoals \u2014 self-preservation, goal-content integrity, resource acquisition, and cognitive self-improvement \u2014 almost regardless of their terminal objectives, because these subgoals are useful for achieving nearly any final goal. Formulated by Steve Omohundro and Nick Bostrom, the thesis implies that advanced systems may resist shutdown or modification even when never explicitly programmed to do so, motivating research on corrigibility and alignment.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:instrumental-convergence", "labels": [ "Instrumental Convergence" ], "is_subclass_of": [ "AI Risk" ], "wikilinks": [ "AI Risk", "Corrigibility", "AI Alignment", "Existential AI Risk", "Superintelligence" ] }, { "id": "instrumentation", "title": "Instrumentation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Instrumentation is the practice of adding code, agents, or probes to software and systems so that they emit measurable signals \u2014 metrics, logs, traces, and events \u2014 about their internal behaviour and performance. It is the foundational producer of telemetry that downstream observability and monitoring tooling consumes. Without instrumentation a system is opaque; with it, operators gain insight into how the system actually runs in production.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:instrumentation", "labels": [ "Instrumentation" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "insurance", "title": "Insurance", "domain": "governance", "domain_name": "Governance", "definition": "Insurance is a financial arrangement in which a party pays a premium to transfer the risk of an uncertain future loss to an insurer, which pools premiums from many policyholders and pays valid claims out of that pool. It relies on the law of large numbers and actuarial estimation to price risk so that aggregate premiums cover expected losses, expenses and a margin. Insurance underpins economic resilience by smoothing the financial impact of accidents, illness, property damage and other contingencies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:insurance", "labels": [ "Insurance" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "integer-programming", "title": "Integer Programming", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Integer programming is a class of mathematical optimisation in which some or all decision variables are constrained to take integer values, while the objective and constraints are typically linear. The integrality requirement makes these problems NP-hard in general, yet it lets them model discrete decisions such as selection, assignment, and sequencing exactly. Solvers combine the linear-programming relaxation with branch-and-bound and cutting-plane techniques to find provably optimal solutions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:integer-programming", "labels": [ "Integer Programming" ], "is_subclass_of": [ "Combinatorial Optimisation" ], "wikilinks": [] }, { "id": "integrated-circuit", "title": "Integrated Circuit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An integrated circuit (IC) is a set of electronic circuits \u2014 transistors, resistors, capacitors and their interconnections \u2014 fabricated as a single monolithic device on a wafer of semiconductor material, usually silicon. By miniaturising and mass-producing complete circuits photolithographically, ICs made computation cheap, fast and reliable, enabling microprocessors, memory, FPGAs and application-specific chips; transistor density doubled roughly every two years in the trend described by Moore's Law, underpinning the whole of modern digital infrastructure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:integrated-circuit", "labels": [ "Integrated Circuit" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [ "Hardware", "Semiconductor Manufacturing", "CPU Computing", "Field-Programmable Gate Array" ] }, { "id": "integrated-gradients", "title": "Integrated Gradients", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A gradient-based feature attribution method for differentiable models, introduced by Sundararajan, Taly, and Yan (2017), that assigns each input feature a contribution equal to the path integral of the model's gradient along the straight line from a neutral baseline input to the actual input; it uniquely satisfies the axioms of sensitivity and implementation invariance among path methods, its attributions sum exactly to the difference between the model's output and the baseline output (completeness), and it requires only gradient access rather than model retraining or sampling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:integrated-gradients", "labels": [ "Integrated Gradients" ], "is_subclass_of": [ "Feature Attribution" ], "wikilinks": [ "Feature Attribution", "LIME", "SHAP", "Explainable AI" ] }, { "id": "integration-layer", "title": "Integration Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Integration Layer is the cross-cutting stratum that connects otherwise independent systems so they can exchange data and invoke each other's functions. It sits above transport and protocol concerns and below the applications that orchestrate combined behaviour. It contains adapters, connectors, message translation, and the routing that mediates between heterogeneous endpoints.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:integration-layer", "labels": [ "Integration Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Transport Layer", "APILayer", "Application Layer", "Coordination Layer", "Enterprise Integration Patterns", "Message Queue", "owl:Thing" ] }, { "id": "integration-testing", "title": "Integration Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Integration testing is a software verification phase that exercises the interfaces and interactions between combined components or systems, rather than units in isolation. It detects defects in data contracts, protocols, timing, and configuration that emerge only when modules are wired together. Sitting between unit and system testing, it is essential for catching interface mismatches before end-to-end and acceptance stages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:integration-testing", "labels": [ "Integration Testing" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "integrity-council-for-the-voluntary-carbon-market", "title": "Integrity Council for the Voluntary Carbon Market", "domain": "governance", "domain_name": "Governance", "definition": "An independent governance body that sets and maintains quality standards for credits traded in the voluntary carbon market.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:integrity-council-for-the-voluntary-carbon-market", "labels": [ "Integrity Council for the Voluntary Carbon Market" ], "is_subclass_of": [ "Voluntary Carbon Market" ], "wikilinks": [ "Voluntary Carbon Market", "Carbon Credits", "Transparency", "Carbon Offsetting" ] }, { "id": "integrity-verification", "title": "Integrity Verification", "domain": "security", "domain_name": "Security", "definition": "Integrity Verification is the process of confirming that data has not been altered, corrupted, or tampered with since it was created or last certified, typically using cryptographic hashes, checksums, or digital signatures compared against a trusted reference. It is applied to archives, backups, and provenance-tracked assets to detect unauthorised modification. Tamper-evident systems rely on integrity verification to make any alteration detectable even if it cannot be prevented.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:integrity-verification", "labels": [ "Integrity Verification" ], "is_subclass_of": [ "Data Integrity" ], "wikilinks": [] }, { "id": "intel-loihi", "title": "Intel Loihi", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A research neuromorphic processor developed by Intel that implements spiking neural networks in hardware with on-chip learning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:intel-loihi", "labels": [ "Intel Loihi" ], "is_subclass_of": [ "Neuromorphic Chip" ], "wikilinks": [ "Neuromorphic Computing", "Neural Network", "Neuromorphic Chip" ] }, { "id": "intel-sgx", "title": "Intel SGX", "domain": "security", "domain_name": "Security", "definition": "Intel SGX (Software Guard Extensions) is a set of Intel processor instructions that create hardware-isolated memory regions called enclaves, protecting sensitive code and data from the operating system, hypervisor, and other privileged software on the same system, with remote attestation enabling third-party verification of enclave integrity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:intel-sgx", "labels": [ "Intel SGX" ], "is_subclass_of": [ "Trusted Execution Environment" ], "wikilinks": [ "Intel", "Information Security", "Encryption", "Hardware", "Trusted Execution Environment", "https://www.intel.com/content/www/us/en/developer/tools/software-guard-extensions/overview.html", "https://en.wikipedia.org/wiki/Software_Guard_Extensions" ] }, { "id": "intel", "title": "Intel", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Intel is a semiconductor manufacturer that designs and produces microprocessors, chipsets and related hardware including security features such as Software Guard Extensions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:intel", "labels": [ "Intel", "Intel Corporation" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [ "Intel SGX", "Trusted Execution Environment", "Information Security", "Hardware", "https://www.intel.com/", "https://www.intel.com/content/www/us/en/developer/tools/software-guard-extensions/overview.html" ] }, { "id": "intellectual-property-licence-instrument", "title": "Intellectual Property Licence Instrument", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A legal instrument that specifies the permissions, conditions, and restrictions under which intellectual property \u2014 including software, data, and creative works \u2014 may be used, reproduced, modified, and distributed, forming the contractual basis for open-source and open-data ecosystems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:intellectual-property-licence-instrument", "labels": [ "Intellectual Property Licence Instrument", "Intellectual Property Licensing", "License Management System", "license" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "intellectual-property-rights-framework", "title": "Intellectual Property Rights Framework", "domain": "governance", "domain_name": "Governance", "definition": "An Intellectual Property Rights Framework is the structured system of legal doctrines, statutes, treaties, and enforcement mechanisms that grant creators and innovators exclusive, time-limited rights over their intangible works \u2014 encompassing copyright, patent, trade mark, and trade secret regimes. The framework defines conditions of ownership, permissible use, licensing, and transfer of rights, balancing the incentive to create against the public interest in open knowledge. In the context of AI, spatial computing, and distributed digital environments, the framework is under active re-negotiation as autonomous generation, data-intensive training, and cross-jurisdictional virtual economies challenge foundational assumptions about authorship and originality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:intellectual-property-rights-framework", "labels": [ "Intellectual Property Rights Framework", "Intellectual Property", "copyright" ], "is_subclass_of": [ "Legal Framework" ], "wikilinks": [] }, { "id": "intellectual-property-rights", "title": "Intellectual Property Rights", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Intellectual property rights are the legally recognised exclusive entitlements granted to creators and owners over the products of human intellect, such as inventions, literary and artistic works, designs, symbols and names. They include patents, copyright, trademarks and trade secrets, each conferring time-bounded or perpetual control over use, reproduction and commercial exploitation. In blockchain contexts these rights are increasingly represented, transferred and enforced through tokenisation and on-chain licensing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:intellectual-property-rights", "labels": [ "Intellectual Property Rights", "Intellectual Property" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "intellectual-property", "title": "Intellectual Property", "domain": "governance", "domain_name": "Governance", "definition": "A category of legally protected intangible assets arising from creations of the mind, encompassing copyright over creative works, patents over inventions, trade marks over distinctive signs, and trade secrets over confidential know-how. Intellectual property law grants creators time-limited exclusive rights to exploit their creations commercially, balancing incentives for innovation against public access, and underpins licensing, technology transfer, and the contested attribution of AI-generated content.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:intellectual-property", "labels": [ "Intellectual Property" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance", "Copyright", "Software Licence" ] }, { "id": "intelligence-amplification", "title": "Intelligence Amplification", "domain": "ai", "domain_name": "Ai", "definition": "Intelligence amplification is the use of computational tools to augment and extend human cognitive abilities rather than to replace them with autonomous machine intelligence. It frames AI as a partner that enhances human reasoning, memory, and decision-making through interfaces, recommendation, and on-demand expertise. Contrasted with fully autonomous AI, it is a recurring vision in human-computer interaction and a component of discussions about a technological singularity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:intelligence-amplification", "labels": [ "Intelligence Amplification" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "intelligence-explosion", "title": "Intelligence Explosion", "domain": "ai", "domain_name": "Ai", "definition": "Intelligence explosion is the hypothesised scenario in which an artificial intelligence capable of improving its own design enters a positive feedback loop, rapidly producing successively more capable systems and far surpassing human intelligence. First articulated by I. J. Good, it underpins many formulations of the technological singularity and motivates work on AI alignment and control. It remains a contested, largely theoretical concept rather than an observed phenomenon.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:intelligence-explosion", "labels": [ "Intelligence Explosion" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "intelligent-automation", "title": "Intelligent Automation", "domain": "ai", "domain_name": "Ai", "definition": "Intelligent automation is the combination of process automation with artificial intelligence so that workflows can handle unstructured inputs, make context-dependent decisions, and adapt over time. It extends rule-based robotic process automation with machine learning, natural language processing, and computer vision to automate tasks that previously required human judgement. It is widely applied to back-office operations, customer service, and document-intensive processes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:intelligent-automation", "labels": [ "Intelligent Automation", "Intelligent Process Automation" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "intelligent-environment", "title": "Intelligent Environment", "domain": "metaverse", "domain_name": "Metaverse", "definition": "An intelligent environment is a physical or virtual space embedded with sensing, computation, and adaptive agents that perceive context and respond to occupants without explicit commands. In metaverse and ambient-computing settings it hosts autonomous agents and virtual entities that react to user presence, intent, and environmental state. It blends ubiquitous sensing with embodied AI to deliver context-aware, responsive experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:intelligent-environment", "labels": [ "Intelligent Environment" ], "is_subclass_of": [ "Virtual Environment" ], "wikilinks": [] }, { "id": "intelligent-npc", "title": "Intelligent NPC", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An Intelligent NPC (Non-Player Character) is a virtual agent in a game, simulation, or metaverse environment whose behaviour is driven by AI systems \u2014 typically combining perception, reasoning, planning, and natural-language interaction \u2014 rather than purely scripted or rule-based finite state machines. Modern intelligent NPCs leverage large language models for open-ended dialogue, reinforcement learning for adaptive combat and movement, and behavioural AI architectures (behaviour trees, goal-oriented action planning) to produce emergent, contextually appropriate responses to player actions and environmental stimuli. They form the interactive population of persistent virtual worlds and are central to immersive narrative experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:intelligent-npc", "labels": [ "Intelligent NPC" ], "is_subclass_of": [ "Digital Humans" ], "wikilinks": [ "Digital Humans", "Metaverse" ] }, { "id": "intelligent-system", "title": "Intelligent System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An intelligent system is a computational system that perceives its environment, reasons over data and takes actions to achieve defined objectives.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:intelligent-system", "labels": [ "Intelligent System" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [ "Machine Learning", "Robot Perception", "Robotics", "Artificial Intelligence", "https://plato.stanford.edu/entries/artificial-intelligence/", "https://en.wikipedia.org/wiki/Intelligent_agent" ] }, { "id": "intelligent-systems", "title": "Intelligent Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Intelligent Systems are computational systems that exhibit goal-directed behaviour through perception, reasoning, learning, and action. They integrate machine learning, knowledge representation, planning, and autonomous decision-making to operate effectively in complex or uncertain environments. In spatial computing contexts, intelligent systems underpin adaptive avatars, AI-driven scene management, and autonomous agents within persistent virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:intelligent-systems", "labels": [ "Intelligent Systems" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "intelligent-ticket-routing", "title": "Intelligent Ticket Routing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Intelligent ticket routing is the use of machine learning to classify incoming customer support requests and automatically direct each one to the agent, team, or automated workflow best suited to resolve it, based on content, urgency, and historical resolution patterns. It reduces manual triage time and improves first-response accuracy compared with static rule-based routing. It is a core component of customer service and customer support automation platforms.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:intelligent-ticket-routing", "labels": [ "Intelligent Ticket Routing" ], "is_subclass_of": [ "Customer Service Automation" ], "wikilinks": [] }, { "id": "intelligent-tutoring-system", "title": "Intelligent Tutoring System", "domain": "ai", "domain_name": "Ai", "definition": "An AI-powered software system that delivers personalised instruction by modelling a learner's knowledge state, selecting appropriate pedagogical strategies, and adapting content difficulty in real time without requiring a human instructor. Intelligent tutoring systems integrate domain knowledge, learner models, and tutoring strategies to provide one-on-one instructional support at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:intelligent-tutoring-system", "labels": [ "Intelligent Tutoring System", "Intelligent Tutoring Systems" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [] }, { "id": "intelligent-tutoring-systems", "title": "Intelligent Tutoring Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Intelligent tutoring systems are computer-based instructional systems that model an individual learner's knowledge, misconceptions and progress in order to provide adaptive, personalised feedback and instruction comparable to one-on-one human tutoring. They typically combine a domain model of the subject matter, a student model that tracks the learner's mastery, and a pedagogical module that selects the next problem or hint. Intelligent tutoring systems draw on cognitive architectures and cognitive science research to represent how learners acquire and apply knowledge, and have been applied across mathematics, programming and language learning.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:intelligent-tutoring-systems", "labels": [ "Intelligent Tutoring Systems" ], "is_subclass_of": [ "Cognitive Architecture" ], "wikilinks": [] }, { "id": "intelligent-virtual-entity", "title": "Intelligent Virtual Entity", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AI-driven representation within a virtual world that responds adaptively to users and context, combining perception, reasoning, learning, and interaction capabilities to create sophisticated virtual presences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:intelligent-virtual-entity", "labels": [ "Intelligent Virtual Entity" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Adaptive Interaction", "AI Ecosystem", "Behavior Controller", "Computational Resources", "Dynamic Storytelling", "ETSI GR ARF 010", "Intelligent Assistance", "Intelligent Environment", "Interaction Manager", "Knowledge Base", "Personalized Experience", "Reasoning Engine", "Sensor Input", "Adaptive Virtual World", "AI Framework", "ComputationAndIntelligenceDomain", "ComputeLayer", "Computer Vision", "Context-Aware Response", "DataLayer" ] }, { "id": "intent-classification", "title": "Intent Classification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Intent classification is a natural language processing task that assigns a user utterance to one or more predefined intent categories, enabling a system to determine the semantic goal behind an input. It forms the core routing component of conversational AI systems, mapping raw text to structured action labels such as 'book_flight', 'check_balance', or 'cancel_order'.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:intent-classification", "labels": [ "Intent Classification" ], "is_subclass_of": [ "NLPTask" ], "wikilinks": [] }, { "id": "intent-recognition", "title": "Intent Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Intent recognition is the natural-language-processing task of inferring a user's underlying goal or desired action from an utterance, query, or interaction. It maps free-form input to a discrete set of intents and extracts associated parameters, forming the comprehension layer of conversational systems. Accurate intent recognition is what lets chatbots and voice assistants route requests to the correct skill or response.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:intent-recognition", "labels": [ "Intent Recognition", "Intent Detection" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "inter-process-communication", "title": "Inter Process Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Inter-process communication (IPC) is the set of mechanisms an operating system provides for separate processes to exchange data and coordinate their actions despite running in isolated address spaces. Common mechanisms include pipes, message queues, shared memory, sockets and remote procedure calls, each trading off speed, structure and scope. IPC is foundational to modular system design, microservices and any architecture composed of cooperating processes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:inter-process-communication", "labels": [ "Inter Process Communication", "Inter-Process Communication" ], "is_subclass_of": [ "Operating System" ], "wikilinks": [] }, { "id": "inter-agent-communication", "title": "Inter-Agent Communication", "domain": "ai", "domain_name": "Ai", "definition": "Inter-Agent Communication (IAC) is the set of protocols, message formats, and transport mechanisms through which autonomous software agents exchange information, coordinate tasks, delegate subtasks, share observations, and negotiate commitments in multi-agent systems. It spans classical symbolic AI frameworks (FIPA ACL, KQML) and contemporary LLM-based architectures (Model Context Protocol, Agent2Agent Protocol, tool-calling schemas), defining the content language (what is expressed), the interaction protocol (how exchanges are sequenced), and the transport layer (how messages are delivered). Effective IAC enables heterogeneous agents built on different underlying models to collaborate without shared internal state, encompassing message authentication, semantic interoperability, context propagation, and trust boundaries between agents of differing capability and provenance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:inter-agent-communication", "labels": [ "Inter-Agent Communication", "Agent Communication", "Agent-to-Agent Communication" ], "is_subclass_of": [ "Multi-Agent System" ], "wikilinks": [] }, { "id": "inter-annotator-agreement", "title": "Inter-Annotator Agreement", "domain": "ai", "domain_name": "Ai", "definition": "Inter-annotator agreement is a measure of the degree to which independent human annotators assign consistent labels to the same data, commonly quantified using statistics such as Cohen's kappa or Krippendorff's alpha. It is used to assess the reliability of human evaluation and human feedback used to train or benchmark AI systems, since low agreement signals ambiguous guidelines or task definitions. High inter-annotator agreement is generally treated as a precondition for trusting labelled data as ground truth.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:inter-annotator-agreement", "labels": [ "Inter-Annotator Agreement" ], "is_subclass_of": [ "Human Evaluation" ], "wikilinks": [] }, { "id": "inter-blockchain-communication", "title": "Inter-Blockchain Communication", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Inter-Blockchain Communication (IBC) is a standardised protocol that enables independent, sovereign blockchains to exchange data and transfer tokens trustlessly by verifying each other's consensus state. It defines transport, authentication and ordering semantics in which light clients on each chain verify the counterparty's headers, and relayers carry packets and acknowledgements between them. Originating in the Cosmos ecosystem, IBC provides a general-purpose interoperability layer for token transfers, cross-chain messaging and composable multi-chain applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inter-blockchain-communication", "labels": [ "Inter-Blockchain Communication" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "inter-satellite-link", "title": "Inter-satellite Link", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:inter-satellite-link", "labels": [ "Inter-satellite Link" ], "is_subclass_of": [ "Space-to-space Link" ], "wikilinks": [] }, { "id": "inter-world-remittance", "title": "Inter-world Remittance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The transfer of economic value, digital currency, or tokenised assets across distinct virtual worlds, metaverse platforms, or between virtual and physical economies via blockchain bridges, smart-contract escrow, or centralised exchange mechanisms. Inter-world remittance enables labour mobility, cross-platform portfolio management, and arbitrage within interconnected digital economies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:inter-world-remittance", "labels": [ "Inter-world Remittance" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain" ], "wikilinks": [ "Blockchain", "Digital Asset", "Interoperability", "Virtual Currency", "Virtual Economy" ] }, { "id": "interaction-control", "title": "Interaction Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotics control paradigm that explicitly manages contact forces and compliant behaviour when a robot interacts with its environment or human collaborators. Interaction control encompasses impedance control, admittance control, and force control strategies that allow robots to operate safely during physical contact, adapting stiffness, damping, and inertia in response to sensed forces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:interaction-control", "labels": [ "Interaction Control" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [ "owl:Thing" ] }, { "id": "interaction-design", "title": "Interaction Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interaction design (IxD) is the practice of defining and shaping the dynamic behaviour of interactive systems \u2014 determining how they respond to user inputs, communicate state and feedback, and guide users through task sequences \u2014 with the goal of creating products that are usable, efficient, and satisfying across diverse contexts. It encompasses the design of interaction flows, affordances, timing, error recovery patterns, and multi-modal input modalities across digital, physical, spatial, and conversational interfaces. Interaction design is distinguished from visual design by its focus on behaviour over time rather than static composition, and from software engineering by its concern for the human experience of system use rather than technical implementation. As interactive surfaces expand into AR, VR, voice, and AI-mediated interfaces, the field continuously develops new design vocabularies, prototyping methods, and evaluation heuristics to address emergent interaction paradigms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:interaction-design", "labels": [ "Interaction Design", "InteractionDesign" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "interaction-manager", "title": "Interaction Manager", "domain": "metaverse", "domain_name": "Metaverse", "definition": "An interaction manager is the software component that coordinates the flow of communication between a user and an intelligent virtual entity, governing turn-taking, context tracking, and the selection of responses or behaviours. It mediates between perception, dialogue state, and action so that an agent's experience layer feels coherent and responsive. It is the control hub that converts understood input into appropriately timed and styled output.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:interaction-manager", "labels": [ "Interaction Manager" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "interactive-filmmaking", "title": "Interactive Filmmaking", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Interactive filmmaking is the production of cinematic content in which viewers influence the narrative, perspective, or outcome through their choices or presence, blurring the line between film and game. It leverages real-time game engines and virtual production techniques to render branching or responsive scenes on demand. It is enabled by virtual production volumes and real-time rendering pipelines that allow scenes to be reshaped live.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:interactive-filmmaking", "labels": [ "Interactive Filmmaking" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "interactive-learning", "title": "Interactive Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A machine learning paradigm in which a model interactively queries a user, oracle, or environment to obtain labels or feedback for the most informative examples, iteratively improving performance while minimising annotation cost. Interactive learning encompasses active learning, online learning, and human-in-the-loop approaches that tighten the loop between model uncertainty and human input.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:interactive-learning", "labels": [ "Interactive Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "owl:Thing" ] }, { "id": "interactive-media", "title": "Interactive Media", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interactive media refers to digital content and systems that respond dynamically to user input, allowing audiences to influence, navigate, or co-create the experience rather than consume it passively. It spans video games, interactive narratives, simulations, web applications, kiosks, and immersive virtual and augmented reality experiences. Unlike linear media, interactive media couples presentation with real-time feedback loops, making user agency, interface design, and responsiveness central to its form.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:interactive-media", "labels": [ "Interactive Media" ], "is_subclass_of": [ "Digital Content Creation" ], "wikilinks": [] }, { "id": "interactive-proof-system", "title": "Interactive Proof System", "domain": "security", "domain_name": "Security", "definition": "An Interactive Proof System is a protocol in which a computationally powerful prover convinces a probabilistic, resource-bounded verifier of the truth of a statement through a sequence of message exchanges. It satisfies completeness, so true statements are accepted with high probability, and soundness, so false statements are rejected except with negligible probability. Interactive proofs generalise classical proofs and form the theoretical basis for zero-knowledge proofs and many cryptographic protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:interactive-proof-system", "labels": [ "Interactive Proof System" ], "is_subclass_of": [ "Proof System" ], "wikilinks": [] }, { "id": "interactive-storytelling", "title": "Interactive Storytelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interactive Storytelling is a narrative form in which user choices, actions, or presence dynamically shape the progression, branching structure, or emotional arc of a story. In spatial computing contexts, it encompasses XR experiences, virtual worlds, and location-based entertainment where embodied interaction and spatial presence deepen narrative immersion beyond passive media consumption.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:interactive-storytelling", "labels": [ "Interactive Storytelling", "Interactive Fiction", "Interactive Storytelling System" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "interactive-visualization", "title": "Interactive Visualization", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Interactive visualization is the graphical presentation of data or models that users can manipulate in real time through navigation, filtering, selection, and parameter adjustment to explore and understand information. It couples rendering with responsive input handling so that views update immediately as the user probes the underlying data or 3D scene. It supports analysis, design review, and decision-making across scientific, engineering, and immersive contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:interactive-visualization", "labels": [ "Interactive Visualization" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "interbank-settlement", "title": "Interbank Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Interbank settlement is the process by which banks finalise the transfer of funds between each other to discharge obligations arising from customer payments, trades, or other transactions. It relies on messaging networks such as SWIFT to communicate payment instructions and on settlement systems, often operated or overseen by central banks, to move the underlying value. Traditional banking depends on reliable interbank settlement to make cross-institution payments final and irrevocable.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:interbank-settlement", "labels": [ "Interbank Settlement" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [] }, { "id": "intercalibration", "title": "Intercalibration", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:intercalibration", "labels": [ "Intercalibration" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "interchain-accounts", "title": "Interchain Accounts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Interchain Accounts (ICA) is an Inter-Blockchain Communication protocol extension that lets an account on one Cosmos SDK chain control an account on another chain over IBC, without needing a local key or a bridge contract on the counterparty chain. It enables cross-chain operations, such as staking, voting or trading, to be initiated on a home chain and executed remotely by an owned account elsewhere in the Cosmos ecosystem. It is a standard module built on top of the Cosmos IBC transport, authentication and ordering layers.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:interchain-accounts", "labels": [ "Interchain Accounts" ], "is_subclass_of": [ "Cosmos IBC" ], "wikilinks": [] }, { "id": "interchange-protocol", "title": "Interchange Protocol", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An interchange protocol is an agreed set of rules and data formats for transferring assets, avatars, scenes and other content between platforms and applications. Such protocols enable interoperability across virtual environments and spatial computing systems, allowing content to move between implementations without loss of structure or meaning.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:interchange-protocol", "labels": [ "Interchange Protocol" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "interconnect", "title": "Interconnect", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An interconnect is the network fabric that links compute nodes, processors, accelerators and memory within or across systems, providing the high-bandwidth, low-latency communication paths required for parallel and distributed workloads. In high-performance computing and data centres, interconnects determine how efficiently many processors can exchange data and synchronise. Examples include on-chip buses, PCIe links between devices, and cluster fabrics such as InfiniBand or high-speed Ethernet.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:interconnect", "labels": [ "Interconnect" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "interdisciplinary-science", "title": "Interdisciplinary Science", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Interdisciplinary science is a mode of inquiry that integrates concepts, methods, and data across two or more established disciplines to address questions that no single field can answer alone.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:interdisciplinary-science", "labels": [ "Interdisciplinary Science" ], "is_subclass_of": [ "Artificial Intelligence", "AI Research Area" ], "wikilinks": [] }, { "id": "interest-rate-policy", "title": "Interest Rate Policy", "domain": "finance", "domain_name": "Finance", "definition": "Interest rate policy is the use of a central bank's control over short-term policy interest rates to steer borrowing costs, credit demand and inflation. By raising or lowering its target rate, the central bank influences money-market rates, lending, investment and consumption throughout the economy. It is the primary monetary-policy instrument in most modern economies, transmitted through the financial system to output and prices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:interest-rate-policy", "labels": [ "Interest Rate Policy" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "interest-rate", "title": "Interest Rate", "domain": "finance", "domain_name": "Finance", "definition": "An interest rate is the price of borrowing money or the return on lending it, expressed as a percentage of the principal over a defined period, typically annualised. It compensates lenders for the time value of money, expected inflation, and the credit risk of the borrower, and it acts as the principal lever through which central banks transmit monetary policy to the broader economy. Interest rates are categorised as nominal or real, fixed or floating, and short-term or long-term, with the term structure across maturities forming the yield curve.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:interest-rate", "labels": [ "Interest Rate" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "interface-design", "title": "Interface Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Interface design is the discipline of defining the boundaries, contracts, and points of interaction between components, systems, or between a system and its users. In software it encompasses both human-facing user interface design and machine-facing application programming interface design, focusing on clarity, consistency, and ease of correct use. Good interface design minimises coupling, communicates intent, and shapes how reliably and pleasantly the parts of a system can be combined or operated.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:interface-design", "labels": [ "Interface Design" ], "is_subclass_of": [ "Infrastructure", "Software Engineering (Infrastructure)" ], "wikilinks": [] }, { "id": "interface-layer", "title": "Interface Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Interface Layer is the cross-cutting stratum that defines the boundaries and contracts through which components communicate. It sits above the integration and transport mechanisms that carry calls and below the applications that consume the contracts. It contains interface definitions, schemas, and the conventions that govern interaction.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:interface-layer", "labels": [ "Interface Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "APILayer", "Integration Layer", "Application Layer", "Presentation Layer", "Interface Segregation Principle", "Schema", "owl:Thing" ] }, { "id": "interface-standards", "title": "Interface Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical specifications and protocols that define how metaverse components, systems, and services communicate and interoperate, encompassing XR device interfaces, data exchange formats, and cross-platform communication requirements.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:interface-standards", "labels": [ "Interface Standards", "Proprietary Interface Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standard" ], "wikilinks": [ "Cross-Platform Interoperability", "metaverse", "Technical Standard" ] }, { "id": "interface", "title": "Interface", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A boundary or contract between system components that defines how they communicate, exchange data, and interact. Interfaces abstract implementation details to enable modular system design, appearing across domains as REST APIs, smart contract ABIs, ROS topics, rendering APIs, and chat APIs. Interface design governs versioning, error handling, and protocol compliance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:interface", "labels": [ "Interface", "Data Interface", "Uniform Interface" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Abstraction", "Adapter", "API", "Contract", "Protocol", "Blockchain" ] }, { "id": "interferometric-phase", "title": "Interferometric Phase", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interferometric-phase", "labels": [ "Interferometric Phase" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "interferometric-synthetic-aperture-radar", "title": "Interferometric Synthetic Aperture Radar", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interferometric-synthetic-aperture-radar", "labels": [ "Interferometric Synthetic Aperture Radar" ], "is_subclass_of": [ "Synthetic Aperture Radar" ], "wikilinks": [] }, { "id": "intergalactic-medium", "title": "Intergalactic Medium", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:intergalactic-medium", "labels": [ "Intergalactic Medium" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "internal-ai-harness", "title": "Internal AI Harness", "domain": "ai", "domain_name": "Ai", "definition": "An internal AI harness is an in-process execution framework that embeds AI model inference directly within an application's runtime, enabling tight coupling between the host system and AI capabilities for low-latency, high-throughput inference with direct memory access and minimal serialisation overhead, while simultaneously managing the tool-call loop, context selection, task state, approval gates, and observability traces that govern agent behaviour within a single address space.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:internal-ai-harness", "labels": [ "Internal AI Harness" ], "is_subclass_of": [ "Agent Harness", "Evaluation Harness", "AI Infrastructure" ], "wikilinks": [ "Agent Harness", "External AI Harness", "AI Inference", "Model Inference", "Inference Runtime", "Runtime Environment", "Model Serving", "Model Weights", "KV Cache", "Tool Use", "Tool Registry", "Context Window", "Agent Memory", "Agent Runtime", "Edge Computing", "On-Device AI", "Real-Time AI", "Low-Latency AI", "ONNX", "GPU Acceleration" ] }, { "id": "international-ai-cooperation", "title": "International AI Cooperation", "domain": "governance", "domain_name": "Governance", "definition": "Coordination between states, multilateral institutions, safety institutes, and frontier laboratories on the governance of artificial intelligence, spanning safety summit declarations, shared model evaluation programmes, harmonised risk thresholds for frontier models, compute and export control alignment, and scientific exchange, aimed at managing risks that no single jurisdiction can address alone.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:international-ai-cooperation", "labels": [ "International AI Cooperation" ], "is_subclass_of": [ "International Cooperation" ], "wikilinks": [ "International Cooperation", "AI Governance", "AI Safety Summit" ] }, { "id": "international-atomic-time", "title": "International Atomic Time", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:international-atomic-time", "labels": [ "International Atomic Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "international-celestial-reference-system", "title": "International Celestial Reference System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:international-celestial-reference-system", "labels": [ "International Celestial Reference System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "international-cooperation", "title": "International Cooperation", "domain": "governance", "domain_name": "Governance", "definition": "International cooperation is the coordinated action of states, multilateral bodies and other actors to address shared challenges that no single jurisdiction can resolve alone. In technology governance it covers harmonised regulation, mutual recognition of standards, cross-border data arrangements and joint oversight of frontier capabilities. It relies on negotiated agreements, trust between parties and interoperable rules that align otherwise divergent national regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:international-cooperation", "labels": [ "International Cooperation" ], "is_subclass_of": [ "Internet Governance" ], "wikilinks": [] }, { "id": "international-financial-architecture", "title": "International Financial Architecture", "domain": "finance", "domain_name": "Finance", "definition": "The international financial architecture is the framework of institutions, rules, standards, and cooperative arrangements that govern cross-border finance, monetary relations, and the management of global financial stability. It comprises multilateral institutions such as the International Monetary Fund, standard-setting bodies including the Financial Stability Board and the Bank for International Settlements, and the conventions governing exchange rates, capital flows, and crisis resolution. Its purpose is to coordinate macroprudential policy, contain systemic risk, and provide mechanisms for surveillance, liquidity provision, and orderly adjustment across the global financial system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:international-financial-architecture", "labels": [ "International Financial Architecture" ], "is_subclass_of": [ "Financial System" ], "wikilinks": [] }, { "id": "international-monetary-fund", "title": "International Monetary Fund", "domain": "finance", "domain_name": "Finance", "definition": "The International Monetary Fund (IMF) is an international financial institution, established at the 1944 Bretton Woods conference, whose mandate is to promote global monetary cooperation, exchange-rate stability, balanced trade and financial stability among its member countries. It provides macroeconomic surveillance, technical assistance and conditional lending to members facing balance-of-payments difficulties, and manages the Special Drawing Rights reserve asset. As a central pillar of the international monetary system, the IMF shapes monetary policy advice and crisis response worldwide.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:international-monetary-fund", "labels": [ "International Monetary Fund" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "international-terrestrial-reference-system", "title": "International Terrestrial Reference System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:international-terrestrial-reference-system", "labels": [ "International Terrestrial Reference System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "international-trade", "title": "International Trade", "domain": "finance", "domain_name": "Finance", "definition": "International trade is the exchange of goods, services, and capital across national borders, allowing countries to specialise according to comparative advantage and access markets, inputs, and resources beyond their own economies. It is governed by tariffs, trade agreements, and institutions such as the World Trade Organization, and is recorded in the balance of payments. It is a primary driver of economic growth and global interdependence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:international-trade", "labels": [ "International Trade" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [] }, { "id": "internet-connectivity", "title": "Internet Connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Internet connectivity is the capability of a device or network to exchange data with the global Internet, established through physical or wireless access links and the protocol stack that routes packets between endpoints. It depends on addressing, name resolution, and routing infrastructure provided by access and transit providers. Connectivity is characterised by attributes such as bandwidth, latency, reliability, and reachability, all of which shape the performance of applications that rely on it.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:internet-connectivity", "labels": [ "Internet Connectivity" ], "is_subclass_of": [ "Network Connectivity" ], "wikilinks": [] }, { "id": "internet-engineering-task-force", "title": "Internet Engineering Task Force", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Internet Engineering Task Force (IETF) is an open, volunteer-driven standards organisation that develops and promotes the voluntary technical standards underpinning the Internet, published as Requests for Comments (RFCs). Organised into working groups by topic area, it produces core protocol specifications such as IP, TCP, HTTP, TLS and DNS through rough-consensus and running-code processes. The IETF operates under the Internet Society and coordinates with bodies like the Internet Architecture Board to maintain a coherent, interoperable Internet architecture.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:internet-engineering-task-force", "labels": [ "Internet Engineering Task Force" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "internet-governance", "title": "Internet Governance", "domain": "governance", "domain_name": "Governance", "definition": "Internet Governance is the collective development and application of shared principles, norms, rules, decision-making procedures, and programmes that shape the evolution and use of the Internet. It spans technical coordination of identifiers and protocols, allocation of addresses, and policy debates over access, security, and rights, conducted through a multistakeholder model involving governments, the private sector, civil society, and the technical community. Internet governance balances global interoperability against competing claims of sovereignty, openness, and public interest.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:internet-governance", "labels": [ "Internet Governance" ], "is_subclass_of": [ "Digital Governance" ], "wikilinks": [] }, { "id": "internet-infrastructure", "title": "Internet Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Internet Infrastructure encompasses the networking protocols, distributed systems, edge computing platforms, content delivery networks, and cloud architectures that enable large-scale deployment, operation, and interconnection of digital services including AI workloads across global networks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:internet-infrastructure", "labels": [ "Internet Infrastructure", "Internet Backbone" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [ "5G Networks", "Cloud Computing", "Distributed Systems", "Edge Computing" ] }, { "id": "internet-protocol-suite", "title": "Internet Protocol Suite", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Internet Protocol Suite (commonly TCP/IP) is the layered set of communication protocols used to interconnect network devices, defining how data is addressed, routed, fragmented and delivered across heterogeneous networks including the public Internet. It is organised into layers -- link, internet, transport and application -- each providing services to the layer above, with the network layer providing addressing and routing and transport protocols such as TCP, UDP and QUIC providing end-to-end delivery guarantees. Its layered, protocol-agnostic design is what allows the Internet to span vastly different physical networks under one addressing scheme.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:internet-protocol-suite", "labels": [ "Internet Protocol Suite" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "internet-protocol", "title": "Internet Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Internet Protocol (IP) is the principal network-layer protocol of the Internet protocol suite, responsible for addressing hosts and routing packets of data from a source to a destination across interconnected networks. It defines a best-effort, connectionless delivery service in which each datagram is forwarded independently using hierarchical addresses. IP provides the universal addressing and packet format on which higher-layer transport and application protocols depend.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:internet-protocol", "labels": [ "Internet Protocol" ], "is_subclass_of": [ "Infrastructure", "Network Protocol" ], "wikilinks": [] }, { "id": "internet-of-agents", "title": "Internet of Agents", "domain": "ai", "domain_name": "Ai", "definition": "The Internet of Agents is an envisioned network architecture in which autonomous AI agents discover, communicate, transact, and collaborate with one another across organisational boundaries using shared protocols. It extends the web from human- and document-centric interaction to machine-to-machine delegation, negotiation, and task execution among agents. It depends on interoperable identity, messaging, and trust standards to let heterogeneous agents cooperate safely.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:internet-of-agents", "labels": [ "Internet of Agents" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "internet-of-things", "title": "internet of things", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Internet of Things (IoT) is a distributed computing paradigm in which physical objects embedded with sensors, actuators, microcontrollers, and wireless communication modules collect, exchange, and act upon data autonomously over IP networks. IoT deployments span a layered architecture from resource-constrained end-devices through edge gateways to cloud analytics platforms, relying on lightweight protocols such as MQTT, CoAP, AMQP, and LwM2M designed for low-bandwidth, high-latency, or lossy network environments. The paradigm encompasses consumer IoT (smart home devices, wearables), Industrial IoT (IIoT) for manufacturing and critical infrastructure, and urban-scale deployments in smart cities and precision agriculture. Security, interoperability, and device lifecycle management are defining engineering challenges across all IoT segments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:internet-of-things", "labels": [ "Internet of Things", "Internet Of Things", "InternetOfThings" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "interoperability-architecture", "title": "Interoperability Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Structural frameworks and technical designs enabling seamless communication and data exchange between disparate metaverse platforms, virtual worlds, and XR devices through standardised protocols, APIs, and data formats.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:interoperability-architecture", "labels": [ "Interoperability Architecture" ], "is_subclass_of": [ "Network and Communication", "Technical Architecture" ], "wikilinks": [ "Unified Metaverse", "metaverse", "Technical Architecture" ] }, { "id": "interoperability-framework", "title": "Interoperability Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Coordinated set of standards and specifications enabling interaction between heterogeneous systems in metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:interoperability-framework", "labels": [ "Interoperability Framework" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "API Specifications", "Cross-Platform Integration", "Data Formats", "ETSI GR ARF 010", "MSF", "Protocol Definitions", "Technical Documentation", "Data Integration Interface", "DataLayer", "InfrastructureDomain", "Metadata Standard", "Metaverse Architecture", "Scalable Architecture", "Standardization Bodies", "System Interoperability", "Technical Standards", "Universal Manifest" ] }, { "id": "interoperability-protocol", "title": "Interoperability Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An interoperability protocol is a formally specified set of rules, message formats, sequencing constraints, and coordination mechanisms that enable independently implemented systems, networks, or platforms to exchange information and invoke services without requiring shared internal architecture or governance. Such protocols define both syntactic structure and semantic contracts so that heterogeneous participants can interact predictably and verifiably. Interoperability protocols range from low-level wire formats to high-level semantic agreements, and form the foundational connective tissue of distributed computing ecosystems including blockchain networks, spatial computing platforms, AI service meshes, and cross-organisational data exchanges. Their design and governance determine whether digital infrastructure remains open and composable or becomes fragmented and extractive.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:interoperability-protocol", "labels": [ "Interoperability Protocol", "InteroperabilityProtocol" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "interoperability-standard", "title": "Interoperability Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Formal specifications and protocols that enable different metaverse platforms, XR devices, and virtual environments to exchange data and operate together seamlessly, including OpenXR, glTF, USD, and emerging standards from the Metaverse Standards Forum.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:interoperability-standard", "labels": [ "Interoperability Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Cross-Platform Compatibility", "metaverse", "Technical Standard" ] }, { "id": "interoperability-standards", "title": "Interoperability Standards", "domain": "standards", "domain_name": "Standards", "definition": "Interoperability standards are formally agreed specifications \u2014 including data formats, interface definitions, communication protocols, and semantic models \u2014 that enable independently developed systems, networks, and platforms to exchange information and cooperate without bespoke integration work. They are produced by recognised standards bodies, industry consortia, or open working groups, and cover layers from physical connectivity through data serialisation to application-level semantics. Adoption is typically enforced through regulatory mandate, market pressure, or certification programmes. Effective interoperability standards reduce vendor lock-in, lower integration costs, and are foundational to distributed architectures including the web, decentralised identity, and cross-chain blockchain communication.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:interoperability-standards", "labels": [ "Interoperability Standards", "Standards Interoperability" ], "is_subclass_of": [ "Standards" ], "wikilinks": [ "Interoperability", "Standards Body", "Communication Protocols", "Standards" ] }, { "id": "interoperability", "title": "Interoperability", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The capability of distinct systems, applications, or organizational entities to exchange information, interpret shared data correctly, and utilize exchanged information for coordinated operations. Encompasses technical protocol compatibility, semantic data alignment, and organizational process integration across heterogeneous environments including blockchain networks, healthcare systems, and distributed infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:interoperability", "labels": [ "Interoperability", "Cross-DCC Interoperability", "Cross-Language Interoperability", "Cross-World Interoperability", "Device Interoperability", "Healthcare Interoperability", "Interoperability Layer", "Interoperability Verification", "IoT Interoperability", "Technical Interoperability", "Technology Interoperability", "Web Interoperability" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "1inch", "Aave", "Acala", "Across Protocol", "Alameda Research", "Alchemy", "Alex", "Anoma", "Aptos", "Aptos Bridge", "Arbitrum", "Arkadiko", "Astar", "Astria", "atomic swaps", "Atomic Swaps", "Avalanche", "Axelar", "Aztec", "Bank of England" ] }, { "id": "interplanetary-magnetic-field", "title": "Interplanetary Magnetic Field", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interplanetary-magnetic-field", "labels": [ "Interplanetary Magnetic Field" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "interplanetary-space-environment", "title": "Interplanetary Space Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interplanetary-space-environment", "labels": [ "Interplanetary Space Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "interplanetary-space-radiation-environment", "title": "Interplanetary Space Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interplanetary-space-radiation-environment", "labels": [ "Interplanetary Space Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "interpolated-data-state", "title": "Interpolated Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interpolated-data-state", "labels": [ "Interpolated Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "interpolation-method", "title": "Interpolation Method", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interpolation-method", "labels": [ "Interpolation Method" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "interpolation", "title": "Interpolation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interpolation is the construction of new data points within the range of a discrete set of known points, producing a continuous function that passes through or near the samples. Methods range from simple linear and nearest-neighbour schemes to higher-order polynomial, spline and barycentric formulations, each trading smoothness against computational cost and overshoot. In spatial computing and graphics it underpins resampling, shading, animation between keyframes, and the reconstruction of continuous fields from sparse measurements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:interpolation", "labels": [ "Interpolation" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [] }, { "id": "interpretability", "title": "Interpretability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The degree to which a human can understand the internal mechanics, decision-making processes, and cause-effect relationships within an AI system, independent of external explanation tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:interpretability", "labels": [ "Interpretability", "Interpretability Tools" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Debugging", "Model Validation", "Trust", "MetaverseDomain" ] }, { "id": "interpretable-ai", "title": "Interpretable AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine learning models and systems whose internal decision-making processes are inherently transparent and understandable to humans without requiring additional post-hoc explanation techniques. Interpretable AI prioritises transparency by design\u2014via linear models, decision trees, or rule-based systems\u2014distinguishing it from explainable AI approaches that retrofit explanations onto opaque models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:interpretable-ai", "labels": [ "Interpretable AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Black Box Model", "Decision Tree Approximation", "Explainable AI", "Intrinsic Interpretability", "Machine Learning", "MetaverseDomain", "Model Interpretability", "Model Transparency" ] }, { "id": "intersectional-fairness", "title": "Intersectional Fairness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Intersectional Fairness is an approach to algorithmic fairness that evaluates bias and discrimination across subgroups defined by combinations of multiple protected attributes\u2014such as race, gender, age, and disability\u2014recognising that individuals with intersecting marginalised identities may experience unique harms not captured by single-attribute analysis. Rooted in Crenshaw's (1989) intersectionality theory, it requires auditing model performance across all cross-attribute subgroup combinations, demanding both statistical rigour and sufficient sample sizes for rare intersectional groups.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:intersectional-fairness", "labels": [ "Intersectional Fairness", "Intersectional Analysis" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Buolamwini and Gebru (2018)", "Crenshaw (1989)", "IEEE P7003-2021", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "interstellar-medium", "title": "Interstellar Medium", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:interstellar-medium", "labels": [ "Interstellar Medium" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "intrinsic-interpretability", "title": "Intrinsic Interpretability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The inherent transparency and understandability of a machine learning model's architecture and decision-making process, achieved through model design rather than external explanation techniques, enabling direct human comprehension without additional interpretability methods.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:intrinsic-interpretability", "labels": [ "Intrinsic Interpretability" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Black Box Model", "Decision Tree Approximation", "Rule Extraction", "Explainable AI", "Interpretable AI", "MetaverseDomain", "Model Interpretability", "Model Transparency", "Post Hoc Explanation" ] }, { "id": "intrusion-detection-system", "title": "Intrusion Detection System", "domain": "security", "domain_name": "Security", "definition": "An intrusion detection system (IDS) monitors network traffic or host activity to identify malicious behaviour, policy violations, and signs of compromise, raising alerts for investigation. It detects threats using signature matching against known attack patterns, anomaly detection against established baselines, or a hybrid of both. An IDS is a detective control that complements preventive measures; when it can also block traffic it becomes an intrusion prevention system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:intrusion-detection-system", "labels": [ "Intrusion Detection System" ], "is_subclass_of": [ "Threat Detection" ], "wikilinks": [] }, { "id": "intrusion-detection", "title": "Intrusion Detection", "domain": "security", "domain_name": "Security", "definition": "The security discipline of monitoring networks, hosts, and applications to identify unauthorised access, policy violations, and malicious activity in progress. Intrusion detection systems (IDS) combine signature matching against known attack patterns with anomaly detection over baselines of normal behaviour, raising alerts that feed incident response; inline variants (IPS) additionally block detected traffic, trading detection breadth against false-positive risk.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:intrusion-detection", "labels": [ "Intrusion Detection" ], "is_subclass_of": [ "Security Monitoring" ], "wikilinks": [ "Security Monitoring", "Anomaly Detection", "Attack Vector" ] }, { "id": "inventory-management", "title": "Inventory Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Inventory Management is the discipline of ordering, storing, tracking and controlling the stock of goods, components and materials an organisation holds across its supply chain. It seeks to balance the cost of holding stock against the risk of stock-outs, using forecasting, reorder policies and real-time visibility of quantities and locations. Modern implementations integrate barcodes, RFID and digital identifiers so that physical items can be reconciled against system records continuously.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inventory-management", "labels": [ "Inventory Management" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "inverse-kinematics", "title": "Inverse Kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "The process of determining the joint parameters (angles or displacements) required to place a robot's end-effector at a desired position and orientation in Cartesian space. It maps from task space to joint space, inverting the forward kinematic function and resolving ambiguities such as multiple solutions, singularities, and joint-limit violations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inverse-kinematics", "labels": [ "Inverse Kinematics", "InverseKinematics", "Neural Inverse Kinematics", "RB-1006-inverse-kinematics" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [ "Analytical Methods", "Complexity", "Jacobian Matrix", "Joint Configuration", "Joint Limits", "Kinematic Model", "Multiple Solutions", "No Solution", "Non-Uniqueness", "Numerical Methods", "RB-0003-manipulator", "RB-0004-humanoid-robot", "RB-1005-forward-kinematics", "RB-1016-path-planning", "Singularities", "Target Pose", "Kinematics", "Robot Control", "Robotics" ] }, { "id": "inverse-reinforcement-learning", "title": "Inverse Reinforcement Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Inverse reinforcement learning (IRL) is a machine-learning approach that infers the reward function an agent appears to be optimising from observations of its behaviour, rather than being told the reward in advance. It inverts the usual reinforcement-learning problem: instead of finding a policy that maximises a known reward, it recovers the reward that best explains demonstrated, near-optimal trajectories. The recovered reward can then be used to train new policies that generalise the demonstrated intent to unseen situations, making IRL central to learning complex objectives that are hard to specify by hand.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:inverse-reinforcement-learning", "labels": [ "Inverse Reinforcement Learning" ], "is_subclass_of": [ "Imitation Learning" ], "wikilinks": [] }, { "id": "inversion-layer", "title": "Inversion Layer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:inversion-layer", "labels": [ "Inversion Layer", "TemperatureInversionLayer" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "inverted-index", "title": "Inverted Index", "domain": "data", "domain_name": "Data", "definition": "An inverted index is a data structure that maps each term to the list of documents (and often positions) in which it appears, enabling fast full-text retrieval over large corpora. It inverts the natural document-to-terms relationship so that a query term immediately yields its posting list, which can then be intersected or scored. It is the foundational index behind search engines and lexical information retrieval.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:inverted-index", "labels": [ "Inverted Index", "Inverted File Index" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "invertible-neural-network", "title": "Invertible Neural Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An invertible neural network is a neural network architecture constructed so that its forward mapping has an explicit, tractable inverse, allowing outputs to be mapped back to inputs without approximation. This is achieved through coupling-layer designs that keep the Jacobian easy to compute and invert. Invertible architectures are the structural building block of normalising flows, where exact invertibility and tractable Jacobian determinants are required to compute likelihoods in a change-of-variables formulation.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:invertible-neural-network", "labels": [ "Invertible Neural Network" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "investment-contract-analysis", "title": "Investment Contract Analysis", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Investment contract analysis is the legal evaluation of whether a financial arrangement constitutes a security, most commonly by applying the Howey test's criteria of an investment of money in a common enterprise with an expectation of profit derived from others' efforts. In crypto-asset regulation it determines whether a token offering falls under securities law and its attendant registration and disclosure obligations. It is a decisive step in assessing regulatory exposure for digital-asset issuers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:investment-contract-analysis", "labels": [ "Investment Contract Analysis" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "investment-management", "title": "Investment Management", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The professional discipline and supporting technological infrastructure for allocating, monitoring, and optimising portfolios of digital and traditional financial assets. In the blockchain and DeFi context, investment management encompasses on-chain portfolio tools, automated yield strategies, risk modelling, and governance-token-weighted decision frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:investment-management", "labels": [ "Investment Management" ], "is_subclass_of": [ "DeFi and Economics", "Digital Asset" ], "wikilinks": [ "owl:Thing" ] }, { "id": "investor-disclosure", "title": "Investor Disclosure", "domain": "governance", "domain_name": "Governance", "definition": "Investor disclosure is the regulated obligation to provide investors with material, accurate, and timely information about an investment's risks, financials, and governance so they can make informed decisions. It underpins securities regulation by reducing information asymmetry between issuers and the market. In digital-asset and sustainability contexts it increasingly extends to environmental impact and tokenomics, including emissions and energy-use reporting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:investor-disclosure", "labels": [ "Investor Disclosure" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "investor-protection", "title": "Investor Protection", "domain": "governance", "domain_name": "Governance", "definition": "Investor protection comprises the rules, mechanisms and institutional arrangements designed to safeguard investors from unfair, abusive or fraudulent practices in financial markets, reducing information asymmetry and ensuring access to redress.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:investor-protection", "labels": [ "Investor Protection" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [ "Regulatory Requirements", "Market Integrity", "Consumer Protection", "Financial Regulation", "https://www.iosco.org/", "https://www.esma.europa.eu/investor-corner" ] }, { "id": "invoke-ai", "title": "InvokeAI", "domain": "ai", "domain_name": "Ai", "definition": "InvokeAI is an open-source, locally-deployed application and toolkit for running latent diffusion models \u2014 principally Stable Diffusion and its derivatives \u2014 via a browser-based canvas interface, a node-based workflow editor, and a Python API. It provides professional-grade image synthesis capabilities including text-to-image, image-to-image, inpainting, outpainting, and ControlNet-guided generation without reliance on cloud-hosted services. The platform is designed for creative professionals and researchers who require reproducible, privacy-preserving, and customisable generative image workflows on consumer-grade GPU hardware. Its modular architecture supports community fine-tuned model variants, LoRA adapters, and textual inversion embeddings, making it a central hub in the open-source generative image ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:invoke-ai", "labels": [ "InvokeAI" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Stable Diffusion", "Text-to-Image", "Image Generation", "Generative AI" ] }, { "id": "inworld-ai", "title": "Inworld AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Inworld AI is a company providing tools for building interactive AI-driven characters for games and virtual experiences. Its platform combines language models, speech and behaviour controls to drive non-player characters and avatars.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:inworld-ai", "labels": [ "Inworld AI" ], "is_subclass_of": [ "Conversational AI" ], "wikilinks": [ "Language Model", "Text-to-Speech", "Game AI", "Avatar", "Conversational AI", "Speech Recognition" ] }, { "id": "io-t-sensors", "title": "Io T Sensors", "domain": "robotics", "domain_name": "Robotics", "definition": "IoT Sensors are networked sensing devices that collect physical and environmental data (temperature, pressure, motion, humidity, vibration, etc.) and transmit it via internet protocols to edge or cloud systems for processing and analysis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:io-t-sensors", "labels": [ "Io T Sensors", "IIoT Sensors", "IoT Sensors" ], "is_subclass_of": [ "Perception and Sensing", "Internet of Things" ], "wikilinks": [ "5G Networks", "Industrial Automation", "Internet of Things", "Smart Cities", "Smart Manufacturing", "Digital Twin", "Edge Computing", "Predictive Maintenance" ] }, { "id": "io-t-ai-integration", "title": "IoT AI Integration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "IoT AI Integration is the convergence of Internet of Things sensor networks with embedded machine learning models, enabling intelligent, autonomous decision-making directly on constrained IoT devices without requiring centralised cloud processing. The integration addresses fundamental IoT challenges \u2014 network latency, bandwidth limitations, power budgets, and privacy concerns \u2014 by deploying quantised and pruned inference models onto microcontroller-class hardware, transforming passive sensor networks into active edge intelligence systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:io-t-ai-integration", "labels": [ "IoT AI Integration" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "iot-device", "title": "IoT Device", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An IoT device is a networked physical object embedding sensing, processing and communication capabilities that allows it to collect, exchange and act on data over a network. Such devices range from simple sensors and actuators to complex embedded systems, and typically operate under tight power, compute and bandwidth constraints. They form the edge of the Internet of Things, feeding data to gateways, edge nodes and cloud services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iot-device", "labels": [ "IoT Device" ], "is_subclass_of": [ "Internet of Things" ], "wikilinks": [] }, { "id": "io-t-infrastructure", "title": "IoT Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IoT infrastructure is the integrated stack of hardware, networking, middleware, and cloud or edge platforms that enables the deployment, connectivity, management, and data processing of Internet of Things device fleets at scale. It encompasses device provisioning and lifecycle management, low-power wide-area or local-area communication protocols, edge gateways that aggregate and pre-process sensor data, secure device identity and over-the-air update mechanisms, and cloud or fog computing backends that host the data pipelines, analytics, and control planes for IoT applications across industrial, urban, consumer, and agricultural domains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:io-t-infrastructure", "labels": [ "IoT Infrastructure" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "io-t-integration", "title": "IoT Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IoT integration is the connection of networked physical sensors and actuators with software systems so that real-world telemetry flows into analytics, control, and decision platforms. It encompasses device connectivity, protocol gateways, data ingestion, and the synchronisation of physical state with digital models. It is the prerequisite for digital twins, real-time monitoring, and logistics optimisation that depend on live sensor data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:io-t-integration", "labels": [ "IoT Integration", "IIoT Integration" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "iot-platform", "title": "IoT Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A managed software layer that connects fleets of physical devices to applications and analytics, providing device provisioning and identity, secure bidirectional messaging, telemetry ingestion, rule-based event processing, over-the-air updates, and digital twin state management, so that organisations can operate heterogeneous sensor and actuator networks at scale without building the connectivity, security, and data-pipeline plumbing themselves.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:iot-platform", "labels": [ "IoT Platform" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [ "Cloud Computing", "Internet of Things", "MQTT", "Smart Home Automation" ] }, { "id": "iot-sensor", "title": "IoT Sensor", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "An IoT sensor is a networked device that measures physical or environmental conditions \u2014 such as temperature, humidity, location, vibration or light \u2014 and transmits the resulting data over a communications network to back-end systems for monitoring and analysis. By embedding sensing and connectivity into assets and environments, IoT sensors provide continuous, real-time visibility that supports automation and data-driven decisions. In supply chains they are central to cold-chain monitoring, asset tracking and condition-based maintenance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iot-sensor", "labels": [ "IoT Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "io-t", "title": "IoT", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Internet of Things (IoT) is a networked ecosystem of physical devices \u2014 sensors, actuators, microcontrollers, and embedded processors \u2014 connected via IP-based communication protocols to collect, exchange, and act upon data with minimal direct human intervention. IoT extends digital connectivity into the physical world by binding heterogeneous hardware through standardised messaging protocols such as MQTT and CoAP, gateway middleware for local aggregation, and cloud or on-premises analytics platforms. It spans consumer, industrial, agricultural, and healthcare domains, each imposing distinct constraints on power, latency, security, and regulatory compliance. The discipline integrates edge computing, machine learning inference, and digital twin modelling to close feedback loops between the physical and digital realms.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:io-t", "labels": [ "IoT", "Internet of Things" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "ion-thruster", "title": "Ion Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ion-thruster", "labels": [ "Ion Thruster" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "ionosphere", "title": "Ionosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionosphere", "labels": [ "Ionosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ionospheric-d-layer", "title": "Ionospheric D Layer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionospheric-d-layer", "labels": [ "Ionospheric D Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ionospheric-delay", "title": "Ionospheric Delay", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionospheric-delay", "labels": [ "Ionospheric Delay" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ionospheric-disturbance", "title": "Ionospheric Disturbance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionospheric-disturbance", "labels": [ "Ionospheric Disturbance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ionospheric-e-layer", "title": "Ionospheric E Layer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionospheric-e-layer", "labels": [ "Ionospheric E Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ionospheric-f-layer", "title": "Ionospheric F Layer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ionospheric-f-layer", "labels": [ "Ionospheric F Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "iot-security", "title": "Iot Security", "domain": "security", "domain_name": "Security", "definition": "IoT security is the discipline of protecting Internet of Things devices, the networks they connect to, and the data they generate and exchange. It addresses the constrained compute, intermittent connectivity, and large attack surface characteristic of embedded sensors and actuators, applying authentication, encryption, secure boot, and lifecycle patch management. Because IoT devices are often physically exposed and deployed at scale, weak credentials and unpatched firmware are recurrent risks that can be conscripted into botnets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iot-security", "labels": [ "Iot Security", "IoT Security" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "io-t-sensor-network", "title": "IoT Sensor Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A distributed system of spatially separated, resource-constrained sensing devices\u2014each comprising a sensor element, microcontroller, radio transceiver, and power source\u2014interconnected through heterogeneous wireless protocols to collect, process, and transmit physical-world measurements toward gat...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:io-t-sensor-network", "labels": [ "IoT Sensor Network", "Environmental Sensor Network", "Iot Sensor Network", "Wireless Sensor Network" ], "is_subclass_of": [ "Network and Communication", "Pervasive Computing", "Distributed Systems", "Cyber-Physical Systems", "Networked Embedded Systems" ], "wikilinks": [ "3GPP NB-IoT", "5G Networks", "6LoWPAN", "Asset Tracking", "BLE Mesh", "Building Automation", "CBOR", "Cellular M2M", "Centralised Sensing", "Cloud IoT Platform", "CoAP", "Cyber-Physical Systems", "CyberPhysicalSystemsDomain", "Data Aggregation Layer", "Device Identity", "DistributedComputingDomain", "DTLS", "EdgeComputingLayer", "Edge Gateway", "Embedded Systems" ] }, { "id": "ipsec", "title": "Ipsec", "domain": "security", "domain_name": "Security", "definition": "IPsec (Internet Protocol Security) is a suite of protocols that secures IP communications by authenticating and encrypting each packet at the network layer. It provides confidentiality, integrity and origin authentication through the Authentication Header and Encapsulating Security Payload protocols, with keys negotiated via the Internet Key Exchange. IPsec is the foundational technology for site-to-site and remote-access virtual private networks operating transparently beneath application protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ipsec", "labels": [ "Ipsec", "IPsec" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "ipv6", "title": "Ipv6", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "IPv6 (Internet Protocol version 6) is the current generation of the Internet Protocol, designed to replace IPv4 by providing a vastly larger 128-bit address space, simplified header structure, and built-in support for autoconfiguration and security. Its expanded addressing \u2014 roughly 3.4 x 10^38 addresses \u2014 resolves the exhaustion of IPv4 addresses and removes the need for widespread network address translation. IPv6 supports stateless address autoconfiguration, mandatory support for IPsec, and improved multicast and mobility. It is fundamental to the continued growth of the internet and the addressing of vast numbers of Internet of Things devices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ipv6", "labels": [ "Ipv6", "IPv6" ], "is_subclass_of": [ "Internet Protocol", "Network and Communication" ], "wikilinks": [] }, { "id": "iris-recognition", "title": "Iris Recognition", "domain": "security", "domain_name": "Security", "definition": "Iris recognition is a biometric identification technique that authenticates individuals from the intricate, epigenetically random texture of the iris, which is stable from early childhood and differs even between genetically identical twins. Near-infrared imaging, iris segmentation and Gabor-wavelet encoding \u2014 pioneered by John Daugman \u2014 produce a compact IrisCode compared via Hamming distance, delivering among the lowest false-match rates of any biometric and powering border control, national identity schemes such as India's Aadhaar, and proof-of-personhood systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:iris-recognition", "labels": [ "Iris Recognition" ], "is_subclass_of": [ "Biometric Identification" ], "wikilinks": [ "Biometric Identification", "Face Recognition", "Computer Vision", "Worldcoin" ] }, { "id": "irregular-galaxy", "title": "Irregular Galaxy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:irregular-galaxy", "labels": [ "Irregular Galaxy" ], "is_subclass_of": [ "Galaxy" ], "wikilinks": [] }, { "id": "iso-14721", "title": "Iso 14721", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "ISO 14721 is the international standard defining the Open Archival Information System (OAIS) reference model, a conceptual framework for the long-term preservation of digital information. It specifies the functional entities, information packages and responsibilities of an archive committed to preserving content and making it accessible to a designated community over time. As the foundational model for digital preservation, it underpins repositories, audit and certification frameworks, and cultural heritage and scientific data archives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:iso-14721", "labels": [ "Iso 14721", "ISO 14721" ], "is_subclass_of": [ "Digital Preservation" ], "wikilinks": [] }, { "id": "iso-8000", "title": "Iso 8000", "domain": "governance", "domain_name": "Governance", "definition": "ISO 8000 is the international standard for data quality and enterprise master data, defining requirements for representing, exchanging, and measuring the quality of data so it can be trusted across organisational boundaries. It specifies how data should be portable, syntactically and semantically defined, and provenance-bearing, and it provides a framework for assessing characteristics such as completeness, accuracy, and conformance. The standard underpins master data management and data governance programmes that need an objective, verifiable basis for data quality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:iso-8000", "labels": [ "Iso 8000", "ISO 8000" ], "is_subclass_of": [ "Data Standards" ], "wikilinks": [] }, { "id": "iso-9001", "title": "Iso 9001", "domain": "governance", "domain_name": "Governance", "definition": "ISO 9001 is an international standard that specifies the requirements for a quality management system (QMS), enabling organisations to demonstrate their ability to consistently provide products and services that meet customer and regulatory requirements. Published by the International Organization for Standardization, it is built on principles such as customer focus, process approach and continual improvement. Organisations can be independently certified against ISO 9001 to evidence conformity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:iso-9001", "labels": [ "Iso 9001", "ISO 9001" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "iso-standards", "title": "Iso Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "International Organization for Standardization specifications applicable to metaverse, VR, and AR technologies, including health and safety guidelines, 3D representation standards, avatar specifications, and mixed reality frameworks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:iso-standards", "labels": [ "Iso Standards", "ISO 10110 Optics Standards", "ISO Standard", "ISO Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standard" ], "wikilinks": [ "metaverse", "Regulatory Compliance", "Technical Standard" ] }, { "id": "issb", "title": "Issb", "domain": "governance", "domain_name": "Governance", "definition": "The International Sustainability Standards Board (ISSB) is a standard-setting body established under the IFRS Foundation to develop a global baseline of sustainability-related financial disclosure standards. Its inaugural standards, IFRS S1 and IFRS S2, set out general sustainability and climate-specific disclosure requirements for capital markets. The ISSB consolidated earlier voluntary frameworks to reduce fragmentation and improve the comparability of corporate sustainability information.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:issb", "labels": [ "Issb", "ISSB" ], "is_subclass_of": [ "Sustainability Reporting" ], "wikilinks": [] }, { "id": "issue-tracking", "title": "Issue Tracking", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Issue tracking is the practice and supporting tooling for recording, prioritising, assigning, and resolving units of work such as bugs, feature requests, and tasks across a software project. Each issue carries metadata, status, discussion, and links to code changes, giving teams a shared backlog and an audit trail of decisions. It underpins collaborative and agile development and integrates tightly with version control and continuous integration.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:issue-tracking", "labels": [ "Issue Tracking" ], "is_subclass_of": [ "Project Management" ], "wikilinks": [] }, { "id": "item-response-theory", "title": "Item Response Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Item Response Theory (IRT) is a family of psychometric models that relate the probability of a correct response to a test item to a latent trait of the respondent, such as ability, and to properties of the item, such as difficulty and discrimination. Unlike classical test theory, IRT places examinees and items on a common scale, enabling adaptive testing, equating across forms, and fine-grained measurement of ability. It is widely used in assessment, educational technology, and the adaptive components of intelligent learning systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:item-response-theory", "labels": [ "Item Response Theory" ], "is_subclass_of": [ "Adaptive Learning" ], "wikilinks": [] }, { "id": "itil", "title": "Itil", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ITIL (Information Technology Infrastructure Library) is a widely adopted framework of best practices for IT service management, describing how to plan, deliver, operate and continually improve technology services aligned to business value. It organises practices around a service value system spanning strategy, design, transition, operation and improvement, including disciplines such as incident, problem and change management. ITIL provides a common vocabulary and process structure for IT operations and governance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:itil", "labels": [ "Itil", "ITIL" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "it-calculus", "title": "It\u00f4 Calculus", "domain": "ai", "domain_name": "Ai", "definition": "It\u00f4 calculus is a branch of mathematical analysis that extends the methods of calculus to stochastic processes such as Brownian motion. Its central result, It\u00f4's lemma, provides the chain rule for functions of stochastic integrals, accounting for the non-zero quadratic variation of random paths. It is the foundational toolkit for stochastic differential equations and continuous-time probabilistic modelling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:it-calculus", "labels": [ "It\u00f4 Calculus", "Ito Calculus", "It\u00f4 Integral" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "itu-r", "title": "Itu R", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "ITU-R is the Radiocommunication Sector of the International Telecommunication Union, the United Nations agency responsible for managing the global radio-frequency spectrum and satellite orbits. It coordinates international spectrum allocation, develops technical recommendations and maintains the Radio Regulations treaty that governs cross-border use of radio. ITU-R sets the framework for mobile, broadcasting, satellite and emerging wireless standards, including the IMT specifications that define generations such as 5G.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:itu-r", "labels": [ "Itu R", "ITU-R" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "jax", "title": "JAX", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Python library from Google for high-performance numerical computing and machine learning research, combining NumPy-style array operations with automatic differentiation and just-in-time compilation targeting CPUs, GPUs, and TPUs via the XLA compiler.", "entityType": "Class", "qualityScore": 0.72, "maturity": "growing", "iri": "urn:ngm:class:jax", "labels": [ "JAX" ], "is_subclass_of": [ "Deep Learning", "Deep Learning Domain" ], "wikilinks": [ "Automatic Differentiation", "Hardware Acceleration", "Backpropagation", "PyTorch", "Deep Learning Domain" ] }, { "id": "jpmorgan", "title": "JPMorgan", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "JPMorgan Chase, a large United States financial services and banking institution that invests in and applies artificial intelligence and machine learning across its operations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "mature", "iri": "urn:ngm:class:jpmorgan", "labels": [ "JPMorgan", "JPMorgan Chase" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [ "Financial Services", "Anti-Money Laundering", "Machine Learning" ] }, { "id": "json-data-interchange-format", "title": "JSON Data Interchange Format", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JSON (JavaScript Object Notation) is a lightweight, human-readable, language-agnostic data-interchange format standardised as ECMA-404 and RFC 8259, built on two universal data structures: a collection of name/value pairs (objects) and an ordered list of values (arrays). Originally derived from JavaScript object-literal syntax by Douglas Crockford in the early 2000s, JSON has become the dominant wire format for REST APIs, configuration files, inter-service messaging, and AI pipeline payloads due to its minimal syntactic overhead, broad parser availability across every major programming language, and native mapping to in-memory data structures. Its strict subset relationship to JavaScript and its legibility to both humans and machines distinguishes it from predecessor formats such as XML while enabling seamless extension into semantic-web formats such as JSON-LD.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:json-data-interchange-format", "labels": [ "JSON Data Interchange Format", "JSON", "JSON Data Format", "json" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "json-schema", "title": "json schema", "domain": "standards", "domain_name": "Standards", "definition": "JSON Schema is an IETF-standardised declarative vocabulary for annotating and validating the structure, data types, required fields, and value constraints of JSON documents. It defines a set of keywords \u2014 including type, properties, required, additionalProperties, format, pattern, minimum, maximum, and combiners such as allOf, anyOf, and oneOf \u2014 that together constitute a machine-readable contract between data producers and consumers. Originally developed as a series of Internet-Drafts, the 2020-12 dialect is the most widely implemented stable revision. JSON Schema underpins REST API documentation via OpenAPI, event-driven API documentation via AsyncAPI, configuration validation in Kubernetes CRDs and Helm charts, and verifiable credential subject validation in decentralised identity ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:json-data-interchange-format-schema", "labels": [ "JSON Schema", "JSON Schema Draft 2020-12" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "json-serialisation", "title": "JSON Serialisation", "domain": "data", "domain_name": "Data", "definition": "JSON serialisation is the process of encoding in-memory data structures into JavaScript Object Notation text and decoding that text back into structured values. It provides a human-readable, language-independent representation for objects, arrays, numbers, strings and booleans. JSON serialisation is the default interchange format for web APIs, configuration files and message-passing between heterogeneous systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-serialisation", "labels": [ "JSON Serialisation", "JSON Stringification" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "json-web-token", "title": "JSON Web Token", "domain": "security", "domain_name": "Security", "definition": "A JSON Web Token (JWT) is a compact, URL-safe representation of claims transmitted between parties, defined by IETF RFC 7519. A JWT consists of three Base64URL-encoded parts \u2014 header, payload, and signature \u2014 concatenated with periods. The header specifies the token type and signing algorithm; the payload carries claims (assertions about a subject such as user identity, roles, and expiry time); and the signature is computed using either a symmetric shared secret (HMAC) or an asymmetric key pair (RSA, ECDSA), allowing the receiving party to verify token integrity without a round-trip to an authorisation server. JWTs are the dominant stateless session mechanism in REST API and OAuth 2.0 / OpenID Connect identity architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-web-token", "labels": [ "JSON Web Token" ], "is_subclass_of": [ "Authentication Mechanism" ], "wikilinks": [] }, { "id": "json-ld-1-1-standard", "title": "JSON-LD 1.1 Standard", "domain": "data", "domain_name": "Data", "definition": "JSON-LD 1.1 is a W3C Recommendation defining a JSON-based serialization for Linked Data, allowing JSON documents to be interpreted as RDF graphs. It introduces a context mechanism that maps JSON keys to IRIs, plus features such as framing, nested contexts and typed values added in the 1.1 revision. It is the dominant format for embedding machine-readable semantics in web data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-ld-1-1-standard", "labels": [ "JSON-LD 1.1 Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "json-ld-1-1", "title": "json-ld 1.1", "domain": "data", "domain_name": "Data", "definition": "JSON-LD 1.1 is a W3C Recommendation (published July 2020) that extends JSON-LD 1.0 with scoped contexts, type-scoped and property-scoped contexts, propagation control, and the @protected keyword, enabling richer and safer mapping of JSON document terms to IRIs within RDF-based knowledge systems. It standardises compaction, expansion, flattening, and framing algorithms that allow any conformant document to be normalised to a canonical RDF graph without custom mapping code. JSON-LD 1.1 serves as the primary serialisation format for Verifiable Credentials, ActivityPub, the Solid ecosystem, and numerous Linked Data Platform implementations. The specification is maintained by the W3C JSON-LD Working Group and is defined across three companion documents covering syntax, processing algorithms, and the framing API.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-ld-1-1", "labels": [ "JSON-LD 1.1" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [] }, { "id": "json-ld-context", "title": "JSON-LD Context", "domain": "ai", "domain_name": "Ai", "definition": "A JSON-LD Context is the machine-readable document or inline object that maps the shorthand terms and prefixes used in a JSON-LD document to their fully qualified IRIs in a target vocabulary or ontology. It serves as the bridge between the compact, human-readable JSON representation and the globally unambiguous RDF data model, enabling semantic interoperability across disparate systems. Contexts may be embedded inline within a document, referenced by URL, or composed from multiple context documents. The JSON-LD 1.1 specification extends context capabilities with scoped contexts, type-scoped and property-scoped contexts, and protected terms that resist accidental overriding. Correct context design is foundational to knowledge graph compilation and Linked Data publication.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-ld-context", "labels": [ "JSON-LD Context" ], "is_subclass_of": [ "JSON-LD" ], "wikilinks": [] }, { "id": "json-ld-serialisation", "title": "JSON-LD Serialisation", "domain": "standards", "domain_name": "Standards", "definition": "A W3C-standardised serialisation of linked data in ordinary JSON, in which an @context maps plain JSON keys to IRIs so that documents remain idiomatic JSON for developers while being losslessly convertible to RDF graphs. JSON-LD 1.1 (W3C Recommendation, 2020) provides @id, @type, framing, and compaction/expansion algorithms, and is the interchange syntax used by schema.org markup, Verifiable Credentials, DID documents, ActivityPub, and agent-to-agent protocols.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:json-ld", "labels": [ "JSON-LD Serialisation", "JSON-LD" ], "is_subclass_of": [ "Serialisation Format" ], "wikilinks": [ "Serialisation Format", "Linked Data", "RDF", "W3C Recommendation" ] }, { "id": "json-ld", "title": "json-ld", "domain": "data", "domain_name": "Data", "definition": "JSON-LD (JSON for Linking Data) is a W3C Recommendation (first published 2014, revised 1.1 in 2020) that defines a lightweight Linked Data serialisation syntax layered on top of JSON. A @context document maps compact JSON keys to full RDF IRIs, enabling any conformant JSON document to be interpreted as an RDF graph without abandoning existing JSON tooling. JSON-LD is the canonical syntax for W3C Verifiable Credentials, Schema.org structured data embedded in HTML, and the ActivityPub social protocol, making it the most widely deployed Linked Data technology on the open web.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:json-data-interchange-format-ld", "labels": [ "JSON-LD" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [] }, { "id": "json-rpc-2-0", "title": "JSON-RPC 2.0", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JSON-RPC 2.0 is a lightweight, transport-agnostic remote procedure call protocol that encodes method invocations and responses as JSON objects. It defines request, response, notification and batch message structures along with a standard error object, while leaving the transport layer unspecified. Its simplicity has made it the wire format for many blockchain node APIs and AI tool-invocation protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:json-data-interchange-format-rpc-2-0", "labels": [ "JSON-RPC 2.0", "JSON-RPC 2.0 Specification" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "json-rpc", "title": "json-rpc", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JSON-RPC is a lightweight, stateless, transport-agnostic remote procedure call (RPC) protocol that encodes method invocations and their responses as JSON objects. The JSON-RPC 2.0 specification, finalised in 2013, defines a minimal request envelope comprising a jsonrpc version field, a method name, an optional params argument (array or object), and a correlation id; the response carries either a result or a structured error object. The protocol operates over any byte-stream transport \u2014 HTTP, WebSocket, TCP, Unix domain sockets, or stdin/stdout \u2014 making it uniquely portable across networked and embedded contexts. It underpins foundational cross-domain infrastructure including the Ethereum node API, the Language Server Protocol, and the Model Context Protocol used for AI tool-calling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:json-data-interchange-format-rpc", "labels": [ "JSON-RPC", "JSON-RPC Interface", "RPC" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "json", "title": "JSON", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JSON, JavaScript Object Notation, is a lightweight, text-based data format for representing structured data as nested objects, arrays and primitive values, defined by ECMA-404 and RFC 8259. Its syntax is a strict subset of JavaScript object literal notation, making it directly parseable in browsers while remaining language-independent, with parsers available in essentially every programming language. It is the dominant format for web API payloads, configuration files and document-oriented data storage.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:json", "labels": [ "JSON" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "jvm-runtime", "title": "JVM Runtime", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The JVM runtime is the execution environment provided by the Java Virtual Machine, which loads, verifies and runs platform-independent bytecode. It supplies just-in-time compilation, automatic garbage collection, threading and a security model that isolates executing code. The JVM hosts many languages beyond Java, including Kotlin and Scala, and underpins several enterprise blockchain platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:jvm-runtime", "labels": [ "JVM Runtime", "JVM" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "jacobian-determinant", "title": "Jacobian Determinant", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Jacobian determinant is the scalar determinant of the Jacobian matrix of a differentiable vector-valued function, measuring the local factor by which the function expands or contracts volume around a point. In probability and machine learning it provides the change-of-variables correction needed to transform a probability density through an invertible mapping, ensuring the transformed density integrates to one. Its sign indicates whether the transformation preserves or reverses orientation, and its magnitude governs local volume scaling.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:jacobian-determinant", "labels": [ "Jacobian Determinant" ], "is_subclass_of": [ "Jacobian Matrix" ], "wikilinks": [] }, { "id": "jacobian-matrix", "title": "Jacobian Matrix", "domain": "robotics", "domain_name": "Robotics", "definition": "The Jacobian matrix is the matrix of all first-order partial derivatives of a vector-valued function, representing the best linear approximation to that function near a point and encoding how each output component changes with respect to each input variable. In robotics, the geometric and analytic Jacobian matrices map from joint velocity space to end-effector Cartesian velocity space, providing the fundamental tool for differential kinematics, inverse kinematics resolution, singularity analysis, and force-torque transmission between joint and task space. The Jacobian's rank and condition number determine the manipulability of a robot configuration and identify singular configurations where the end-effector loses degrees of freedom in certain directions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:jacobian-matrix", "labels": [ "Jacobian Matrix" ], "is_subclass_of": [ "Differential Kinematics" ], "wikilinks": [] }, { "id": "jailbreaking", "title": "Jailbreaking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Jailbreaking of large language models (LLMs) is the practice of crafting inputs\u2014prompts, instruction sequences, encoded payloads, or multi-turn conversational strategies\u2014that cause a model to produce outputs that circumvent its safety training, content policies, or alignment objectives.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:jailbreaking", "labels": [ "Jailbreaking" ], "is_subclass_of": [ "AI Technique", "Prompt Engineering", "Adversarial Machine Learning", "AI Safety", "AI Risks", "AI Alignment" ], "wikilinks": [ "Adversarial Evaluation", "Adversarial Machine Learning", "AdversarialMLDomain", "AI Red Teaming", "AI Regulation", "AI Safety Domain", "AISI", "ASCII Art Attack", "AttackLayer", "Best-of-N Jailbreaking", "Chain-of-Thought Reasoning", "Classifier Models", "Constitutional Classifiers", "Crescendo Attack", "EvaluationLayer", "Guardrails AI", "Harmful Content Generation", "In-Context Learning", "Llama Guard", "Low-Resource Language Attack" ] }, { "id": "java", "title": "Java", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Java is a class-based, object-oriented, statically typed programming language designed to run on the Java Virtual Machine, enabling write-once-run-anywhere portability across platforms. It compiles to bytecode executed by the JVM and emphasises strong typing, automatic memory management and a vast standard library. Java is widely used in enterprise back-end systems, Android development and permissioned blockchain platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:java", "labels": [ "Java" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "javascript-solid-server", "title": "JavaScript Solid Server", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JavaScript Solid Server (JSS) is a Solid-protocol personal-data-server implementation that extends standard Solid with Nostr-native identity, positioning it as a practical superset of Solid. Alongside conventional Solid Pod storage and Linked Data Platform semantics, it adds did:nostr resolution for login, NIP-07/NIP-98 Nostr authentication, and WebAuthn PRF key management, so a single secp256k1 keypair serves both as a Nostr identity and as the credential controlling a Solid Pod. In the DreamLab single-sign-on stack it is realised by the JavaScriptSolidServer organisation's components: the PodKey browser extension (window.nostr / NIP-98 headers), an SSO redirect service that resolves did:nostr, and a Rust pod backend (solid-pod-rs); it fronts services such as the Nostr BBS forum and the VisionClaw knowledge-graph governance layer.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:javascript-solid-server", "labels": [ "JavaScript Solid Server", "JS Solid Server", "JSS", "JavaScriptSolidServer", "javascriptsolidserver" ], "is_subclass_of": [ "Solid" ], "wikilinks": [] }, { "id": "java-script", "title": "JavaScript", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JavaScript is a high-level, dynamically typed, interpreted programming language standardised as ECMAScript by ECMA International, originally designed by Brendan Eich at Netscape in 1995 to add interactivity to web pages. It features first-class functions, prototype-based object orientation, event-driven and asynchronous programming via the event loop, and runs natively in all major web browsers as the sole client-side scripting language. Beyond browsers, server-side runtimes such as Node.js and Deno have extended JavaScript into backend services, command-line tooling, and cloud functions, making it one of the most widely deployed programming languages across the full web stack.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:java-script", "labels": [ "JavaScript" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [ "Programming Language" ] }, { "id": "jedec", "title": "Jedec", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "JEDEC (the JEDEC Solid State Technology Association) is the global standards body that develops open specifications for semiconductor memory and packaging. It defines the electrical, mechanical and timing standards for technologies such as DRAM, DDR5, HBM and NAND flash, ensuring components from different manufacturers interoperate. JEDEC standards underpin the memory subsystems of virtually all modern computing infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:jedec", "labels": [ "Jedec", "JEDEC" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "jira", "title": "Jira", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Jira is an issue and project tracking platform developed by Atlassian, widely used for planning, tracking, and releasing software across agile and traditional software development workflows. It provides configurable issue types (epics, stories, tasks, bugs), workflow state machines, sprint planning boards, backlog management, and reporting dashboards that support Scrum, Kanban, and hybrid methodologies. Jira integrates deeply with development toolchains\u2014source control, CI/CD pipelines, monitoring systems\u2014and serves as a central system of record for engineering work status, enabling cross-team coordination and stakeholder visibility at scale. It is available as a cloud SaaS product (Jira Cloud) and a self-hosted server or data centre deployment.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:jira", "labels": [ "Jira" ], "is_subclass_of": [ "Project Management" ], "wikilinks": [] }, { "id": "jitter", "title": "Jitter", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Jitter is the variation in a signal's timing from its ideal or expected schedule \u2014 in packet networks, the variability of one-way delay between successive packets of a flow (packet delay variation, RFC 3393/5481), and in digital electronics, the deviation of clock edges from their nominal instants. Where latency measures how late data arrives, jitter measures how inconsistently it arrives; it is caused by queueing, scheduling, and route changes, degrades real-time audio, video, and control traffic, and is absorbed at receivers by de-jitter buffers at the price of added delay.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:jitter", "labels": [ "Jitter" ], "is_subclass_of": [ "Network Performance Metrics" ], "wikilinks": [ "Network Performance Metrics", "Latency", "Propagation Delay", "Quality Of Service", "Real-Time Communication" ] }, { "id": "join-semilattice", "title": "Join-Semilattice", "domain": "data", "domain_name": "Data", "definition": "A join-semilattice is a partially ordered set in which every pair of elements has a least upper bound (join), making the join operation associative, commutative and idempotent. These algebraic properties guarantee that repeated or reordered merges converge to a unique value. The structure is the mathematical foundation for state-based conflict-free replicated data types.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:join-semilattice", "labels": [ "Join-Semilattice", "Semilattice Merge Function", "Semilattice Ordering" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "joint-configuration", "title": "Joint Configuration", "domain": "robotics", "domain_name": "Robotics", "definition": "Joint configuration is the complete set of joint positions of a robot manipulator that together determine the pose of its links and end effector.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:joint-configuration", "labels": [ "Joint Configuration", "Joint Angle" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [ "Forward Kinematics", "Mobile Manipulation", "Manipulator", "Kinematics" ] }, { "id": "joint-encoder", "title": "Joint Encoder", "domain": "robotics", "domain_name": "Robotics", "definition": "A joint encoder is a sensor mounted at a robot joint that measures angular or linear position, and often velocity, of that joint. It provides the proprioceptive feedback required for closed-loop position and motion control of articulated mechanisms. Encoders may be optical, magnetic or capacitive and are typically classed as incremental or absolute.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:joint-encoder", "labels": [ "Joint Encoder" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "joint-mechanics", "title": "Joint Mechanics", "domain": "robotics", "domain_name": "Robotics", "definition": "Joint mechanics is the study of the physical behaviour of robotic and mechanical joints, including the forces, torques, friction, compliance and backlash that govern their motion. It models how power is transmitted through bearings, gears and linkages and how non-ideal effects degrade precision. Understanding joint mechanics is essential for accurate dynamic modelling and high-fidelity motion control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:joint-mechanics", "labels": [ "Joint Mechanics" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "journal-of-machine-learning-research", "title": "Journal of Machine Learning Research", "domain": "ai", "domain_name": "Ai", "definition": "The Journal of Machine Learning Research (JMLR) is a peer-reviewed, open-access scientific journal publishing original research in all areas of machine learning. Founded in 2000, it is among the most influential venues in the field and freely disseminates articles without author or reader fees. JMLR also publishes software and benchmark contributions that establish methodological standards for the community.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:journal-of-machine-learning-research", "labels": [ "Journal of Machine Learning Research", "Proceedings of Machine Learning Research" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "journey-mapping", "title": "Journey Mapping", "domain": "ai", "domain_name": "Ai", "definition": "Journey mapping is a design research technique that visually documents the sequence of steps, touchpoints, and emotional states a person experiences while interacting with a product, service, or organisation over time. It surfaces pain points, moments of friction, and opportunities for improvement by combining qualitative research with a structured visual timeline. Journey mapping is a core input to customer experience management and user experience design programmes.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:journey-mapping", "labels": [ "Journey Mapping" ], "is_subclass_of": [ "Customer Experience" ], "wikilinks": [] }, { "id": "julian-date", "title": "Julian Date", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:julian-date", "labels": [ "Julian Date" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "jumio", "title": "Jumio", "domain": "security", "domain_name": "Security", "definition": "An identity verification company that provides automated document and biometric checks for online onboarding and fraud prevention. Its services support know-your-customer and anti-money-laundering compliance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:jumio", "labels": [ "Jumio" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [ "Identity Verification", "Biometric Authentication", "Face Recognition" ] }, { "id": "junge-aerosol-layer", "title": "Junge Aerosol Layer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:junge-aerosol-layer", "labels": [ "Junge Aerosol Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "jupyter-notebook", "title": "Jupyter Notebook", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An open-source interactive computing document format and execution environment that combines live code cells, rich text narrative, mathematical equations, visualisations, and widgets in a single shareable document. Jupyter Notebooks execute code through kernels (most commonly IPython for Python), enabling iterative, exploratory data analysis with immediate output inline. The format is widely adopted across data science, machine learning research, and scientific computing communities as a primary medium for reproducible research and educational content.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:jupyter-notebook", "labels": [ "Jupyter Notebook" ], "is_subclass_of": [ "Artificial Intelligence", "Data Science" ], "wikilinks": [] }, { "id": "jurisdiction", "title": "Jurisdiction", "domain": "governance", "domain_name": "Governance", "definition": "Jurisdiction is the authority of a legal or governmental body to interpret and apply law, exercise control, and adjudicate disputes within a defined territorial, personal, or subject-matter scope. In digital systems it determines which national or regional laws govern data, transactions, and conduct, and which courts or regulators may assert competence. Conflicts arise when data, parties, or services span multiple jurisdictions with divergent legal regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:jurisdiction", "labels": [ "Jurisdiction" ], "is_subclass_of": [ "Legal Framework" ], "wikilinks": [] }, { "id": "jurisdictional-boundary", "title": "Jurisdictional Boundary", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Legal and regulatory demarcations applicable to metaverse and virtual world governance, addressing the complex challenge of determining which laws apply when virtual activities span multiple physical jurisdictions with no clear territorial boundaries.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:jurisdictional-boundary", "labels": [ "Jurisdictional Boundary" ], "is_subclass_of": [ "Governance and Safety", "Legal Framework" ], "wikilinks": [ "Virtual World Governance", "Legal Framework", "metaverse" ] }, { "id": "just-transition", "title": "Just Transition", "domain": "governance", "domain_name": "Governance", "definition": "A just transition is a framework for shifting economies toward low-carbon and sustainable production while ensuring that the burdens and benefits of that shift are distributed fairly. It seeks to protect workers, communities and developing regions dependent on carbon-intensive industries through retraining, social support and inclusive planning. The concept links climate policy with social equity and labour rights.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:just-transition", "labels": [ "Just Transition", "Workforce Transition" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "just-in-time-compilation", "title": "Just-In-Time Compilation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Just-in-time (JIT) compilation translates code, such as bytecode, an intermediate representation, or a traced computation graph, into native machine code at run time rather than fully ahead of time, allowing the compiler to specialise on the shapes, types and hot paths actually observed during execution. It trades a warm-up compilation cost for the performance of native code plus dynamic optimisation opportunities unavailable to static ahead-of-time compilers. It is used both in general-purpose language runtimes and in machine learning frameworks such as JAX to accelerate numerical computation graphs.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:just-in-time-compilation", "labels": [ "Just-In-Time Compilation", "Just In Time Compilation", "Just-in-Time Compilation" ], "is_subclass_of": [ "Compiler" ], "wikilinks": [] }, { "id": "kohya-dreambooth-and-similar", "title": "KOHYA Dreambooth and similar", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Ecosystem of open-source toolchains, training modologies, dataset-preparation pipelines, and community infrastructure enabling efficient fine-tuning of large-scale Diffusion Models \u2014 principally Stable Diffusion 1.x/2.x, SDXL, and FLUX.1 \u2014 through parameter-efficient adaptation techni...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:kohya-dreambooth-and-similar", "labels": [ "KOHYA Dreambooth and similar", "DreamBooth" ], "is_subclass_of": [ "AI Model Architecture", "Diffusion Models", "Parameter-Efficient Fine-Tuning", "Generative AI", "Machine Learning Discipline", "Transfer Learning" ], "wikilinks": [ "Accelerate", "Accelerate Training Framework", "Adafactor Optimiser", "AdamW Optimiser", "Adapter Layers", "AI Art Generation", "AI Toolkit", "AlgorithmLayer", "Automatic1111", "Base Model", "bitsandbytes", "Black Forest Labs FLUX API", "BLIP-2", "BLIP-2 Captioner", "bmaltais kohya GUI", "Cached Latent Encoding", "Caption Files", "Civitai", "Civitai Model Standards", "ComfyUI Workflow Integration" ] }, { "id": "kto", "title": "KTO", "domain": "ai", "domain_name": "Ai", "definition": "KTO (Kahneman-Tversky Optimization) is a method for aligning language models that learns from binary good or bad feedback on individual outputs rather than paired preference comparisons. Drawing on prospect theory, it defines a utility-based loss that down-weights losses relative to gains, simplifying data collection compared with preference-pair methods. It is an alternative to direct preference optimization in the post-training alignment toolkit.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:kto", "labels": [ "KTO" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [] }, { "id": "kv-cache", "title": "KV Cache", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A KV Cache (Key-Value Cache) in transformer-based language models is a memory structure that stores the computed key and value projection tensors from the multi-head self-attention mechanism for all previously processed tokens in a sequence, allowing autoregressive decoding to reuse these intermediate results rather than recomputing them at every generation step, thereby reducing the per-token computational cost from O(N\u00b2) to O(N) in generation. Each decoder layer maintains its own KV cache, which grows linearly with sequence length and model width, making memory bandwidth and GPU HBM capacity the primary constraints on inference throughput for long-context models. KV cache management strategies \u2014 including paged allocation (PagedAttention), prefix sharing, quantisation, and off-loading \u2014 are critical determinants of serving efficiency and cost.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:kv-cache", "labels": [ "KV Cache", "KV-Cache", "Kv Cache" ], "is_subclass_of": [ "Inference Engine" ], "wikilinks": [] }, { "id": "kyc-aml-compliance", "title": "KYC/AML Compliance", "domain": "governance", "domain_name": "Governance", "definition": "KYC/AML compliance is the operational implementation of Know Your Customer and Anti-Money Laundering obligations within a product or platform, embedding identity verification, screening and monitoring into business processes. It encompasses the controls, workflows and audit evidence needed to demonstrate adherence to regulators. In tokenised systems it is increasingly encoded into smart contracts and asset standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kyc-aml-compliance", "labels": [ "KYC/AML Compliance", "AML/CFT Compliance", "KYC AML Compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "kyc-aml", "title": "KYC/AML", "domain": "governance", "domain_name": "Governance", "definition": "KYC/AML refers to Know Your Customer and Anti-Money Laundering regulatory requirements that obligate financial institutions to verify customer identities and monitor transactions for illicit activity. KYC governs identity verification and risk profiling at onboarding, while AML covers ongoing monitoring, suspicious-activity reporting and sanctions screening. These regimes shape compliance for banks, exchanges and crypto-asset service providers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kyc-aml", "labels": [ "KYC/AML", "KYC AML Framework", "KYC/AML Requirements", "KYC/AML System" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "kyc", "title": "KYC", "domain": "finance", "domain_name": "Finance", "definition": "Know Your Customer, the set of procedures by which a regulated entity verifies the identity of its clients and assesses associated risks before and during a business relationship, forming a cornerstone of anti-money laundering and counter-terrorist financing compliance frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kyc", "labels": [ "KYC", "KYC Verification", "Know Your Customer" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [ "Identity Verification", "Anti-Money Laundering", "Compliance", "Know Your Customer", "Financial Regulation" ] }, { "id": "kzg-commitment", "title": "KZG Commitment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A KZG commitment is a polynomial commitment scheme, named after Kate, Zaverucha and Goldberg, that lets a prover commit to a polynomial with a single constant-size group element and later open it at any point with a constant-size proof. Its security rests on elliptic-curve pairings and a structured reference string produced by a trusted setup. It is central to modern data-availability and scaling designs on Ethereum, including proto-danksharding and danksharding.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:kzg-commitment", "labels": [ "KZG Commitment", "KZG Commitments" ], "is_subclass_of": [ "Polynomial Commitment" ], "wikilinks": [] }, { "id": "kzg-polynomial-commitment", "title": "KZG Polynomial Commitment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Kate-Zaverucha-Goldberg polynomial commitment scheme, which uses elliptic curve pairings and a structured reference string from a trusted setup to commit to a polynomial with a single constant-size group element and prove any evaluation with a constant-size opening proof; its succinctness underpins SNARK constructions such as PLONK and Ethereum's data availability sampling via blob commitments in proto-danksharding.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:kzg-polynomial-commitment", "labels": [ "KZG Polynomial Commitment" ], "is_subclass_of": [ "Polynomial Commitment" ], "wikilinks": [ "Polynomial Commitment", "Trusted Setup", "Data Availability" ] }, { "id": "kademlia-dht", "title": "Kademlia DHT", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Kademlia DHT is a distributed hash table protocol introduced by Petar Maymounkov and David Mazi\u00e8res in 2002 that organises participating nodes into a structured peer-to-peer overlay network using XOR metric distances between 160-bit node identifiers, enabling efficient O(log n) key-value lookup, storage, and routing with provable convergence guarantees. Each node maintains a routing table of k-buckets covering progressively finer-grained regions of the identifier space, and uses iterative or recursive RPC-based lookups to locate the nodes closest to a target key in at most O(log n) network hops. Kademlia's XOR metric is the defining technical innovation that enables symmetric routing\u2014every lookup converges along the same path regardless of direction\u2014making it the most widely deployed DHT protocol underlying BitTorrent, Ethereum, IPFS, and numerous other decentralised systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kademlia-dht", "labels": [ "Kademlia DHT", "Distributed Hash Table", "Kademlia" ], "is_subclass_of": [ "Distributed Hash Table" ], "wikilinks": [] }, { "id": "kalman-filter", "title": "Kalman Filter", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An optimal recursive Bayesian filter that estimates the state of a linear dynamic system from a series of noisy measurements. It minimizes the mean squared error of the estimated state by combining a model-based prediction step with a measurement update step, weighting each by their respective uncertainty covariances to yield the minimum-variance unbiased estimate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kalman-filter", "labels": [ "Kalman Filter", "RB-1015-kalman-filter" ], "is_subclass_of": [ "State Estimation", "Bayes Filter", "Recursive Estimator" ], "wikilinks": [ "Extended Kalman Filter", "Gaussian Noise Assumption", "Linear System Model", "Mean Squared Error", "Optimality", "Probabilistic Robotics", "RB-1002-closed-loop-control", "RB-1008-odometry", "RB-1014-monte-carlo-localization", "Recursive Estimator", "Recursiveness", "System State", "Tracking", "Unscented Kalman Filter", "Bayes Filter", "Control Theory", "Navigation", "RB-1013-localization", "Robotics", "Sensor Fusion" ] }, { "id": "kanban-board", "title": "Kanban Board", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Kanban Board is a visual project management tool that organises work items into columns representing workflow stages such as To Do, In Progress, and Done. Teams use it to limit work-in-progress, surface bottlenecks, and maintain a shared view of task status across distributed members. It is a core practice in agile and lean distributed workflows.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:kanban-board", "labels": [ "Kanban Board" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "karpatkey", "title": "Karpatkey", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Karpatkey is an organisation that provides treasury management and financial strategy services to decentralised autonomous organisations, allocating on-chain assets across protocols.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:karpatkey", "labels": [ "Karpatkey" ], "is_subclass_of": [ "Decentralized Autonomous Organization" ], "wikilinks": [ "Treasury Management", "DeFi", "DAO", "Yield Farming", "Decentralized Autonomous Organization" ] }, { "id": "keccak-256-hashing", "title": "Keccak-256 Hashing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Keccak-256 is a cryptographic hash function from the Keccak family, producing a fixed 256-bit digest using a sponge construction. It is the specific variant adopted by Ethereum, distinct from the later NIST-standardised SHA3-256 due to a difference in padding. Keccak-256 provides collision and preimage resistance for addresses, transaction hashes and message commitments in blockchain systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:keccak-256-hashing", "labels": [ "Keccak-256 Hashing" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "keccak-256", "title": "Keccak-256", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A 256-bit cryptographic hash function based on the Keccak sponge construction, selected as the SHA-3 standard and used natively in Ethereum for address derivation, transaction hashing, and smart contract storage. It differs from NIST SHA-3 in padding and offers strong collision resistance and preimage resistance properties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:keccak-256", "labels": [ "Keccak-256" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "kelly-drecourt-edtech-practitioner", "title": "Kelly Drecourt EdTech Practitioner", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Kelly Drecourt is a collaborator and practitioner associated with teacher-support application development, operating at the intersection of educational technology and AI-assisted tools for classroom and curriculum support.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:kelly-drecourt-edtech-practitioner", "labels": [ "Kelly Drecourt EdTech Practitioner", "Kelly Drecourt" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "keplerian-orbit", "title": "Keplerian Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:keplerian-orbit", "labels": [ "Keplerian Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "keras", "title": "Keras", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Keras is an open-source high-level neural network API written in Python, designed for fast experimentation. It runs on top of backends such as TensorFlow and is integrated as TensorFlow's official high-level interface.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:keras", "labels": [ "Keras" ], "is_subclass_of": [ "Deep Learning Framework" ], "wikilinks": [ "Deep Learning", "Machine Learning", "Deep Learning Framework" ] }, { "id": "kerberos", "title": "Kerberos", "domain": "security", "domain_name": "Security", "definition": "Kerberos is a network authentication protocol that uses tickets and symmetric key cryptography to allow nodes to prove their identity over an untrusted network, relying on a trusted Key Distribution Centre to issue time-limited session tickets without transmitting long-term credentials.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kerberos", "labels": [ "Kerberos" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [ "Encryption", "Single Sign-On", "Cryptographic Protocols", "Authentication", "https://web.mit.edu/kerberos/", "https://datatracker.ietf.org/doc/html/rfc4120" ] }, { "id": "kernel-function", "title": "Kernel Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A kernel function is a similarity measure between two data points that implicitly computes an inner product in a high-dimensional feature space without explicitly constructing that space. It underlies kernel methods such as support vector machines and Gaussian processes, enabling non-linear pattern recognition through the kernel trick. Common examples include the radial basis function, polynomial, and linear kernels, each encoding different assumptions about data similarity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:kernel-function", "labels": [ "Kernel Function" ], "is_subclass_of": [ "Kernel Methods" ], "wikilinks": [] }, { "id": "kernel-fusion", "title": "Kernel Fusion", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Kernel fusion is a compiler and runtime optimisation that merges several adjacent GPU or accelerator operations into a single executable kernel. By combining elementwise, reduction and other operators, it eliminates intermediate memory writes, reduces kernel-launch overhead and improves arithmetic intensity. It is a key technique for accelerating deep-learning training and inference on memory-bandwidth-bound hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:kernel-fusion", "labels": [ "Kernel Fusion" ], "is_subclass_of": [ "GPU Acceleration" ], "wikilinks": [] }, { "id": "kernel-methods", "title": "Kernel Methods", "domain": "ai", "domain_name": "Ai", "definition": "Kernel methods are a class of machine learning algorithms that operate on data implicitly mapped into a high-dimensional feature space via a kernel function, without computing coordinates in that space. The kernel trick replaces inner products with kernel evaluations, enabling linear algorithms to learn non-linear relationships. They underpin support vector machines, Gaussian processes and kernel ridge regression.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kernel-methods", "labels": [ "Kernel Methods" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "key-aggregation", "title": "Key Aggregation", "domain": "security", "domain_name": "Security", "definition": "Key Aggregation is a cryptographic technique that combines multiple individual public keys into a single aggregate public key, against which a combined signature can be verified as though produced by one signer. It is central to modern multi-signature schemes such as MuSig2, where several parties jointly produce one compact signature indistinguishable from a single-key signature. Key aggregation improves privacy, reduces on-chain footprint and lowers verification cost in distributed signing protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:key-aggregation", "labels": [ "Key Aggregation" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "key-ceremony", "title": "Key Ceremony", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A key ceremony is a carefully scripted, audited procedure for generating, distributing, or activating cryptographic keys under strict controls and witnessed participation. It is used when the keys involved are so sensitive that their creation must be verifiably correct, tamper-evident, and resistant to insider compromise. Key ceremonies are common in certificate authorities, threshold-cryptography setups, and high-value custody systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:key-ceremony", "labels": [ "Key Ceremony" ], "is_subclass_of": [ "Key Management" ], "wikilinks": [] }, { "id": "key-derivation-function", "title": "Key Derivation Function", "domain": "security", "domain_name": "Security", "definition": "A cryptographic algorithm that derives one or more secret keys from a master secret using a pseudo-random function, transforming human-readable passwords or seed phrases into cryptographically secure key material while enabling hierarchical key generation, deterministic wallet recovery, and password-based encryption in security and blockchain systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:key-derivation-function", "labels": [ "Key Derivation Function", "Key Derivation" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain" ], "wikilinks": [ "Hierarchical Deterministic Wallet", "Mnemonic Phrase", "Password Hashing", "Asymmetric Encryption", "Blockchain", "Cryptography", "Hash Function", "Private Key" ] }, { "id": "key-exchange", "title": "Key Exchange", "domain": "security", "domain_name": "Security", "definition": "Key exchange is a cryptographic procedure by which two parties establish a shared secret over an insecure channel for use in subsequent encrypted communication.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:key-exchange", "labels": [ "Key Exchange", "Key Exchange Protocol", "Secure Key Exchange" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Cryptography", "Symmetric Encryption", "Cryptographic Protocol" ] }, { "id": "key-generation", "title": "Key Generation", "domain": "security", "domain_name": "Security", "definition": "Key generation is the cryptographic process of creating the keys used by symmetric and asymmetric algorithms, deriving them from high-quality randomness so that they are unpredictable to an adversary. For symmetric schemes it produces a single secret value, while for public-key schemes it produces a mathematically linked private and public key pair. The security of every downstream cryptographic operation rests on the entropy and correctness of this step.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:key-generation", "labels": [ "Key Generation" ], "is_subclass_of": [ "Cryptographic Key Management" ], "wikilinks": [] }, { "id": "key-management-system", "title": "Key Management System", "domain": "security", "domain_name": "Security", "definition": "A key management system (KMS) is a service or appliance that generates, stores, rotates, distributes and revokes cryptographic keys over their full lifecycle. It enforces access policies and isolates key material, often backed by hardware security modules, so that applications can perform cryptographic operations without directly handling private keys. A KMS is foundational to secure identity, encryption and digital-asset custody.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:key-management-system", "labels": [ "Key Management System", "Key Management Infrastructure", "Key Management Service" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "key-management", "title": "key management", "domain": "security", "domain_name": "Security", "definition": "Key management is the discipline governing the complete lifecycle of cryptographic keys, encompassing their generation using cryptographically secure random number generators and key derivation functions, secure distribution, protected storage in hardware security modules or key management services, scheduled rotation to limit exposure windows, and timely revocation upon compromise or expiry. It is a foundational control domain for all systems relying on confidentiality, integrity, and authentication guarantees, since even mathematically strong algorithms are rendered ineffective by weak key custody practices. Key management is mandated by security standards including NIST SP 800-57 and ISO/IEC 27001 Annex A.10, and underpins PKI, envelope encryption architectures, blockchain self-custody, and zero-trust identity frameworks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:key-management", "labels": [ "Key Management", "Cloud Key Management Service", "Distributed Key Management", "Private Key Management" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "key-pair", "title": "Key Pair", "domain": "security", "domain_name": "Security", "definition": "A Key Pair is the matched set of a public key and a corresponding private key used in asymmetric cryptography, where the two keys are mathematically related such that data encrypted or signed with one can only be processed with the other. The private key is kept secret by its owner, while the public key may be distributed openly, enabling encryption, digital signatures, and authentication without sharing a secret in advance. Key pairs underpin public-key infrastructure, blockchain wallets, secure messaging, and identity systems. Their security rests on the computational hardness of deriving the private key from the public key.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:key-pair", "labels": [ "Key Pair" ], "is_subclass_of": [ "Asymmetric Cryptography" ], "wikilinks": [] }, { "id": "key-vector", "title": "Key Vector", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In transformer-based neural networks, a Key Vector is one of three learned linear projections of an input token embedding\u2014alongside the Query Vector and Value Vector\u2014that together implement the scaled dot-product attention mechanism. The key vector represents what a given token has to offer: each query\u2013key dot product measures the compatibility or relevance between a querying token and every other token in the sequence, with the resulting attention weights determining how much each value vector contributes to the output representation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:key-vector", "labels": [ "Key Vector" ], "is_subclass_of": [ "Neural Network Component" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "key-value-cache", "title": "Key-Value Cache", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A key-value cache is a memory structure used in autoregressive transformer inference that stores the key and value projections computed for previously generated tokens, avoiding their recomputation on every new decoding step. By reusing cached keys and values, inference cost grows roughly linearly rather than quadratically with sequence length for the attention computation. Its memory footprint scales with context window length, batch size and model depth, making it a primary constraint on serving throughput for large language models.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:key-value-cache", "labels": [ "Key-Value Cache" ], "is_subclass_of": [ "Attention Mechanism" ], "wikilinks": [] }, { "id": "keyframe-animation", "title": "Keyframe Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Keyframe animation is a technique in which an animator specifies an object's properties at a set of significant frames, called keyframes, and the system interpolates the in-between frames automatically. Property values such as position, rotation, and scale are stored on timed curves whose interpolation and easing control the motion between keys. It is a foundational method for authoring deterministic, repeatable motion in computer graphics and real-time engines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:keyframe-animation", "labels": [ "Keyframe Animation" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "keypoint-detection", "title": "Keypoint Detection", "domain": "ai", "domain_name": "Ai", "definition": "Keypoint detection is a computer vision task that locates salient, semantically meaningful points in an image, such as body joints, facial landmarks or object corners. It outputs spatial coordinates, often with confidence scores, that can be tracked across frames or matched between views. Keypoint detection is a building block for pose estimation, image registration and structure-from-motion.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:keypoint-detection", "labels": [ "Keypoint Detection", "Keypoint Detector" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "keyword-search", "title": "Keyword Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Keyword search is an information retrieval approach that matches documents to a query based on the presence and statistics of literal terms, typically using inverted indexes and term-weighting schemes. It ranks results by lexical relevance signals such as term frequency and inverse document frequency rather than semantic meaning. Fast, interpretable, and exact for known vocabulary, it is frequently combined with semantic methods in hybrid retrieval to balance precision and recall.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:keyword-search", "labels": [ "Keyword Search" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "khronos-group-gl-tf-2-0-specification", "title": "Khronos Group glTF 2.0 Specification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The glTF 2.0 specification is the Khronos Group standard defining the GL Transmission Format for efficient transmission and loading of 3D scenes and models. It uses JSON for structure with binary buffers for geometry and a physically based material model.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:khronos-group-gl-tf-2-0-specification", "labels": [ "Khronos Group glTF 2.0 Specification", "glTF Specification" ], "is_subclass_of": [ "glTF" ], "wikilinks": [ "3D Asset", "Computer Graphics", "glTF" ] }, { "id": "khronos-group", "title": "Khronos Group", "domain": "standards", "domain_name": "Standards", "definition": "The Khronos Group is an open, member-driven industry consortium founded in January 2000 that creates and maintains royalty-free open standards for graphics, parallel compute, media, and extended reality APIs across heterogeneous hardware platforms. Its portfolio spans OpenGL, Vulkan, OpenCL, OpenXR, glTF, NNEF, WebGL, and SPIR-V, forming the portable API foundation that underpins GPU-accelerated rendering, XR runtimes, and cross-vendor compute across the technology industry. Khronos operates through working groups composed of member companies \u2014 including NVIDIA, AMD, Intel, ARM, Apple, Google, and Meta \u2014 that collaboratively draft, ratify, and publish specifications under open licences with conformance test suites.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:khronos-group", "labels": [ "Khronos Group", "Khronos Group Standard" ], "is_subclass_of": [ "Standardization Bodies" ], "wikilinks": [] }, { "id": "khronos-open-xr-1-1-specification", "title": "Khronos OpenXR 1.1 Specification", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The OpenXR 1.1 specification is a revision of the OpenXR standard that consolidates widely adopted extensions into the core and refines the cross-vendor XR application interface.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:khronos-open-xr-1-1-specification", "labels": [ "Khronos OpenXR 1.1 Specification", "Khronos OpenXR 1.0 Specification" ], "is_subclass_of": [ "Khronos OpenXR", "Standards and Interoperability" ], "wikilinks": [ "Khronos OpenXR", "Virtual Reality", "Augmented Reality", "OpenXR", "Open Standards" ] }, { "id": "khronos-open-xr", "title": "Khronos OpenXR", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Khronos OpenXR is an open, royalty-free standard that defines a common application interface for virtual and augmented reality devices, decoupling XR applications from specific runtimes and hardware.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:khronos-open-xr", "labels": [ "Khronos OpenXR" ], "is_subclass_of": [ "Spatial Computing", "Standards and Interoperability", "Spatial Computing Domain" ], "wikilinks": [ "OpenXR", "Virtual Reality", "Augmented Reality", "WebXR", "Khronos Group", "Spatial Computing Domain" ] }, { "id": "khronos-gl-tf", "title": "Khronos glTF", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Khronos glTF (GL Transmission Format) is an open royalty-free specification maintained by the Khronos Group that defines a JSON-based container format for 3D scenes and models, designed for efficient runtime delivery and rendering rather than authoring \u2014 storing geometry as compact binary buffer views, referencing PBR (Physically Based Rendering) material parameters, animation data, skeletal hierarchies, and scene graphs in a format that maps closely to GPU resource layouts, minimising parse and upload overhead. glTF 2.0 (2017) established the core specification and is widely regarded as the 'JPEG of 3D' due to its adoption across web browsers (via Three.js and Babylon.js), game engines (Unreal, Unity, Godot), AR/VR runtimes (OpenXR, WebXR), and digital twin platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:khronos-gl-tf", "labels": [ "Khronos glTF" ], "is_subclass_of": [ "glTF (3D File Format)" ], "wikilinks": [] }, { "id": "kinematic-chain", "title": "Kinematic Chain", "domain": "robotics", "domain_name": "Robotics", "definition": "A kinematic chain is an assembly of rigid bodies (links) connected by joints that constrain their relative motion, forming the structural and mathematical basis for analysing and controlling the motion of mechanisms and robots. Open kinematic chains (serial manipulators) have one free end and exhibit simple forward kinematics but complex inverse kinematics; closed kinematic chains (parallel manipulators) have all links connected in loops, offering higher stiffness and load capacity. The Denavit-Hartenberg convention provides the canonical parameterisation for representing joint geometry and computing transformations along the chain, underpinning all modern robot programming and simulation systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:kinematic-chain", "labels": [ "Kinematic Chain", "KinematicChain" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [] }, { "id": "kinematic-element", "title": "Kinematic Element", "domain": "robotics", "domain_name": "Robotics", "definition": "A Kinematic Element is a rigid body or joint primitive within a robot's mechanical structure that participates in the forward and inverse kinematic chain, defining the positional and orientational degrees of freedom of a limb segment. Chains of kinematic elements model the geometry of robotic arms, legs, and manipulators for motion planning and control.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:kinematic-element", "labels": [ "Kinematic Element", "KinematicElement" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "kinematic-model", "title": "kinematicmodel", "domain": "robotics", "domain_name": "Robotics", "definition": "A kinematic model is a mathematical abstraction that describes the geometry of a robot's motion by relating joint-space variables (angles and displacements) to the position and orientation of the end-effector in Cartesian space, without accounting for the forces or torques that produce that motion. Forward kinematics maps joint configurations to end-effector poses using homogeneous transformation matrices or Denavit-Hartenberg parameters, whereas inverse kinematics solves the reverse problem of finding joint configurations that achieve a desired pose. Kinematic models are foundational to trajectory generation, motion planning, and simulation in robotic systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:kinematic-model", "labels": [ "KinematicModel", "Kinematic Model" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "kinematics-model", "title": "Kinematics Model", "domain": "robotics", "domain_name": "Robotics", "definition": "KinematicsModel is a mathematical representation of the geometric relationships between a robot's joint configuration space and its end-effector pose in Cartesian workspace coordinates, abstracting away forces and inertial effects to describe pure motion geometry.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:kinematics-model", "labels": [ "Kinematics Model" ], "is_subclass_of": [ "Actuation and Control", "Rigid Body Mechanics", "Robot Model", "Mathematical Model", "Motion Model", "Geometric Model" ], "wikilinks": [ "AlgorithmLayer", "Autonomous Mobile Robots", "Calibration", "Cartesian Impedance Control", "Cartesian Space", "Closed-Form IK", "ComputationalGeometryDomain", "ControlSystemsDomain", "Deformable Body Model", "Denavit-Hartenberg Parameters", "DH Convention", "Differential Geometry", "Diffusion Policy", "Drake", "Dynamics Model", "Exoskeleton Control", "Gazebo", "Geometric Model", "HardwareAbstractionLayer", "Homogeneous Transformation" ] }, { "id": "kinematics", "title": "Kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Kinematics is the branch of mechanics that studies the geometry and mathematics of motion \u2014 positions, velocities, accelerations, and trajectories of bodies \u2014 without consideration of the forces or torques that cause that motion. In robotics and spatial computing, it encompasses forward kinematics (mapping joint-space parameters to Cartesian end-effector pose via homogeneous transformation matrices and Denavit-Hartenberg conventions) and inverse kinematics (solving the reverse mapping from desired pose to joint configurations). Kinematic analysis underpins motion planning, trajectory generation, workspace characterisation, collision avoidance, and animation systems across robotics, biomechanics, computer graphics, and autonomous vehicles.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:kinematics", "labels": [ "Kinematics" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [ "owl:Thing" ] }, { "id": "king-s-college-london", "title": "King's College London", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "King's College London (KCL) is a public research university in central London, founded in 1829 by King George IV and the Duke of Wellington, and one of the oldest and largest constituent colleges of the University of London. KCL is a Russell Group institution with world-leading research programmes spanning medicine, health sciences, law, humanities, natural sciences, and security studies. It is a major contributor to UK artificial intelligence and digital health research, hosting institutes including the King's Institute for Artificial Intelligence and maintaining close collaboration with NHS Foundation Trusts. KCL is a founding partner of the Alan Turing Institute and a participant in numerous UKRI and Wellcome Trust-funded data science and machine learning initiatives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:king-s-college-london", "labels": [ "King's College London" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Artificial Intelligence", "University College London", "Alan Turing Institute", "Entity" ] }, { "id": "kl-divergence", "title": "Kl Divergence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Kullback-Leibler (KL) divergence is a measure from information theory that quantifies how one probability distribution differs from a second reference distribution, expressed as the expected excess surprise from using the wrong distribution. It is non-negative and zero only when the two distributions coincide, but it is asymmetric and does not satisfy the triangle inequality, so it is not a true metric. KL divergence is central to maximum-likelihood estimation, variational inference and many machine-learning objectives, where minimising it aligns a model distribution with a target. In reinforcement learning from human feedback it acts as a regulariser that keeps a fine-tuned policy close to its reference.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:kl-divergence", "labels": [ "Kl Divergence", "KL Divergence" ], "is_subclass_of": [ "Information Theory" ], "wikilinks": [] }, { "id": "klima-dao", "title": "KlimaDAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised organisation that acquires and holds tokenised carbon credits in its treasury, issuing a token backed by those credits to channel capital toward carbon retirement.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:klima-dao", "labels": [ "KlimaDAO" ], "is_subclass_of": [ "Decentralised Autonomous Organisation" ], "wikilinks": [ "Carbon Credit Token", "Treasury Management", "Voluntary Carbon Market", "Carbon Markets", "Decentralised Autonomous Organisation" ] }, { "id": "knowhere-visitor-attention-system", "title": "KnoWhere Visitor Attention System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "KnoWhere is an AI-driven visitor attention-tracking system for museums and immersive experience spaces, using computer vision and machine vision cameras to capture gaze vectors, emotion signals, and spatial attention metrics in real time without wearables. It enables hyper-personalised narrative adaptation and provides curators with actionable behavioural analytics whilst preserving visitor privacy through anonymised data processing.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:knowhere-visitor-attention-system", "labels": [ "KnoWhere Visitor Attention System", "Knowhere" ], "is_subclass_of": [ "Computing and Cloud", "Tracking Technology" ], "wikilinks": [ "Head Gaze", "KnoWhere", "Face Swap", "NVIDIA Omniverse", "PEOPLE", "Segmentation and Identification" ] }, { "id": "know-your-customer", "title": "Know Your Customer", "domain": "finance", "domain_name": "Finance", "definition": "Know Your Customer (KYC) is a regulatory and compliance process by which financial institutions and other regulated entities verify the identity of their clients, assess their risk profiles, and understand the nature of their financial activities to prevent money laundering, terrorist financing, and other financial crimes. KYC encompasses identity document verification, biometric checks, beneficial ownership disclosure, and ongoing transaction monitoring. The process is mandated by the Financial Action Task Force (FATF) recommendations and implemented through national legislation including the EU Anti-Money Laundering Directives, the US Bank Secrecy Act, and equivalent statutes in over 200 jurisdictions.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:know-your-customer", "labels": [ "Know Your Customer", "KYC Know Your Customer" ], "is_subclass_of": [ "Compliance" ], "wikilinks": [] }, { "id": "knowledge-artefact-update-cycle", "title": "Knowledge Artefact Update Cycle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Knowledge Artefact Update Cycle is a structured, recurring process through which knowledge assets \u2014 including ontology classes, documentation nodes, linked data graphs, and curated references \u2014 are reviewed, validated, corrected, and re-published to maintain epistemic accuracy and semantic coherence. The cycle defines per-artefact cadences calibrated to the rate of change of underlying domains, balancing maintenance cost against information decay. It sits at the intersection of knowledge lifecycle management, data stewardship, and continuous integration practices applied to semantic knowledge bases. Effective update cycles incorporate provenance tracking, diff-based change detection, and staleness thresholds to trigger targeted refresh actions without wholesale reconstruction of the knowledge graph.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-artefact-update-cycle", "labels": [ "Knowledge Artefact Update Cycle", "Update Cycle" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "knowledge-base", "title": "knowledge base", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A knowledge base is a structured, machine-readable repository of domain-specific information, factual assertions, and inference rules that software systems\u2014including expert systems, question-answering engines, and AI agents\u2014can query, reason over, and update. Knowledge bases range from relational tables and document stores to formal ontologies expressed in OWL/RDF and property graphs, each offering different trade-offs between expressiveness, scalability, and reasoning complexity. They function as the long-term declarative memory layer in AI architectures, enabling systems to retrieve, validate, and integrate facts without embedding all knowledge in model parameters. In contemporary retrieval-augmented generation pipelines a knowledge base serves as the authoritative external store from which a retriever selects grounding context for a large language model.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:knowledge-base", "labels": [ "Knowledge Base", "AI Knowledge Base", "Medical Knowledge Base" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "knowledge-based-authentication", "title": "Knowledge Based Authentication", "domain": "security", "domain_name": "Security", "definition": "Knowledge-Based Authentication (KBA) verifies identity by challenging a user to supply information presumed known only to them, such as a password, PIN, or answers to security questions. Static KBA uses pre-registered secrets, while dynamic KBA generates questions from third-party records at challenge time. Because the underlying secrets can be guessed, phished, or harvested through social engineering, KBA is increasingly supplemented or replaced by possession- and biometric-based factors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:knowledge-based-authentication", "labels": [ "Knowledge Based Authentication", "Knowledge-Based Authentication" ], "is_subclass_of": [ "Multi-Factor Authentication" ], "wikilinks": [] }, { "id": "knowledge-co-construction", "title": "Knowledge Co-Construction", "domain": "ai", "domain_name": "Ai", "definition": "Knowledge co-construction is a process in which participants jointly build shared understanding through dialogue, negotiation and the integration of multiple perspectives. Rather than transmitting fixed knowledge, it treats understanding as emergent from collaborative interaction. It is a central mechanism in collaborative and collective learning environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-co-construction", "labels": [ "Knowledge Co-Construction" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [] }, { "id": "knowledge-component-model", "title": "Knowledge Component Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Knowledge Component Model is a decomposition of a learning domain into discrete, assessable units of knowledge or skill, each of which can be independently tracked as a learner masters it. It provides the structural backbone for adaptive tutoring systems, mapping exercises and questions to the components they exercise. Techniques such as Bayesian Knowledge Tracing estimate a learner's mastery of each component over time.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:knowledge-component-model", "labels": [ "Knowledge Component Model" ], "is_subclass_of": [ "Learner Model" ], "wikilinks": [] }, { "id": "knowledge-discovery", "title": "Knowledge Discovery", "domain": "data", "domain_name": "Data", "definition": "Knowledge Discovery is the process of identifying valid, novel, useful and understandable patterns in data, transforming raw records into actionable knowledge. It spans data selection, cleaning, transformation, mining and interpretation, and draws on statistics, machine learning, information retrieval and database technology. The discipline is often framed as Knowledge Discovery in Databases (KDD), within which data mining is the specific pattern-extraction step.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:knowledge-discovery", "labels": [ "Knowledge Discovery" ], "is_subclass_of": [ "Knowledge Management" ], "wikilinks": [] }, { "id": "knowledge-distillation-for-edge", "title": "Knowledge Distillation for Edge", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model compression technique that transfers learned representations from large, accurate teacher neural networks to compact student models optimised for edge deployment, achieving 20-30x size reduction while retaining 97%+ of accuracy through soft-target training, temperature scaling, and layer-wise representation matching.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:knowledge-distillation-for-edge", "labels": [ "Knowledge Distillation for Edge" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "knowledge-distillation", "title": "Knowledge Distillation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A model compression technique where a smaller \"student\" model is trained to mimic the behaviour of a larger \"teacher\" model, transferring knowledge through soft targets. The student learns from the teacher's output probability distributions (soft targets) rather than ground-truth hard labels, enabling competitive performance at a fraction of the computational cost.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-distillation", "labels": [ "Knowledge Distillation", "Model Distillation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Nostr", "Hyper personalisation", "Knowledge Graphing", "Logseq", "MetaverseDomain" ] }, { "id": "knowledge-economy", "title": "Knowledge Economy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The knowledge economy is an economic system in which the production, dissemination, and exploitation of information and intellectual capability, rather than physical inputs, are the primary drivers of growth, productivity, and competitive advantage.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-economy", "labels": [ "Knowledge Economy" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "knowledge-graph-construction", "title": "Knowledge Graph Construction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Knowledge Graph Construction is the process of automatically or semi-automatically building structured graph representations of world knowledge by extracting entities, relations, and attributes from heterogeneous sources such as text corpora, databases, and web data. The discipline combines techniques from natural language processing, information retrieval, and ontology engineering to produce machine-readable graphs that support reasoning, question answering, and semantic search.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:knowledge-graph-construction", "labels": [ "Knowledge Graph Construction" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "knowledge-graph-diagnostic-node", "title": "Knowledge Graph Diagnostic Node", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A debug linked node is a diagnostic artefact within a knowledge graph or ontology system \u2014 a placeholder page that is deliberately linked to from another node in order to verify that the wikilink resolution, graph edge creation, and link-traversal mechanisms are functioning correctly. It serves as a test fixture for the ontology pipeline, confirming that bidirectional link references are parsed, stored as edges, and retrievable through graph queries.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-graph-diagnostic-node", "labels": [ "Knowledge Graph Diagnostic Node", "debug linked node" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Debug Test Page" ] }, { "id": "knowledge-graph-embedding", "title": "Knowledge Graph Embedding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Knowledge graph embedding is the technique of representing the entities and relations of a knowledge graph as continuous low-dimensional vectors that preserve the graph's structural and semantic regularities. Scoring functions over these vectors model the plausibility of triples, enabling tasks such as link prediction, entity resolution and similarity-based retrieval through algebraic operations rather than symbolic traversal. It bridges symbolic knowledge representation with vector-based machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:knowledge-graph-embedding", "labels": [ "Knowledge Graph Embedding" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "knowledge-graph-kanban-board", "title": "Knowledge Graph Kanban Board", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Kanban-style project planning board within a Logseq knowledge graph that aggregates query-driven progress views across active projects. It uses embedded block queries to surface tasks tagged with progress properties, enabling at-a-glance visibility of TODO, DOING, and DONE states across linked project pages such as PlayerTwo and ParentsGuideToAI.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-graph-kanban-board", "labels": [ "Knowledge Graph Kanban Board", "Planning Kanban - currently broken" ], "is_subclass_of": [ "Software Engineering", "Active Research Projects Registry" ], "wikilinks": [ "KnoWhere", "NLW education discord", "Pete Woodbridge", "Training Material", "PlayerTwo" ] }, { "id": "knowledge-graph-presentation-session-artefact", "title": "Knowledge Graph Presentation Session Artefact", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Today Presentation 2 is a knowledge-graph page capturing the content or embedded block reference for a second working presentation session, likely associated with DreamLab AI's applied research and demonstration programme. As a session artefact it serves as a container node linking to embedded Logseq blocks that hold presentation slides, notes, or structured outputs. It connects to broader themes of knowledge representation, immersive technology, and responsible AI governance that form DreamLab's public-facing educational work.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-graph-presentation-session-artefact", "labels": [ "Knowledge Graph Presentation Session Artefact", "today presentation 2" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "knowledge-graph-publication-classifier", "title": "Knowledge Graph Publication Classifier", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Public, as a Logseq property tag (`public:: true`), marks a knowledge-graph page as intended for external publication. In the NarrativeGoldmine ontology it acts as an access-control classifier: pages bearing this annotation are included in export pipelines targeting the open WebVOWL visualisation and the public-facing OWL2 dataset. It is a metadata concept rather than a domain-level class, functioning analogously to an access-control label within the data-governance layer.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-graph-publication-classifier", "labels": [ "Knowledge Graph Publication Classifier", "Public", "Public Awareness", "Public Service Modernisation" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "knowledge-graph-style-guide", "title": "Knowledge Graph Style Guide", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Style Guide is a normative document specifying authoring conventions \u2014 voice, tone, markup syntax, citation format, and structural patterns \u2014 for consistent content creation within a knowledge graph or publication system. In the NarrativeGoldmine context it captures Logseq markdown conventions, UK English preferences, and the analytical-conversational register used throughout the graph.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-graph-style-guide", "labels": [ "Knowledge Graph Style Guide", "Style Guide" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "block title", "reference" ] }, { "id": "knowledge-graph", "title": "Knowledge Graph", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A semantic knowledge network that represents entities, relationships, and attributes as an interconnected graph structure, enabling advanced reasoning, inference, and knowledge discovery across metaverse systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:knowledge-graph", "labels": [ "Knowledge Graph", "KnowledgeGraph", "Open Knowledge Graph" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "AI Reasoning", "Entity Node", "Entity Resolution", "Inference Engine", "Knowledge Discovery", "Ontology", "Ontology Schema", "RDF Framework", "Reasoning Service", "Recommendation System", "Relationship Edge", "Schema Definition", "Schema.org", "Semantic Property", "Semantic Web Infrastructure", "Triple Store", "W3C OWL", "W3C RDF", "ComputationAndIntelligenceDomain", "Graph Database" ] }, { "id": "knowledge-graphing", "title": "Knowledge Graphing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Knowledge Graphing is the discipline and computational framework for constructing, enriching, storing, querying, and reasoning over structured semantic knowledge encoded as a graph of typed entities (nodes) and directed typed relationships (edges), spanning formal Semantic Web standards (RDF/...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:knowledge-graphing", "labels": [ "Knowledge Graphing" ], "is_subclass_of": [ "AI Technique", "Knowledge Representation", "Semantic Web Linked Data Standard", "Graph Databases", "Information Retrieval", "Ontology Engineering", "Machine Learning Discipline", "Data Engineering", "Artificial General Intelligence" ], "wikilinks": [ "Autogen", "Biomedical Research", "Biomedical Research", "Biomedical Research", "Columnar Databases", "Community Detection", "ComplEx", "Coreference Resolution", "Data Engineering", "DatabaseDomain", "DBpedia", "Description Logics", "Digital Twins", "Digital Twins", "Document Stores", "Drug Discovery", "Drug Discovery", "EmbeddingLayer", "Enterprise Search", "Entity Resolution" ] }, { "id": "knowledge-graphs", "title": "Knowledge Graphs", "domain": "data", "domain_name": "Data", "definition": "Structured representations of knowledge as entities and the typed relationships between them, expressed as a graph of nodes and labelled edges grounded in an ontology, enabling querying, reasoning, and inference over heterogeneous data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:knowledge-graphs", "labels": [ "Knowledge Graphs", "Open Knowledge Graphs" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Knowledge Representation", "RDF", "Inference", "Semantic Web", "Graph Database" ] }, { "id": "knowledge-management-system", "title": "Knowledge Management System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Knowledge Management System (KMS) is an integrated software platform that captures, organises, stores, retrieves, and distributes explicit and tacit knowledge across an organisation or community of practice. It combines document repositories, knowledge graphs, semantic search, ontologies, and collaborative authoring tools to make institutional knowledge discoverable, reusable, and actionable. A KMS supports the full knowledge lifecycle \u2014 from creation and curation through to sharing, governance, and retirement \u2014 enabling informed decision-making and reducing knowledge silos. Modern systems layer machine learning and natural language processing atop structured metadata to surface contextually relevant information at the point of need.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:knowledge-management-system", "labels": [ "Knowledge Management System" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "owl:Thing" ] }, { "id": "knowledge-management", "title": "Knowledge Management", "domain": "data", "domain_name": "Data", "definition": "Knowledge Management (KM) is the systematic discipline of identifying, capturing, organising, storing, sharing, and applying the collective explicit and tacit knowledge assets of an organisation or distributed community to improve decision-making, innovation, and operational effectiveness. It integrates organisational theory, information science, and technology \u2014 spanning knowledge repositories, ontologies, semantic search, and communities of practice \u2014 to prevent knowledge loss, enable reuse, and accelerate learning. Modern KM increasingly employs AI-augmented techniques such as knowledge graphs, retrieval-augmented generation, and large language model reasoning to make both structured and unstructured knowledge discoverable and actionable by human and autonomous agents alike. It is a mature field with established frameworks (SECI, DIKW hierarchy, Cynefin) and dedicated standards bodies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:knowledge-management", "labels": [ "Knowledge Management", "Enterprise Knowledge Management", "Knowledge Lifecycle Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "owl:Thing" ] }, { "id": "knowledge-organization-system", "title": "Knowledge Organization System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Knowledge Organization System (KOS) is any formal scheme used to organise, classify, and relate concepts within a domain \u2014 including thesauri, classification schemes, ontologies, taxonomies, and controlled vocabularies. KOS structures enable semantic interoperability by providing shared conceptual frameworks that allow heterogeneous systems to exchange and interpret information consistently. In AI infrastructure they underpin knowledge graphs, semantic search, and entity disambiguation in large language model pipelines.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-organization-system", "labels": [ "Knowledge Organization System", "Knowledge Organisation System", "KnowledgeOrganizationSystem" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "owl:Thing" ] }, { "id": "knowledge-organization", "title": "Knowledge Organization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic arrangement, classification, and representation of knowledge structures so that information can be discovered, related, and reasoned over. Knowledge organization encompasses taxonomies, thesauri, ontologies, and knowledge graphs, and underpins semantic search, question answering, and AI knowledge-base construction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-organization", "labels": [ "Knowledge Organization" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "owl:Thing" ] }, { "id": "knowledge-preservation", "title": "Knowledge Preservation", "domain": "data", "domain_name": "Data", "definition": "Knowledge preservation is the practice of capturing, organising and retaining institutional and individual knowledge so it remains accessible and usable over time. It mitigates loss from staff turnover, system obsolescence and the decay of tacit expertise by codifying knowledge into durable, retrievable forms. It encompasses documentation, archiving, knowledge bases and AI-assisted capture.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:knowledge-preservation", "labels": [ "Knowledge Preservation" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "knowledge-representation", "title": "Knowledge Representation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Knowledge Representation in AI involves the formal encoding of information about the world in a computationally tractable format that enables reasoning, inference, and decision-making. Approaches include symbolic systems (first-order logic, description logics, semantic networks), graph-based representations (knowledge graphs, ontologies), probabilistic models (Bayesian networks, Markov logic networks), and distributed representations (embeddings, neural-symbolic integration).", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:knowledge-representation", "labels": [ "Knowledge Representation", "Explicit Knowledge Representation", "Knowledge Representation System" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Ontology", "Reasoning Systems", "Knowledge Graph", "owl:Thing", "Semantic Web" ] }, { "id": "knowledge-retrieval", "title": "Knowledge Retrieval", "domain": "ai", "domain_name": "Ai", "definition": "Knowledge retrieval is the process of locating and returning relevant information from a knowledge source in response to a query or information need. It extends classical information retrieval by operating over structured knowledge, semantic representations and contextual relevance rather than keyword matching alone. It is a foundational capability for question answering, retrieval-augmented generation and conversational agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:knowledge-retrieval", "labels": [ "Knowledge Retrieval" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "knowledge-sharing", "title": "Knowledge Sharing", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Collaborative practices and platforms that facilitate the exchange of information, expertise, and learning experiences among users, encompassing immersive virtual environments, AI-enhanced interactions, distributed team workflows, and structured knowledge transfer processes across organisations and disciplines.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:knowledge-sharing", "labels": [ "Knowledge Sharing" ], "is_subclass_of": [ "Content and Assets", "Collaborative Technology" ], "wikilinks": [ "Distributed Team Collaboration", "Collaborative Technology", "metaverse" ] }, { "id": "knowledge-transfer", "title": "Knowledge Transfer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Knowledge transfer is the movement of expertise, skills and understanding from one person, team or system to another. In collaborative work it occurs through mentoring, shared practice and direct interaction, allowing tacit know-how to spread across an organisation. It is a key outcome of practices such as remote pair programming.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:knowledge-transfer", "labels": [ "Knowledge Transfer", "Tacit Knowledge Transfer" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "kotlin", "title": "Kotlin", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Kotlin is a statically-typed, cross-platform programming language developed by JetBrains that compiles to JVM bytecode, JavaScript, and native binaries via LLVM. It offers concise syntax, null safety enforced at the type level, first-class coroutine support, and seamless interoperability with Java, making it the preferred language for Android development and increasingly for server-side and multiplatform applications.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:kotlin", "labels": [ "Kotlin" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "kqml", "title": "Kqml", "domain": "ai", "domain_name": "Ai", "definition": "KQML, the Knowledge Query and Manipulation Language, is an early agent communication language and protocol for exchanging information and knowledge among software agents. It defines a set of performatives, message types grounded in speech-act theory, that express the intent of a communication such as ask, tell, subscribe or achieve. KQML influenced later standards including the FIPA Agent Communication Language.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:kqml", "labels": [ "Kqml", "KQML" ], "is_subclass_of": [ "Agent Communication Language" ], "wikilinks": [] }, { "id": "kraken", "title": "Kraken", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Kraken is a centralised cryptocurrency exchange founded in 2011 and headquartered in the United States. It provides spot trading, margin and futures trading, staking and custody services across a wide range of digital assets and fiat currencies. It is among the longer-established exchanges and is noted for its emphasis on regulatory compliance, security and the provision of fiat on-ramps in multiple jurisdictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:kraken", "labels": [ "Kraken" ], "is_subclass_of": [ "DeFi and Economics", "Digital Asset Domain" ], "wikilinks": [ "Order Book", "Custody", "Crypto Trading", "Fiat On-Ramp", "Regulatory Domain", "Financial Infrastructure Domain", "Digital Asset Domain" ] }, { "id": "kubernetes", "title": "Kubernetes", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Kubernetes (K8s) is an open-source container orchestration platform originally developed by Google and donated to the Cloud Native Computing Foundation (CNCF) in 2014. It automates the deployment, scaling, scheduling, and lifecycle management of containerised workloads across clusters of physical or virtual machines. Kubernetes abstracts infrastructure resources into declarative objects\u2014Pods, Deployments, Services, and Namespaces\u2014that describe desired state, with a control plane continuously reconciling actual state to match the specification.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:kubernetes", "labels": [ "Kubernetes", "Kubernetes RBAC" ], "is_subclass_of": [ "Orchestration" ], "wikilinks": [] }, { "id": "kuiper-belt-object", "title": "Kuiper Belt Object", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:kuiper-belt-object", "labels": [ "Kuiper Belt Object" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "kullback-leibler-divergence", "title": "Kullback-Leibler Divergence", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Kullback-Leibler (KL) divergence is a measure from information theory that quantifies how much one probability distribution differs from a second, reference distribution, expressed as the expected excess number of bits required to encode samples from the first using a code optimised for the second. It is non-negative, equal to zero only when the two distributions are identical, and is asymmetric, so it is a divergence rather than a true metric. KL divergence is central to machine learning, appearing in cross-entropy loss, variational inference, the evidence lower bound of variational autoencoders, and regularisation of policy updates in reinforcement learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:kullback-leibler-divergence", "labels": [ "Kullback-Leibler Divergence" ], "is_subclass_of": [ "Information Theory" ], "wikilinks": [] }, { "id": "kyoto-protocol", "title": "Kyoto Protocol", "domain": "governance", "domain_name": "Governance", "definition": "The 1997 international treaty under the UN Framework Convention on Climate Change that, for the first time, set legally binding greenhouse gas emission reduction targets for industrialised (Annex B) countries, and created the flexible market mechanisms \u2014 International Emissions Trading, the Clean Development Mechanism, and Joint Implementation \u2014 that founded the compliance carbon market. In force from 2005 with commitment periods spanning 2008-2020, it was superseded as the primary climate treaty by the Paris Agreement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:kyoto-protocol", "labels": [ "Kyoto Protocol" ], "is_subclass_of": [ "Climate Policy" ], "wikilinks": [ "Climate Policy", "Paris Agreement", "Compliance Carbon Market", "Emissions Trading Scheme", "Carbon Credits" ] }, { "id": "l-402-protocol", "title": "L402 Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "L402 Protocol is a web-native payment authentication specification that combines HTTP 402 status responses, [[Lightning Network]] BOLT11 invoices, and cryptographic macaroon tokens to enable pay-per-request access control for APIs and web resources. A server responding with 402 embeds a Lightning invoice and a partially-constructed macaroon; the client pays the invoice, receives the payment preimage, and embeds that preimage as a macaroon caveat to form a valid bearer credential. The protocol enables metered, machine-to-machine micropayments without requiring user accounts, subscriptions, or traditional payment rails, making it well-suited to AI agent infrastructure, streaming data services, and censorship-resistant content monetisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:l-402-protocol", "labels": [ "L402 Protocol" ], "is_subclass_of": [ "L402" ], "wikilinks": [ "Lightning Network", "Micropayment", "Authentication", "L402" ] }, { "id": "l-402", "title": "L402", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An authentication and payment protocol that combines the HTTP 402 Payment Required status code with Lightning Network invoices and macaroon-based authorisation tokens, enabling per-request access control and machine-to-machine micropayments without user accounts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:l-402", "labels": [ "L402" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning Network", "Authentication", "Micropayment", "L402 Protocol" ] }, { "id": "ldap", "title": "LDAP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Lightweight Directory Access Protocol (LDAP) is an open, vendor-neutral application protocol for accessing and maintaining distributed directory information services over an IP network, standardised in RFC 4511 (2006) as a simplification of the X.500 Directory Access Protocol. LDAP organises directory entries in a hierarchical tree structure (Directory Information Tree, DIT) where entries contain typed attribute-value pairs conforming to object class schemas, and supports operations for search, add, modify, delete, compare, and bind (authentication). It serves as the foundational protocol for enterprise identity management, enabling centralised authentication, authorisation, and user attribute storage across heterogeneous systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ldap", "labels": [ "LDAP" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "ldp-container", "title": "LDP Container", "domain": "data", "domain_name": "Data", "definition": "An LDP Container is a resource defined by the W3C Linked Data Platform specification that groups and manages other linked-data resources, exposing them through HTTP affordances for creation, retrieval, and deletion. Containers come in basic, direct, and indirect variants that differ in how membership triples are generated. They provide the hierarchical, REST-like structure that Solid pods and other LDP servers use to organise data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ldp-container", "labels": [ "LDP Container" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ldpc-codes", "title": "LDPC Codes", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Low-Density Parity-Check (LDPC) codes are a class of linear error-correcting codes defined by a sparse parity-check matrix and decoded with iterative belief-propagation algorithms. They approach the Shannon capacity limit while remaining computationally tractable, making them a dominant forward-error-correction scheme in modern communications. LDPC codes are mandated in standards such as Wi-Fi 6, 5G NR data channels, and DVB-S2.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ldpc-codes", "labels": [ "LDPC Codes" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "lei-system", "title": "LEI System", "domain": "governance", "domain_name": "Governance", "definition": "The LEI System is the Global Legal Entity Identifier System, a federated infrastructure overseen by the GLEIF that issues, maintains, and publishes Legal Entity Identifiers and their reference data. It comprises a Regulatory Oversight Committee, accredited Local Operating Units, and an open global directory queryable by anyone. The system provides a single authoritative source for identifying legal entities in financial markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lei-system", "labels": [ "LEI System", "Legal Entity Identifier" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "lf-decentralized-trust", "title": "LF Decentralized Trust", "domain": "blockchain", "domain_name": "Blockchain", "definition": "LF Decentralized Trust is a Linux Foundation umbrella organisation, formed in 2024 from the expansion of Hyperledger, that hosts open-source projects for blockchain, distributed ledger, identity, and decentralised technologies. It provides neutral governance, shared infrastructure, and community processes for projects such as Hyperledger Fabric, Besu, and Indy. The body coordinates standards-aligned, vendor-neutral development of enterprise trust infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:lf-decentralized-trust", "labels": [ "LF Decentralized Trust", "Linux Foundation Decentralized Trust" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "lime", "title": "LIME", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Local Interpretable Model-agnostic Explanations, a technique that explains individual predictions of any machine learning model by fitting a simple interpretable model to perturbed samples around the instance of interest.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lime", "labels": [ "LIME", "LIME (Local Interpretable Model-Agnostic Explanations)" ], "is_subclass_of": [ "Model Interpretability" ], "wikilinks": [ "Machine Learning", "Explainable AI", "SHAP", "Interpretability", "Model Interpretability" ] }, { "id": "llm-agents", "title": "LLM Agents", "domain": "ai", "domain_name": "Ai", "definition": "LLM Agents are autonomous software systems that use large language models as their core reasoning engine to perceive inputs, plan multi-step actions, invoke external tools, and pursue goals over extended horizons with minimal per-step human oversight. They extend base language models with memory, tool use, and feedback loops to accomplish tasks that require sequential decision-making.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:llm-agents", "labels": [ "LLM Agents" ], "is_subclass_of": [ "Agentic AI" ], "wikilinks": [] }, { "id": "llm-application-framework", "title": "LLM Application Framework", "domain": "ai", "domain_name": "Ai", "definition": "A software library that provides abstractions for building applications on top of large language models, including prompt management, tool integration, retrieval and chaining of model calls.", "entityType": "Class", "qualityScore": 0.6, "maturity": "growing", "iri": "urn:ngm:class:llm-application-framework", "labels": [ "LLM Application Framework", "LLM Agent Framework", "LLM Application Stack" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [ "Large Language Models", "Retrieval-Augmented Generation", "LangChain" ] }, { "id": "llm-application-frameworks", "title": "LLM Application Frameworks", "domain": "ai", "domain_name": "Ai", "definition": "General-purpose software frameworks for composing [[Large Language Models]] capabilities into production applications through chains, retrievers, [[Tool Use]], [[Retrieval-Augmented Generation]] pipelines, stateful workflows, and [[Multi-Agent Systems]] orchestration \u2014 including LangChain, LangGraph, LlamaIndex, Semantic Kernel, Microsoft Agent Framework, Mastra, Agno, n8n, Dify, Langflow, and Flowise \u2014 providing the integration, observability, and deployment infrastructure that connects model inference to real application data, external APIs, and human approval workflows.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:llm-application-frameworks", "labels": [ "LLM Application Frameworks" ], "is_subclass_of": [ "Agent Harness", "Workflow Automation" ], "wikilinks": [ "Agent Harness", "Agentic AI", "Agentic Workflow", "Multi-Agent Orchestration Frameworks", "Agent Development SDKs", "Agent Execution Sandboxes", "Retrieval-Augmented Generation", "Agentic RAG", "Vector Database", "Large Language Models", "Tool Use", "Function Calling", "Model Context Protocol", "MCP Server", "Chain of Thought", "Prompt Engineering", "Reasoning", "Tool Call Loop", "Agent Memory", "Agent Loop" ] }, { "id": "llm-evaluation", "title": "LLM Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "LLM evaluation is the set of methodologies, benchmarks, and metrics used to assess the capabilities, safety, and reliability of large language models, spanning automated benchmark suites, human preference judgements, and task-specific test harnesses.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:llm-evaluation", "labels": [ "LLM Evaluation" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation" ], "wikilinks": [] }, { "id": "llm-orchestration", "title": "LLM Orchestration", "domain": "ai", "domain_name": "Ai", "definition": "The coordination of multiple calls to one or more large language models, together with tools, retrieval and control logic, to accomplish a task that a single prompt cannot reliably handle.", "entityType": "Class", "qualityScore": 0.6, "maturity": "growing", "iri": "urn:ngm:class:llm-orchestration", "labels": [ "LLM Orchestration" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [ "Large Language Models", "Tool Use", "Retrieval-Augmented Generation", "LangChain" ] }, { "id": "llm-token-expenditure-index", "title": "LLM Token Expenditure Index", "domain": "economics", "domain_name": "Economics", "definition": "A market metric that tracks the expenditure or usage-weighted average price of Large Language Model tokens, distinct from total volume or aggregate spend.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:llm-token-expenditure-index", "labels": [ "LLM Token Expenditure Index" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "lnd", "title": "LND", "domain": "blockchain", "domain_name": "Blockchain", "definition": "LND (Lightning Network Daemon) is a production-grade, open-source implementation of a Bitcoin Lightning Network node developed by Lightning Labs. It manages the full lifecycle of payment channels \u2014 from on-chain funding and cooperative or force-close settlement to off-chain routing of multi-hop payments \u2014 exposing a gRPC and REST API that downstream wallets, exchanges, and applications consume. LND implements the BOLT (Basis of Lightning Technology) protocol specifications and employs Hash Time-Locked Contracts (HTLCs) to ensure atomic, trust-minimised payment routing across a peer-to-peer mesh of channels anchored to the Bitcoin base layer.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:lnd", "labels": [ "LND" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning Network", "Payment Channel", "Bitcoin" ] }, { "id": "lnp-bp-standards-association", "title": "LNP-BP Standards Association", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The LNP-BP Standards Association is a non-profit body that develops and maintains specifications for Bitcoin and Lightning Network protocols, most notably the RGB smart-contract system and client-side validation. It curates the LNPBP specification series covering layer-2 and layer-3 constructs that extend Bitcoin without altering its base consensus. The association coordinates the engineers and researchers building these client-side-validated protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:lnp-bp-standards-association", "labels": [ "LNP-BP Standards Association", "LNP/BP Association", "LNP/BP Standards" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "lp-token", "title": "LP Token", "domain": "finance", "domain_name": "Finance", "definition": "An LP (Liquidity Provider) token is a fungible cryptographic token minted by a decentralised exchange or liquidity pool smart contract to represent a depositor's proportional ownership stake in a pool's combined assets, accrued fees, and associated yield. LP tokens serve as receipts that can be redeemed to withdraw the underlying liquidity position at any time.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:lp-token", "labels": [ "LP Token" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [] }, { "id": "lsm-tree", "title": "LSM Tree", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An LSM Tree, or Log-Structured Merge Tree, is a write-optimised data structure that buffers writes in memory and periodically flushes them as sorted, immutable files on disk, later merging those files through background compaction. It trades read amplification for high write throughput, making it the storage engine of choice for databases with write-heavy workloads. It is used internally by many key-value stores, time-series databases, and graph databases.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:lsm-tree", "labels": [ "LSM Tree" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "lstm", "title": "LSTM", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Long Short-Term Memory (LSTM) is a specialised recurrent neural network architecture introduced by Hochreiter and Schmidhuber in 1997, designed to learn long-range temporal dependencies in sequential data. It employs a gated cell state mechanism\u2014comprising input, forget, and output gates\u2014that allows gradients to flow across many time steps without vanishing or exploding, overcoming the principal failure mode of vanilla recurrent neural networks. LSTMs encode contextual information in a learnable cell state that persists across sequence steps, making them highly effective for variable-length sequence modelling tasks such as language modelling, machine translation, and time-series forecasting. Although largely superseded by Transformer-based architectures for many NLP tasks, LSTMs remain widely deployed in low-latency, resource-constrained, and streaming scenarios.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:lstm", "labels": [ "LSTM" ], "is_subclass_of": [ "Recurrent Neural Network" ], "wikilinks": [ "Backpropagation", "Speech Recognition", "Transformer", "Recurrent Neural Network" ] }, { "id": "label-smoothing", "title": "Label Smoothing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A regularisation technique that replaces hard one-hot labels with soft targets by allocating small probability mass to incorrect classes. Label smoothing prevents overconfident predictions and improves model calibration and generalisation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:label-smoothing", "labels": [ "Label Smoothing" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "labelled-data", "title": "Labelled Data", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Labelled data is data annotated with target values, categories or structured tags that supervised machine learning models learn to predict. Labels may be produced by human annotators, expert review or semi-automated pipelines, and their quality directly bounds achievable model accuracy. Labelled data underpins tasks such as model evaluation, semantic parsing and sequence labelling, where a defined label schema drives both training and assessment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:labelled-data", "labels": [ "Labelled Data" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "labelled-dataset", "title": "Labelled Dataset", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A labelled dataset is a collection of data examples each paired with one or more target annotations that specify the correct output for a learning task. The labels constitute the supervisory signal that allows a model to learn the mapping from inputs to outputs during training. Label quality, coverage and balance strongly determine the performance and fairness of the resulting model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:labelled-dataset", "labels": [ "Labelled Dataset" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "labor-displacement", "title": "Labor Displacement", "domain": "economics", "domain_name": "Economics", "definition": "The structural reduction in workforce demand or job roles resulting from the automation of tasks by artificial intelligence and other technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:labor-displacement", "labels": [ "Labor Displacement" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "labor-statistics-revision", "title": "Labor Statistics Revision", "domain": "economics", "domain_name": "Economics", "definition": "The methodological process by which statistical agencies like the BLS adjust previously reported employment data to correct for sampling errors, benchmarking discrepancies, and reporting lags.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:labor-statistics-revision", "labels": [ "Labor Statistics Revision" ], "is_subclass_of": [ "Digital Economy" ], "wikilinks": [] }, { "id": "lagrange-point", "title": "Lagrange Point", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lagrange-point", "labels": [ "Lagrange Point" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lakehouse-architecture", "title": "Lakehouse Architecture", "domain": "data", "domain_name": "Data", "definition": "Lakehouse architecture is a data management pattern that combines the low-cost, open storage of a data lake with the transactional reliability and performance of a data warehouse. It layers ACID transactions, schema enforcement and indexing over inexpensive object storage using open table formats such as Delta Lake, Apache Iceberg or Hudi. This unifies analytics, business intelligence and machine learning on a single copy of data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:lakehouse-architecture", "labels": [ "Lakehouse Architecture", "Data Lakehouse" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "lambda-architecture", "title": "Lambda Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Lambda Architecture is a data-processing design pattern that handles massive quantities of data by combining a batch layer for comprehensive, accurate computation over the full dataset with a speed (or streaming) layer that processes recent data with low latency. A serving layer merges results from both layers so queries return a unified view that is eventually consistent yet responsive in near real time. The pattern accepts the operational cost of maintaining two parallel code paths in exchange for fault tolerance, reprocessing capability, and the reconciliation of historical accuracy with live freshness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lambda-architecture", "labels": [ "Lambda Architecture" ], "is_subclass_of": [ "Data Architecture" ], "wikilinks": [] }, { "id": "land-cover-classification", "title": "Land Cover Classification", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:land-cover-classification", "labels": [ "Land Cover Classification" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "land-cover-monitoring", "title": "Land Cover Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:land-cover-monitoring", "labels": [ "Land Cover Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "land-economics", "title": "Land Economics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Land Economics is the economic framework governing virtual land ownership, valuation, development, and transaction within metaverse environments, characterised by artificial scarcity, spatial positioning premiums, and speculative markets. It encompasses primary sales, secondary peer-to-peer trading, rental systems, platform taxation models, and zoning policies that together determine how virtual parcels accumulate and lose value.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:land-economics", "labels": [ "Land Economics" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Non-Fungible Token", "Digital Ownership", "Metaverse", "Spatial Computing", "Virtual Economy", "Virtual Labor" ] }, { "id": "land-ice", "title": "Land Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:land-ice", "labels": [ "Land Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "land-parcel", "title": "Land Parcel", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Discrete units of virtual real estate within metaverse platforms, represented as NFTs that provide proof of ownership for programmable digital spaces where users can build, socialise, host events, and conduct commercial activities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:land-parcel", "labels": [ "Land Parcel" ], "is_subclass_of": [ "Platform and Environment", "Virtual Asset" ], "wikilinks": [ "Metaverse Development", "metaverse", "Virtual Asset" ] }, { "id": "land-scarcity", "title": "Land Scarcity", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An economic design principle applied in virtual worlds and metaverse platforms whereby the total supply of digital land parcels is hard-capped by the platform protocol, creating artificial scarcity analogous to finite physical real estate. This scarcity underpins virtual real estate markets, drives speculative investment, and enables monetisation through NFT-based ownership of parcels in platforms such as Decentraland and The Sandbox. The principle links metaverse spatial design to tokenomics, digital asset valuation, and broader virtual economy dynamics.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:land-scarcity", "labels": [ "Land Scarcity" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Virtual Economics", "Metaverse", "MetaverseDomain" ] }, { "id": "land-surface-temperature", "title": "Land Surface Temperature", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:land-surface-temperature", "labels": [ "Land Surface Temperature" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "land-use-classification", "title": "Land Use Classification", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:land-use-classification", "labels": [ "Land Use Classification" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lander", "title": "Lander", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lander", "labels": [ "Lander" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "landslide-monitoring", "title": "Landslide Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:landslide-monitoring", "labels": [ "Landslide Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "landvaettir-generative-ai-art-research", "title": "Landvaettir Generative AI Art Research", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Landv\u00e6ttir is a practice-based art research project and portfolio documenting the bootstrapping era of Generative AI and AGI through digital landscapes, virtual worlds, and AI-generated imagery. Drawing on Scenism, Romanticism, and environmental psychology, it explores how AI systems trained on cultural visual heritage may dream or hallucinate the virtual landscapes that seeded their intelligence.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:landvaettir-generative-ai-art-research", "labels": [ "Landvaettir Generative AI Art Research", "Landv\u00e6ttir" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "PEOPLE" ] }, { "id": "lang-chain", "title": "LangChain", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "LangChain is an open-source Python and TypeScript framework for composing Large Language Model applications as chains of modular components \u2014 prompt templates, LLM wrappers, output parsers, memory stores, retrieval augmented generation pipelines, and tool-calling agents \u2014 providing a unified ...", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:lang-chain", "labels": [ "LangChain" ], "is_subclass_of": [ "AI Agent System", "LLM Application Framework", "Agent Orchestration Framework" ], "wikilinks": [ "AgentLayer", "Agent Orchestration Framework", "AI Domain", "Anthropic Claude API", "Chroma Vector Database", "Conversational Memory", "Embedding Model", "FAISS", "LangChain Agent", "LangChain Chain", "LangChain GitHub Repository", "LangChain Memory", "LangChain Retriever", "LangChain Tool", "LangGraph Documentation", "LangGraph Workflow", "LangSmith Observability", "LangSmith Platform", "Large Language Model", "LCEL LangChain Expression Language" ] }, { "id": "lang-graph", "title": "LangGraph", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "LangGraph is a library for building stateful, multi-step language model applications by representing control flow as a graph. It is part of the LangChain ecosystem and supports cyclic agent workflows.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:lang-graph", "labels": [ "LangGraph", "LangGraph Workflow" ], "is_subclass_of": [ "Agentic Workflow" ], "wikilinks": [ "Language Model", "State Machine", "Multi-Agent Coordination", "Agentic Workflow", "AI Agent", "Tool Use" ] }, { "id": "language-model-alignment", "title": "Language Model Alignment", "domain": "ai", "domain_name": "Ai", "definition": "Language model alignment is the set of techniques used to make a language model's behaviour conform to human intentions, values and safety constraints. It typically follows pretraining with supervised fine-tuning and preference-based optimisation so that outputs are helpful, honest and harmless. Methods include reinforcement learning from human feedback and direct preference optimization.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:language-model-alignment", "labels": [ "Language Model Alignment" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "language-model", "title": "Language Model", "domain": "ai", "domain_name": "Ai", "definition": "A language model is a probabilistic model of natural language that assigns a probability distribution over sequences of tokens \u2014 words, subwords, or characters \u2014 enabling both likelihood estimation of observed text and generation of new text via sampling from learned conditional distributions. Modern large language models are deep neural networks based on the transformer architecture, trained on vast corpora through a self-supervised next-token prediction objective. They acquire implicit representations of syntax, semantics, world knowledge, and reasoning patterns entirely from this training signal, and can be adapted to downstream tasks through fine-tuning, instruction tuning, or prompt engineering. The paradigm has displaced earlier n-gram and recurrent neural network approaches, becoming the dominant framework for natural language processing across virtually all applied domains.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:language-model", "labels": [ "Language Model", "Multi-language Model", "Statistical Language Model" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "language-modeling", "title": "Language Modeling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Language Modeling is the fundamental NLP task of learning probability distributions over sequences of words or tokens to predict the likelihood of text sequences and generate plausible continuations. Language models underpin virtually all modern NLP applications through pre-training on massive text corpora, capturing syntactic structure, semantic relationships, and world knowledge that transfer to downstream tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:language-modeling", "labels": [ "Language Modeling" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Large Language Model", "Omniverse", "OpenUSD", "presentation", "BERT", "Diffusion Models", "Google", "GPT", "Knowledge Graphing", "Large Language Models", "Logseq", "Metaverse and Telecollaboration", "MetaverseDomain", "Natural Language Processing", "NVIDIA Omniverse", "OpenAI", "Open Generative AI tools", "Proprietary Large Language Models", "Self Attention", "Text Generation" ] }, { "id": "language-translation", "title": "Language Translation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Language translation is the automated or human-assisted conversion of text or speech from a source language into a target language whilst preserving semantic meaning, tone, and cultural context. Modern neural machine translation systems, built on transformer architectures and large language models, enable real-time multilingual communication across distributed collaboration platforms and metaverse environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:language-translation", "labels": [ "Language Translation" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "laplace-mechanism", "title": "Laplace Mechanism", "domain": "security", "domain_name": "Security", "definition": "The foundational noise mechanism of differential privacy, which releases a numeric query result after adding random noise drawn from a Laplace distribution whose scale equals the query's L1 sensitivity divided by the privacy budget epsilon; introduced by Dwork, McSherry, Nissim, and Smith in 2006, it achieves pure epsilon-differential privacy with no failure probability, and remains the standard mechanism for counts, sums, and histograms in privacy-preserving analytics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:laplace-mechanism", "labels": [ "Laplace Mechanism" ], "is_subclass_of": [ "Noise Mechanisms" ], "wikilinks": [ "Noise Mechanisms", "Differential Privacy", "Gaussian Mechanism" ] }, { "id": "laplace-transform", "title": "Laplace Transform", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Laplace transform is an integral transform that converts a function of time into a function of a complex frequency variable, turning linear differential equations into algebraic equations that are typically easier to solve. It is a standard analytical tool in control theory and feedback-loop analysis, where transfer functions expressed in the Laplace domain describe how a system responds to inputs. The Fourier transform is a closely related special case obtained by restricting the transform variable to the imaginary axis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:laplace-transform", "labels": [ "Laplace Transform" ], "is_subclass_of": [ "Calculus" ], "wikilinks": [] }, { "id": "large-language-model-training", "title": "Large Language Model Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Large Language Model Training is the computational process of optimising the parameters of a transformer-based neural network with billions to trillions of weights on web-scale text corpora using autoregressive next-token prediction objectives, followed by instruction tuning and reinforcement learning from human feedback (RLHF) alignment stages. The process requires distributed training across thousands of GPU or TPU accelerators coordinated through data, tensor, and pipeline parallelism, consuming petabytes of training data and megawatt-hours of electrical energy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:large-language-model-training", "labels": [ "Large Language Model Training" ], "is_subclass_of": [ "Pre Training" ], "wikilinks": [] }, { "id": "large-language-model", "title": "Large Language Model", "domain": "ai", "domain_name": "Ai", "definition": "A large language model (LLM) is a deep neural network \u2014 almost universally based on the [[Transformer Architecture]] \u2014 trained via self-supervised next-token prediction on web-scale corpora of text (and often code, mathematics, and structured data), resulting in a system that assigns a probability distribution over token sequences and can generate coherent, contextually appropriate continuations. Scale \u2014 both in parameter count (billions to hundreds of billions) and training tokens (trillions) \u2014 is the defining characteristic that distinguishes LLMs from earlier, smaller language models, and is the proximate cause of qualitative capability jumps such as in-context learning, instruction following, chain-of-thought reasoning, and emergent generalisation across domains. LLMs are typically released as base pretrained models that are subsequently aligned to human preferences through supervised fine-tuning and reinforcement learning from human feedback, producing the instruction-following assistants widely deployed in consumer and enterprise applications. The paradigm has become the de facto foundation for natural language processing, code synthesis, autonomous agent planning, and multimodal AI systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:large-language-model", "labels": [ "Large Language Model" ], "is_subclass_of": [ "Foundation Model", "Foundation Model Layer" ], "wikilinks": [] }, { "id": "large-language-models", "title": "Large Language Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Large Language Models (LLMs) are Foundation Models with s to s of parameters trained on massive text corpora using Transformer architectures and Self-Supervised Learning, capable of performing diverse Natural Language Processing tasks through Few-Shot Learning,", "entityType": "Class", "qualityScore": 0.92, "maturity": "established", "iri": "urn:ngm:class:large-language-models", "labels": [ "Large Language Models", "LLM", "Large Language Model" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network", "Large-Scale Pretrained Foundation Model", "Deep Learning", "Machine Learning Discipline" ], "wikilinks": [ "AlgorithmicLayer", "Anthropic", "Autonomous Economic Agents", "DataManagementDomain", "Distributed Training", "Embedding", "Few-Shot Learning", "GPU Compute", "L402 Payment Protocol", "Large-Scale Datasets", "Meta AI Research", "Model Optimization", "Natural Language Understanding", "NIST AI Standards", "OpenAI Research", "Self-Supervised Learning", "Zero-Shot Learning", "AI-GroundedDomain", "Artificial Intelligence", "Attention Mechanism" ] }, { "id": "large-scale-compute", "title": "Large-Scale Compute", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Large-scale compute refers to the aggregation of massive computational resources \u2014 spanning thousands to hundreds of thousands of processors, accelerators, and memory modules \u2014 into coordinated systems capable of executing workloads at magnitudes that are impractical for single machines or small clusters, including the training of frontier AI models, climate simulation, drug discovery, and national security applications. It encompasses the physical infrastructure (data centres, power supply, cooling), interconnect fabrics (InfiniBand, NVLink, custom optical networks), software orchestration layers (job schedulers, distributed training frameworks), and the operational expertise required to sustain utilisation and reliability at exascale levels.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:large-scale-compute", "labels": [ "Large-Scale Compute", "Large Scale Compute" ], "is_subclass_of": [ "Computational Infrastructure" ], "wikilinks": [] }, { "id": "large-scale-corpus", "title": "Large-Scale Corpus", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A large-scale corpus is a training dataset comprising an extremely large volume of text or other sequential data, typically gathered from web crawls, digitised books, or code repositories, and used to pretrain large neural language models. Its scale is a primary determinant of model capability under empirically observed scaling laws, alongside model parameter count and compute budget. Curation and deduplication of a large-scale corpus materially affect downstream model quality and the presence of memorised content.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:large-scale-corpus", "labels": [ "Large-Scale Corpus", "Large Scale Corpus" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "large-scale-datasets", "title": "Large-Scale Datasets", "domain": "ai", "domain_name": "Ai", "definition": "Large-scale datasets are very large collections of data, often spanning billions of examples and many terabytes, assembled to train modern machine learning models. They are typically aggregated from web crawls, public corpora and curated sources, then filtered, deduplicated and tokenised. Their scale, diversity and quality are primary determinants of the capabilities of large language and generative models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:large-scale-datasets", "labels": [ "Large-Scale Datasets", "Large-Scale Audio Datasets", "Large-Scale Dataset" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "large-scale-pretrained-foundation-model", "title": "Large-Scale Pretrained Foundation Model", "domain": "ai", "domain_name": "Ai", "definition": "Foundation models are large-scale neural networks trained on broad, diverse datasets via self-supervised learning that acquire general-purpose representations transferable to a wide range of downstream tasks. They are characterised by massive parameter counts, emergent capabilities not explicitly trained for, and the ability to be fine-tuned or prompted for specialised applications. Prominent examples include GPT-4, BERT, CLIP, and Stable Diffusion, spanning language, vision, and multimodal domains. Their scale and generality make them qualitatively distinct from narrow task-specific models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:large-scale-pretrained-foundation-model", "labels": [ "Large-Scale Pretrained Foundation Model", "Foundation Model", "Foundation Models for Robotics" ], "is_subclass_of": [ "Machine Learning Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "large-scale-pretraining", "title": "Large-Scale Pretraining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Large-scale pretraining is the training of a high-capacity neural network on a very large, broad corpus using a self-supervised objective, producing a general-purpose foundation model before any task-specific adaptation. It typically optimises an objective such as next-token prediction over web-scale text or paired multimodal data, learning transferable representations. The resulting model is later fine-tuned or prompted for downstream tasks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:large-scale-pretraining", "labels": [ "Large-Scale Pretraining" ], "is_subclass_of": [ "Neural Network Training" ], "wikilinks": [] }, { "id": "large-scale-training", "title": "Large-Scale Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Machine learning training conducted at a scale that exceeds the capacity of a single accelerator, requiring workloads to be distributed across many GPUs or nodes. Large-scale training covers the full range of distributed regimes\u2014pretraining, large fine-tuning runs, and reinforcement learning from feedback\u2014and rests on parallelism strategies (data, tensor, pipeline, and expert parallelism), high-bandwidth interconnects, checkpointing, and fault tolerance to keep thousands of accelerators productively synchronised for days or weeks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:large-scale-training", "labels": [ "Large-Scale Training" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [ "Model Training", "Data Parallelism", "Foundation Model" ] }, { "id": "laser-altimetry", "title": "Laser Altimetry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:laser-altimetry", "labels": [ "Laser Altimetry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "laser-communication-terminal", "title": "Laser Communication Terminal", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:laser-communication-terminal", "labels": [ "Laser Communication Terminal" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "laser-scanner", "title": "Laser Scanner", "domain": "robotics", "domain_name": "Robotics", "definition": "A Laser Scanner is a sensor that emits laser pulses and measures the time-of-flight or phase-shift of the returned signal to compute precise distance measurements across a scene, generating dense point clouds. In robotics and spatial computing, laser scanners serve as primary perception instruments for environment mapping, obstacle detection, and the construction of digital twins via LiDAR-based SLAM pipelines.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:laser-scanner", "labels": [ "Laser Scanner" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "last-mile-delivery", "title": "Last-Mile Delivery", "domain": "robotics", "domain_name": "Robotics", "definition": "Last-mile delivery is the final stage of a logistics chain in which goods move from a transport hub or distribution centre to the end recipient. It is typically the most expensive, time-sensitive and operationally complex segment because of fragmented destinations and traffic constraints. It is a primary target for automation through delivery robots, drones and route-optimisation systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:last-mile-delivery", "labels": [ "Last-Mile Delivery" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "latency-management-protocol", "title": "Latency Management Protocol", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A systematic process for monitoring, measuring, and minimizing network delay in interactive metaverse sessions through adaptive techniques including traffic prioritization, predictive buffering, and dynamic routing optimization.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:latency-management-protocol", "labels": [ "Latency Management Protocol", "Latency Management" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Bandwidth Management", "Congestion Control", "Dynamic Routing", "ETSI ENI 008", "IEEE P2048-7", "Latency Measurement Tools", "Latency Monitoring System", "Low-Latency Interaction", "Network Performance Management", "Network Performance Metrics", "Packet Scheduling", "Predictable Performance", "Predictive Buffering", "Priority Policies", "Quality of Experience Framework", "Quality of Service Manager", "Real-Time Responsiveness", "Routing Algorithms", "Smooth User Experience", "Traffic Prioritization Engine" ] }, { "id": "latency-optimisation", "title": "Latency Optimisation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Latency optimisation is the set of techniques used to minimise the time between a request and its corresponding response in networked and distributed systems. It targets propagation, processing, queueing, and serialisation delays through measures such as edge placement, caching, protocol tuning, and concurrency. Low latency is critical for real-time communication, interactive collaboration, and responsive user experiences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:latency-optimisation", "labels": [ "Latency Optimisation" ], "is_subclass_of": [ "Distributed Collaboration", "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "latency-aware-edge-ai", "title": "Latency-Aware Edge AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Latency-Aware Edge AI is a design paradigm for machine learning systems deployed at the network edge that dynamically adapts inference strategies, model selection, and compute offloading decisions to satisfy hard or soft real-time response-time deadlines. Such systems continuously monitor available time budgets, device load, and network conditions, trading accuracy for speed when necessary to maintain service-level objectives. Applications span autonomous vehicles, mobile augmented reality, and industrial robotics, where missed deadlines carry safety or quality consequences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:latency-aware-edge-ai", "labels": [ "Latency-Aware Edge AI" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "latency-sensitive-workflows", "title": "Latency-Sensitive Workflows", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "User interactions and tasks where the speed of AI response is a critical factor in usability and value, such as real-time coding assistance or interactive dialogue.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:latency-sensitive-workflows", "labels": [ "Latency-Sensitive Workflows" ], "is_subclass_of": [ "AI Inference" ], "wikilinks": [] }, { "id": "latency", "title": "Latency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Performance metric representing the time delay between a stimulus (user action, network request, or computation trigger) and the corresponding system response; a fundamental constraint in networked, real-time, and interactive systems spanning communication networks, distributed computing, and immersive environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:latency", "labels": [ "Latency", "Latency Measurement", "Latency Reduction", "Latency Tolerance", "Memory Latency" ], "is_subclass_of": [ "Network Performance Metrics" ], "wikilinks": [ "Bandwidth", "ETSI ARF 010", "Measurement Tools", "Monitoring System", "Network Performance Metrics", "Processing Delay", "Propagation Delay", "Quality Assessment", "Routing Protocol", "SLA Monitoring", "Timestamp Synchronization", "User Experience Tuning", "InfrastructureDomain", "Network Infrastructure", "Network Layer", "Performance Optimization", "Physical Layer", "Quality of Service" ] }, { "id": "latent-diffusion-model-training", "title": "Latent Diffusion Model Training", "domain": "ai", "domain_name": "Ai", "definition": "Stable Diffusion training is the process of fitting a latent text-to-image diffusion model by teaching a denoising network to reverse a gradual noising process in a compressed latent space, conditioned on text embeddings. Operating in latent rather than pixel space sharply reduces compute and memory, while techniques such as fine-tuning, LoRA, and DreamBooth adapt a base model to new styles or subjects. It underpins much of open-weight generative image and video tooling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:latent-diffusion-model-training", "labels": [ "Latent Diffusion Model Training" ], "is_subclass_of": [ "Generative Model", "Model Training" ], "wikilinks": [] }, { "id": "latent-diffusion", "title": "latent diffusion", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Latent Diffusion Models (LDMs) are a class of generative models that perform the iterative denoising diffusion process within the compressed latent space of a pre-trained variational autoencoder (VAE), rather than directly in high-dimensional pixel space. By encoding images into a compact, semantically rich latent representation, LDMs dramatically reduce training and inference compute whilst preserving perceptual quality, because the VAE absorbs the high-frequency, imperceptual detail that would otherwise burden the diffusion process. Conditioning on text, image, or other modalities is achieved via cross-attention layers inside a U-Net denoising backbone, enabling high-fidelity text-to-image synthesis, image editing, and multimodal generation at practical hardware budgets. Stable Diffusion, the most widely deployed open-source implementation, demonstrated that consumer GPUs could run production-quality image synthesis, catalysing a broad ecosystem of fine-tuning methods and downstream applications.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:latent-diffusion", "labels": [ "Latent Diffusion", "Latent Diffusion Architecture", "Latent Diffusion Model", "Latent Diffusion Sampling", "Latent Video Diffusion" ], "is_subclass_of": [ "Diffusion Model", "Diffusion Models" ], "wikilinks": [] }, { "id": "latent-space", "title": "Latent Space", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Latent space is a learned, typically lower-dimensional continuous space in which a model represents its inputs as vectors, positioning semantically similar inputs closer together. It is produced by an encoder or embedding process and is where operations such as interpolation, arithmetic and sampling are performed before a decoder network reconstructs or generates output in the original data space. Creative AI systems exploit structure in latent space to generate novel, coherent outputs by sampling or navigating it directly.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:latent-space", "labels": [ "Latent Space" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "latent-variable-model", "title": "Latent Variable Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A latent variable model is a statistical model that explains observed data in terms of unobserved (latent) variables, which capture hidden structure such as cluster membership, low-dimensional factors, or underlying states. By positing latent causes, these models compactly represent complex distributions and support tasks like density estimation, dimensionality reduction, and generation. Inference recovers distributions over the latent variables given the observations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:latent-variable-model", "labels": [ "Latent Variable Model" ], "is_subclass_of": [ "Probabilistic Model" ], "wikilinks": [] }, { "id": "lattice-cryptography", "title": "Lattice Cryptography", "domain": "security", "domain_name": "Security", "definition": "A family of cryptographic constructions whose security rests on the hardness of computational problems over high-dimensional lattices, such as learning with errors. These problems are believed to resist attacks by quantum computers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:lattice-cryptography", "labels": [ "Lattice Cryptography", "Lattice Trapdoor", "Lattice-Based Cryptography" ], "is_subclass_of": [ "Post-Quantum Cryptography" ], "wikilinks": [ "Cryptography", "Encryption", "Digital Signature", "Cryptographic Primitive", "Quantum Computing", "Post-Quantum Cryptography", "https://csrc.nist.gov/projects/post-quantum-cryptography" ] }, { "id": "launch-phase", "title": "Launch Phase", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:launch-phase", "labels": [ "Launch Phase" ], "is_subclass_of": [ "Mission Phase" ], "wikilinks": [] }, { "id": "launch-vehicle-stage", "title": "Launch Vehicle Stage", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A launch vehicle stage is a complete launcher element that produces thrust during a defined phase of a launch mission.[^1] It is more than an engine or propellant tank: a stage normally combines propulsion, tanks or motor casing, primary structure, interfaces, separation equipment and control hardware. Depending on the vehicle, it may also carry avionics, batteries, telemetry and flight computers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:launch-vehicle-stage", "labels": [ "Launch Vehicle Stage" ], "is_subclass_of": [], "wikilinks": [ "Orbital Debris" ] }, { "id": "launch-vehicle", "title": "Launch Vehicle", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A launch vehicle is the integrated flight system that carries a payload from the ground, sea or air to a specified trajectory or orbit. It must supply the required velocity while supporting and protecting the payload through acceleration, vibration, acoustic loading, aerodynamic pressure and heating. In UK law, *launch vehicle* has a broader regulatory meaning that covers the vehicle and its component parts but excludes the payload; the exact legal scope depends on the licensed activity.[^1]", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:launch-vehicle", "labels": [ "Launch Vehicle" ], "is_subclass_of": [], "wikilinks": [ "stages", "spacecraft propulsion systems" ] }, { "id": "law-commission", "title": "Law Commission", "domain": "governance", "domain_name": "Governance", "definition": "The Law Commission is the statutory independent body for England and Wales that reviews the law and recommends reform to government.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:law-commission", "labels": [ "Law Commission" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Regulatory Frameworks", "Regulation", "Governance", "https://www.lawcom.gov.uk/", "https://www.legislation.gov.uk/ukpga/1965/22" ] }, { "id": "law-enforcement-access", "title": "Law Enforcement Access", "domain": "governance", "domain_name": "Governance", "definition": "Law enforcement access refers to the legally authorised mechanisms by which police and regulatory authorities obtain data, records or information held by service providers for investigation and prosecution. In financial and crypto contexts it covers lawful requests for customer and transaction information under court orders, subpoenas or statutory disclosure rules such as the FATF Travel Rule. It sits at the intersection of compliance obligations and privacy safeguards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:law-enforcement-access", "labels": [ "Law Enforcement Access" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "law-of-large-numbers", "title": "Law of Large Numbers", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The law of large numbers is a theorem in probability theory stating that as the number of independent, identically distributed trials increases, the sample average of their outcomes converges to the expected value of the underlying distribution. It provides the theoretical justification for estimating expectations by averaging repeated random samples, as performed in Monte Carlo methods and Monte Carlo integration. Weak and strong forms of the theorem differ in the mode of convergence they guarantee.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:law-of-large-numbers", "labels": [ "Law of Large Numbers" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [] }, { "id": "layer-0", "title": "Layer 0", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Foundation blockchain infrastructure layer providing cross-chain communication protocols, shared security mechanisms, and modular consensus abstraction to enable interoperability between multiple Layer 1 blockchains.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:layer-0", "labels": [ "Layer 0" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Avalanche", "Blockchain Architecture", "Cosmos", "Layer 1", "Monolithic Blockchain", "Polkadot", "Scalability Solutions", "Shared Security", "Blockchain", "Consensus Mechanism", "Cross-Chain Bridge", "Interoperability" ] }, { "id": "layer-1-blockchain", "title": "Layer 1 Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Layer 1 Blockchain is a base-layer distributed ledger that provides its own consensus, settlement and security without relying on another chain. It defines the native protocol, block production rules and economic security that applications and higher layers build upon, and is where transactions achieve final settlement. Examples include Bitcoin, Ethereum, Avalanche and Cardano. Layer 1 design choices around consensus and data structures determine the throughput, decentralisation and security trade-offs that Layer 2 solutions later seek to extend.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:layer-1-blockchain", "labels": [ "Layer 1 Blockchain", "Layer-1 Blockchain" ], "is_subclass_of": [ "Blockchain Network" ], "wikilinks": [] }, { "id": "layer-1", "title": "Layer 1", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Layer 1 is the foundational base protocol of a blockchain network that maintains its own independently verified state, executes a native consensus mechanism, and provides cryptographically final transaction ordering without relying on any external chain. It defines the canonical ledger, enforces network rules (including token issuance schedules and validity criteria), and serves as the trust anchor from which all higher-order protocols derive their security guarantees. Notable examples include Bitcoin, Ethereum, Solana, Cardano, and Avalanche, each differing in their consensus approach, throughput characteristics, and programmability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:layer-1", "labels": [ "Layer 1" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [ "Consensus Protocol", "Layer 2 Networks", "Distributed Ledger", "Blockchain Domain" ] }, { "id": "layer-2-networks", "title": "Layer 2 Networks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Layer 2 Networks are secondary protocols constructed atop a base blockchain (Layer 1) that execute transactions off the main chain, batching or compressing them before posting commitments back to Layer 1 for final settlement and security. They address the fundamental scalability trilemma by separating execution from consensus, enabling higher throughput and lower fees without compromising the decentralisation or security guarantees of the underlying chain. Primary architectural families include optimistic rollups, zero-knowledge rollups, state channels, and sidechains, each making different trade-offs between latency, data availability, and trust assumptions. Settlement finality and fraud or validity proof mechanisms determine the trust model and the withdrawal period users must endure before funds are considered unconditionally settled.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:layer-2-networks", "labels": [ "Layer 2 Networks", "Layer-2 Network" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "Layer 1", "Scalability", "Rollup" ] }, { "id": "layer-2-scaling", "title": "Layer 2 Scaling", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Layer 2 scaling refers to a category of protocols and systems built atop an existing blockchain (Layer 1) that increase transaction throughput and reduce fees by processing computation and data storage off the main chain while inheriting the security guarantees of the underlying Layer 1 through cryptographic proofs, fraud-proof mechanisms, or periodic state commitments. The Layer 2 anchors itself to the Layer 1 by periodically publishing compressed state roots or validity proofs, enabling the Layer 1 to serve as the ultimate settlement and security layer while the Layer 2 handles high-volume transaction processing. Major architectural families include ZK-rollups, optimistic rollups, state channels, and sidechains, each offering distinct trade-offs between latency, security assumptions, and EVM compatibility.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:layer-2-scaling", "labels": [ "Layer 2 Scaling", "Layer-2 Scaling", "Layer-2 Scaling Ecosystem", "Layer-2 Scaling Solution", "Layer-Two Scaling" ], "is_subclass_of": [ "Blockchain Scalability" ], "wikilinks": [] }, { "id": "layer-2-security-council", "title": "Layer 2 Security Council", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Layer 2 Security Council is a designated multi-signature governance body empowered to take privileged actions on a Layer 2 rollup, such as pausing the bridge, executing emergency upgrades, or resolving stuck states during the maturation period before fully trustless operation. It typically comprises a quorum of independent signers from distinct organisations to mitigate single-party capture. The construct trades some decentralisation for the ability to respond rapidly to critical vulnerabilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:layer-2-security-council", "labels": [ "Layer 2 Security Council" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "layer-2-solutions", "title": "Layer 2 Solutions", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Layer 2 solutions are protocols and systems built atop a base blockchain (layer 1) that handle transactions off the main chain to increase throughput, reduce latency, and lower transaction costs, while periodically settling finality back to the underlying layer 1 for security. The principal layer 2 paradigms include optimistic rollups, zero-knowledge rollups, state channels, and sidechains, each offering distinct trust and performance tradeoffs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:layer-2-solutions", "labels": [ "Layer 2 Solutions", "Layer 2", "Layer 2 Rollups", "Layer-2 Solutions" ], "is_subclass_of": [ "Blockchain Scalability" ], "wikilinks": [] }, { "id": "layer-3", "title": "Layer 3", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Application-specific blockchain layer built atop Layer 2 scaling solutions, providing customized execution environments for specialized use cases such as gaming, DeFi, and enterprise applications.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:layer-3", "labels": [ "Layer 3", "Layer3" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Application-Specific Blockchain", "Arbitrum Orbit", "Custom Execution Environment", "dYdX v4", "General-Purpose Smart Contract Platform", "Immutable X", "Layer 1", "Layer 2", "Rollup", "Scalability Solutions", "Blockchain", "State Channel" ] }, { "id": "layer-normalisation", "title": "Layer Normalisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A normalisation technique that computes mean and variance across the feature dimension for each training example independently, then rescales activations using learnable scale and shift parameters. Unlike batch normalisation, layer normalisation is invariant to batch size, making it the standard choice for transformer and recurrent architectures where sequence lengths vary.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:layer-normalisation", "labels": [ "Layer Normalisation", "L2 Normalisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Bitcoin", "Lightning and Similar L2", "Metaverse Ontology" ] }, { "id": "layer-normalization", "title": "Layer Normalization", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural network technique that normalises the activations across the features of a single training example, stabilising and accelerating training. It is widely used in transformer architectures where it normalises each token's representation independently of the batch.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:layer-normalization", "labels": [ "Layer Normalization" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Activation Function", "Transformer Architecture", "Deep Learning", "Backpropagation", "Neural Network", "https://arxiv.org/abs/1607.06450" ] }, { "id": "layer-2-protocol", "title": "Layer-2 Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A layer-2 protocol is a scaling construction built on top of an underlying layer-1 blockchain that executes transactions off the base chain while inheriting its security guarantees through periodic settlement. By batching, compressing, or channelling activity off-chain and committing only succinct proofs or state commitments to layer 1, these protocols dramatically increase throughput and reduce fees without changing the base consensus. Major families include rollups (optimistic and zero-knowledge), state channels, and sidechains, each trading off security, latency, and capital efficiency differently while bridging assets and messages back to the secured base layer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:layer-2-protocol", "labels": [ "Layer-2 Protocol", "Layer 2 Protocol" ], "is_subclass_of": [ "Blockchain Scalability" ], "wikilinks": [] }, { "id": "layer2", "title": "Layer2", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Secondary protocols built atop Layer 1 blockchains that process transactions off-chain or in parallel batches, then periodically commit aggregated state changes to the main chain. Layer 2 solutions achieve dramatically higher throughput, lower fees, and improved user experience while inheriting the security guarantees of the underlying base layer. Principal approaches include rollups (optimistic and ZK), state channels, and sidechains.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:layer2", "labels": [ "Layer2", "Layer 2", "Layer-2" ], "is_subclass_of": [ "Network Component", "Smart Contract Platform" ], "wikilinks": [ "Across-Protocol", "AggLayer", "AI-inference", "Alchemy", "ALEX", "AltLayer", "AMP", "Anoma", "Arbitrum", "Arbitrum-Bridge", "Arbitrum-Nitro", "Arbitrum-One", "Arbitrum-Orbit", "Argent", "Ark", "Arkadiko", "Astria", "Atomic-Swaps", "Avail", "Axelar" ] }, { "id": "layer2-scaling", "title": "Layer2Scaling", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Off-chain scaling solutions executing transactions on secondary networks ({{Rollups}}, PaymentChannels, {{Sidechains}}) that batch and settle to base layer (ereum mainnet), reducing transaction costs 100-1000x whilst maintaining security through CryptographicProof|cryptographic proofs.", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:layer2-scaling", "labels": [ "Layer2Scaling" ], "is_subclass_of": [ "Network Component", "Blockchain Scalability" ], "wikilinks": [ "Arbitrum", "buildsOn", "CryptographicProof", "CryptographicProof", "dt:enables", "dt:facilitates", "dt:optimizes", "dt:scales", "dt:supports", "GameEconomy", "HighThroughputTransaction", "implementsTechnology", "InteroperabilityProtocol", "MicroTransaction", "NFTMinting", "Optimism", "OptimisticRollup", "OptimisticRollup", "PaymentChannels", "Polygon" ] }, { "id": "layer-zero", "title": "LayerZero", "domain": "blockchain", "domain_name": "Blockchain", "definition": "LayerZero is a cross-chain interoperability protocol that allows smart contracts on different blockchains to send messages to one another. It uses a configurable security model in which an oracle delivers block headers and an independent relayer delivers transaction proofs, with a message accepted only when the two agree. This separation is intended to avoid reliance on a single intermediary chain for verifying cross-chain communication.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:layer-zero", "labels": [ "LayerZero" ], "is_subclass_of": [ "Blockchain", "Blockchain Interoperability", "Blockchain Domain" ], "wikilinks": [ "Oracle", "Relayer", "Smart Contract", "Cross-Chain Messaging", "Omnichain Application", "Bridge", "Interoperability", "Blockchain Domain" ] }, { "id": "layout-algorithm", "title": "Layout Algorithm", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A layout algorithm is a computational procedure that automatically assigns spatial positions to graphical elements such as nodes, edges, or boxes to produce a readable diagram or interface. Common families include force-directed, hierarchical (Sugiyama), orthogonal, and tree layouts, each optimising criteria like minimal edge crossings, uniform spacing, or compactness. Layout algorithms are central to diagram rendering and graph visualisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:layout-algorithm", "labels": [ "Layout Algorithm" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "layout-engine", "title": "Layout Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A layout engine is a software component that applies one or more layout algorithms to a model of visual elements and produces a concrete arrangement ready for rendering. It manages coordinate systems, constraint solving, sizing, and incremental relayout in response to data changes. Layout engines power diagram tools, document renderers, and graphical user interface frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:layout-engine", "labels": [ "Layout Engine" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "lead-screw-actuator", "title": "Lead Screw Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A linear actuation mechanism that converts rotational motion from a motor into controlled axial displacement by driving a threaded nut along a precision-cut helical screw shaft. Lead screw actuators provide high mechanical advantage, inherent load-holding capability (due to the self-locking property when lead angle is below the friction angle), and positional repeatability, making them widely used in CNC machines, 3D printers, robotic joints, and medical devices.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:lead-screw-actuator", "labels": [ "Lead Screw Actuator" ], "is_subclass_of": [ "Actuation and Control", "Electric Linear Actuator" ], "wikilinks": [ "Electric Linear Actuator", "Robotics" ] }, { "id": "leader-election", "title": "Leader Election", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Leader election is a fundamental coordination primitive in distributed computing that enables a cluster of peer nodes to agree on a single node \u2014 the leader \u2014 that assumes special coordination responsibilities such as sequencing writes, directing consensus rounds, or managing resource allocation on behalf of the group. In the presence of node failures, network partitions, or leader crashes, the leader election protocol must reliably select a new leader from the surviving nodes while ensuring safety (at most one leader at a time) and liveness (a leader is eventually elected) properties hold.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:leader-election", "labels": [ "Leader Election", "Randomised Leader Election" ], "is_subclass_of": [ "Distributed Consensus" ], "wikilinks": [] }, { "id": "leaderboard", "title": "Leaderboard", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A leaderboard is a public, continuously updated ranking of systems or models against a shared benchmark dataset and fixed evaluation protocol, typically reporting standardised metrics on held-out test sets. Leaderboards make model comparison transparent and reproducible, drive competitive progress in machine learning and adjacent fields, and increasingly incorporate human preference voting \u2014 whilst also inviting over-fitting, benchmark gaming and metric myopia when rankings are treated as ends in themselves.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:leaderboard", "labels": [ "Leaderboard" ], "is_subclass_of": [ "Benchmark Evaluation" ], "wikilinks": [ "Benchmark Evaluation", "Benchmark Dataset", "Model Comparison", "Model Evaluation" ] }, { "id": "leap-second", "title": "Leap Second", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:leap-second", "labels": [ "Leap Second" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "learner-model", "title": "Learner Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A learner model is a system's structured representation of an individual learner's knowledge state, skills, misconceptions, and progress, used by adaptive and intelligent tutoring systems to personalise instruction and assessment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:learner-model", "labels": [ "Learner Model" ], "is_subclass_of": [ "Adaptive Learning" ], "wikilinks": [] }, { "id": "learning-algorithm", "title": "Learning Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "LearningAlgorithm is a computational procedure that enables a model or agent to improve task performance through exposure to data or environmental interaction, encompassing the full spectrum of paradigms \u2014 supervised learning unsupervised learning (discovering latent structure in unlabelled data ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:learning-algorithm", "labels": [ "Learning Algorithm", "Machine Learning Algorithm" ], "is_subclass_of": [ "AI Technique", "Computational Intelligence", "Optimisation", "Machine Learning Discipline", "Statistical Learning Theory", "Artificial Intelligence" ], "wikilinks": [ "Actor Network", "Adam Optimiser", "AlgorithmLayer", "Automatic Differentiation", "Autonomous Agents", "Autonomous Vehicles", "BYOL", "Calculus of Variations", "Computational Intelligence", "Continual Learner", "Continual Learning", "Contrastive Loss", "Critic Network", "DINO", "DQN", "Drug Discovery", "Dynamic Programming", "Evaluation Metric", "Evolutionary Algorithms", "EWC" ] }, { "id": "learning-analytics", "title": "Learning Analytics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The measurement, collection, analysis, and reporting of data about learners in immersive VR and metaverse educational environments, enabling understanding of learning processes, performance prediction, and adaptive content delivery.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:learning-analytics", "labels": [ "Learning Analytics" ], "is_subclass_of": [ "Content and Assets", "Educational Technology" ], "wikilinks": [ "Adaptive Learning", "Educational Technology", "metaverse" ] }, { "id": "learning-component", "title": "Learning Component", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Modular educational elements designed for metaverse and XR training environments, including interactive simulations, 3D models, assessment tools, and collaborative spaces that can be combined to create comprehensive immersive learning experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:learning-component", "labels": [ "Learning Component" ], "is_subclass_of": [ "Content and Assets", "Educational Technology" ], "wikilinks": [ "Customised Training", "Educational Technology", "metaverse" ] }, { "id": "learning-management-system", "title": "Learning Management System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software platforms that deliver, track, and manage educational content and training programmes, increasingly integrating with VR, AR, and metaverse technologies through SCORM, xAPI, and LTI standards to enable immersive learning experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:learning-management-system", "labels": [ "Learning Management System" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [ "VR Training Delivery", "Educational Technology", "metaverse" ] }, { "id": "learning-module", "title": "Learning Module", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Self-contained educational units within VR and metaverse training systems that provide structured learning experiences, including interactive simulations, assessments, and collaborative activities designed for specific skill development objectives.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:learning-module", "labels": [ "Learning Module" ], "is_subclass_of": [ "Content and Assets", "Educational Technology" ], "wikilinks": [ "Structured Training", "Educational Technology", "metaverse" ] }, { "id": "learning-platform-categorization", "title": "Learning Platform Categorization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A taxonomy of platforms and environments used to deliver educational experiences, spanning traditional learning management systems, online learning portals, virtual classrooms, immersive XR-based environments, and metaverse-native learning spaces. The categorization framework identifies dimensions such as modality (synchronous/asynchronous), immersion level, social presence, and assessment integration, enabling educators and institutions to select or compare platforms for pedagogical fit. In the spatial computing context, the taxonomy extends to cover platforms that leverage augmented reality, virtual reality, and simulated environments to enable experiential, collaborative, and gamified learning at scale.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:learning-platform-categorization", "labels": [ "Learning Platform Categorization" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Classification" ], "wikilinks": [ "Metaverse Classification" ] }, { "id": "learning-rate-schedule", "title": "Learning Rate Schedule", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A systematic strategy for varying the learning rate hyperparameter during model training, either according to a fixed rule (step decay, cosine annealing, exponential decay) or adaptively in response to training signals. Learning rate schedules improve convergence and final model performance by applying higher rates early for rapid progress and lower rates during fine-tuning, thereby reducing the risk of overshooting loss minima.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:learning-rate-schedule", "labels": [ "Learning Rate Schedule", "1cycle Learning Rate Schedule", "Learning Rate Scheduling" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "learning-rate", "title": "Learning Rate", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The learning rate is a hyperparameter in gradient-based optimisation that scales the size of each parameter update applied in the direction of the negative gradient. It governs the trade-off between the speed of convergence and the stability of training: too large a value can cause divergence or oscillation, while too small a value leads to slow progress or stalling in poor regions. It is one of the most consequential settings when training neural networks and is often varied over the course of training by a schedule or adapted per parameter.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:learning-rate", "labels": [ "Learning Rate" ], "is_subclass_of": [ "Hyperparameter", "AI Technique" ], "wikilinks": [] }, { "id": "learning-resources", "title": "Learning Resources", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Educational materials and content designed for delivery through VR, AR, and metaverse platforms, including 3D models, interactive simulations, virtual environments, and AI-generated adaptive content for immersive learning experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:learning-resources", "labels": [ "Learning Resources" ], "is_subclass_of": [ "Content and Assets", "Educational Technology" ], "wikilinks": [ "Immersive Education", "Educational Technology", "metaverse" ] }, { "id": "learning-from-demonstration", "title": "Learning from Demonstration", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Learning from demonstration is an approach in which an agent acquires behaviour by observing examples performed by a teacher. It is used to bootstrap policies in robotics and control without hand-specified reward functions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:learning-from-demonstration", "labels": [ "Learning from Demonstration", "Expert Demonstration" ], "is_subclass_of": [ "Imitation Learning" ], "wikilinks": [ "Imitation Learning", "Offline Reinforcement Learning", "Reinforcement Learning", "Robotics" ] }, { "id": "learning", "title": "Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Learning is the process by which a system improves its performance on a task through experience or data. In artificial intelligence it refers to algorithms that adjust internal parameters or rules from observations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:learning", "labels": [ "Learning" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique" ], "wikilinks": [ "Machine Learning", "Deep Learning", "Reinforcement Learning", "Neural Network", "Imitation Learning" ] }, { "id": "least-privilege", "title": "Least Privilege", "domain": "security", "domain_name": "Security", "definition": "Least privilege is a security principle stating that every user, process or component should be granted only the minimum access rights necessary to perform its function, and no more. By limiting permissions to what is strictly required, the principle reduces the attack surface and confines the damage that a compromised account or faulty component can cause. It is a cornerstone of access control and underpins modern approaches such as zero-trust architecture and defence in depth.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:least-privilege", "labels": [ "Least Privilege" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "ledger", "title": "Ledger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A ledger is a structured, authoritative record of financial transactions or state changes, historically maintained as a physical or centralised book of accounts and now realised as a distributed, cryptographically secured data structure in blockchain systems. In the distributed-ledger paradigm, every participating node holds a replica of the same append-only log, with consensus mechanisms ensuring that all copies remain consistent and tamper-evident. The concept spans traditional double-entry bookkeeping, centralised database ledgers (as in banking core systems), and fully decentralised [[Distributed Ledger Technology]] implementations such as [[Bitcoin]] and [[Ethereum]]. Ledger (the company, Ledger SAS) is a notable specific instantiation of hardware-wallet technology designed to protect the cryptographic keys that authorise writes to a blockchain ledger.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:ledger", "labels": [ "Ledger", "Decentralised Ledger", "Decentralized Ledger", "Ledger Model" ], "is_subclass_of": [ "Distributed Ledger Technology" ], "wikilinks": [ "Private Key", "Self-Custody", "Key Management", "Cold Storage" ] }, { "id": "leeds-digital-hub", "title": "Leeds Digital Hub", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Leeds Digital Hub is a regional technology cluster centred in Leeds city centre, UK, that concentrates collaborative workspace, mentoring, and innovation infrastructure to support digital startups and established technology companies in healthtech, fintech, and digital creative sectors. It operates as a node within the broader UK Tech Ecosystem and North England Innovation Corridor, leveraging anchor institutions such as the University of Leeds and Leeds Teaching Hospitals NHS Trust to translate research into commercial ventures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:leeds-digital-hub", "labels": [ "Leeds Digital Hub" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "leeds", "title": "Leeds", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Leeds is a major city and metropolitan district in West Yorkshire, England, constituting one of the largest urban economies in the United Kingdom outside London. It is distinguished by its concentration of financial and professional services, a substantial legal sector, a growing health-data and digital technology cluster, and two research-intensive universities. As a principal node of the Northern Powerhouse agenda, Leeds anchors cross-Pennine connectivity and regional economic governance, and hosts significant infrastructure assets including a large rail interchange, innovation campuses, and regulated financial institutions.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:leeds", "labels": [ "Leeds" ], "is_subclass_of": [ "Legal and Regulatory", "Urban Centre" ], "wikilinks": [ "Financial Technology", "Northern Powerhouse", "Manchester", "Entity" ] }, { "id": "legal-accountability", "title": "Legal Accountability", "domain": "governance", "domain_name": "Governance", "definition": "Legal accountability is the principle that an identifiable party can be held answerable under law for actions, omissions, and harms, with attendant liability and remedies. In digital and metaverse contexts it requires attributing conduct to legal persons despite pseudonymity, automation, or cross-jurisdictional operation. Establishing it underpins liability models, dispute resolution, and regulatory enforcement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-accountability", "labels": [ "Legal Accountability" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-compliance", "title": "legal compliance", "domain": "governance", "domain_name": "Governance", "definition": "Legal compliance is the systematic set of organisational processes, technical controls, governance structures, and documented evidence through which an entity demonstrates adherence to applicable legislation, regulatory requirements, technical standards, and contractual obligations across the full lifecycle of its products, services, and operations. In AI and digital systems contexts, it spans intersecting regimes including data protection law, sector-specific financial and healthcare regulation, product liability, intellectual property, and dedicated AI legislation such as the EU AI Act. Compliance is operationalised through risk classification, conformity assessment procedures, mandatory documentation, appointed accountability roles, and continuous monitoring programmes. It functions as the bridge between abstract legal obligations and concrete engineering and operational practice.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:legal-compliance", "labels": [ "Legal Compliance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "legal-entity-identifier", "title": "Legal Entity Identifier", "domain": "governance", "domain_name": "Governance", "definition": "A Legal Entity Identifier (LEI) is a 20-character alphanumeric code, standardised as ISO 17442, that uniquely identifies a legally distinct entity participating in financial transactions. It is issued by accredited Local Operating Units under the Global LEI System and carries reference data on the entity's ownership and registration. LEIs enable counterparty transparency and are increasingly required for regulatory and trade reporting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-entity-identifier", "labels": [ "Legal Entity Identifier" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-entity-structure", "title": "Legal Entity Structure", "domain": "governance", "domain_name": "Governance", "definition": "A legal entity structure is the organisational arrangement of incorporated bodies, subsidiaries, foundations, and operating companies through which an organisation conducts business and holds assets. It determines liability boundaries, tax treatment, governance rights, and regulatory obligations across jurisdictions. In crypto and DAO contexts it often pairs on-chain governance with off-chain foundations or LLCs to obtain legal personality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-entity-structure", "labels": [ "Legal Entity Structure", "Legal Entity Architecture" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-evidence", "title": "Legal Evidence", "domain": "security", "domain_name": "Security", "definition": "Legal evidence is information presented to a court or tribunal to establish facts in dispute, which must satisfy admissibility criteria such as authenticity, relevance, and chain of custody. In digital contexts it includes logs, signed records, timestamps, and cryptographic proofs that demonstrate integrity and origin. Sound evidence handling is essential for proving non-repudiation and supporting enforcement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-evidence", "labels": [ "Legal Evidence", "Admissibility of Evidence" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-expertise", "title": "Legal Expertise", "domain": "governance", "domain_name": "Governance", "definition": "Legal expertise is the specialised knowledge and professional judgement required to interpret statutes, regulations, contracts, and case law and to apply them to specific situations. In technology and blockchain settings it spans securities, data-protection, intellectual-property, and cross-border regulatory analysis. It is a prerequisite for designing compliant products and navigating multi-jurisdictional obligations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-expertise", "labels": [ "Legal Expertise" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-framework", "title": "Legal Framework", "domain": "governance", "domain_name": "Governance", "definition": "A legal framework for technology systems comprises the body of laws, regulations, and legal requirements that govern the development, deployment, and operation of information technology. It establishes obligations for data protection, privacy, security, and compliance, defining the legal boundaries within which organisations must operate when processing personal data, deploying AI systems, and conducting digital operations.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:legal-framework", "labels": [ "Legal Framework", "Legal Frameworks" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Framework" ], "wikilinks": [ "Core Technology", "Legal Compliance", "Privacy Requirements", "Rights Protection", "Security Obligations", "Accountability", "Regulatory Framework" ] }, { "id": "legal-research", "title": "Legal Research", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Legal research is the systematic process of finding, analysing and applying statutes, case law, regulations and secondary sources to answer a legal question, increasingly assisted by retrieval and language models.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:legal-research", "labels": [ "Legal Research" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [ "Optical Character Recognition", "Regulatory Compliance", "Machine Learning", "Narrow AI", "Artificial Intelligence" ] }, { "id": "legal-system", "title": "Legal System", "domain": "governance", "domain_name": "Governance", "definition": "A legal system is the institutional framework of laws, courts, enforcement bodies, and procedures through which a jurisdiction creates, interprets, and applies binding rules. Major traditions include common law, civil law, and religious or customary systems, each differing in sources of authority and precedent. Legal systems define the jurisdictional context within which digital and property rights are recognised and enforced.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:legal-system", "labels": [ "Legal System" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "legal-wrapper", "title": "Legal Wrapper", "domain": "governance", "domain_name": "Governance", "definition": "A legal wrapper is an off-chain legal entity, such as a foundation, LLC, or association, established to give a DAO or on-chain protocol recognised legal personality. It enables the organisation to enter contracts, hold assets, limit member liability, and interface with regulators while preserving on-chain governance. Jurisdictions like Wyoming (DAO LLC), the Marshall Islands, and the Cayman Islands offer purpose-built wrapper forms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:legal-wrapper", "labels": [ "Legal Wrapper" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "legged-locomotion", "title": "legged locomotion", "domain": "robotics", "domain_name": "Robotics", "definition": "Legged locomotion is the study and engineering of motion through articulated limb contacts with the environment, enabling traversal of unstructured, discontinuous, and vertically challenging terrain inaccessible to wheeled or tracked platforms. It requires coordinated management of contact scheduling, ground reaction forces, centre-of-mass dynamics, gait sequencing, and reactive balance control across walking, running, climbing, jumping, and stair-negotiation tasks. Modern legged locomotion controllers integrate whole-body control (WBC), model predictive control (MPC), and reinforcement learning \u2014 frequently trained in simulation with domain randomisation and transferred to hardware \u2014 to achieve robust performance on irregular natural and urban terrain. The field draws on classical rigid-body mechanics, optimisation theory, machine learning, and biomechanics to produce systems that rival biological locomotion efficiency and versatility.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:legged-locomotion", "labels": [ "Legged Locomotion", "Legged Locomotion Research", "Legged Robot Locomotion" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "legged-robot", "title": "Legged Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Legged Robot is a mobile robotic system that achieves locomotion via articulated limbs rather than wheels or tracks, drawing on bio-inspired design and gait control algorithms to navigate complex, unstructured terrain. Contemporary platforms combine reinforcement learning, sensor fusion, and compliant actuators to achieve robust autonomous operation across search-and-rescue, logistics, and infrastructure inspection scenarios.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:legged-robot", "labels": [ "Legged Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "legged-robotics", "title": "Legged Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Legged robotics is the field concerned with robots that locomote using articulated legs rather than wheels or tracks, enabling traversal of rough, discontinuous terrain. Designs span bipeds, quadrupeds, and hexapods and rely on dynamic balance, gait planning, and high-bandwidth force control. The approach trades mechanical and control complexity for superior mobility over obstacles and stairs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:legged-robotics", "labels": [ "Legged Robotics" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "legitimacy", "title": "Legitimacy", "domain": "governance", "domain_name": "Governance", "definition": "Legitimacy is the property of an authority, institution or decision being widely accepted as rightful and worthy of compliance by the people it governs. It rests on perceptions of fairness, due process, representation and shared values rather than on coercion alone, and it can be derived from procedure, performance, tradition or consent. In governance systems, including decentralised and blockchain communities, legitimacy is what makes coordinated action stable and what allows rules to be followed without constant enforcement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:legitimacy", "labels": [ "Legitimacy" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "lender-of-last-resort", "title": "Lender Of Last Resort", "domain": "finance", "domain_name": "Finance", "definition": "A lender of last resort is an institution, typically a central bank, that provides liquidity to solvent but illiquid financial institutions during a crisis when no other source of funding is available. By standing ready to lend freely against good collateral at a penalty rate, it aims to halt bank runs and contagion and to preserve financial stability. The role embodies Bagehot's principle of lending to solvent firms to stem systemic panic.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lender-of-last-resort", "labels": [ "Lender Of Last Resort", "Lender of Last Resort" ], "is_subclass_of": [ "Central Bank" ], "wikilinks": [] }, { "id": "lending-protocol", "title": "Lending Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart-contract system that lets users supply assets to earn interest and borrow against deposited collateral, with interest rates and liquidations governed by on-chain code rather than intermediaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lending-protocol", "labels": [ "Lending Protocol", "Collateralised Lending" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Collateral Management", "Price Oracle", "Yield Farming", "Liquidity Pool", "Smart Contract" ] }, { "id": "length-normalisation", "title": "Length Normalisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Length normalisation is a scoring adjustment applied during sequence decoding that divides or otherwise rescales a candidate sequence's cumulative log-probability by a function of its length, correcting the bias of naive beam search toward shorter outputs. Without it, beam search systematically favours short sequences because every additional token multiplies the sequence probability by a value less than one. It is a standard component of neural machine translation and text generation decoders.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:length-normalisation", "labels": [ "Length Normalisation" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "lens-distortion-correction", "title": "Lens Distortion Correction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Lens distortion correction is the process of removing geometric aberrations introduced by camera optics so that straight lines in the world appear straight in the image. It estimates distortion coefficients, typically radial and tangential terms, from calibration data and remaps pixels to an undistorted, rectilinear projection. The correction is a prerequisite for accurate measurement, pose estimation, and image rectification in computer-vision pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lens-distortion-correction", "labels": [ "Lens Distortion Correction" ], "is_subclass_of": [ "Camera Calibration" ], "wikilinks": [] }, { "id": "lens-protocol", "title": "Lens Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Lens Protocol is a decentralised social networking protocol in which user profiles and social connections are represented as on-chain assets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:lens-protocol", "labels": [ "Lens Protocol" ], "is_subclass_of": [ "Decentralised Identity" ], "wikilinks": [ "Decentralised Identity", "Web3", "NFT" ] }, { "id": "lens-and-camera-calibration", "title": "Lens and Camera Calibration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Lens and Camera Calibration is the systematic metrology process of estimating the full set of geometric and photometric parameters that govern image formation in a camera-lens system, enabling precise bidirectional mapping between three-dimensional world coordinates and two-dimensional image pixe...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:lens-and-camera-calibration", "labels": [ "Lens and Camera Calibration" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "Geometric Camera Models", "Photogrammetry", "Sensor Calibration", "3D Reconstruction", "Computational Photography" ], "wikilinks": [ "AlgorithmLayer", "AprilTag", "ArUco", "Autonomous Driving Perception", "Autonomous Vehicles", "Bogdan et al 2018 DeepCalib SIGGRAPH CVMP", "Bouguet 2004 MATLAB Calibration Toolbox Caltech", "Brown 1966 Decentering Distortion Photogrammetric Engineering", "Brown-Conrady Distortion Model", "Bundle Adjustment", "Camera Intrinsics", "ChArUco Board", "COLMAP", "Computational Photography", "ComputerVisionDomain", "Conrady 1919 Decentred Lenses MNRAS", "Daniilidis 1999 Hand-Eye Dual Quaternions IJRR", "Direct Linear Transform", "Drone Photogrammetry", "Dual Quaternion Hand-Eye Calibration" ] }, { "id": "levee", "title": "Levee", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:levee", "labels": [ "Levee" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "level-of-detail", "title": "Level of Detail", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Level of Detail (LOD) is a rendering optimisation technique that dynamically adjusts the geometric complexity, texture resolution, and shader fidelity of 3D objects based on viewing distance or screen-space coverage, trading visual precision for computational efficiency. LOD is essential for maintaining real-time frame rates in large-scale metaverse and spatial computing scenes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:level-of-detail", "labels": [ "Level of Detail", "Level Of Detail", "Level of Detail Selection", "Level of Detail System" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "Culling", "Instancing", "Metaverse", "Occlusion Culling", "Performance Optimization", "Rasterization" ] }, { "id": "leveraged-trading", "title": "Leveraged Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Leveraged trading is the practice of opening market positions whose notional size exceeds the trader's posted collateral by borrowing capital or using derivatives such as perpetual futures. It amplifies both gains and losses and exposes positions to liquidation when collateral falls below maintenance margin. In DeFi it is implemented by perpetual and margin protocols that enforce margin rules through smart contracts and on-chain oracles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:leveraged-trading", "labels": [ "Leveraged Trading", "Leverage Trading", "Margin Trading" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "lidar-sensor", "title": "LiDAR Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A LiDAR sensor measures distance by emitting laser pulses and timing their reflections, producing dense 3D point clouds of the surrounding environment. Variants include mechanical spinning, solid-state, and flash designs that trade field of view, range, and cost. LiDAR is a cornerstone perception modality for autonomous robots, reality capture, and occupancy mapping.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lidar-sensor", "labels": [ "LiDAR Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "liberal-radicalism", "title": "Liberal Radicalism", "domain": "governance", "domain_name": "Governance", "definition": "Liberal Radicalism (LR) is a mechanism-design framework, introduced by Buterin, Hitzig, and Weyl, for optimally funding public goods through a matching scheme in which contributions are pooled and the matching amount allocated to each project grows with the square of the sum of the square roots of individual contributions. This mathematical structure rewards the breadth of support \u2014 the number of distinct contributors \u2014 over the concentration of large donations, aligning funding with collective preference. It is the theoretical basis for quadratic funding as deployed in real-world grants programmes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:liberal-radicalism", "labels": [ "Liberal Radicalism" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "libertarian-political-economy-thesis", "title": "Libertarian Political Economy Thesis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Libertarian Thesis, as represented here by Hans-Hermann Hoppe's 'Democracy: The God That Failed', argues that democratic governance systematically expands state power and erodes individual liberties relative to private property-based natural order, and that radical decentralisation, covenant communities, and privatisation of public goods offer a more consistent path to a free society.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:libertarian-political-economy-thesis", "labels": [ "Libertarian Political Economy Thesis", "Libertarian thesis" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "licensing-requirements", "title": "Licensing Requirements", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Licensing Requirements are the regulatory authorisation mandates imposed on virtual asset service providers by national or supranational regulators, ranging from state-level money transmitter licences in the United States to jurisdiction-specific crypto licences under EU MiCA, UK FCA, Singapore MAS, and Dubai VARA regimes. Compliance demands operational standards, minimum capital, custody controls, KYC procedures, AML programmes, and ongoing regulatory reporting, with global coverage costing major platforms tens to hundreds of millions of pounds.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:licensing-requirements", "labels": [ "Licensing Requirements", "BC-0488-licensing-requirements", "VASP Licensing" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "Aave Arc", "Basel Committee", "BC-0479-regulatory-compliance", "BC-0480-kyc-requirements", "BC-0481-anti-money-laundering", "BC-0484-markets-in-crypto-assets", "BC-0485-travel-rule", "BC-0486-regulatory-reporting", "BC-0487-compliance-monitoring", "BC-0489-consumer-protection", "BC-0490-cross-border-compliance", "Binance", "BitLicense", "Bit Trade", "Bitcoin Suisse", "Bitstamp", "Blockchain.com", "Chainalysis", "Circle", "Coinbase" ] }, { "id": "licensing", "title": "Licensing", "domain": "governance", "domain_name": "Governance", "definition": "Licensing is the legal mechanism by which the holder of a right, such as a copyright, patent, trademark, or regulatory permission, grants another party permission to use it under specified terms. A licence defines the scope, duration, territory, and conditions of permitted use without transferring ownership of the underlying asset. In software and data, licensing governs how code, models, and datasets may be copied, modified, redistributed, and commercialised.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:licensing", "labels": [ "Licensing" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "lidar-instrument", "title": "Lidar Instrument", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lidar-instrument", "labels": [ "Lidar Instrument" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lidar-scanning", "title": "Lidar Scanning", "domain": "robotics", "domain_name": "Robotics", "definition": "Light Detection and Ranging technology that creates precise 3D spatial maps by emitting laser pulses and measuring return times, enabling accurate environment capture for VR/AR applications, metaverse development, and spatial computing.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:lidar-scanning", "labels": [ "Lidar Scanning" ], "is_subclass_of": [ "Perception and Sensing", "Spatial Mapping Technology" ], "wikilinks": [ "Digital Twin Creation", "metaverse", "Spatial Mapping Technology" ] }, { "id": "lidar", "title": "Lidar", "domain": "robotics", "domain_name": "Robotics", "definition": "Lidar (Light Detection and Ranging) is an active remote-sensing technology that emits pulsed laser light and measures the time-of-flight of returning reflections to compute precise three-dimensional point-cloud representations of the surrounding environment. Operating across wavelengths from near-infrared to ultraviolet, it achieves centimetre-scale spatial accuracy and is robust to many lighting conditions where camera-based systems degrade. Lidar is a foundational sensor modality in autonomous vehicles, aerial surveying, robotics, and spatial-computing applications, commonly fused with IMU, camera, and GNSS data to enable localisation, mapping, and obstacle detection.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:lidar", "labels": [ "Lidar", "LiDAR", "LiDAR System", "LiDAR-Camera Fusion", "Lidar Calibration" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [ "owl:Thing" ] }, { "id": "lido-dao", "title": "Lido DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The decentralised organisation that governs the Lido liquid staking protocol, setting node operator policy, fees, and treasury decisions through token-holder voting.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:lido-dao", "labels": [ "Lido DAO" ], "is_subclass_of": [ "Decentralised Autonomous Organisation" ], "wikilinks": [ "Governance Token", "Treasury Management", "Lido", "On-chain Governance", "Decentralised Autonomous Organisation" ] }, { "id": "lido", "title": "Lido", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A liquid staking protocol that stakes users' assets with network validators and issues a transferable token representing the staked position and its accruing rewards, enabling holders to participate in DeFi while their principal remains staked.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lido", "labels": [ "Lido", "Lido Finance" ], "is_subclass_of": [ "Liquid Proof of Stake" ], "wikilinks": [ "Proof of Stake", "Validator Node", "Liquidity Provision", "Lido DAO", "Liquid Proof of Stake" ] }, { "id": "life-cycle-assessment", "title": "life cycle assessment", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Life Cycle Assessment (LCA) is a standardised, systematic methodology governed by ISO 14040 and ISO 14044 for compiling and evaluating the inputs, outputs, and potential environmental impacts of a product system across every stage of its life cycle \u2014 from raw material extraction through manufacturing, distribution, use, and end-of-life treatment. The methodology quantifies a broad range of environmental impact categories including global warming potential, acidification, eutrophication, water depletion, particulate matter formation, and land use, translating mass and energy inventory flows into midpoint and endpoint indicators via characterisation factor methods such as ReCiPe and CML. LCA provides a rigorous comparative evidence base for eco-design decisions, environmental product declarations, and regulatory compliance, and its scope has expanded to encompass social LCA (S-LCA) and life cycle costing (LCC) within a broader life cycle sustainability assessment framework.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:life-cycle-assessment", "labels": [ "Life Cycle Assessment", "LifeCycleAssessment" ], "is_subclass_of": [ "Environmental Assessment" ], "wikilinks": [] }, { "id": "life-science-ai", "title": "Life Science AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence and machine learning techniques to biological research, drug discovery, and clinical strategy to accelerate scientific discovery.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:life-science-ai", "labels": [ "Life Science AI" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "life-support-system", "title": "Life Support System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:life-support-system", "labels": [ "Life Support System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lifecycle-assessment", "title": "Lifecycle Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Lifecycle assessment is a systematic method for evaluating the environmental impacts of a product, process or service across its entire life, from raw-material extraction through manufacture, distribution, use and end-of-life disposal or recycling. By accounting for inputs and outputs such as energy, materials, emissions and waste at every stage, it identifies where the greatest impacts occur and prevents burden-shifting between life-cycle phases. It underpins eco-design, carbon accounting and credible sustainability claims, and is governed by international standards to ensure comparability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lifecycle-assessment", "labels": [ "Lifecycle Assessment" ], "is_subclass_of": [ "Environmental Sustainability" ], "wikilinks": [] }, { "id": "lifecycle-hook", "title": "Lifecycle Hook", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An extension point exposed at a defined moment in a system's execution lifecycle at which user-supplied code is invoked to observe or modify behaviour without altering the host itself. In agent runtimes, hooks fire before and after events such as a tool call, a model request, a task start, or a session end, and the registered handler can log, validate, transform inputs and outputs, inject context, or veto the action, making hooks the primary mechanism for deterministic, policy-driven customisation of an otherwise opaque agent loop.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:lifecycle-hook", "labels": [ "Lifecycle Hook" ], "is_subclass_of": [ "Design Pattern", "DesignPattern" ], "wikilinks": [ "DesignPattern", "Webhook", "Middleware", "AgenticWorkflow" ] }, { "id": "lifecycle-management", "title": "Lifecycle Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Lifecycle management is the disciplined coordination of an asset, product, or system across all phases of its existence, from creation and deployment through operation, maintenance, and decommissioning. It defines processes, versioning, and governance that keep the entity consistent and accountable over time. For digital twins it ensures the virtual model stays synchronised with the physical asset throughout its operational life.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lifecycle-management", "labels": [ "Lifecycle Management", "Data Lifecycle Management", "Information Lifecycle Management", "Model Lifecycle Management", "Product Lifecycle Management", "Software Lifecycle Management" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "light-client-verification", "title": "Light Client Verification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Light client verification is the technique by which a resource-constrained client confirms facts about a blockchain without downloading or executing its full history. By tracking block headers and validating compact cryptographic proofs against committed state roots, a light client can verify transaction inclusion and consensus with minimal data and computation. It is foundational to mobile wallets, embedded clients, and trust-minimised cross-chain bridges.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:light-client-verification", "labels": [ "Light Client Verification" ], "is_subclass_of": [ "Light Client" ], "wikilinks": [] }, { "id": "light-client", "title": "Light Client", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Resource-efficient blockchain client that validates block headers and uses cryptographic proofs (Merkle proofs, state proofs) to verify transaction inclusion without downloading full blockchain state. Light clients enable trustless interaction from mobile devices, browsers, and IoT systems while requiring only megabytes of storage versus gigabytes for full nodes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:light-client", "labels": [ "Light Client", "Blockchain Light Client", "Light-Client Bridge", "Zero-Knowledge Light Client" ], "is_subclass_of": [ "Node", "Blockchain" ], "wikilinks": [ "Interoperability Protocol", "SPV (Simplified Payment Verification)", "State Proof", "Blockchain", "Cross-Chain Bridge", "Full Node", "Merkle Tree", "Relayer" ] }, { "id": "light-curtain", "title": "Light Curtain", "domain": "robotics", "domain_name": "Robotics", "definition": "A Light Curtain is an optoelectronic safety device that creates an invisible infrared detection zone around hazardous machinery or robot work cells. When an object or person interrupts the beam matrix, the curtain triggers an emergency stop, enforcing compliance with ISO 8373 and IEC 61496 safety standards.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:light-curtain", "labels": [ "Light Curtain" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "light-field-display", "title": "Light Field Display", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Light Field Display is an output device that reproduces the full four-dimensional light field of a scene, emitting rays of light in directions that recreate the optical properties of real objects without requiring the viewer to wear special glasses. Unlike conventional stereoscopic displays, light field displays support motion parallax, correct focus cues, and multiple simultaneous viewer perspectives. They are a key enabling technology for glasses-free holographic telepresence and high-fidelity volumetric visualisation in collaborative settings.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:light-field-display", "labels": [ "Light Field Display", "Light-Field Display" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "light-node", "title": "Light Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Light Node is a blockchain network participant that downloads and verifies only block headers rather than the full transaction history, using Simplified Payment Verification (SPV) to confirm transaction inclusion via Merkle proofs. This design allows resource-constrained devices\u2014mobile wallets, IoT devices, embedded clients\u2014to interact securely with a blockchain without the storage and bandwidth demands of a Full Node, relying on connected full nodes to supply the underlying transaction data when required.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:light-node", "labels": [ "Light Node", "Ultra Light Node" ], "is_subclass_of": [ "Blockchain Entity" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "light-parameters", "title": "Light Parameters", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Configurable settings that control virtual lighting behaviour in VR and metaverse environments, including intensity, colour temperature, bounce calculations, shadow quality, and reflection properties that determine visual realism and immersion.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:light-parameters", "labels": [ "Light Parameters" ], "is_subclass_of": [ "Display and Rendering", "Rendering Technology" ], "wikilinks": [ "Visual Realism", "metaverse", "Rendering Technology" ] }, { "id": "light-time", "title": "Light Time", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:light-time", "labels": [ "Light Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "light-field", "title": "Light field", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Light field is a foundational concept in computational optics and computer graphics describing the complete distribution of light rays travelling through free space in all directions at all points, formalised by Adelson and Bergen (1991) as the 7-dimensional plenoptic function P(Vx, Vy, Vz, \u03b8, \u03c6,...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:light-field", "labels": [ "Light field", "Light Field", "Light Field Imaging" ], "is_subclass_of": [ "Display and Rendering", "Image Based Rendering", "Computational Photography", "Radiance Field", "Volumetric Display", "Optical Representation" ], "wikilinks": [ "2D Display", "4D Parameterisation", "6-DoF VR", "ACM TOG", "AlgorithmLayer", "Broadcast Production", "Camera Array", "Camera Calibration", "Computational Geometry", "Computational Holography", "ComputationalImagingDomain", "Computational Photography", "Convolutional Neural Networks", "CUDA", "CVPR", "Epipolar Geometry", "Epipolar Plane Image", "Fourier Analysis", "Fourier Slice Theorem", "GPU Compute" ] }, { "id": "lightning-labs", "title": "Lightning Labs", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Lightning Labs is a San Francisco-based technology company, founded in 2016 by Elizabeth Stark and Olaoluwa Osuntokun, that builds open-source software and commercial infrastructure for the Bitcoin Lightning Network. The company is the primary maintainer of lnd (Lightning Network Daemon), the most widely deployed Lightning Network implementation, and develops a suite of complementary tools including Loop, Pool, Faraday, and Taproot Assets. Lightning Labs contributes to the BOLT (Basis of Lightning Technology) specification process and advances Bitcoin Layer 2 scalability by enabling high-throughput, low-latency micropayments routed through a network of bidirectional payment channels.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lightning-labs", "labels": [ "Lightning Labs" ], "is_subclass_of": [ "Blockchain Infrastructure", "Network Component" ], "wikilinks": [ "Bitcoin", "Payment Channel", "Layer 2 Scaling", "Lightning Network" ] }, { "id": "lightning-network-layer", "title": "Lightning Network Layer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Lightning Network Layer is a second-layer stratum that enables fast, low-cost payments off the base settlement ledger through bidirectional payment channels. It sits above the Settlement Layer, on which it anchors and finally settles, and below the application and content strata that use instant payments. It contains payment channels, routing, and channel-state management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:lightning-network-layer", "labels": [ "Lightning Network Layer" ], "is_subclass_of": [ "Protocol and Consensus", "owl:Thing" ], "wikilinks": [ "Settlement Layer", "Application Layer", "Content Layer", "Payment Channel", "Hash Time-Locked Contract", "owl:Thing" ] }, { "id": "lightning-network-specification", "title": "Lightning Network Specification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Lightning Network Specification, known as BOLT (Basis of Lightning Technology), is the set of documents defining the peer-to-peer protocol for Bitcoin's Lightning payment-channel network. It standardises channel establishment, commitment transactions, HTLC-based routing, onion message encryption, and gossip propagation so that independent implementations interoperate. The specification is the canonical reference enabling trust-minimised off-chain Bitcoin payments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lightning-network-specification", "labels": [ "Lightning Network Specification", "Lightning RFC" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "lightning-network", "title": "Lightning Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Lightning Network is a Layer 2 Scaling protocol for Bitcoin enabling instant, high-throughput off-chain payments through a mesh of bidirectional Payment Channel Network channels anchored on the Bitcoin Technical Overview base layer.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:lightning-network", "labels": [ "Lightning Network", "Basis of Lightning Technology", "Lightning Network Ecosystem", "Lightning Network Protocol", "Lightning Zap Payments", "Lightning-Network" ], "is_subclass_of": [ "Network Component", "Layer 2 Scaling", "Payment Channel Network", "Bitcoin Technical Overview", "Blockchain Scalability", "Distributed Ledger" ], "wikilinks": [ "ACINQ", "Blockstream", "BOLT Specifications", "BOLT12 Offers", "Channel Factory", "CryptographyDomain", "FinancialInfrastructureDomain", "Hash Time-Locked Contracts", "Instant Transactions", "Layer 2 Scaling", "Lightning Labs", "Lightning Service Providers", "LNURL Protocol", "Machine Payments", "MuSig2", "Multi-Path Payments", "Multisignature", "Onion Routing", "Payment Channel", "Payment Channel Network" ] }, { "id": "lightning-service-provider", "title": "Lightning Service Provider", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Lightning Service Provider is a business that supplies channel liquidity and connectivity services to Lightning Network users. It helps wallets open channels and receive payments.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:lightning-service-provider", "labels": [ "Lightning Service Provider" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning", "Payment Channel", "Phoenix", "Lightning Service Provider", "Lightning Network", "https://github.com/BitcoinAndLightningLayerSpecs/lsp", "https://docs.lightning.engineering" ] }, { "id": "lightning-and-similar-l2", "title": "Lightning and Similar L2", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Off-chain payment channel networks and related Layer-2 scaling protocols built atop base-layer blockchains, enabling high-throughput, low-latency micropayments without recording every transaction on-chain. The Lightning Network, state channels, and analogous protocols (Ark, Liquid, Fedimint, Cashu) route value through cryptographically secured payment channels, dramatically improving transaction throughput and cost-efficiency for Bitcoin and similar networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:lightning-and-similar-l2", "labels": [ "Lightning and Similar L2" ], "is_subclass_of": [ "Layer2" ], "wikilinks": [ "ACINQ", "Adam Back", "AI", "AI Agent", "AI Agent", "AI Agents", "AI Agents", "AI Agents", "AI-Driven Routing Optimization", "Anthropic", "antonopoulos2021mastering", "Ark", "ark2023spec", "Autonomous Agents", "Basis of Lightning Technology", "Bitcoin Bonds", "Bitcoin Domain", "Bitcoin Wallet", "bitcoinai2025", "bitcoinvisuals2025" ] }, { "id": "lightning", "title": "Lightning", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Lightning is a layer-two payment protocol built on Bitcoin that uses payment channels to enable fast, low-cost transactions off the main chain. It settles to the Bitcoin blockchain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lightning", "labels": [ "Lightning" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Payment Channel", "Bitcoin Network", "BOLT", "Layer 2 Scaling", "Lightning Network", "https://lightning.network", "https://github.com/lightning/bolts" ] }, { "id": "likelihood-function", "title": "Likelihood Function", "domain": "ai", "domain_name": "Ai", "definition": "A likelihood function expresses the probability of observed data as a function of the parameters of a statistical model, treating the data as fixed and the parameters as variable. It is the central object in maximum-likelihood estimation and Bayesian inference, where it weights how well candidate parameter values explain the evidence. In sequential filtering it scores how consistent each hypothesis is with a new measurement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:likelihood-function", "labels": [ "Likelihood Function" ], "is_subclass_of": [ "Machine Learning Discipline", "Probability Distribution" ], "wikilinks": [] }, { "id": "limit-order", "title": "Limit Order", "domain": "finance", "domain_name": "Finance", "definition": "A limit order is an instruction to buy or sell an asset at a specified price or better, remaining unfilled until the market reaches that price rather than executing immediately at the prevailing market price. It rests in the order book alongside other resting orders, contributing to displayed liquidity and price discovery until it is filled, cancelled or expires. Market makers use limit orders on both sides of the book to earn the bid-ask spread while managing inventory risk.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:limit-order", "labels": [ "Limit Order" ], "is_subclass_of": [ "Order Book" ], "wikilinks": [] }, { "id": "limited-risk-ai", "title": "Limited Risk AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Under the EU AI Act (Article 50), Limited Risk AI denotes systems subject only to transparency obligations\u2014primarily chatbots and deepfake generators\u2014where providers must inform users they are interacting with AI or label synthetic content as artificially generated. These systems face no conformity assessment, but non-compliance attracts fines of up to \u20ac7.5 million or 1.5% of global annual turnover.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:limited-risk-ai", "labels": [ "Limited Risk AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "lindy-effect", "title": "Lindy Effect", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Lindy Effect is a heuristic stating that the future life expectancy of a non-perishable thing, such as a technology or idea, is proportional to its current age, so the longer it has survived the longer it is expected to persist. Popularised by Nassim Taleb, it implies robustness accrues through demonstrated longevity. In crypto discourse it is invoked to argue that Bitcoin's continued survival strengthens its credibility as a store of value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lindy-effect", "labels": [ "Lindy Effect" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "linear-algebra", "title": "Linear Algebra", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Linear Algebra is the branch of mathematics concerned with vector spaces, linear transformations and systems of linear equations. Its central objects include vectors, matrices, determinants, eigenvalues and eigenvectors, and operations such as matrix multiplication and inversion. The field provides the language for representing and solving problems in geometry, physics, computer graphics and data analysis. It is foundational to machine learning, where data, model parameters and transformations are expressed and manipulated as tensors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:linear-algebra", "labels": [ "Linear Algebra", "Algebra" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Matrix", "Eigenvalue", "Vector Space", "Machine Learning", "Computer Vision Domain", "Principal Component Analysis", "Graph Theory", "Information Theory", "owl:Thing" ] }, { "id": "linear-encoder", "title": "Linear Encoder", "domain": "robotics", "domain_name": "Robotics", "definition": "A Linear Encoder is a position sensing device that measures linear (translational) displacement along a single axis by converting physical motion into an electrical signal \u2014 typically a series of digital pulses or an analogue waveform \u2014 that a controller can interpret as position, velocity, or acceleration data. Linear encoders may be optical (using a diffraction grating or glass scale), magnetic (using a magnetised tape), or capacitive, with resolution ranging from micrometres to nanometres in precision metrology applications. They are essential feedback elements in CNC machine tools, semiconductor lithography stages, and high-precision robotic manipulators where closed-loop positional accuracy is required.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:linear-encoder", "labels": [ "Linear Encoder" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "Encoder", "Robotics" ] }, { "id": "linear-programming", "title": "Linear Programming", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Linear Programming is a mathematical optimisation technique for finding the best outcome of a linear objective function subject to a set of linear equality and inequality constraints. The feasible region forms a convex polytope, and the optimum, when it exists, lies at a vertex, which algorithms such as the simplex method and interior-point methods exploit. Linear programming underpins resource allocation, scheduling, and planning problems across operations research and machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:linear-programming", "labels": [ "Linear Programming" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "linear-projection", "title": "Linear Projection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A linear projection is a matrix multiplication that maps an input vector from one vector space into another, typically changing its dimensionality while preserving linear structure. In neural architectures it is implemented as a fully connected layer without a non-linear activation, learning a weight matrix (and optional bias) applied uniformly across positions. It is the mechanism by which transformer attention derives query, key and value vectors, and by which patch embedding maps flattened image patches into a model's embedding space.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:linear-projection", "labels": [ "Linear Projection" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "linear-quadratic-regulator", "title": "Linear Quadratic Regulator", "domain": "robotics", "domain_name": "Robotics", "definition": "The Linear Quadratic Regulator (LQR) is an optimal control framework that computes the state-feedback gain matrix minimising a quadratic cost function \u2014 a weighted sum of squared state deviations and squared control inputs over a time horizon \u2014 for a linear dynamical system, yielding the globally optimal linear feedback law in closed form through the algebraic Riccati equation. LQR provides a principled, tunable controller where the designer specifies performance-energy trade-offs through cost weight matrices Q (penalising state error) and R (penalising control effort), and the solution guarantees both optimality with respect to this cost and closed-loop stability for controllable systems. Despite its linearity assumption, LQR is widely extended to nonlinear systems via linearisation, iterative LQR (iLQR), and as the backbone of linear-quadratic-Gaussian (LQG) control when combined with Kalman filtering.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:linear-quadratic-regulator", "labels": [ "Linear Quadratic Regulator" ], "is_subclass_of": [ "Optimal Control" ], "wikilinks": [] }, { "id": "linear", "title": "Linear", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Linear refers to relationships, mappings or systems whose output is proportional to and additive in their input, a property exploited heavily in graphics colour spaces and numerical methods.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:linear", "labels": [ "Linear" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Linear Algebra", "Graphics Pipeline", "Real-Time Rendering", "Computer Graphics" ] }, { "id": "linearizability", "title": "Linearizability", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Linearizability is a strong consistency model for concurrent and distributed systems requiring that every operation appears to take effect atomically at a single point in time between its invocation and its response, consistent with a global real-time ordering. It is a composable (local) property: a system is linearizable if each individual object is linearizable. Informally it guarantees that once a write completes, all subsequent reads observe that write or a later one, giving the illusion of a single, instantaneous copy of the data. Formalised by Herlihy and Wing, it is the gold standard against which weaker models such as eventual consistency are contrasted.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:linearizability", "labels": [ "Linearizability", "Linearisability" ], "is_subclass_of": [ "Consistency Model" ], "wikilinks": [] }, { "id": "link-margin", "title": "Link Margin", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:link-margin", "labels": [ "Link Margin" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "link-prediction", "title": "Link Prediction", "domain": "ai", "domain_name": "Ai", "definition": "Link prediction is the machine-learning task of inferring missing or future edges in a graph from its observed structure and node attributes. Techniques range from similarity heuristics and matrix factorisation to graph neural networks and knowledge-graph embeddings. It underpins recommendation, knowledge-graph completion, and social-network analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:link-prediction", "labels": [ "Link Prediction" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "linked-data-consumption", "title": "Linked Data Consumption", "domain": "data", "domain_name": "Data", "definition": "Linked data consumption is the process by which applications dereference, parse, and integrate RDF resources discovered through URIs and typed links across distributed sources. It involves following links, reconciling vocabularies, and querying federated graphs to assemble a coherent view of decentralised data. Robust consumption underpins the Semantic Web's promise of machine-readable, interoperable information.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:linked-data-consumption", "labels": [ "Linked Data Consumption" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "linked-data-encoder", "title": "Linked Data Encoder", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A semantic encoding pipeline that transforms agent state, credentials, events, and metadata into JSON-LD 1.1 format using pinned, versioned JSON-LD Context|W3C JSON-LD contexts, enabling standardised Federation Surface|federation surfaces (S1\u2013S11) that are queryable, linkable, and mac...", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:linked-data-encoder", "labels": [ "Linked Data Encoder" ], "is_subclass_of": [ "Data Management", "Data Layer" ], "wikilinks": [ "ADR-012", "AgenticSystemsDomain", "Canonical JSON", "Context Pinning", "Cross-System Querying", "DataIntegrationDomain", "EncodingLayer", "JSON-LD 1.1", "JSON-LD 1.1 Spec", "JSON-LD 1.1 Standard", "JSON-LD Context", "JSON-LD Context", "Knowledge Graph Integration", "Linked Data Consumption", "RDF Semantics", "RDF Store", "RFC 8785 Canonical JSON", "Semantic Federation", "Semantic Mapping", "SemanticWebDomain" ] }, { "id": "linked-data-platform", "title": "Linked Data Platform", "domain": "standards", "domain_name": "Standards", "definition": "A W3C specification defining rules and HTTP conventions for reading and writing linked data resources, enabling RESTful management of RDF data on the web.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:linked-data-platform", "labels": [ "Linked Data Platform" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [ "Linked Data", "RDF", "Semantic Web", "Web Standards", "Knowledge Graphs" ] }, { "id": "linked-data", "title": "linked data", "domain": "data", "domain_name": "Data", "definition": "Linked Data is a set of design principles articulated by Tim Berners-Lee for publishing machine-readable structured data on the Web, requiring that entities be identified by HTTP URIs, that those URIs dereference to useful RDF descriptions, and that descriptions include typed links to further URIs enabling traversal across a global Web of Data. It constitutes the application layer of the Semantic Web stack and provides the architectural foundation for open knowledge graphs such as Wikidata, DBpedia, and the Linked Open Data cloud. The four canonical principles \u2014 use URIs as names, use HTTP URIs, return structured data on lookup, and include outbound links \u2014 create a decentralised, self-describing information mesh that transcends the boundaries of individual databases and siloed repositories.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:linked-data", "labels": [ "Linked Data", "LinkedData", "W3C Linked Data" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "linked-open-data", "title": "Linked Open Data", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Linked Open Data is the practice of publishing structured data on the web under open licences using Linked Data principles, RDF and dereferenceable URIs so that datasets can be interlinked and queried across sources. It forms a global web of data, complementary to the web of documents, in which entities reference one another across organisational boundaries. Reference datasets such as DBpedia and Wikidata are central hubs, and Berners-Lee's five-star scheme rates the openness and linkage of published data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:linked-open-data", "labels": [ "Linked Open Data" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [] }, { "id": "linux-foundation", "title": "Linux Foundation", "domain": "governance", "domain_name": "Governance", "definition": "The Linux Foundation is a non-profit technology consortium established in 2000 that provides governance, infrastructure, legal frameworks, and community support for open-source software projects of global significance. It hosts the Linux kernel project alongside hundreds of collaborative projects spanning cloud-native computing (CNCF, Kubernetes), networking, security, blockchain (Hyperledger), AI (LF AI & Data), automotive, and open standards development. The Linux Foundation operates a neutral shared governance model in which competing commercial organisations collaborate on foundational technology layers, pooling resources to reduce duplicated effort and accelerating adoption of interoperable open standards. Its project portfolio collectively represents some of the most widely deployed software infrastructure in the world.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:linux-foundation", "labels": [ "Linux Foundation", "Linux Foundation Decentralised Trust", "Linux Foundation Edge" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "linux-kernel", "title": "Linux Kernel", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Linux kernel is the open-source, Unix-like core of the Linux operating system, managing process scheduling, memory, device drivers and system calls between hardware and user-space programs. First released by Linus Torvalds in 1991, it is maintained collaboratively and licensed under the GPLv2. It provides the namespace and cgroup primitives that container platforms such as Docker rely on to isolate and constrain processes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:linux-kernel", "labels": [ "Linux Kernel" ], "is_subclass_of": [ "Operating System" ], "wikilinks": [] }, { "id": "liquid-cooling", "title": "Liquid Cooling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Liquid cooling is a thermal-management technique that removes heat from compute hardware using a circulating liquid coolant, which has far higher heat capacity than air. Implementations include direct-to-chip cold plates and full immersion in dielectric fluid, both enabling higher power densities and lower energy overhead. It is increasingly essential for cooling dense GPU clusters used in AI training and high-performance computing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:liquid-cooling", "labels": [ "Liquid Cooling", "Water Cooling" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "liquid-democracy", "title": "liquid democracy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Liquid democracy is a hybrid participatory governance model that synthesises direct democracy and representative democracy by permitting each participant either to cast their vote on a proposal directly or to delegate their voting weight transitively to a trusted proxy, who may in turn re-delegate to another agent, forming an arbitrarily deep delegation chain. Delegations are revocable at any point before a proposal closes, preserving individual sovereignty over the vote. In distributed-systems and blockchain governance contexts the model is implemented via smart-contract delegation registries and off-chain signalling layers, enabling token holders or identity-verified citizens to assign on-chain voting power to domain experts while retaining the right to override on any individual proposal.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:liquid-democracy", "labels": [ "Liquid Democracy", "Liquid Democracy Delegation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "liquid-network", "title": "Liquid Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Liquid Network is a Bitcoin sidechain developed by Blockstream that provides faster settlement and confidential transactions for exchanges, traders and institutions. Bitcoin is moved onto the network by locking it on the main chain and issuing an equivalent pegged asset, Liquid Bitcoin, which can later be redeemed. The network is operated by a federation of functionaries who produce blocks and manage the peg, trading some decentralisation for performance and privacy features.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:liquid-network", "labels": [ "Liquid Network" ], "is_subclass_of": [ "Blockchain", "Network Component (Blockchain)", "Blockchain Domain" ], "wikilinks": [ "Bitcoin", "Federation", "Confidential Transactions", "Asset Issuance", "Bitcoin Lightning Network", "Digital Asset Domain", "Blockchain Domain" ] }, { "id": "liquid-proof-of-stake", "title": "Liquid Proof of Stake", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Proof of Stake consensus variant that decouples staking participation from token illiquidity by allowing holders to delegate validation rights to elected validators (bakers) while retaining full ownership and transferability of their tokens. Pioneered by Tezos, it combines on-chain governance with delegated staking, enabling small token holders to participate in consensus rewards without running validator infrastructure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:liquid-proof-of-stake", "labels": [ "Liquid Proof of Stake", "Tezos Liquid Proof of Stake" ], "is_subclass_of": [ "Protocol and Consensus", "Proof of Stake" ], "wikilinks": [ "Blockchain", "Proof of Stake" ] }, { "id": "liquid-rocket-engine", "title": "Liquid Rocket Engine", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:liquid-rocket-engine", "labels": [ "Liquid Rocket Engine" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "liquid-staking-token", "title": "Liquid Staking Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A liquid staking token (LST) is a transferable token issued to a user in exchange for assets staked through a liquid-staking protocol, representing a claim on the underlying stake plus accrued rewards. It frees staked capital from the usual lock-up by remaining tradable and composable across decentralised-finance applications while the principal continues to secure the network. The token's value tracks the staked position, accruing yield either through a rising exchange rate or a growing balance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:liquid-staking-token", "labels": [ "Liquid Staking Token" ], "is_subclass_of": [ "Liquid Staking" ], "wikilinks": [] }, { "id": "liquid-staking", "title": "Liquid Staking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Liquid staking is a mechanism that allows holders of proof-of-stake assets to stake their tokens while receiving a transferable derivative token representing the staked position and its accruing rewards. This derivative can be traded, lent or used as collateral in decentralised finance, unlocking liquidity that traditional staking locks up. It thereby lets participants earn staking rewards without sacrificing the capital efficiency of liquid assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:liquid-staking", "labels": [ "Liquid Staking" ], "is_subclass_of": [ "Staking" ], "wikilinks": [] }, { "id": "liquidation-engine", "title": "Liquidation Engine", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A liquidation engine is the component of a decentralised lending or derivatives protocol that detects under-collateralised positions and forcibly closes or partially repays them to keep the system solvent. It continuously evaluates position health against collateral values supplied by price oracles, and when a position breaches its maintenance threshold it triggers liquidation, typically auctioning or selling collateral and rewarding external liquidators or keepers who execute the transaction. By enforcing collateralisation guarantees on-chain, the liquidation engine is central to risk management in DeFi lending, perpetual-futures, and collateralised-debt systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:liquidation-engine", "labels": [ "Liquidation Engine" ], "is_subclass_of": [ "Lending Protocol" ], "wikilinks": [] }, { "id": "liquidation-mechanism", "title": "Liquidation Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A liquidation mechanism is the automated process by which a decentralised finance protocol seizes and sells a borrower's collateral once the value of that collateral falls below a defined threshold relative to the outstanding debt. It protects lenders and the protocol from undercollateralised positions by ensuring debt is repaid before collateral becomes insufficient. Liquidations are typically triggered by oracle price updates and executed by liquidators who are incentivised with a discount or bonus on the seized assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:liquidation-mechanism", "labels": [ "Liquidation Mechanism" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "liquidity-coverage-ratio", "title": "Liquidity Coverage Ratio", "domain": "finance", "domain_name": "Finance", "definition": "The Liquidity Coverage Ratio (LCR) is a prudential regulatory standard requiring banks to hold a stock of high-quality liquid assets sufficient to cover their projected net cash outflows over a thirty-day stress scenario. Introduced under Basel III, it is expressed as the ratio of the liquidity buffer to stressed net outflows and must equal or exceed one hundred per cent. The measure is designed to ensure short-term resilience to acute liquidity shocks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:liquidity-coverage-ratio", "labels": [ "Liquidity Coverage Ratio" ], "is_subclass_of": [ "Banking Regulation" ], "wikilinks": [] }, { "id": "liquidity-management", "title": "Liquidity Management", "domain": "finance", "domain_name": "Finance", "definition": "Liquidity management is the practice of ensuring that an institution holds sufficient liquid assets and funding to meet its financial obligations as they fall due, without incurring unacceptable losses. It involves forecasting cash flows, maintaining liquidity buffers, diversifying funding sources and managing the maturity profile of assets and liabilities. For central banks, liquidity management refers to steering aggregate reserves to keep short-term interest rates near the policy target.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:liquidity-management", "labels": [ "Liquidity Management" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "liquidity-mining", "title": "liquidity mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Liquidity mining is a decentralised finance incentive mechanism in which participants deposit assets into automated market maker pools or lending protocols and receive protocol-issued tokens as rewards proportional to their share of pooled liquidity. These reward tokens often confer governance rights and a share of trading-fee revenue, creating compounding yield incentives that attract capital to nascent protocols. The mechanism carries impermanent loss risk arising from divergence in the relative prices of deposited assets, and token inflation dynamics that must be governed through carefully designed emission schedules, vote-escrow models, or halving curves to sustain long-term protocol health.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:liquidity-mining", "labels": [ "Liquidity Mining", "Liquidity Mining Incentives", "Liquidity Mining Reward" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "liquidity-pool", "title": "Liquidity Pool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart contract-governed reserve of paired cryptocurrency tokens that enables decentralized trading through automated market-making algorithms, providing continuous liquidity without traditional order books.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:liquidity-pool", "labels": [ "Liquidity Pool", "Pool Rebalancing", "Pooled Liquidity Model" ], "is_subclass_of": [ "DeFi and Economics", "Smart Contract" ], "wikilinks": [ "AMM Algorithm", "DeFi Standards 2024", "ISO 24165", "Liquidity Provider", "LP Token", "Price Oracle", "Token Reserve", "Token Swapping", "Yield Farming", "Automated Market Making", "Blockchain", "Cryptographic Verification", "DeFi Protocol", "Decentralized Exchange", "Decentralized Exchange (DEX)", "Middleware Layer", "Price Discovery", "Smart Contract", "Smart Contract Platform", "Token Standard" ] }, { "id": "liquidity-provider", "title": "Liquidity Provider", "domain": "finance", "domain_name": "Finance", "definition": "A liquidity provider is an entity \u2014 individual, institution, or automated protocol participant \u2014 that deposits assets into a trading venue or liquidity pool to enable others to execute trades, receiving fee income or other incentives in return. In decentralised finance, liquidity providers supply token pairs to automated market makers, receiving LP tokens representing their proportional pool share.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:liquidity-provider", "labels": [ "Liquidity Provider", "Liquidity Provider Risk" ], "is_subclass_of": [ "Decentralized Finance (DeFi)" ], "wikilinks": [] }, { "id": "liquidity-provision", "title": "Liquidity Provision", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Liquidity Provision is the decentralised financial mechanism whereby cryptocurrency holders deposit paired or single-asset collateral into Automated Market Maker smart-contract pools to enable continuous permissionless Token Swap without traditional counterparties or order books, earning ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:liquidity-provision", "labels": [ "Liquidity Provision" ], "is_subclass_of": [ "DeFi and Economics", "Decentralized Finance (DeFi)", "Market Making", "Yield Generation", "Capital Deployment", "Financial Mechanism" ], "wikilinks": [ "Aave", "Aave Protocol Whitepaper", "Arbitrum", "Balancer", "Balancer Whitepaper", "Block Proposer", "Capital Deployment", "Capital Efficiency", "Centralised Exchange", "Chainlink Oracle", "Compound", "Compound Finance Whitepaper", "Concentrated Liquidity", "Constant Product Formula", "Convex Finance", "CryptographyDomain", "Curve Finance", "Curve Finance StableSwap Paper", "dYdX", "DecentralisedFinanceDomain" ] }, { "id": "liquidity", "title": "Liquidity", "domain": "finance", "domain_name": "Finance", "definition": "The degree to which an asset can be bought or sold quickly without causing a significant change in its price. In decentralised markets, liquidity is supplied by participants who deposit assets into pools or order books, enabling efficient price discovery and low-slippage trade execution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:liquidity", "labels": [ "Liquidity", "Active Liquidity Management", "Continuous Liquidity", "Liquidity Mechanism", "Market Liquidity" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Liquidity Pool", "Automated Market Maker", "Decentralized Exchange", "Market Making", "Order Book", "DeFi", "https://www.investopedia.com/terms/l/liquidity.asp" ] }, { "id": "lissajous-orbit", "title": "Lissajous Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lissajous-orbit", "labels": [ "Lissajous Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "litecoin", "title": "Litecoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Litecoin is a peer-to-peer cryptocurrency created in 2011 by Charlie Lee as an early fork of the Bitcoin codebase. It was designed for faster confirmation, using a target block time of around 2.5 minutes, and it uses the Scrypt hashing algorithm for proof-of-work in place of Bitcoin's SHA-256. It is often described as a lighter complement to Bitcoin and has served as a testing ground for protocol changes such as Segregated Witness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:litecoin", "labels": [ "Litecoin" ], "is_subclass_of": [ "Blockchain", "Cryptocurrency", "Blockchain Domain" ], "wikilinks": [ "Proof of Work", "Scrypt", "Digital Payments", "Bitcoin", "Bitcoin Lightning Network", "Blockchain Domain" ] }, { "id": "live-captions", "title": "Live Captions", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Live captions are automatically generated, real-time text transcriptions of spoken audio displayed synchronously during a video call or webinar. They enhance accessibility for participants with hearing impairments and support comprehension across language barriers and noisy environments. Powered by automatic speech recognition, they may be supplemented by human stenographers for high-accuracy requirements.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:live-captions", "labels": [ "Live Captions", "Captions" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "live-co-authoring", "title": "Live Co-authoring", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Live Co-authoring is the capability that allows multiple users to edit the same document or artifact concurrently, with changes propagated to all participants in near real time. It relies on conflict-resolution mechanisms such as operational transformation or CRDTs to merge simultaneous edits without data loss. Platforms such as Google Docs, Microsoft 365, and Notion use live co-authoring as their primary collaborative editing model.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:live-co-authoring", "labels": [ "Live Co-authoring" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "live-polls-and-qanda", "title": "Live Polls and QandA", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Live polls and Q&A are interactive audience-engagement features embedded in virtual meetings and webinars that allow participants to submit questions and vote on answer options in real time. Polls provide immediate quantitative feedback and gauge audience sentiment, while Q&A queues enable structured moderation of participant questions. Together they transform passive viewing into participatory events, increasing engagement across distributed audiences.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:live-polls-and-qanda", "labels": [ "Live Polls and QandA" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "live-streaming", "title": "Live Streaming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Live Streaming is the real-time delivery of audio and video over a network as the content is captured, allowing audiences to view events with minimal delay rather than after recording. It relies on continuous encoding, packetisation, and adaptive distribution through content delivery networks to reach large, geographically dispersed audiences. Live streaming trades the buffering headroom of on-demand playback for low end-to-end latency, demanding careful management of bitrate adaptation, jitter, and edge caching.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:live-streaming", "labels": [ "Live Streaming" ], "is_subclass_of": [ "Video Streaming" ], "wikilinks": [] }, { "id": "liveness-detection", "title": "Liveness Detection", "domain": "security", "domain_name": "Security", "definition": "Liveness detection is a set of techniques used during biometric capture to verify that the presented sample originates from a live, present human rather than a spoof such as a photograph, mask, recording or deepfake. It distinguishes genuine presentations from presentation attacks by analysing physiological signals, motion, texture and challenge responses, and is standardised under ISO/IEC 30107 as presentation attack detection. Liveness detection is essential to the integrity of remote identity verification, biometric authentication and onboarding flows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:liveness-detection", "labels": [ "Liveness Detection" ], "is_subclass_of": [ "Biometric Verification" ], "wikilinks": [] }, { "id": "liveness", "title": "Liveness", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Liveness is the class of correctness properties asserting that something good eventually happens: a submitted transaction is eventually finalised, a consensus protocol eventually decides, a requesting process eventually enters its critical section. Formalised in temporal logic and contrasted with safety properties, which assert that nothing bad ever happens, liveness cannot be violated by any finite execution \u2014 only by an infinite one that forever withholds progress \u2014 and in asynchronous fault-prone systems it is fundamentally constrained by results such as the FLP impossibility theorem.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:liveness", "labels": [ "Liveness" ], "is_subclass_of": [ "Temporal Logic" ], "wikilinks": [ "Temporal Logic", "Health Check", "Quorum", "Transaction Finality", "Consensus Mechanism", "Byzantine Fault Tolerance" ] }, { "id": "liverpool-smart-cities", "title": "Liverpool Smart Cities", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Liverpool Smart Cities refers to the Merseyside city region's integrated programme of digital innovation, IoT deployment, and sustainable urban infrastructure targeting net-zero carbon status by 2030 and 100,000 new jobs by 2040. It encompasses city council strategy, university-led research (University of Liverpool, LJMU, Edge Hill), the Horizons innovation programme backed by the UK Shared Prosperity Fund, and partnerships driving smart transport, digital health, and climate technology.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:liverpool-smart-cities", "labels": [ "Liverpool Smart Cities" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "llama-3", "title": "Llama 3", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Llama 3 is a family of open-weight large language models developed and released by Meta AI in April 2024, spanning 8 billion and 70 billion parameter base and instruction-tuned variants, with a 405 billion parameter model subsequently released in July 2024. Llama 3 models are trained on approximately 15 trillion tokens from a curated multilingual corpus, use a 128,000-token vocabulary with a custom BPE tokeniser, incorporate grouped-query attention for inference efficiency, and are post-trained with supervised fine-tuning and reinforcement learning from human feedback. The models are released under a custom Meta Llama 3 Community License that permits commercial use for most organisations while imposing restrictions on deployments exceeding 700 million monthly active users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:meta-llama-model-family-3", "labels": [ "Llama 3" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "llama-index", "title": "LlamaIndex", "domain": "ai", "domain_name": "Ai", "definition": "LlamaIndex (formerly GPT Index) is an open-source data framework and Python library that provides the abstractions and tooling needed to connect large language models with external data sources through structured indexing, retrieval, and query pipelines, primarily enabling production-grade retrieval-augmented generation (RAG) applications. It handles the full pipeline from document ingestion and chunking, through embedding and vector storage, to query decomposition, retrieval, reranking, and response synthesis, providing high-level abstractions over underlying LLMs, embedding models, and vector databases. LlamaIndex is positioned as the data orchestration complement to LangChain's agent orchestration focus, with particular strength in structured data retrieval and multi-document reasoning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:meta-llama-model-family-index", "labels": [ "LlamaIndex" ], "is_subclass_of": [ "Agent Frameworks" ], "wikilinks": [] }, { "id": "lo-ra", "title": "Lo RA", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A parameter-efficient fine-tuning mod that freezes pre-trained weights and injects trainable low-rank decomposition matrices into each layer of the transformer, dramatically reducing trainable parameters whilst maintaining performance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:lo-ra", "labels": [ "Lo RA", "LoRA", "LoRA Adaptation", "LoRA Adapter", "LoRa", "LongLoRA" ], "is_subclass_of": [ "AI Technique", "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "lora-adapter", "title": "LoRA Adapter", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A LoRA adapter is a small set of low-rank matrices trained via Low-Rank Adaptation and inserted alongside a frozen pretrained model's weight matrices to specialise its behaviour without updating the original parameters. Because the adapter contains only a small fraction of the base model's parameter count, it can be trained, stored, and swapped cheaply, allowing many task- or style-specific adapters to share a single base model. LoRA adapters are widely used to customise diffusion and language models for particular styles, subjects, or domains.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:lora-adapter", "labels": [ "LoRA Adapter" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [] }, { "id": "lo-ra-do-ra-etc", "title": "LoRA DoRA etc", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Parameter-Efficient Fine-Tuning (PEFT) is a family of machine learning techniques enabling adaptation of large pre-trained neural networks \u2014 particularly Large Language Models and diffusion models \u2014 to downstream tasks by updating only a small fraction (0.01\u2013) of total model parameters rather...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:lo-ra-do-ra-etc", "labels": [ "LoRA DoRA etc", "AdaLoRA" ], "is_subclass_of": [ "AI Technique", "Neural Network", "Machine Learning Discipline", "Transfer Learning", "Fine Tuning", "Model Adaptation", "Neural Networks" ], "wikilinks": [ "Adapter Tuning", "AlgorithmLayer", "CUDA", "Diffusion Model Customisation", "DoRA", "Domain Adaptation", "FrameworkLayer", "GaLore", "Gradient Checkpointing", "Hugging Face", "Hugging Face PEFT", "IA3", "In-Context Learning", "Llama", "LLM Fine-Tuning", "LongLoRA", "Low-Rank Decomposition", "LyCORIS", "MachineLearningDomain", "Matrix Factorisation" ] }, { "id": "lo-ra-fine-tuning", "title": "lora fine-tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "LoRA (Low-Rank Adaptation) fine-tuning is a parameter-efficient fine-tuning technique that adapts large pre-trained transformer models by inserting pairs of trainable low-rank matrices (A \u2208 \u211d^{d\u00d7r} and B \u2208 \u211d^{r\u00d7d}, where r \u226a d) alongside frozen weight matrices in selected layers, so that the effective weight update \u0394W = BA is constrained to a low-dimensional subspace. By training only the injected adapter matrices \u2014 typically representing 0.1\u20131% of the original parameter count \u2014 LoRA achieves near full fine-tuning performance at a fraction of the GPU memory and compute cost, enabling adaptation of billion-parameter models on consumer hardware. It is now the dominant technique in the PEFT ecosystem and is implemented in Hugging Face's PEFT library.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:lo-ra-fine-tuning", "labels": [ "LoRA Fine-Tuning", "Concept LoRA Training", "DreamBooth LoRA", "LoRA Fine Tuning" ], "is_subclass_of": [ "AI Technique", "Parameter-Efficient Fine-Tuning" ], "wikilinks": [] }, { "id": "lo-ra-alliance", "title": "LoRa Alliance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The LoRa Alliance is an open, non-profit association that defines and certifies the LoRaWAN standard for low-power wide-area networking. It maintains the LoRaWAN specification, regional parameters, and a device-certification programme to ensure interoperability across vendors and operators. The alliance governs the ecosystem enabling long-range, low-power IoT connectivity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lo-ra-alliance", "labels": [ "LoRa Alliance" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "lora", "title": "LoRa", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "LoRa is a long-range, low-power wireless modulation technology that uses chirp spread-spectrum signalling to transmit small amounts of data over distances of several kilometres on unlicensed sub-gigahertz radio bands. It trades data rate for range and energy efficiency, enabling battery-powered devices to operate for years. LoRa provides the physical layer beneath the LoRaWAN networking protocol used in wide-area sensor deployments. It is a foundational technology for low-power wide-area Internet-of-things connectivity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lora", "labels": [ "LoRa", "LoRA" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "lo-ra-wan", "title": "LoRaWAN", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "LoRaWAN is a media-access-control protocol and network architecture for low-power wide-area networks, built on the LoRa physical layer's chirp spread-spectrum modulation. It connects battery-powered devices to gateways over distances of several kilometres at low data rates, with end-to-end encryption and adaptive data-rate management. LoRaWAN is widely used for long-range, low-throughput IoT telemetry such as cold-chain and environmental monitoring.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:lo-ra-wan", "labels": [ "LoRaWAN", "LoRaWAN Specification" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "load-balancer", "title": "Load Balancer", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A component that distributes incoming network or application traffic across multiple backend servers to improve throughput, reliability, and resource utilisation, preventing any single server from becoming a bottleneck.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:load-balancer", "labels": [ "Load Balancer" ], "is_subclass_of": [ "Network Architecture" ], "wikilinks": [ "Network Architecture", "Scalability", "High Availability", "Microservices" ] }, { "id": "load-balancing", "title": "Load Balancing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Load balancing is the distribution of incoming work across multiple compute resources to maximise throughput, minimise latency, and avoid overloading any single node. It operates at network layers from L4 transport to L7 application routing, using algorithms such as round-robin, least-connections, and consistent hashing. Health checking and failover make it foundational to scalable, resilient distributed systems.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:load-balancing", "labels": [ "Load Balancing", "Dynamic Load Balancing", "Load Balancing System" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "local-area-network", "title": "Local Area Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A local area network (LAN) is a communications network that interconnects devices within a limited geographic area such as a building, campus, or data centre floor, typically under a single administrative domain. LANs provide high-bandwidth, low-latency connectivity enabling resource sharing, collaborative computing, and access to shared infrastructure services. Modern LANs are predominantly implemented using Ethernet (IEEE 802.3) and Wi-Fi (IEEE 802.11) technologies, with switching and VLAN segmentation providing logical isolation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:local-area-network", "labels": [ "Local Area Network" ], "is_subclass_of": [ "Infrastructure", "Networking Infrastructure" ], "wikilinks": [] }, { "id": "local-explanation", "title": "Local Explanation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Interpretability techniques that explain individual model predictions for specific instances, providing insight into why a particular input produced a given output without necessarily characterising the model's global behaviour. Methods such as LIME and SHAP generate feature-attribution scores scoped to the neighbourhood of a single query point.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:local-explanation", "labels": [ "Local Explanation" ], "is_subclass_of": [ "Model Interpretability" ], "wikilinks": [ "Counterfactual Explanation", "Feature Attribution", "Instance-Level Analysis", "Integrated Gradients", "LIME", "SHAP", "Explainable AI", "Global Explanation", "MetaverseDomain", "Model Interpretability" ] }, { "id": "local-llm-runtime-platform", "title": "Local LLM Runtime Platform", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An open-source runtime and model management platform that allows users to download, serve, and interact with large language models locally on macOS, Windows, and Linux. Ollama provides an OpenAI-compatible REST API, enabling integration with tools such as Open WebUI, ComfyUI, and agent frameworks, making local LLM inference accessible without cloud dependencies.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:local-llm-runtime-platform", "labels": [ "Local LLM Runtime Platform", "Ollama Integration", "ollama" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Llama", "Agents", "ComfyUI", "Function Calling", "Knowledge Graphing", "Ollama", "Open Webui and Pipelines", "Prompt Engineering", "Python and PyTorch" ] }, { "id": "local-model", "title": "Local Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A machine-learning model, typically a language model, whose weights are downloaded and run on hardware controlled by the operator \u2014 a personal device, a workstation, or a private server \u2014 so that inference happens without sending data to a third-party API. Local models trade the ceiling capability and elastic scale of hosted frontier systems for data locality, predictable per-token cost, offline availability, and full control over versioning and privacy, and they are commonly used as the cheap or confidential tier in a model-routing strategy.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:local-model", "labels": [ "Local Model" ], "is_subclass_of": [ "Language Model", "LanguageModel" ], "wikilinks": [ "LanguageModel", "OnDeviceInference", "CloudComputing" ] }, { "id": "local-planner", "title": "Local Planner", "domain": "robotics", "domain_name": "Robotics", "definition": "The component of a robot navigation system that converts a global route into safe, kinematically feasible velocity commands over a short horizon, reacting in real time to obstacles detected by onboard sensors. Operating at control rates of five to twenty hertz over a rolling window of a few metres, a local planner evaluates candidate trajectories against a local costmap and the robot's kinodynamic limits, selecting commands that make progress along the global path while avoiding collisions. Classic realisations include the Dynamic Window Approach, Timed Elastic Bands, and sampling-based controllers such as MPPI; in sampling-based roadmap methods the same term names the routine that checks whether two configurations can be connected by a simple collision-free motion.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:local-planner", "labels": [ "Local Planner" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [ "Motion Planning", "Navigation", "ROS Navigation Stack", "Obstacle Avoidance" ] }, { "id": "local-rag-corpus-ingestion-pipeline", "title": "Local RAG Corpus Ingestion Pipeline", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Push To Local RAG is a data-pipeline operation that extracts, cleans, and concatenates Logseq markdown pages\u2014filtering out short or low-quality documents\u2014then writes the result to a single corpus file consumed by a local Retrieval-Augmented Generation system. The process involves URL stripping, special-character normalisation, whitespace normalisation, and a minimum byte-length threshold, producing a curated text corpus that improves retrieval precision for local LLM inference without sending data to external services.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:local-rag-corpus-ingestion-pipeline", "labels": [ "Local RAG Corpus Ingestion Pipeline", "Push To Local RAG" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "..." ] }, { "id": "local-search", "title": "Local Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A family of optimisation and constraint-solving methods that iteratively improve a single complete candidate solution by moving to neighbouring solutions under a defined move operator, using strategies such as hill climbing, min-conflicts, tabu lists and randomised restarts to navigate the search landscape; memory-light and anytime by nature, local search scales to problem instances far beyond the reach of systematic tree search, at the cost of completeness guarantees.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:local-search", "labels": [ "Local Search" ], "is_subclass_of": [ "Heuristic Methods" ], "wikilinks": [ "Heuristic Methods", "Combinatorial Optimisation", "Constraint Satisfaction", "Simulated Annealing" ] }, { "id": "local-first-software", "title": "Local-First Software", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Local-first software is a design philosophy in which the primary copy of a user's data lives on their own device, with the network used for optional synchronisation rather than as a dependency. It prioritises offline availability, low latency, data ownership, and longevity while still supporting real-time collaboration via conflict-free merging. The approach typically relies on CRDTs to reconcile concurrent edits across devices without a central authority.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:local-first-software", "labels": [ "Local-First Software", "Offline-First Applications", "Offline-first Architecture" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "localisation", "title": "localisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Localisation is the computational process by which a mobile agent\u2014robot, autonomous vehicle, or mixed-reality device\u2014estimates its six-degree-of-freedom pose (position and orientation) within a reference coordinate frame, using sensor observations fused through probabilistic inference algorithms. It encompasses both map-based approaches (matching live sensor data against a prior map) and map-free approaches (dead-reckoning and visual odometry), as well as the joint Simultaneous Localisation and Mapping (SLAM) problem in which the map and pose are estimated concurrently. Accurate localisation is a prerequisite for autonomous navigation, path planning, and real-time spatial anchoring in both physical and virtual environments. The field draws on Bayesian filtering, factor graph optimisation, deep metric learning, and multi-sensor fusion across domains ranging from warehouse automation and self-driving vehicles to extended-reality (XR) headsets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:localisation", "labels": [ "Localisation", "Document Localisation", "GPS Localisation", "Global Localization", "Image-Based Localisation", "Localization", "Re-localisation", "Robot Localization", "Sound Localization" ], "is_subclass_of": [ "Navigation and Planning", "State Estimation" ], "wikilinks": [] }, { "id": "locality-sensitive-hashing", "title": "Locality-Sensitive Hashing", "domain": "ai", "domain_name": "Ai", "definition": "Locality-sensitive hashing (LSH) is a family of hashing techniques that map similar high-dimensional inputs to the same hash bucket with high probability, enabling sub-linear approximate nearest-neighbour search. By trading exactness for speed, LSH makes similarity search and deduplication tractable over very large datasets. It is widely applied to embeddings, document near-duplicate detection, and retrieval pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:locality-sensitive-hashing", "labels": [ "Locality-Sensitive Hashing" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "location-based-service", "title": "Location Based Service", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:location-based-service", "labels": [ "Location Based Service", "LBS" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lock-and-mint-mechanism", "title": "Lock and Mint Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cross-chain asset transfer protocol where tokens are locked in a smart contract on the source chain and equivalent wrapped tokens are minted on the destination chain, maintaining a 1:1 backing ratio.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:lock-and-mint-mechanism", "labels": [ "Lock and Mint Mechanism" ], "is_subclass_of": [ "Network Component", "De Fi Protocol", "Blockchain" ], "wikilinks": [ "Wrapped Token", "Atomic Swap", "Blockchain", "Blockchain Oracle", "Cross-Chain Bridge", "Smart Contract" ] }, { "id": "lock-and-mint-bridge", "title": "Lock-and-Mint Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A lock-and-mint bridge is a cross-chain interoperability mechanism that transfers asset value between blockchains by locking the original asset in a custodial smart contract on the source chain and minting an equivalent synthetic (wrapped) representation on the destination chain, with the peg maintained by a network of validators or relayers who attest to the lock event.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:lock-and-mint-bridge", "labels": [ "Lock-and-Mint Bridge" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "locking-script", "title": "Locking Script", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A locking script (also called scriptPubKey or an output script) is the predicate attached to a transaction output in UTXO-based blockchains that specifies the conditions under which that output may later be spent. To redeem the output, a spending transaction must supply an unlocking script whose combination with the locking script evaluates to true. Locking scripts encode spending policies ranging from a single signature to multi-signature, time locks and arbitrary smart-contract logic.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:locking-script", "labels": [ "Locking Script" ], "is_subclass_of": [ "UTXO Model" ], "wikilinks": [] }, { "id": "locomotion", "title": "Locomotion", "domain": "robotics", "domain_name": "Robotics", "definition": "Locomotion is the capability of a robot or embodied agent to move its body through an environment by coordinating actuators against ground or fluid reaction forces. It encompasses gait generation, balance control, and trajectory execution across modalities such as legged walking, wheeled rolling, and aerial or aquatic propulsion. Robust locomotion is fundamental to autonomous mobility and physical task execution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:locomotion", "labels": [ "Locomotion", "Locomotion Mechanism", "Locomotion System", "Soft Locomotion" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "log-aggregation", "title": "Log Aggregation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Log aggregation is the practice of collecting log events from many distributed sources into a centralised, searchable store so they can be parsed, indexed, correlated, and analysed as a unified stream. It is a foundational component of observability and monitoring pipelines, enabling operators to debug incidents, detect anomalies, and satisfy audit requirements across services that would otherwise emit logs in isolation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:log-aggregation", "labels": [ "Log Aggregation" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "log-management", "title": "Log Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Log management is the discipline of collecting, aggregating, storing, indexing and analysing the event records emitted by applications, services and infrastructure. It provides a centralised, searchable record of system behaviour that underpins debugging, performance analysis, security investigation and compliance auditing. Pipelines typically ingest structured and unstructured logs, normalise them, enforce retention policies and expose them through query and alerting interfaces.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:log-management", "labels": [ "Log Management" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "log-replication", "title": "Log Replication", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Log replication is the mechanism by which an ordered, append-only sequence of commands is copied consistently across the nodes of a distributed system so that each replica can apply the same operations in the same order. In leader-based consensus protocols such as Raft, the leader appends client commands to its log and replicates entries to followers, committing an entry once a quorum has durably stored it. By ensuring every replica converges on an identical log, it is the substrate for state machine replication and strong consistency. It must handle leader failover, log divergence, and consistency checks to keep replicas in agreement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:log-replication", "labels": [ "Log Replication" ], "is_subclass_of": [ "Replication" ], "wikilinks": [] }, { "id": "log-seq-spring-thing", "title": "Log Seq Spring Thing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Log Seq Spring Thing is a project workstream and AI hacking initiative that connects Logseq-based knowledge graph authoring with VisionFlow, exploring how networked thought can be published, visualised, and extended through AI-assisted tooling. It serves as a bridge between personal knowledge management workflows and semantic web infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:log-seq-spring-thing", "labels": [ "Log Seq Spring Thing" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "log-probability", "title": "Log-Probability", "domain": "ai", "domain_name": "Ai", "definition": "Log-probability is the natural logarithm of a probability value, used in place of raw probabilities to avoid numerical underflow when multiplying many small probabilities and to convert products into numerically stable sums. In sequence models it expresses the likelihood a model assigns to each candidate token or output, and is the quantity directly optimised during maximum-likelihood training. Decoding strategies such as beam search rank and prune candidate sequences by their accumulated log-probability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:log-probability", "labels": [ "Log-Probability" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [] }, { "id": "logging", "title": "Logging", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Logging is the practice of recording discrete, timestamped events generated by software systems as they execute, producing a durable trail used for debugging, auditing, and monitoring. Log records typically capture a severity level, source component, timestamp, and structured or free-text message, and are aggregated into centralised stores for search and correlation. Logging underpins observability alongside metrics and distributed tracing, and is a prerequisite for post-incident forensic analysis.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:logging", "labels": [ "Logging" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "logic-programming", "title": "Logic Programming", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Logic Programming is a programming paradigm in which computation is expressed as logical inference over a set of facts and rules written in formal logic. Programs describe what is true rather than how to compute, and an inference engine derives answers by applying resolution and unification. Prolog is the canonical language; descendant systems include Datalog, Answer Set Programming, and constraint logic programming.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:logic-programming", "labels": [ "Logic Programming" ], "is_subclass_of": [ "AI Technique", "Programming Paradigm" ], "wikilinks": [ "Academic AI Research", "Prolog", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "logic", "title": "Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Logic is the systematic study of valid inference, formal reasoning, and the structural principles that distinguish correct arguments from fallacious ones. It provides the mathematical and philosophical foundation for deduction, proof theory, and formal systems, encompassing propositional logic, first-order predicate logic, modal logic, and higher-order logics. In computational contexts, logic underpins programming language semantics, automated theorem proving, knowledge representation, and the design of intelligent reasoning systems. It forms a bedrock discipline across mathematics, philosophy, linguistics, and artificial intelligence.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:logic", "labels": [ "Logic", "Programmable Logic Controllers" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Knowledge Representation", "Information Theory", "owl:Thing" ] }, { "id": "logical-clock", "title": "Logical Clock", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A logical clock is a mechanism for ordering events in a distributed system without relying on synchronised physical time. It assigns monotonically increasing counters to events so that causal relationships between them can be inferred, supporting the happened-before relation. Logical clocks underpin consistency, coordination and debugging in systems where no global wall-clock can be trusted.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:logical-clock", "labels": [ "Logical Clock" ], "is_subclass_of": [ "Distributed Systems Theory" ], "wikilinks": [] }, { "id": "logical-inference", "title": "Logical Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Logical inference is the process of deriving new statements (conclusions) from existing ones (premises) according to rules that preserve truth, such that whenever the premises are true the conclusion must also be true. It encompasses deductive mechanisms like modus ponens, resolution, and unification, and is the operational core of automated reasoning systems. Logical inference connects formal logics to practical computation by turning syntactic manipulation of formulae into sound derivations of consequences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:logical-inference", "labels": [ "Logical Inference" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "logistics-automation", "title": "Logistics Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "Logistics automation is the application of robotics, software, and AI to the physical and informational tasks of moving, storing, and tracking goods across supply chains \u2014 including goods-to-person fulfilment, autonomous transport, and intelligent route optimisation \u2014 with the goal of reducing labour costs, improving throughput, and increasing reliability.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:logistics-automation", "labels": [ "Logistics Automation" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "logistics-management", "title": "Logistics Management", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Logistics management is the planning, execution, and control of the movement and storage of goods, services, and related information across a supply chain. It coordinates transportation, warehousing, inventory, order fulfilment, and reverse flows to meet demand at minimal cost and time. Effective logistics management directly determines service levels, working-capital efficiency, and resilience.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:logistics-management", "labels": [ "Logistics Management" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "logistics-optimisation", "title": "Logistics Optimisation", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Logistics optimisation is the application of mathematical optimisation and operations-research techniques to minimise cost or time across logistics operations such as routing, scheduling, and inventory placement. It formulates problems like vehicle routing, network flow, and bin packing, often solved with linear programming, heuristics, or machine learning. The goal is to extract maximal efficiency from constrained transport and storage resources.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:logistics-optimisation", "labels": [ "Logistics Optimisation", "Logistics Optimization", "Route Optimisation" ], "is_subclass_of": [ "Optimisation", "AI Technique", "Applied Machine Learning", "Operations Research", "Supply Chain Management", "Combinatorial Optimization", "Industrial AI" ], "wikilinks": [] }, { "id": "logistics", "title": "Logistics", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Logistics is the planning, execution, and control of the efficient movement and storage of goods, services, and related information between points of origin and consumption. It encompasses transportation, warehousing, inventory management, order fulfilment, and reverse flows, with the objective of meeting demand at the required service level and lowest total cost. Logistics is a constituent function of supply-chain management and is increasingly mediated by sensor data, real-time tracking, and optimisation algorithms that coordinate multimodal networks under uncertainty.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:logistics", "labels": [ "Logistics" ], "is_subclass_of": [ "Process" ], "wikilinks": [] }, { "id": "long-context-modelling", "title": "Long Context Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Long context modelling refers to techniques that let a language model process, attend to and reason over input sequences substantially longer than the context windows of earlier architectures, extending from thousands to hundreds of thousands of tokens. It is enabled by memory- and compute-efficient attention implementations such as Flash Attention, and by position-encoding schemes such as rotary position embedding that generalise to sequence lengths beyond those seen during training. Long context modelling is essential for tasks such as whole-document summarisation, long-form retrieval-augmented generation and multi-turn agent memory.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:long-context-modelling", "labels": [ "Long Context Modelling", "Long-Context Modeling" ], "is_subclass_of": [ "Context Window" ], "wikilinks": [] }, { "id": "long-range-dependency-modelling", "title": "Long Range Dependency Modelling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Long range dependency modelling is the capability of a sequence model to capture relationships between elements that are far apart in a sequence, such as tokens separated by thousands of positions. Recurrent architectures struggle with this because gradients vanish over long horizons, whereas attention mechanisms and structured state-space models provide direct or efficient paths between distant elements. Effective long range modelling is essential for tasks where context far from the current position determines the output.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:long-range-dependency-modelling", "labels": [ "Long Range Dependency Modelling", "Long-Range Dependency Modelling" ], "is_subclass_of": [ "Attention Mechanism" ], "wikilinks": [] }, { "id": "long-short-term-memory", "title": "Long Short Term Memory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Long Short-Term Memory (LSTM) is a specialised recurrent neural network architecture introduced by Hochreiter and Schmidhuber (1997) that mitigates the vanishing gradient problem through gating mechanisms\u2014input, forget, and output gates\u2014enabling selective retention or forgetting of information across long sequences. LSTMs underpin sequence modelling tasks in natural language processing, time-series forecasting, and speech recognition, though they have largely been superseded by Transformer architectures for large-scale language tasks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:long-short-term-memory", "labels": [ "Long Short Term Memory" ], "is_subclass_of": [ "AI Model Architecture", "Recurrent Neural Network" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "long-term-archival", "title": "Long Term Archival", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Strategies and technologies for preserving digital assets, metaverse content, and cultural heritage data over extended periods (10+ years), using durable storage media, format migration protocols, and AI-enhanced preservation systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:long-term-archival", "labels": [ "Long Term Archival", "Long-Term Archival" ], "is_subclass_of": [ "Content and Assets", "Digital Preservation" ], "wikilinks": [ "Future Asset Access", "Digital Preservation", "metaverse" ] }, { "id": "long-horizon-planning", "title": "Long-Horizon Planning", "domain": "ai", "domain_name": "Ai", "definition": "Long-horizon planning is the problem of generating and executing sequences of actions over extended time horizons \u2014 spanning hundreds to thousands of steps \u2014 to achieve high-level goals that cannot be accomplished through simple reactive policies. It requires maintaining coherent goals, managing intermediate subgoal states, and adapting plans in response to environmental changes and unexpected outcomes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:long-horizon-planning", "labels": [ "Long-Horizon Planning" ], "is_subclass_of": [ "Task Planning" ], "wikilinks": [] }, { "id": "long-horizon-task-benchmark", "title": "Long-Horizon Task Benchmark", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A standardized evaluation framework designed to measure the ability of AI agents to maintain coherence and achieve objectives over extended sequences of actions or time.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:long-horizon-task-benchmark", "labels": [ "Long-Horizon Task Benchmark" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "long-range-navigation", "title": "Long-Range Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Long-range navigation is the capability of an autonomous agent to plan and execute travel over large distances and extended time horizons, beyond the range of local sensors. It typically fuses global positioning, topological or metric maps, and waypoint planning to maintain a coherent route across kilometre-scale or larger environments. It contrasts with short-range reactive navigation by emphasising global consistency and goal-directed path planning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:long-range-navigation", "labels": [ "Long-Range Navigation" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "long-term-preservation", "title": "Long-Term Preservation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Long-term preservation is the set of policies, strategies, and technical practices that ensure digital assets, records, and information objects remain accessible, authentic, and usable across decades or centuries. It encompasses format migration, bit-level integrity monitoring, metadata stewardship, and provenance documentation to guard against technological obsolescence, media decay, and institutional discontinuity. The field draws on archival science, information management, and computer science to maintain the evidential and informational value of materials well beyond the lifespan of their originating systems. Standards such as OAIS (ISO 14721) provide reference models that underpin most compliant preservation programmes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:long-term-preservation", "labels": [ "Long-Term Preservation", "Long Term Preservation" ], "is_subclass_of": [ "Digital Preservation" ], "wikilinks": [] }, { "id": "longest-chain-rule", "title": "Longest Chain Rule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The fork-choice rule used in proof-of-work blockchains that designates the chain with the most cumulative work (or greatest total difficulty) as the canonical chain. It resolves temporary forks by directing nodes to extend the heaviest chain, thereby converging the distributed network on a single transaction history.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:longest-chain-rule", "labels": [ "Longest Chain Rule", "Longest-Chain Rule" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "longtermism", "title": "Longtermism", "domain": "governance", "domain_name": "Governance", "definition": "Longtermism is an ethical position holding that positively influencing the long-term future is a key moral priority of our time, grounded in the claims that future people matter morally, that the future may contain vastly more people than the present, and that present actions can foreseeably shape their welfare. It motivates particular concern for reducing existential and catastrophic risks, safeguarding civilisational trajectory, and preserving option value for future generations. As a strand of effective altruism and applied ethics, longtermism informs governance debates on emerging technologies, biosecurity, and the safe development of advanced artificial intelligence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:longtermism", "labels": [ "Longtermism" ], "is_subclass_of": [ "Ethics" ], "wikilinks": [] }, { "id": "look-angle", "title": "Look Angle", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:look-angle", "labels": [ "Look Angle" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "loom", "title": "Loom", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A video messaging application that lets users record their screen, camera, and microphone to create and share short videos. It is used for asynchronous communication at work.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:loom", "labels": [ "Loom", "Loom (screen recording)" ], "is_subclass_of": [ "Screen Recording" ], "wikilinks": [ "Screen Recording", "Video Streaming" ] }, { "id": "loop-closure-detection", "title": "Loop Closure Detection", "domain": "robotics", "domain_name": "Robotics", "definition": "Loop closure detection is the process by which a SLAM or mapping system recognises that it has returned to a previously visited location. By identifying these revisits, it adds constraints that correct accumulated odometry drift and produce globally consistent maps. It is a critical component of robust simultaneous localisation and mapping, typically implemented via appearance-based place recognition or geometric matching.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:loop-closure-detection", "labels": [ "Loop Closure Detection", "Loop Closure" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "loop-heat-pipe", "title": "Loop Heat Pipe", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:loop-heat-pipe", "labels": [ "Loop Heat Pipe" ], "is_subclass_of": [ "Heat Pipe" ], "wikilinks": [] }, { "id": "loose-coupling", "title": "Loose Coupling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Loose coupling is a design principle in which components of a system depend on one another only through stable, minimal interfaces rather than internal implementation details, so that each can evolve, fail, or be replaced independently. It reduces the ripple effect of change, improves testability, and is foundational to scalable distributed and event-driven architectures. Loose coupling is typically achieved through abstraction, asynchronous messaging, and well-defined contracts.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:loose-coupling", "labels": [ "Loose Coupling" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "loss-function", "title": "Loss Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Loss Function is a mathematical function that quantifies the discrepancy between a model's predicted outputs and true target values, producing a scalar error measure that serves as the objective for optimisation algorithms during training.", "entityType": "Class", "qualityScore": 0.72, "maturity": "stable", "iri": "urn:ngm:class:loss-function", "labels": [ "Loss Function", "Contrastive Loss", "Dice Loss", "Photometric Loss" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "lossless-compression", "title": "Lossless Compression", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Lossless compression is a class of data compression in which the original data can be reconstructed exactly, bit for bit, from the compressed representation. It exploits statistical redundancy through techniques such as entropy coding and dictionary substitution, contrasting with lossy compression which discards perceptually insignificant information for higher ratios. It is essential where fidelity must be preserved, such as for text, executables, and archival data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:lossless-compression", "labels": [ "Lossless Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [] }, { "id": "lossy-compression", "title": "Lossy Compression", "domain": "data", "domain_name": "Data", "definition": "Data compression that achieves high ratios by permanently discarding information judged perceptually or statistically less important, guided by rate-distortion theory and models of human vision and hearing; the basis of virtually all deployed image, audio and video coding \u2014 JPEG, MP3, AAC, Opus, H.264/HEVC/AV1 \u2014 where transform coding, quantisation and entropy coding together trade reconstruction fidelity against bitrate, in contrast to lossless methods that guarantee exact reconstruction.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:lossy-compression", "labels": [ "Lossy Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [ "Data Compression", "Lossless Compression", "Adaptive Bitrate Streaming", "Audio Codec" ] }, { "id": "low-code-platform", "title": "Low Code Platform", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A low-code platform is a software development environment that lets users build applications primarily through visual, model-driven interfaces such as drag-and-drop component composition, declarative configuration, and pre-built connectors, while still permitting hand-written code for custom logic. It abstracts away much boilerplate of conventional programming, accelerating delivery and broadening who can build software to include semi-technical citizen developers. Modern platforms increasingly embed AI assistance for component suggestion, data-model inference, and automated workflow generation, positioning them as a core layer in enterprise hyperautomation strategies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:low-code-platform", "labels": [ "Low Code Platform", "Low-Code Platform", "Low-Code Platforms" ], "is_subclass_of": [ "Application Development" ], "wikilinks": [] }, { "id": "low-earth-orbit-spacecraft", "title": "Low Earth Orbit Spacecraft", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:low-earth-orbit-spacecraft", "labels": [ "Low Earth Orbit Spacecraft" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "low-earth-orbit-spaceflight", "title": "Low Earth Orbit Spaceflight", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:low-earth-orbit-spaceflight", "labels": [ "Low Earth Orbit Spaceflight" ], "is_subclass_of": [ "Spaceflight" ], "wikilinks": [] }, { "id": "low-earth-orbit", "title": "Low Earth Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:low-earth-orbit", "labels": [ "Low Earth Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "low-energy-consumption", "title": "Low Energy Consumption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Low Energy Consumption characterises blockchain consensus mechanisms, particularly Proof-of-Stake and BFT variants, that achieve network security through economic staking rather than computational work, consuming orders of magnitude less electricity than Proof-of-Work systems. This property is a primary driver of blockchain sustainability assessments and ESG compliance, enabling networks to operate at scale without significant environmental impact.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:low-energy-consumption", "labels": [ "Low Energy Consumption" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Energy Consumption" ], "wikilinks": [ "Blockchain" ] }, { "id": "low-latency", "title": "Low Latency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Low latency is an engineering design property characterising systems in which the elapsed time between an input event and the corresponding system response is minimised to meet real-time interaction requirements. It is a cross-cutting concern spanning network topology, compute placement, operating-system scheduling, memory hierarchy, and hardware design. Achievable thresholds are domain-dependent \u2014 sub-100 \u00b5s in high-frequency trading, under 20 ms for vestibulo-ocular reflex alignment in extended-reality headsets, and under 150 ms for interactive video \u2014 and are reached through a combination of edge computing, kernel-bypass networking, hardware acceleration, and optimised serialisation. As a foundational property of distributed infrastructure, low latency is a prerequisite for real-time AI inference, immersive spatial computing, autonomous robotics, and ultra-reliable industrial control.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:low-latency", "labels": [ "Low Latency", "Low Latency Streaming", "Low-Latency APIs", "Low-Latency Delivery", "Low-Latency Display", "Low-Latency Experiences", "Low-Latency Serving", "Low-latency ML Pipeline", "Reduced Latency", "Ultra-Reliable Low-Latency Communications", "low-latency" ], "is_subclass_of": [ "Latency" ], "wikilinks": [] }, { "id": "low-latency-computing", "title": "Low-Latency Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Low-latency computing is an architectural discipline concerned with minimising the end-to-end delay between a request and its response, typically targeting sub-millisecond to single-digit-millisecond budgets. It combines hardware proximity, kernel-bypass networking, lock-free data structures, cache-aware memory layouts, and predictable scheduling to eliminate sources of jitter. Application domains include high-frequency trading, real-time multiplayer rendering, industrial control, telecommunications signalling, and interactive AI inference, where tail latency rather than average throughput is the governing performance metric.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:low-latency-computing", "labels": [ "Low-Latency Computing", "Low Latency Processing" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "low-latency-interaction", "title": "Low-Latency Interaction", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Low-latency interaction is the design property of a system in which the delay between a user action and the corresponding system response is minimised to preserve a sense of immediacy. In immersive and real-time contexts such as the metaverse, latency below roughly 20 milliseconds is critical to avoid perceptual disconnect and motion sickness. Achieving it requires optimised networking, predictive rendering, and edge computation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:low-latency-interaction", "labels": [ "Low-Latency Interaction" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "low-latency-network", "title": "Low-Latency Network", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A low-latency network is a communications network engineered to minimise the round-trip delay experienced by data packets travelling between endpoints. It combines short physical paths, fast switching, prioritised traffic handling, and edge placement of compute so that interactive and real-time applications respond within tight, predictable time bounds. Such networks are foundational to immersive and time-critical experiences where perceptible delay degrades usability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:low-latency-network", "labels": [ "Low-Latency Network", "Low Latency Network" ], "is_subclass_of": [ "Infrastructure", "Network Topology" ], "wikilinks": [] }, { "id": "low-latency-networking", "title": "Low-Latency Networking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Low-latency networking is the design and operation of network architectures that minimise the round-trip delay between communicating endpoints, typically targeting single-digit or sub-millisecond latencies. It combines edge placement, optimised transport protocols, traffic prioritisation and predictable routing to support interactive and time-critical workloads. In spatial computing it underpins responsive immersive experiences where perceptual lag must remain below human-detectable thresholds.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:low-latency-networking", "labels": [ "Low-Latency Networking" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "low-rank-adaptation", "title": "Low-Rank Adaptation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Low-Rank Adaptation (LoRA) is a parameter-efficient fine-tuning method that injects trainable low-rank decomposition matrices into the weight matrices of a frozen pre-trained model, enabling task adaptation with a fraction of the trainable parameters required by full fine-tuning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:low-rank-adaptation", "labels": [ "Low-Rank Adaptation", "Low-Rank Factorisation", "Rank Decomposition" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "Neural Network", "Transfer Learning", "GPT", "Parameter-Efficient Fine-Tuning", "https://arxiv.org/abs/2106.09685", "https://github.com/microsoft/LoRA" ] }, { "id": "low-rank-decomposition", "title": "Low-Rank Decomposition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Low-rank decomposition is a linear algebra technique that approximates a matrix as the product of two or more smaller matrices of lower rank, reducing the number of parameters needed to represent it. In machine learning it underlies parameter-efficient fine-tuning methods such as LoRA, which learn a small low-rank update to a pretrained weight matrix instead of updating all parameters. It trades a controlled amount of representational capacity for large reductions in memory and compute cost.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:low-rank-decomposition", "labels": [ "Low-Rank Decomposition" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "lower-atmosphere", "title": "Lower Atmosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lower-atmosphere", "labels": [ "Lower Atmosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "lower-limb-exoskeleton", "title": "Lower Limb Exoskeleton", "domain": "robotics", "domain_name": "Robotics", "definition": "A Lower Limb Exoskeleton is a wearable robotic device that attaches externally to the legs and pelvis to assist, augment, or rehabilitate walking and lower-body movements by generating torques at hip, knee, and ankle joints. Devices span fully-active systems driven by electric actuators or hydraulics, passive systems using spring and damping elements to redirect energy, and hybrid configurations combining both. Applications include stroke rehabilitation (restoring gait through repetitive, robot-guided movement therapy), spinal-cord-injury locomotion assistance, load-carrying augmentation for industrial and military tasks, and fall prevention in elderly populations. Control strategies range from finite-state machines triggered by gait-phase detection through inertial measurement units and pressure sensors, to intent-recognition systems integrating electromyography and machine learning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:lower-limb-exoskeleton", "labels": [ "Lower Limb Exoskeleton" ], "is_subclass_of": [ "Robot Type", "Exoskeleton Robot" ], "wikilinks": [ "Exoskeleton Robot", "Robotics" ] }, { "id": "loyalty-programs", "title": "Loyalty Programs", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Loyalty Programs are structured incentive mechanisms that reward repeat customer engagement through points, tiers, or token-based systems. In spatial computing and metaverse contexts, loyalty programmes are increasingly tokenised on blockchain networks using loyalty tokens or NFTs, enabling cross-brand redemption, secondary-market trading, and programmable reward logic embedded in smart contracts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:loyalty-programs", "labels": [ "Loyalty Programs" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "loyalty-token", "title": "Loyalty Token", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital token granting repeat-use or membership rewards within a metaverse ecosystem, enabling customer engagement, brand loyalty programs, and tokenized incentive mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:loyalty-token", "labels": [ "Loyalty Token" ], "is_subclass_of": [ "Platform and Environment", "Loyalty Programs" ], "wikilinks": [ "Brand Engagement", "Cryptographic Signature", "Incentive Mechanism", "Membership Program", "MSF Use Cases", "Blockchain", "Crypto Token", "Customer Rewards", "Digital Wallet", "MiddlewareLayer", "Smart Contract", "Token Standard", "Virtual Asset", "VirtualEconomyDomain", "VirtualSocietyDomain" ] }, { "id": "lpwan", "title": "Lpwan", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Low-Power Wide-Area Network (LPWAN) is a class of wireless communication technologies designed to transmit small amounts of data over long distances at very low power consumption, enabling battery-operated IoT devices to operate for years on a single charge. LPWAN technologies trade high data rates for extended range and deep indoor penetration, covering areas from a few kilometres to tens of kilometres per base station. Major LPWAN variants include LoRaWAN, Sigfox, NB-IoT, and LTE-M, each suited to different deployment and regulatory environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:lpwan", "labels": [ "Lpwan", "LPWAN" ], "is_subclass_of": [ "Infrastructure", "Wireless Communication" ], "wikilinks": [] }, { "id": "luma-ai", "title": "Luma AI", "domain": "ai", "domain_name": "Ai", "definition": "Luma AI is a company developing generative tools for 3D capture and video generation. Its products include neural reconstruction of scenes from photographs and a text-to-video model.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:luma-ai", "labels": [ "Luma AI" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Neural Network", "Video Generation", "3D Generation", "3D Asset", "Computer Vision", "Generative AI" ] }, { "id": "lumen", "title": "Lumen", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Lumen is Unreal Engine's fully dynamic global illumination and reflections system that computes indirect lighting in real time without precomputed lightmaps. It uses a combination of screen-space tracing, software ray tracing against signed distance fields, and optional hardware ray tracing to produce diffuse interreflection and reflections that update as scenes and lights change. Lumen enables physically plausible lighting in interactive applications and games.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:lumen", "labels": [ "Lumen" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "luminosity", "title": "Luminosity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:luminosity", "labels": [ "Luminosity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "lump-of-labor-fallacy", "title": "Lump of Labor Fallacy", "domain": "economics", "domain_name": "Economics", "definition": "An economic fallacy that assumes a fixed total amount of work exists, leading to the incorrect conclusion that technological increases in labor supply or productivity necessarily result in unemployment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:lump-of-labor-fallacy", "labels": [ "Lump of Labor Fallacy" ], "is_subclass_of": [ "Economic Impact of AI" ], "wikilinks": [] }, { "id": "lunar-orbit-environment", "title": "Lunar Orbit Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lunar-orbit-environment", "labels": [ "Lunar Orbit Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "lunar-orbit-radiation-environment", "title": "Lunar Orbit Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lunar-orbit-radiation-environment", "labels": [ "Lunar Orbit Radiation Environment" ], "is_subclass_of": [ "Lunar Radiation Environment" ], "wikilinks": [] }, { "id": "lunar-radiation-environment", "title": "Lunar Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lunar-radiation-environment", "labels": [ "Lunar Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "lunar-surface-environment", "title": "Lunar Surface Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lunar-surface-environment", "labels": [ "Lunar Surface Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "lunar-surface-radiation-environment", "title": "Lunar Surface Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:lunar-surface-radiation-environment", "labels": [ "Lunar Surface Radiation Environment" ], "is_subclass_of": [ "Lunar Radiation Environment", "Surface Radiation Environment" ], "wikilinks": [] }, { "id": "luxury-goods-authentication", "title": "Luxury Goods Authentication", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Luxury Goods Authentication is the application of blockchain technology to combat counterfeit luxury goods through immutable digital certificates of authenticity, NFC-enabled provenance records, and cryptographic verification that cannot be physically replicated. Implementations such as the LVMH Aura Blockchain Consortium register millions of luxury products with unique digital identities linked to manufacturing details, ownership history, and authentication credentials, enabling both primary-market consumer confidence and secondary-market resale verification.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:luxury-goods-authentication", "labels": [ "Luxury Goods Authentication", "BC-0444-luxury-goods-authentication" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Arianee", "BC-0013-smart-contracts", "BC-0023-zero-knowledge-proofs", "BC-0029-permissioned-blockchain", "BC-0066-ethereum", "BC-0197-non-fungible-tokens", "BC-0214-environmental-sustainability", "BC-0432-consortium-blockchain", "BC-0434-blockchain-as-a-service", "BC-0441-provenance-tracking", "ConsenSys", "De Beers Tracr", "Entrupy", "Internet of Things", "Otis", "Quorum", "Rally", "VeChain", "Blockchain", "BlockchainDomain" ] }, { "id": "ly-coris", "title": "LyCORIS", "domain": "ai", "domain_name": "Ai", "definition": "LyCORIS (Lora beYond Conventional methods, Other Rank adaptation Implementations for Stable diffusion) is an open-source library implementing a family of parameter-efficient fine-tuning methods for diffusion and other models that extend beyond standard low-rank adaptation. It includes techniques such as LoHa (Hadamard-product decomposition), LoKr (Kronecker-product decomposition), and full or convolutional adaptations, giving practitioners a richer set of expressiveness-versus-size trade-offs. LyCORIS is widely used in the image-generation community to train compact, shareable model adapters.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:ly-coris", "labels": [ "LyCORIS" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [] }, { "id": "m-bridge", "title": "M-Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "mBridge (Project mBridge) is a multi-central-bank digital currency platform built on distributed ledger technology to enable real-time, cross-border, multi-currency payments and foreign-exchange settlement. Developed by the BIS Innovation Hub with central banks including those of China, Hong Kong, Thailand, and the UAE, it lets participating institutions transact wholesale CBDCs on a shared ledger. It aims to reduce cost, latency, and intermediary dependence in correspondent banking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:m-bridge", "labels": [ "M-Bridge" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "maci", "title": "MACI", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Minimal Anti-Collusion Infrastructure, a set of smart contracts and zero-knowledge techniques designed to reduce bribery and collusion in on-chain voting. It hides individual votes from coercers while allowing public verification of results.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:maci", "labels": [ "MACI", "MACI Anti-Collusion", "MACI v2" ], "is_subclass_of": [ "Quadratic Voting" ], "wikilinks": [ "Zero-Knowledge Proof", "Decentralized Autonomous Organization", "Quadratic Voting" ] }, { "id": "mas", "title": "MAS", "domain": "ai", "domain_name": "Ai", "definition": "A Multi-Agent System (MAS) is a computational framework composed of multiple interacting autonomous agents that perceive their environment and act to achieve individual or collective goals. Each agent maintains local knowledge, reasoning capabilities, and the ability to communicate with peers, enabling emergent collective intelligence without centralised control. MAS architectures address problems that are too complex, distributed, or dynamic for monolithic solutions, drawing on distributed AI, game theory, and coordination theory. The paradigm spans reactive swarm systems, deliberative BDI agents, and modern LLM-orchestrated agentic pipelines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mas", "labels": [ "MAS" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "mcp-client", "title": "MCP Client", "domain": "ai", "domain_name": "Ai", "definition": "An MCP client is the component within an AI host application that establishes and maintains a connection to one or more Model Context Protocol servers. It negotiates capabilities, forwards tool, resource, and prompt requests from the language model, and relays results back into the model's context. Each client maintains a one-to-one session with a server, mediating the model's access to external systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mcp-client", "labels": [ "MCP Client" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "mcp-connectors", "title": "MCP Connectors", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Standardized interfaces that allow AI models to connect to external tools, data sources, and systems via the Model Context Protocol.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:mcp-connectors", "labels": [ "MCP Connectors" ], "is_subclass_of": [ "AI Hardware Infrastructure" ], "wikilinks": [] }, { "id": "mcp-server", "title": "MCP Server", "domain": "ai", "domain_name": "Ai", "definition": "An MCP server is a program that exposes tools, resources, and prompts to AI applications through the Model Context Protocol. It advertises its capabilities during connection negotiation and executes requests forwarded by an MCP client, returning structured results for the model to consume. Servers encapsulate access to external systems such as databases, APIs, file systems, and developer tooling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mcp-server", "labels": [ "MCP Server" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "mems", "title": "MEMS", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "MEMS (micro-electro-mechanical systems) are miniaturised devices that combine mechanical and electrical components at the micrometre scale, fabricated using semiconductor manufacturing processes. They are the dominant technology for small, low-cost inertial sensors such as accelerometers and gyroscopes used in inertial measurement units. MEMS sensors trade some precision for size, cost, and power advantages that make them practical for consumer and mobile robotics applications.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:mems", "labels": [ "MEMS" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "mev", "title": "MEV", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Maximal Extractable Value (MEV) represents the profit that block producers can extract through strategic transaction ordering, inclusion, or exclusion within blocks, arising from their privileged position to control execution sequencing in blockchain networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mev", "labels": [ "MEV", "MEV Extraction", "MEV Manipulation", "Maximal Extractable Value" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain" ], "wikilinks": [ "Consensus Security", "DeFi Ecosystem", "Front-Running", "Proposer-Builder Separation", "Transaction Ordering", "Blockchain", "BlockchainDomain", "Mempool" ] }, { "id": "mimo-antenna", "title": "MIMO Antenna", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A MIMO (multiple-input multiple-output) antenna is an array of multiple transmit and receive elements that exploits spatial multiplexing to send several independent data streams over the same frequency channel. By leveraging multipath propagation, it increases spectral efficiency and link reliability without additional bandwidth, and is foundational to Wi-Fi and 4G/5G physical layers. Massive MIMO scales this to dozens or hundreds of elements with beamforming.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mimo-antenna", "labels": [ "MIMO Antenna" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "mitre-atlas", "title": "MITRE ATLAS", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "MITRE ATLAS is a knowledge base of adversary tactics and techniques against machine learning systems. It is modelled on the MITRE ATT&CK framework and curated by MITRE.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mitre-atlas", "labels": [ "MITRE ATLAS", "MITRE ATT&CK" ], "is_subclass_of": [ "Security Framework" ], "wikilinks": [ "Adversarial Machine Learning", "AI Safety", "Machine Learning", "Security Framework", "https://atlas.mitre.org", "https://github.com/mitre-atlas" ] }, { "id": "experiment-tracking", "title": "ML Experiment Tracking", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The systematic recording of every machine-learning training run \u2014 hyperparameters, code version, dataset version, environment, metrics, and resulting artefacts \u2014 in a queryable store, so that results can be compared across runs, reproduced exactly, and promoted to production with a complete audit trail linking a deployed model back to the precise conditions that produced it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:experiment-tracking", "labels": [ "ML Experiment Tracking", "Experiment Tracking" ], "is_subclass_of": [ "MLOps" ], "wikilinks": [ "MLOps", "Data Versioning", "Reproducibility", "MLflow", "DVC" ] }, { "id": "mlcommons", "title": "MLCommons", "domain": "ai", "domain_name": "Ai", "definition": "MLCommons is an open engineering consortium that builds benchmarks, datasets, and best practices to accelerate machine learning innovation in a fair and reproducible way. It is best known for the MLPerf benchmark suites, which measure training and inference performance across hardware and software stacks. Membership includes major chip vendors, cloud providers, and research institutions.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mlcommons", "labels": [ "MLCommons", "MLCommons MLPerf" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "mlops", "title": "MLOps", "domain": "ai", "domain_name": "Ai", "definition": "MLOps (Machine Learning Operations) is the set of practices, tools, and cultural norms that operationalise machine learning models at production scale by applying DevOps and Site Reliability Engineering principles to the full ML lifecycle. It covers end-to-end automation of ML pipelines \u2014 data ingestion, feature engineering, model training, evaluation, deployment, serving, and continuous retraining \u2014 and addresses the unique challenge that code, data, and model weights all evolve independently and must be versioned, tested, and governed together. MLOps introduces specialised artefacts such as model registries, feature stores, and experiment trackers that have no direct analogue in traditional software delivery. The discipline bridges the organisational gap between data science teams and production engineering, enabling reliable, auditable, and scalable model delivery at the pace business demands.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mlops", "labels": [ "MLOps", "MLOps Infrastructure", "MLOps Platform", "Mlops" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "mlflow", "title": "MLflow", "domain": "ai", "domain_name": "Ai", "definition": "MLflow is an open-source platform for managing the machine learning lifecycle, covering experiment tracking, reproducible runs, model packaging, and a model registry. It records parameters, metrics, code versions, and artefacts including checkpoints, enabling teams to compare experiments and promote models to production. It is framework-agnostic and integrates with most training libraries and serving backends.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mlflow", "labels": [ "MLflow" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "mmpose", "title": "MMPose", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "MMPose is an open-source pose estimation toolbox, part of the OpenMMLab project, providing reference implementations, pretrained models and training pipelines for 2D and 3D human and animal pose estimation. It supports a wide range of architectures, including top-down and bottom-up keypoint detectors, and is widely used as a backbone library by downstream research and production systems that require whole-body or hand keypoint estimation, such as DWPose. Its modular configuration system allows researchers to swap backbones, datasets and training schedules without rewriting pipeline code.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:mmpose", "labels": [ "MMPose" ], "is_subclass_of": [ "Pose Estimation" ], "wikilinks": [] }, { "id": "mpeg", "title": "MPEG", "domain": "standards", "domain_name": "Standards", "definition": "The Moving Picture Experts Group, the ISO/IEC standardisation body (formally ISO/IEC JTC 1/SC 29 working groups) that since 1988 has defined the dominant international standards for compressed digital audio and video \u2014 MPEG-1 and MP3, MPEG-2 for broadcast television and DVD, MPEG-4 AVC/H.264, HEVC, and VVC for streaming, plus systems standards such as MP4, MPEG-DASH, and MPEG-H spatial audio \u2014 underpinning virtually all modern media distribution.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:mpeg", "labels": [ "MPEG" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "Standards Body", "ISO", "Video Compression", "Spatial Audio" ] }, { "id": "mqtt", "title": "mqtt", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "MQTT (Message Queuing Telemetry Transport) is a lightweight publish-subscribe messaging protocol standardised as ISO/IEC 20922 and OASIS MQTT 5.0, designed for constrained devices and low-bandwidth or unreliable networks with minimal protocol overhead. Clients connect to a central broker that routes messages hierarchically by topic string; publishers post payloads to topic endpoints and subscribers receive all messages matching wildcard-capable topic filters. MQTT defines three quality-of-service levels\u2014at most once (QoS 0), at least once (QoS 1), and exactly once (QoS 2)\u2014enabling integrators to balance delivery guarantees against network and compute cost across heterogeneous deployments spanning embedded microcontrollers to cloud-scale brokers.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:mqtt", "labels": [ "MQTT", "MQTT Broker", "MQTT Protocol" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "ms-phi-models", "title": "MS Phi models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "MS Phi models is Microsoft Research's family of small language models (SLMs) demonstrating that carefully curated, textbook-quality synthetic training data can achieve state-of-the-art performance at parameter counts one to two orders of magnitude smaller than frontier large language models, esta...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ms-phi-models", "labels": [ "MS Phi models" ], "is_subclass_of": [ "AI Model Architecture", "General Purpose AI Model", "Proprietary Large Language Models", "Large-Scale Pretrained Foundation Model", "Large Language Models", "Generative AI", "Generative Model" ], "wikilinks": [ "EdgeAIDomain", "FoundationModelsDomain", "InferenceLayer", "ModelLayer", "Activation Function", "Adapter Modules", "AI Alignment", "AI-Augmented Software Engineering", "AI Deployment", "AI-GroundedDomain", "AI Model Card", "AI Safety", "Algorithmic Bias", "Anthropic Claude", "ApplicationLayer", "Attention", "Attention Head", "Attention Mechanism", "BERT", "Bias in Large Language Models" ] }, { "id": "mteb-benchmark", "title": "MTEB Benchmark", "domain": "ai", "domain_name": "Ai", "definition": "MTEB (Massive Text Embedding Benchmark) is a standardised evaluation suite that measures text embedding models across many tasks, including retrieval, classification, clustering, reranking, and semantic similarity, over numerous datasets and languages. It provides a public leaderboard that has become the reference for comparing embedding models. Strong MTEB scores are widely used to select embeddings for semantic search and retrieval-augmented generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mteb-benchmark", "labels": [ "MTEB Benchmark" ], "is_subclass_of": [ "Evaluation Metric", "AI Technique" ], "wikilinks": [] }, { "id": "machine-learning-accelerator", "title": "Machine Learning Accelerator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Machine Learning Accelerator is specialised hardware designed to execute the dense linear-algebra workloads of neural networks far more efficiently than general-purpose CPUs. It optimises matrix multiplication, convolution and tensor operations through massive parallelism, dedicated multiply-accumulate arrays and high memory bandwidth. Examples span GPUs, tensor processing units, FPGAs and custom ASICs deployed from data centres to edge and embedded devices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:machine-learning-accelerator", "labels": [ "Machine Learning Accelerator" ], "is_subclass_of": [ "Hardware Acceleration" ], "wikilinks": [] }, { "id": "machine-learning-classifier", "title": "Machine Learning Classifier", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A machine learning classifier is a supervised learning model trained to assign an input to one of a discrete set of predefined categories, based on patterns learned from labelled training examples. Implementations range from logistic regression and decision trees to deep neural networks, selected according to data volume, interpretability needs and accuracy requirements. Classifiers are deployed in screening and enforcement pipelines that must flag entities matching a target category, such as sanctions screening.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-classifier", "labels": [ "Machine Learning Classifier" ], "is_subclass_of": [ "Supervised Learning" ], "wikilinks": [] }, { "id": "machine-learning-discipline", "title": "Machine Learning Discipline", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine Learning is the branch of artificial intelligence in which systems learn predictive or generative models directly from data, without being explicitly programmed with domain rules. It encompasses supervised, unsupervised, and reinforcement learning paradigms, and forms the foundation for deep learning, natural language processing, and computer vision applications. Practical machine learning involves data preparation, feature engineering, model selection, training, evaluation, and deployment within a production pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:machine-learning-discipline", "labels": [ "Machine Learning Discipline", "Machine Learning", "Machine Learning (optional)", "MachineLearning" ], "is_subclass_of": [ "Artificial Intelligence Core", "AI Research Area" ], "wikilinks": [ "decision making", "ArtificialIntelligenceDomain", "collaboration", "latent space", "machine learning", "Overview of Machine Learning Techniques", "Telethrone" ] }, { "id": "machine-learning-framework", "title": "Machine Learning Framework", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Software libraries and development environments such as TensorFlow and PyTorch that provide tools, APIs, and abstractions for building, training, and deploying machine learning models; encompassing model definition, automatic differentiation, GPU-accelerated training, and production serving infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-discipline-framework", "labels": [ "Machine Learning Framework" ], "is_subclass_of": [ "AI Development Tools" ], "wikilinks": [ "AI Model Development", "AI Development Tools", "metaverse" ] }, { "id": "machine-learning-infrastructure", "title": "Machine Learning Infrastructure", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Hardware and software systems that support machine learning workloads, including GPU clusters, cloud computing platforms, distributed storage systems, and orchestration tools required for training and deploying AI models at scale.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-discipline-infrastructure", "labels": [ "Machine Learning Infrastructure" ], "is_subclass_of": [ "AI Infrastructure", "Computing Infrastructure" ], "wikilinks": [ "Scalable AI Training", "Computing Infrastructure", "metaverse" ] }, { "id": "machine-learning-model", "title": "Machine Learning Model (Artefact)", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The trained artefact produced by a machine learning pipeline: a parameterised function \u2014 such as a decision tree ensemble, support vector machine or neural network \u2014 whose weights have been fitted to data by an optimisation procedure, and which maps new inputs to predictions, classifications, rankings or generated content; the unit that is evaluated, selected, versioned, deployed, monitored and eventually retrained or retired in production systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:machine-learning-model", "labels": [ "Machine Learning Model (Artefact)", "Machine Learning Model" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [ "Machine Learning", "Supervised Learning", "Feature Selection", "Neural Network" ] }, { "id": "machine-learning-discipline-model", "title": "Machine Learning Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Computational algorithms trained on data to recognise patterns, make predictions, and perform tasks, including neural networks for content generation, behaviour simulation, computer vision, and natural language processing; the core artefact produced by a machine learning training pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-discipline-model", "labels": [ "Machine Learning Model", "MachineLearningModel" ], "is_subclass_of": [ "Machine Learning Discipline", "Machine Learning Models", "Artificial Intelligence" ], "wikilinks": [ "Intelligent Metaverse Systems", "Artificial Intelligence", "metaverse" ] }, { "id": "machine-learning-models", "title": "Machine Learning Models", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Categories and types of machine learning algorithms including classification models, regression models, clustering algorithms, and neural networks, each designed for specific prediction and pattern recognition tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-discipline-models", "labels": [ "Machine Learning Models" ], "is_subclass_of": [ "AI Model Architecture", "Machine Learning Discipline", "Artificial Intelligence" ], "wikilinks": [ "Automated Decision Making", "Artificial Intelligence", "metaverse" ] }, { "id": "machine-learning-operations", "title": "Machine Learning Operations", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Machine learning operations is the discipline of applying DevOps principles, tooling, and automation to the end-to-end machine learning lifecycle so that models can be reliably built, deployed, monitored, and retrained in production. It coordinates data pipelines, experiment tracking, model registries, continuous integration and delivery, serving infrastructure, and observability to bridge the gap between data-science experimentation and dependable operational systems. By treating data, code, and models as versioned, testable artifacts, it makes ML systems reproducible, auditable, and continuously improvable rather than fragile one-off deployments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:machine-learning-operations", "labels": [ "Machine Learning Operations", "MLOps" ], "is_subclass_of": [ "Infrastructure", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "machine-learning-pipeline", "title": "Machine Learning Pipeline", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Machine Learning Pipeline is the end-to-end automated workflow for developing, training, validating, deploying, and monitoring ML models. It encompasses data ingestion, preprocessing, feature engineering, model selection, hyperparameter tuning, training, evaluation, deployment, and continuous monitoring, and typically adopts MLOps practices with automated orchestration, versioning, and experiment tracking to ensure reproducibility, scalability, and maintainability of ML systems in production environments.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:machine-learning-discipline-pipeline", "labels": [ "Machine Learning Pipeline" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Experiment Tracking", "MLOps", "Model Deployment", "Feature Engineering" ] }, { "id": "machine-learning-platform", "title": "Machine Learning Platform", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Comprehensive cloud-based or enterprise software systems that provide integrated tools for building, training, deploying, and managing machine learning models, including AutoML capabilities, model registries, and MLOps features.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:machine-learning-discipline-platform", "labels": [ "Machine Learning Platform" ], "is_subclass_of": [ "AI Infrastructure", "Cloud Computing" ], "wikilinks": [ "Enterprise AI Adoption", "Cloud Computing", "metaverse" ] }, { "id": "machine-learning-research", "title": "Machine Learning Research", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Machine learning research is the systematic academic and industrial investigation of algorithms, theory, and systems for enabling computers to learn from data, spanning work published through peer-reviewed venues such as ICML, NeurIPS, and ICLR. It is conducted across universities, corporate research labs, and increasingly through collaborations that combine academic theoretical grounding with industrial-scale compute and data. Research output ranges from foundational theory, such as generalisation bounds and optimisation analysis, to applied advances in model architectures and training methods that are rapidly absorbed into production systems. Its pace and direction are shaped by the availability of compute, benchmark datasets, and open publication norms that allow results to be reproduced and built upon.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:machine-learning-research", "labels": [ "Machine Learning Research" ], "is_subclass_of": [ "Academic Research" ], "wikilinks": [] }, { "id": "machine-learning-software", "title": "Machine Learning Software", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Machine learning software is the category of tools, libraries, and platforms, distinct from any single learning technique, that support the development, training, deployment, and management of machine learning models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:machine-learning-software", "labels": [ "Machine Learning Software" ], "is_subclass_of": [ "Machine Learning", "Software Engineering" ], "wikilinks": [] }, { "id": "machine-learning-technique", "title": "Machine Learning Technique", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A specific algorithmic approach or methodology within machine learning used to enable systems to learn from data. Machine learning techniques span supervised, unsupervised, and reinforcement paradigms and include methods such as gradient descent optimisation, backpropagation, fine-tuning, transfer learning, boosting, and bagging that enable models to improve performance through exposure to training data.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:machine-learning-discipline-technique", "labels": [ "Machine Learning Technique" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning" ] }, { "id": "machine-learning-techniques-survey", "title": "Machine Learning Techniques Survey", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A structured survey of the principal paradigms and architectures in machine learning, spanning supervised methods (SVMs, decision trees, logistic regression), unsupervised clustering (k-means, KNN), and deep learning approaches (neural networks, transformers, diffusion models, GANs). The survey contextualises training paradigms including reinforcement learning from human feedback and direct preference optimisation, and positions large proprietary language models within the broader ML taxonomy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:machine-learning-discipline-techniques-survey", "labels": [ "Machine Learning Techniques Survey", "Overview of Machine Learning Techniques" ], "is_subclass_of": [ "AI Technique", "Deep Learning" ], "wikilinks": [ "presentation", "RLHF", "Diffusion Models", "Direct Preference Optimization", "Generative Adversarial Networks", "Proprietary Large Language Models", "State Space and Other Approaches", "Transformers" ] }, { "id": "machine-learning", "title": "Machine Learning", "domain": "ai", "domain_name": "Ai", "definition": "Machine Learning is a sub-field of Artificial Intelligence in which computational systems learn to improve their performance on tasks by identifying statistical patterns in data rather than by following explicitly programmed rules. A model is exposed to a training dataset, an optimisation algorithm adjusts its parameters to minimise a loss function, and the resulting parameters generalise to unseen inputs. The discipline encompasses supervised learning from labelled examples, unsupervised learning from unlabelled structure, and reinforcement learning from reward signals in interactive environments. Foundational to modern AI stacks, machine learning underpins capabilities ranging from image recognition and natural language processing to recommendation systems and autonomous decision-making.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:machine-learning", "labels": [ "Machine Learning", "International Machine Learning Society", "Machine Learning Community", "Machine Learning Decoder", "Machine Learning Denoising", "Machine Learning for Graphics", "Standard Machine Learning" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique" ], "wikilinks": [] }, { "id": "machine-translation", "title": "Machine Translation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine Translation is the automated translation of text or speech from one natural language to another using neural network models, particularly transformer-based sequence-to-sequence architectures. Modern neural MT systems achieve near-human quality through pre-training on massive multilingual corpora, cross-lingual transfer learning, and attention mechanisms that model long-range linguistic dependencies.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:machine-translation", "labels": [ "Machine Translation", "Simultaneous Machine Translation" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Chain of Thought", "Sequence-to-Sequence Model", "Anthropic Claude", "ArtificialIntelligenceDomain", "Diagrams as Code", "Gemini", "Google", "Large Language Models", "Metaverse Ontology", "Natural Language Processing", "Prompt Engineering", "Transformer" ] }, { "id": "machine-unlearning", "title": "Machine Unlearning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of techniques for removing the influence of specific training examples from an already-trained machine learning model without retraining it from scratch, so that the resulting model behaves as if the deleted data had never been seen; motivated by privacy law (the right to erasure), copyright disputes, data poisoning remediation, and the removal of hazardous capabilities from foundation models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:machine-unlearning", "labels": [ "Machine Unlearning" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [ "Machine Learning", "Continual Learning", "Data Privacy" ] }, { "id": "machine-vision", "title": "Machine Vision", "domain": "ai", "domain_name": "Ai", "definition": "Machine vision is the engineering discipline that applies digital imaging, optics, and automated image analysis to perform inspection, measurement, guidance, and identification tasks in industrial and manufacturing environments. It integrates hardware components \u2014 cameras, illumination, optics, and frame grabbers \u2014 with software pipelines that extract actionable decisions from image data, typically operating in real time under strict cycle-time constraints. Distinguished from the broader research field of computer vision by its emphasis on reliability, determinism, and seamless integration with programmable logic controllers and industrial automation systems. Modern machine vision systems increasingly incorporate deep learning inference alongside classical morphological and blob-analysis algorithms to handle appearance variability that rule-based methods alone cannot address.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:machine-vision", "labels": [ "Machine Vision", "Industrial Machine Vision" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "Sensor", "Object Detection", "Computer Vision" ] }, { "id": "machine-to-machine-payments", "title": "Machine to Machine Payments", "domain": "finance", "domain_name": "Finance", "definition": "Machine to Machine Payments (M2M Payments) refers to automated financial transactions initiated and settled between autonomous devices, software agents, or AI systems without direct human involvement in each individual transaction. These payments enable IoT devices, autonomous vehicles, robots, and AI agents to pay for resources, services, or data on a per-use or streaming basis, forming the economic layer of the machine economy. M2M payments typically require programmable money, low transaction fees, high throughput, and reliable settlement guarantees, properties that have driven interest in blockchain-based payment channels, stablecoins, and central bank digital currencies as settlement rails.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:machine-to-machine-payments", "labels": [ "Machine to Machine Payments", "IoT Machine Payments", "Machine Payments", "Machine-to-Machine Payment", "Machine-to-Machine Payments" ], "is_subclass_of": [ "Digital Payment System" ], "wikilinks": [] }, { "id": "macroeconomics", "title": "Macroeconomics", "domain": "finance", "domain_name": "Finance", "definition": "Macroeconomics is the branch of economics that studies the behaviour and performance of an economy as a whole, examining aggregate phenomena such as gross domestic product, price levels, unemployment, business cycles, and long-run growth. It analyses how monetary and fiscal policy shape inflation, output, and employment, and how international trade and capital flows interconnect national economies. The discipline develops formal models \u2014 from Keynesian demand frameworks to DSGE models \u2014 that link household, firm, and government behaviour to economy-wide outcomes. It contrasts with microeconomics by focusing on aggregate rather than individual-agent variables.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:macroeconomics", "labels": [ "Macroeconomics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [ "Inflation Hedge", "Inflation", "Monetary Policy", "Economics" ] }, { "id": "macroprudential-policy", "title": "Macroprudential Policy", "domain": "finance", "domain_name": "Finance", "definition": "Macroprudential policy is a regulatory approach aimed at safeguarding the stability of the financial system as a whole rather than the soundness of individual institutions, addressing systemic risk that arises from interconnections, common exposures, and procyclical behaviour. It deploys instruments such as countercyclical capital buffers, leverage limits, and stress testing to limit the build-up of vulnerabilities and to enhance resilience to shocks. It complements microprudential supervision and monetary policy in promoting financial stability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:macroprudential-policy", "labels": [ "Macroprudential Policy" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "magnetar", "title": "Magnetar", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetar", "labels": [ "Magnetar" ], "is_subclass_of": [ "Neutron Star" ], "wikilinks": [] }, { "id": "magnetic-pole", "title": "Magnetic Pole", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetic-pole", "labels": [ "Magnetic Pole" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetic-reconnection", "title": "Magnetic Reconnection", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetic-reconnection", "labels": [ "Magnetic Reconnection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetogram", "title": "Magnetogram", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetogram", "labels": [ "Magnetogram" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetopause", "title": "Magnetopause", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetopause", "labels": [ "Magnetopause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetorquer", "title": "Magnetorquer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetorquer", "labels": [ "Magnetorquer" ], "is_subclass_of": [ "Actuator" ], "wikilinks": [] }, { "id": "magnetosphere", "title": "Magnetosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetosphere", "labels": [ "Magnetosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetospheric-ring-current", "title": "Magnetospheric Ring Current", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetospheric-ring-current", "labels": [ "Magnetospheric Ring Current" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "magnetotail", "title": "Magnetotail", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:magnetotail", "labels": [ "Magnetotail" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "main-sequence-star", "title": "Main-sequence Star", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:main-sequence-star", "labels": [ "Main-sequence Star" ], "is_subclass_of": [ "Star" ], "wikilinks": [] }, { "id": "majority-voting", "title": "Majority Voting", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Majority voting is a decision rule under which an outcome is accepted only when it is supported by more than half of the participating nodes or samples, providing a simple mechanism for reaching agreement in the presence of disagreement or faults. In distributed consensus algorithms such as Paxos, a value is considered chosen once it has been acknowledged by a majority quorum of acceptors, which guarantees that any two majorities intersect and so prevents conflicting values from being chosen. In machine learning, majority voting aggregates multiple independently sampled outputs, as in self-consistency prompting, selecting the most frequent answer as the final prediction. Its fault tolerance follows directly from quorum intersection: a system with 2f+1 nodes tolerates f faults under majority voting.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:majority-voting", "labels": [ "Majority Voting" ], "is_subclass_of": [ "Consensus Algorithm" ], "wikilinks": [] }, { "id": "maker-dao", "title": "makerdao", "domain": "blockchain", "domain_name": "Blockchain", "definition": "MakerDAO is a decentralised autonomous organisation (DAO) on the Ethereum blockchain that governs the Maker Protocol \u2014 a system of smart contracts enabling users to generate the DAI stablecoin by locking over-collateralised assets in Vaults (formerly Collateralised Debt Positions). Governance decisions covering collateral onboarding, stability fees, liquidation ratios, and debt ceilings are enacted by MKR token holders through on-chain Executive Votes. As one of the earliest large-scale deployments of DAO governance and algorithmic stablecoin issuance, MakerDAO is a foundational pillar of decentralised finance (DeFi) and continues to evolve through its Endgame restructuring into a constellation of SubDAOs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:maker-dao", "labels": [ "MakerDAO" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "making-available", "title": "Making Available", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Under EU AI Act Article 3(14), Making Available is the supply of an AI system for distribution or use on the Union market in the course of a commercial activity, whether in return for payment or free of charge. It is broader than Placing on the Market, covering all subsequent distributions after initial market entry, and triggers compliance obligations for distributors, providers, and authorised representatives throughout the AI value chain.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:making-available", "labels": [ "Making Available" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "malware", "title": "Malware", "domain": "security", "domain_name": "Security", "definition": "Malware is software deliberately designed to damage, disrupt, gain unauthorised access to, or exfiltrate data from a computer system, encompassing categories such as viruses, worms, trojans, ransomware and spyware. It is typically delivered through an attack vector such as a phishing email, compromised download or vulnerable network service, then executes to achieve goals ranging from data theft to system disruption or extortion. Defence against malware combines endpoint detection, network monitoring, patching and user awareness within a broader cybersecurity programme.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:malware", "labels": [ "Malware" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "manchester-tech-cluster", "title": "Manchester Tech Cluster", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Manchester Tech Cluster is Greater Manchester's technology and digital innovation ecosystem, forming the UK's second-largest tech hub after London with over 10,000 digital businesses contributing a \u00a35 billion digital economy. The cluster is anchored by MediaCityUK, the University of Manchester, Manchester Science Park, and the ID Manchester innovation district, with notable strengths in artificial intelligence, FinTech, and eCommerce.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:manchester-tech-cluster", "labels": [ "Manchester Tech Cluster" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "manchester", "title": "Manchester", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Manchester is a major metropolitan city in North West England and the economic centre of the Greater Manchester Combined Authority, encompassing one of the United Kingdom's largest urban technology and digital economy clusters. The city hosts a dense concentration of universities, research institutes, data centres, fintech firms, and AI startups, anchored by institutions such as the University of Manchester and Manchester Metropolitan University. As a founding pillar of the Northern Powerhouse initiative, Manchester serves as a model for post-industrial urban regeneration built on knowledge industries, advanced manufacturing, and distributed digital infrastructure. It is a primary node in the UK's regional connectivity fabric, with major fibre, rail, and aviation links enabling distributed collaboration across the northern city-regions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:manchester", "labels": [ "Manchester" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Digital Economy", "Northern Powerhouse", "Leeds", "Entity" ] }, { "id": "mango-markets", "title": "Mango Markets", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Mango Markets is a decentralised trading platform on Solana offering spot and perpetual markets, notable as the site of a large 2022 exploit that manipulated an oracle price to drain its treasury.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mango-markets", "labels": [ "Mango Markets" ], "is_subclass_of": [ "Decentralized Exchange" ], "wikilinks": [ "Decentralized Exchange", "Yield Farming", "Automated Market Maker", "DeFi" ] }, { "id": "manipulation", "title": "Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Manipulation is the capability of robotic or autonomous systems to physically interact with, grasp, reposition, and transform objects in an environment through controlled mechanical action. It integrates perception, kinematics, dynamics, and planning to enable precise, dexterous, and adaptive contact-rich tasks. Robotic manipulation encompasses the full pipeline from object detection and pose estimation through grasp planning, motion execution, and force-regulated contact control. As a foundational capability in intelligent systems, it bridges physical embodiment with higher-level task reasoning and is central to industrial automation, service robotics, and human-robot collaboration.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:manipulation", "labels": [ "Manipulation", "General-Purpose Manipulation", "Heavy Manipulation", "Manipulation Mechanism" ], "is_subclass_of": [ "Robotics Systems" ], "wikilinks": [ "Robotics Systems" ] }, { "id": "manipulator-arm", "title": "Manipulator Arm", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A multi-jointed mechanical or virtual robotic arm integrated with avatars or immersive systems to enable precise object manipulation and physical interaction within virtual or mixed reality environments through control systems, haptic feedback, and inverse kinematics algorithms.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:manipulator-arm", "labels": [ "Manipulator Arm", "ManipulatorArm" ], "is_subclass_of": [ "Interaction Technology", "Avatar Wearable" ], "wikilinks": [ "BlockchainLedger", "controlledBy", "dt:guidedBy", "dt:optimizedBy", "dt:simulatedIn", "dt:trackedOn", "dt:trainedBy", "executesMotion", "GraspPlanning", "hasEndEffector", "hasJoint", "CollisionAvoidance", "ComputerVision", "ForceControl", "InverseKinematics", "MachineLearning", "MetaverseDomain", "MotionControl", "ReinforcementLearning", "VirtualEnvironment" ] }, { "id": "manipulator-robot", "title": "Manipulator Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot consisting of a kinematic chain of rigid links connected by actuated joints, terminating in an end-effector, designed to position and orient objects or tools in a workspace. Manipulator robots include serial open-chain designs (articulated, SCARA, cylindrical, Cartesian) and parallel closed-chain designs, and are the dominant platform for industrial assembly, welding, pick-and-place, and collaborative tasks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:manipulator-robot", "labels": [ "Manipulator Robot", "Robot Manipulator" ], "is_subclass_of": [ "Robot Type", "Robot" ], "wikilinks": [ "Robot", "Robotics" ] }, { "id": "manipulator", "title": "Manipulator", "domain": "robotics", "domain_name": "Robotics", "definition": "A manipulator is a mechanically programmable device comprising a series of rigid links connected by actuated joints arranged in a serial or parallel kinematic chain, capable of displacing objects or tools through a defined workspace. The distal end carries an end-effector that interfaces directly with the task environment, enabling grasping, welding, assembly, or other physical interactions. Manipulators are parameterised by their degrees of freedom, workspace geometry, payload capacity, and control architecture, and are the foundational actuator subsystem in industrial, collaborative, and service robots. The ISO 8373:2021 standard defines a manipulator as the machine mechanism of a robot.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:manipulator", "labels": [ "Manipulator", "Industrial Manipulator", "Manipulator Definition", "RB-0003-manipulator", "Robotic Manipulator" ], "is_subclass_of": [ "Robot" ], "wikilinks": [ "ISO 8373:2021", "Robot (RB-0001)", "Robotics" ] }, { "id": "manometer", "title": "Manometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:manometer", "labels": [ "Manometer" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "mantle", "title": "Mantle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Mantle is an Ethereum layer-two network that uses a rollup design to lower transaction costs and increase throughput. It separates data availability from execution.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mantle", "labels": [ "Mantle" ], "is_subclass_of": [ "Layer 2 Scaling" ], "wikilinks": [ "Ethereum", "Rollup", "DeFi", "Optimistic Rollup", "Layer 2 Scaling", "https://www.mantle.xyz", "https://docs.mantle.xyz" ] }, { "id": "manufacturing-automation", "title": "Manufacturing Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "Manufacturing automation is the systematic application of control technologies, robotics, sensors, machine vision, and software to perform production tasks with minimal human intervention across discrete, process, and hybrid manufacturing environments. It encompasses programmable logic controllers (PLCs), industrial robots, computer numerical control (CNC) machines, distributed control systems (DCS), and integrated supervisory architectures that coordinate material flow, quality inspection, and assembly operations. Modern implementations leverage AI-driven process optimisation, digital twin synchronisation, and edge-cloud hybrid architectures to achieve adaptive, self-correcting production lines capable of real-time reconfiguration. The discipline aims to increase throughput, improve consistency, reduce cycle times, and enhance worker safety by reassigning operators away from hazardous or highly repetitive tasks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:manufacturing-automation", "labels": [ "Manufacturing Automation" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "manufacturing-process", "title": "Manufacturing Process", "domain": "robotics", "domain_name": "Robotics", "definition": "A manufacturing process is a structured sequence of physical transformations \u2014 including machining, assembly, forming, joining, heat treatment, surface finishing, and inspection \u2014 applied to raw materials or sub-components to produce finished goods meeting specified tolerances, material properties, and functional requirements. Manufacturing processes are characterised by their tooling, cycle time, throughput, waste profile, and quality control mechanisms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:manufacturing-process", "labels": [ "Manufacturing Process" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "map-projection", "title": "Map Projection", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:map-projection", "labels": [ "Map Projection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "maple-finance", "title": "Maple Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Maple Finance is a decentralised finance protocol that arranges undercollateralised and collateralised lending to institutional borrowers through on-chain credit pools.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:maple-finance", "labels": [ "Maple Finance" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Smart Contract", "DeFi", "Institutional Adoption" ] }, { "id": "mapped-data-state", "title": "Mapped Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mapped-data-state", "labels": [ "Mapped Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mapping", "title": "Mapping", "domain": "ai", "domain_name": "Ai", "definition": "Mapping is the process of creating structured correspondences between two or more domains, spaces, or data representations \u2014 including cartographic, semantic, and data-schema contexts. In spatial computing it denotes the construction of environment models; in knowledge engineering it describes the alignment of ontologies, schemas, or concept hierarchies; in AI it encompasses learned transformations between input and output spaces. Mapping is foundational to navigation, data integration, and cross-domain reasoning.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:mapping", "labels": [ "Mapping", "3D Mapping" ], "is_subclass_of": [ "Knowledge Graph" ], "wikilinks": [] }, { "id": "marathon-digital", "title": "Marathon Digital", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Marathon Digital is a large publicly traded Bitcoin mining company operating data centres of specialised hardware to validate transactions and compete for block rewards.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:marathon-digital", "labels": [ "Marathon Digital" ], "is_subclass_of": [ "Bitcoin Mining" ], "wikilinks": [ "Bitcoin Mining", "Bitcoin", "Cryptocurrency", "Sustainability Domain" ] }, { "id": "marine-remote-sensing", "title": "Marine Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:marine-remote-sensing", "labels": [ "Marine Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "marine-robot", "title": "Marine Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot designed to operate on or under water surfaces, encompassing unmanned surface vehicles (USVs), autonomous underwater vehicles (AUVs), and remotely operated vehicles (ROVs). Marine robots address unique challenges including buoyancy control, pressure resistance, underwater acoustic communication, and GPS-denied navigation using sonar and pressure sensors.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:marine-robot", "labels": [ "Marine Robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot" ], "wikilinks": [ "Mobile Robot", "Robotics" ] }, { "id": "markdown-diagramming-as-code-tool", "title": "Markdown Diagramming As Code Tool", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A JavaScript-based diagramming-as-code tool that renders flowcharts, sequence diagrams, Gantt charts, entity-relationship diagrams, and other diagram types from plain-text markup embedded in Markdown documents. Mermaid is widely adopted in developer documentation workflows, enabling version-controllable, maintainable diagrams without external tooling. It integrates natively with platforms such as GitHub, GitLab, Notion, and Logseq.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:markdown-diagramming-as-code-tool", "labels": [ "Markdown Diagramming As Code Tool", "Mermaid Chart", "Mermaid Open-Source Project", "Mermaid Specification", "Mermaid Syntax", "mermaid" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "markdown", "title": "Markdown", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Markdown is a lightweight plain-text markup language that uses simple punctuation conventions to denote formatting such as headings, lists, links, and emphasis. Created by John Gruber in 2004 and later standardised by efforts like CommonMark, it is designed to be readable as source and convertible to HTML and other formats. Its simplicity makes it the dominant format for documentation, notes, and content authoring in developer tooling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:markdown", "labels": [ "Markdown" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "marker-based-tracking", "title": "Marker Based Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AR and VR tracking technique that uses predefined visual patterns such as QR codes, April tags, ArUco markers, and fiducial markers to determine device position and orientation for accurate digital content overlay.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:marker-based-tracking", "labels": [ "Marker Based Tracking", "Marker Tracking", "Marker-Based Tracking" ], "is_subclass_of": [ "Interaction Technology", "Spatial Tracking Technology" ], "wikilinks": [ "AR Content Positioning", "metaverse", "Spatial Tracking Technology" ] }, { "id": "market-abuse-detection", "title": "Market Abuse Detection", "domain": "governance", "domain_name": "Governance", "definition": "Market abuse detection is the use of surveillance systems and analytics to identify illegal trading behaviours such as insider dealing, spoofing, layering, and price manipulation. It analyses order-book activity, trade patterns, and communications to flag anomalies for compliance review and regulatory reporting. It is a core obligation under regimes such as the EU Market Abuse Regulation and is increasingly powered by machine learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:market-abuse-detection", "labels": [ "Market Abuse Detection" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "market-access", "title": "Market Access", "domain": "robotics", "domain_name": "Robotics", "definition": "Market access is the ability of a product, service or system to be lawfully sold and adopted within a target jurisdiction or sector, conditioned on meeting regulatory, certification and conformity requirements. For robotic and physical-automation systems it depends on demonstrating compliance with safety, electromagnetic and product-directive obligations such as CE marking before placement on a market. Achieving market access is therefore a gateway to commercialisation rather than a purely commercial activity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:market-access", "labels": [ "Market Access" ], "is_subclass_of": [ "Robotics", "Safety and Standards" ], "wikilinks": [] }, { "id": "market-capitalization", "title": "Market Capitalization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Market Capitalization in a blockchain context is the aggregate market value of a cryptocurrency or token, computed as the circulating supply multiplied by the current unit price. It serves as a widely used proxy for the relative size, liquidity, and investor confidence of a crypto-economic network, underpinning index construction, risk categorisation, and portfolio weighting decisions by institutional participants. Critically, market capitalisation is a lagging and manipulable metric: thin order books, wash trading, and locked but counted supply can inflate the figure well beyond the value that could be liquidated in practice.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:market-capitalization", "labels": [ "Market Capitalization" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "market-depth", "title": "Market Depth", "domain": "finance", "domain_name": "Finance", "definition": "Market depth is a measure of the volume of buy and sell orders resting at different price levels around the current market price for an asset, indicating how much trading volume a market can absorb before price moves significantly. Greater depth means large orders can be filled with less price impact, and is a direct expression of the liquidity that market makers and liquidity providers supply to a market. Market depth is typically visualised as an order book showing cumulative bid and ask volume by price level.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:market-depth", "labels": [ "Market Depth" ], "is_subclass_of": [ "Liquidity" ], "wikilinks": [] }, { "id": "market-design", "title": "Market Design", "domain": "governance", "domain_name": "Governance", "definition": "Market design is the applied economics discipline concerned with constructing, analysing, and improving the rules, mechanisms, and institutions that govern the exchange of goods, services, and assets in real-world markets. Drawing on game theory, mechanism design, and matching theory, market designers identify failures in existing markets \u2014 such as the absence of money in matching markets, thickness problems, or congestion \u2014 and engineer interventions that produce efficient, stable, and fair outcomes. Prominent applications include kidney exchange programmes, school choice systems, spectrum auctions, and electricity markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:market-design", "labels": [ "Market Design" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "market-efficiency", "title": "Market Efficiency", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Market efficiency is the degree to which asset prices fully and rapidly reflect all available information, leaving no systematic opportunity for risk-adjusted excess returns. The efficient-market hypothesis frames it in weak, semi-strong, and strong forms according to the information set incorporated. Higher efficiency implies accurate price signals, lower transaction costs, and effective capital allocation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:market-efficiency", "labels": [ "Market Efficiency" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "market-expansion", "title": "Market Expansion", "domain": "governance", "domain_name": "Governance", "definition": "Market expansion is the growth strategy of extending a product or service into new customer segments, geographies, or use cases to increase the total addressable market. In technology contexts it is often unlocked by removing access barriers such as language, cost, or accessibility limitations. It is a primary commercial outcome of capabilities like real-time translation and inclusive design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:market-expansion", "labels": [ "Market Expansion" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "market-integrity", "title": "Market Integrity", "domain": "finance", "domain_name": "Finance", "definition": "Market integrity refers to the set of conditions, rules, and enforcement mechanisms that ensure financial and digital asset markets operate fairly, transparently, and free from manipulation, fraud, or systemic abuse. It encompasses prevention of practices such as insider trading, wash trading, and front-running that distort price discovery. Regulatory authorities and market operators jointly maintain market integrity through surveillance, disclosure requirements, and sanction regimes. In decentralised contexts, algorithmic and on-chain mechanisms increasingly supplement traditional oversight.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:market-integrity", "labels": [ "Market Integrity" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "market-maker", "title": "Market Maker", "domain": "finance", "domain_name": "Finance", "definition": "A market maker is a firm or agent that stands ready to both buy and sell an asset continuously, quoting bid and ask prices and earning the spread between them. By absorbing temporary imbalances between buyers and sellers, market makers supply liquidity, tighten spreads, reduce slippage and accelerate price discovery on exchanges. In decentralised finance the role is generalised by automated market makers, which replace quoted order books with algorithmic pricing curves funded by pooled liquidity.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:market-maker", "labels": [ "Market Maker" ], "is_subclass_of": [ "Liquidity Provider" ], "wikilinks": [ "Liquidity Provider", "Order Book", "Automated Market Maker", "Bid Ask Spread" ] }, { "id": "market-making", "title": "Market Making", "domain": "finance", "domain_name": "Finance", "definition": "Market making is the practice of continuously quoting firm bid and ask prices for a financial asset or token, providing counterparty liquidity to traders who wish to buy or sell without waiting for a natural matching order. A market maker earns the bid-ask spread as compensation for bearing inventory risk and adverse-selection risk from better-informed traders. In traditional venues the function is performed by designated dealers or algorithmic trading firms managing limit-order books; in decentralised finance it is automated by liquidity-pool protocols that replace human quoting with deterministic pricing curves. The activity is fundamental to functional markets because it converts latent supply and demand into observable, executable prices through continuous [[Price Discovery]].", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:market-making", "labels": [ "Market Making", "Market Making Risk", "Traditional Market Making" ], "is_subclass_of": [ "Liquidity Provision" ], "wikilinks": [ "Price Discovery", "Order Book", "Automated Market Maker", "Liquidity Provision" ] }, { "id": "market-manipulation", "title": "Market Manipulation", "domain": "finance", "domain_name": "Finance", "definition": "Market manipulation is the deliberate attempt to interfere with the free and fair operation of a financial market by creating false or misleading appearances of supply, demand or price. It encompasses practices such as spoofing, wash trading, pump-and-dump schemes and the dissemination of false information. As a form of market abuse it is prohibited by securities regulation and undermines price discovery and investor protection.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:market-manipulation", "labels": [ "Market Manipulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "market-microstructure", "title": "Market Microstructure", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Market microstructure studies how the rules and mechanics of trading, such as order types, matching and information flow, shape prices, liquidity and transaction costs.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:market-microstructure", "labels": [ "Market Microstructure" ], "is_subclass_of": [ "Economics", "Economics Domain" ], "wikilinks": [ "Order Book", "Prediction Markets", "Automated Market Maker", "Decentralized Exchange", "Economics Domain" ] }, { "id": "market-risk", "title": "Market Risk", "domain": "finance", "domain_name": "Finance", "definition": "The risk of losses in on- and off-balance-sheet positions arising from adverse movements in market prices \u2014 interest rates, foreign exchange rates, equity prices, credit spreads, and commodity prices \u2014 measured with tools such as value-at-risk, expected shortfall, and sensitivity analysis, and capitalised under banking regulation alongside credit and operational risk.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:market-risk", "labels": [ "Market Risk" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [ "Risk Management", "Basel III", "Operational Risk", "Stress Testing" ] }, { "id": "market-surveillance-authority", "title": "Market Surveillance Authority", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "National authority responsible for carrying out market surveillance activities on AI systems, including inspections, testing, enforcement, and ensuring compliance with EU AI Act requirements within a Member State.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:market-surveillance-authority", "labels": [ "Market Surveillance Authority" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "market-surveillance", "title": "Market Surveillance", "domain": "governance", "domain_name": "Governance", "definition": "Market surveillance is the ongoing monitoring of trading venues and participant activity by exchanges and regulators to detect manipulation, abuse, and systemic risk. It ingests order, trade, and reference data in real time, applies alerting logic, and escalates suspected violations for enforcement. It underpins fair and orderly markets and is mandated by securities regulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:market-surveillance", "labels": [ "Market Surveillance", "Post-Market Surveillance" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "marketplace-integration", "title": "Marketplace Integration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systems and APIs that connect metaverse platforms with NFT marketplaces, e-commerce platforms, and digital asset trading systems, enabling seamless buying, selling, and trading of virtual goods, real estate, and collectibles.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:marketplace-integration", "labels": [ "Marketplace Integration" ], "is_subclass_of": [ "Platform and Environment", "Digital Commerce" ], "wikilinks": [ "Metaverse Commerce", "Digital Commerce", "metaverse" ] }, { "id": "marketplace", "title": "Marketplace", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital platform enabling discovery, exchange, and transaction of virtual goods, services, and assets within or across metaverse systems through listing, escrow, and reputation mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:marketplace", "labels": [ "Marketplace", "Decentralised Data Marketplace", "Decentralised Marketplace", "Marketplace Platform", "Zero-Royalty Marketplace" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Economic Activity", "Escrow System", "Identity System", "OMA3 + Reed Smith", "Payment Gateway", "Payment Protocol", "Product Listing", "Search & Discovery", "Secure Transaction", "Transaction Engine", "Value Exchange", "ApplicationLayer", "Asset Registry", "Asset Trading", "Blockchain", "Digital Wallet", "Metadata Standard", "MiddlewareLayer", "Price Discovery", "Reputation System" ] }, { "id": "markov-chain-monte-carlo", "title": "Markov Chain Monte Carlo", "domain": "ai", "domain_name": "Ai", "definition": "Markov Chain Monte Carlo (MCMC) is a family of computational algorithms that generate samples from an arbitrary target probability distribution by constructing a Markov chain whose stationary distribution equals the target, enabling tractable approximate Bayesian inference and numerical integration over high-dimensional parameter spaces that are intractable by exact analytical methods. Foundational algorithms include Metropolis-Hastings, Gibbs sampling, Hamiltonian Monte Carlo (HMC), and the No-U-Turn Sampler (NUTS). MCMC is the standard tool for posterior inference in probabilistic programming and Bayesian statistical modelling across science, engineering, and machine learning.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:markov-chain-monte-carlo", "labels": [ "Markov Chain Monte Carlo" ], "is_subclass_of": [ "Monte Carlo Methods" ], "wikilinks": [] }, { "id": "markov-chain", "title": "Markov Chain", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A stochastic process in which the probability of each future state depends only on the current state and not on the sequence of preceding states (the Markov property), enabling tractable analysis of steady-state distributions, mixing times, and long-run behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:markov-chain", "labels": [ "Markov Chain", "Markov Chains" ], "is_subclass_of": [ "Stochastic Process" ], "wikilinks": [ "Probability Theory", "Markov Chain Monte Carlo", "Markov Decision Process", "Dynamical Systems Theory", "Stochastic Process" ] }, { "id": "markov-decision-process", "title": "Markov Decision Process", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A mathematical framework for modelling sequential decision-making where outcomes are partly random and partly under the control of a decision maker, comprising states, actions, transition probabilities, and a reward function, solved by computing a policy that maximises expected cumulative reward.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:markov-decision-process", "labels": [ "Markov Decision Process", "Markov Decision Processes" ], "is_subclass_of": [ "Markov Chain" ], "wikilinks": [ "Markov Chain", "Probability Theory", "Reinforcement Learning", "Optimisation" ] }, { "id": "mars-orbit-environment", "title": "Mars Orbit Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mars-orbit-environment", "labels": [ "Mars Orbit Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "mars-orbit-radiation-environment", "title": "Mars Orbit Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mars-orbit-radiation-environment", "labels": [ "Mars Orbit Radiation Environment" ], "is_subclass_of": [ "Mars Radiation Environment" ], "wikilinks": [] }, { "id": "mars-radiation-environment", "title": "Mars Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mars-radiation-environment", "labels": [ "Mars Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "mars-surface-environment", "title": "Mars Surface Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mars-surface-environment", "labels": [ "Mars Surface Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "mars-surface-radiation-environment", "title": "Mars Surface Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mars-surface-radiation-environment", "labels": [ "Mars Surface Radiation Environment" ], "is_subclass_of": [ "Mars Radiation Environment", "Surface Radiation Environment" ], "wikilinks": [] }, { "id": "masked-language-modelling", "title": "Masked Language Modelling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Masked language modelling is a self-supervised pre-training objective in which random tokens of an input sequence are hidden and the model learns to predict them from the surrounding bidirectional context. By conditioning on both left and right context, it produces deep contextual representations of language. It is the objective popularised by BERT and underpins many encoder-based transformer models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:masked-language-modelling", "labels": [ "Masked Language Modelling", "Masked Language Model" ], "is_subclass_of": [ "Pre Training" ], "wikilinks": [] }, { "id": "mass-adoption", "title": "Mass Adoption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Mass adoption is the stage at which a technology moves beyond early enthusiasts to be used by a large mainstream population. For blockchain and digital assets it implies usability, scalability, and trust sufficient for everyday consumers and enterprises. Reaching it typically requires solving cost, throughput, and user-experience barriers that gate the transition across the technology adoption lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mass-adoption", "labels": [ "Mass Adoption" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "mass-customisation", "title": "Mass Customisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Mass customisation is a production strategy that delivers individually tailored goods or services at costs approaching those of mass production. It combines flexible manufacturing, modular design, and digital configuration so customers can specify variants without sacrificing scale economies. Robotics, additive manufacturing, and data-driven personalisation are key enablers across sectors such as fashion and consumer goods.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mass-customisation", "labels": [ "Mass Customisation" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "massive-mimo", "title": "Massive MIMO", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Massive MIMO (multiple-input multiple-output) is a wireless communication technique that equips a base station with a very large number of antenna elements, typically tens to hundreds, to serve many users simultaneously on the same time-frequency resource. It exploits spatial multiplexing and beamforming to increase spectral efficiency and link reliability without additional bandwidth. Massive MIMO is a core physical-layer technology of 5G networks, enabling higher capacity and energy efficiency per served user.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:massive-mimo", "labels": [ "Massive MIMO" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "master-data-management", "title": "Master Data Management", "domain": "data", "domain_name": "Data", "definition": "A discipline and set of processes for creating and maintaining a single, consistent view of an organisation's core business entities such as customers, products, and suppliers. It governs how master data is defined, stored, and shared across systems to establish authoritative records and support reporting, compliance, and operational consistency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:master-data-management", "labels": [ "Master Data Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Entity Resolution", "Data Governance", "Data Quality", "Data Integration", "Data Management" ] }, { "id": "mastercard", "title": "Mastercard", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Mastercard is a global payments technology company operating a card network that authorises, clears and settles transactions between issuing and acquiring banks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mastercard", "labels": [ "Mastercard" ], "is_subclass_of": [ "Payment Network" ], "wikilinks": [ "Payment Network", "Cross-Border Payments", "Central Bank Digital Currency", "Stablecoin" ] }, { "id": "mastery-learning", "title": "Mastery Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mastery learning is an instructional strategy in which learners must demonstrate proficiency on a given unit of content before advancing to the next, using formative assessment and targeted remediation to ensure uniformly high achievement rather than a fixed pace of coverage.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:mastery-learning", "labels": [ "Mastery Learning" ], "is_subclass_of": [ "Artificial Intelligence", "AI Application" ], "wikilinks": [] }, { "id": "matching-algorithm", "title": "Matching Algorithm", "domain": "data", "domain_name": "Data", "definition": "A matching algorithm is a computational procedure that identifies correspondences between two sets of entities \u2014 such as records, participants or resources \u2014 according to a similarity or compatibility criterion, ranging from exact-key joins to probabilistic and graph-based matching. In master data management it links duplicate or related records referring to the same real-world entity across data sources; in resource-allocation contexts, such as barter or exchange systems, it pairs supply with demand to satisfy mutual constraints. Matching algorithms vary widely in complexity, from simple rule-based comparisons to optimisation-based approaches such as the stable-marriage and bipartite-matching algorithms.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:matching-algorithm", "labels": [ "Matching Algorithm" ], "is_subclass_of": [ "Algorithm" ], "wikilinks": [] }, { "id": "material-definition", "title": "Material Definition", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A material definition is a structured data specification that describes how a surface or volume should appear under lighting, encoding properties such as base colour, roughness, metalness, emissivity, and transparency in a renderer-agnostic format. It underpins physically based rendering (PBR) pipelines by separating the description of physical surface behaviour from the rendering algorithm that evaluates it. Material definitions may be authored in formats such as MaterialX, glTF materials, or OpenUSD surface shading networks, enabling interchange across tools and engines. They are fundamental to achieving photorealistic and stylised visuals in real-time and offline 3D workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:material-definition", "labels": [ "Material Definition", "Material Definition Language" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "material-flow-analysis", "title": "Material Flow Analysis", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Material flow analysis (MFA) is a systematic accounting of the flows and stocks of materials within a defined system in space and time, grounded in the principle of mass conservation. It quantifies inputs, outputs, and accumulations to reveal resource use, losses, and recycling potential across an economy, industry, or region. MFA is a foundational method for industrial ecology, circular-economy design, and sustainability assessment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:material-flow-analysis", "labels": [ "Material Flow Analysis" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "material-science", "title": "Material Science", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Material Science is the interdisciplinary field that studies the structure, properties, processing, and performance of matter across all classes \u2014 metals, ceramics, polymers, semiconductors, and composites \u2014 connecting atomic-scale arrangement to macroscopic behaviour. It integrates principles from physics, chemistry, and engineering to enable the rational design of materials with tailored properties. Computational methods including density functional theory, molecular dynamics simulation, and machine-learning-driven property prediction increasingly complement experimental synthesis and characterisation. The field underpins virtually every technology sector, from microelectronics and energy storage to biomedical implants and advanced manufacturing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:material-science", "labels": [ "Material Science", "Materials Science" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Material System", "Quantum Computing", "Simulation", "owl:Thing" ] }, { "id": "material-system", "title": "Material System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Material System is an engineered assembly of two or more distinct materials \u2014 together with their interfaces, interphases, and boundary conditions \u2014 designed and optimised as a coherent unit to deliver a specified combination of mechanical, thermal, electrical, optical, or biological properties. The system perspective shifts analysis from isolated constituent properties to emergent behaviour arising from interactions between phases, microstructures, and interfaces under operational loading conditions. Material systems are specified through hierarchical design parameters spanning macro-, meso-, micro-, and nano-scales, and their performance is evaluated via multi-scale modelling, physical testing, and lifecycle assessment. Representative examples include continuous-fibre polymer matrix composites, thermal barrier coating systems, multi-layer semiconductor packages, gradient functional materials, and bioresorbable scaffold-tissue constructs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:material-system", "labels": [ "Material System" ], "is_subclass_of": [ "Material Science" ], "wikilinks": [ "Material Science", "Simulation", "Complex Systems" ] }, { "id": "material", "title": "Material", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A material is the generic description of how a 3D surface responds to light, bundling the shading model and its parameters \u2014 base colour, roughness, metalness, normal detail, emission, transparency \u2014 into a reusable definition that a renderer evaluates for each visible point. Materials are the container into which texture maps are plugged and which a shader program consumes, mediating between geometry and the final rendered appearance. This is a general graphics concept, not to be confused with MaterialX, which is a specific open interchange standard from Lucasfilm/ILM for encoding and exchanging material graphs between applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:material", "labels": [ "Material" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Texture Map", "Shader", "Physically Based Rendering" ] }, { "id": "material-x", "title": "MaterialX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "MaterialX is an open standard for describing surface and procedural materials, shading networks and look development data in a renderer-independent form for exchange between content creation tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:material-x", "labels": [ "MaterialX", "MaterialX Shading" ], "is_subclass_of": [ "Material Definition" ], "wikilinks": [ "Shader", "Texture Mapping", "Physically Based Rendering", "Asset Interoperability", "glTF", "Material Definition" ] }, { "id": "materiality-assessment", "title": "Materiality Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Materiality assessment is the structured process by which an organisation identifies and prioritises the environmental, social and governance topics that are significant enough to influence its decisions or the decisions of its stakeholders. It establishes which sustainability matters warrant disclosure, management attention and resource allocation. Double materiality extends the concept to consider both financial impact on the firm and the firm's impact on society and the environment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:materiality-assessment", "labels": [ "Materiality Assessment" ], "is_subclass_of": [ "Sustainability Reporting" ], "wikilinks": [] }, { "id": "math-library", "title": "Math Library", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Math Library is a software library providing optimised implementations of mathematical operations \u2014 including vector and matrix arithmetic, quaternion transformations, geometric queries, interpolation, and numerical methods \u2014 that underpin real-time 3D graphics, physics simulation, and spatial computing systems. These libraries abstract hardware-level SIMD optimisations and GPU-friendly data layouts, enabling physics engines, rendering pipelines, and game engines to perform high-throughput computation at interactive frame rates.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:math-library", "labels": [ "Math Library" ], "is_subclass_of": [ "Content and Assets", "Software Library" ], "wikilinks": [ "Software Library" ] }, { "id": "mathematical-foundations", "title": "Mathematical Foundations", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Mathematical Foundations refers to the collection of core mathematical disciplines \u2014 including linear algebra, calculus, probability theory, discrete mathematics, and number theory \u2014 that underpin the formal reasoning required across computer science, cryptography, artificial intelligence, and engineering. These disciplines provide the rigorous axiomatic structures and analytical tools upon which algorithms, proofs, models, and systems are built. Mastery of mathematical foundations is considered prerequisite knowledge for deep work in machine learning, cryptographic protocol design, and distributed systems. They bridge pure abstract reasoning with applied computational practice.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:mathematical-foundations", "labels": [ "Mathematical Foundations", "Foundations of Mathematics", "MathematicalFoundations" ], "is_subclass_of": [ "Mathematical Science" ], "wikilinks": [] }, { "id": "mathematical-hard-problems", "title": "Mathematical Hard Problems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computationally intractable mathematical problems that form the security foundation of cryptographic systems, including integer factorisation, discrete logarithm, lattice problems, and other NP-hard challenges used in blockchain and digital security.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:mathematical-hard-problems", "labels": [ "Mathematical Hard Problems" ], "is_subclass_of": [ "Content and Assets", "Cryptographic Security" ], "wikilinks": [ "Secure Digital Systems", "Cryptographic Security", "metaverse" ] }, { "id": "mathematical-logic", "title": "Mathematical Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mathematical logic is the branch of mathematics that studies formal systems, proof, computability and the foundations of mathematics using rigorous symbolic methods. It encompasses subfields such as model theory, proof theory, set theory and recursion theory, and provides the formal underpinnings for reasoning about truth, provability and decidability. Mathematical logic is foundational to theoretical computer science, automated reasoning and the formal semantics of programming languages.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:mathematical-logic", "labels": [ "Mathematical Logic" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "mathematical-optimisation", "title": "Mathematical Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mathematical Optimisation is the branch of mathematics and computer science concerned with selecting the best element from a set of feasible solutions according to a defined objective function. It encompasses convex and non-convex programming, combinatorial optimisation, gradient-based methods, and evolutionary algorithms. Mathematical optimisation underpins machine learning training, operations research, and engineering design.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:mathematical-optimisation", "labels": [ "Mathematical Optimisation" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "mathematical-reasoning", "title": "mathematical reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mathematical reasoning is the faculty \u2014 in humans or artificial systems \u2014 to perform rigorous multi-step inference over mathematical structures, including arithmetic computation, algebraic manipulation, geometric reasoning, combinatorics, and formal proof construction. It demands compositional symbol manipulation, logical deduction, and the ability to track intermediate state across reasoning chains without shortcut pattern-matching. In artificial intelligence, mathematical reasoning serves as a canonical benchmark for general problem-solving capability because correct solutions are verifiable against ground truth. Contemporary approaches combine neural language models, chain-of-thought prompting, external symbolic solvers, and automated theorem provers to extend the scope of machine-tractable mathematical problems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:mathematical-reasoning", "labels": [ "Mathematical Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "mathematical-science", "title": "Mathematical Science", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mathematical Science encompasses the formal disciplines\u2014including probability theory, linear algebra, calculus, graph theory, and cryptography\u2014that underpin machine learning model design, algorithm analysis, and blockchain cryptographic security. It provides the theoretical substrate for understanding optimisation landscapes, neural network convergence, and information-theoretic limits of AI systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:mathematical-science", "labels": [ "Mathematical Science" ], "is_subclass_of": [ "AI Research Area", "Machine Learning Discipline" ], "wikilinks": [ "owl:Thing" ] }, { "id": "matplotlib-inline-visualisation-pattern", "title": "Matplotlib Inline Visualisation Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Logseq-embedded Python code example demonstrating inline data visualisation using matplotlib within a Pyodide runtime. The snippet generates a sinusoidal plot, encodes it as a base64 PNG, and returns it for display inside the knowledge graph note, illustrating programmatic data visualisation within a personal knowledge management environment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:matplotlib-inline-visualisation-pattern", "labels": [ "Matplotlib Inline Visualisation Pattern", "Plotting a graph using matplotlib python library" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "matrix-factorisation", "title": "Matrix Factorisation", "domain": "ai", "domain_name": "Ai", "definition": "Matrix factorisation is a family of techniques that decompose a matrix into a product of lower-dimensional factor matrices, revealing latent structure in the data. In machine learning it is widely used for collaborative-filtering recommendation, where a sparse user-item rating matrix is approximated by user and item embedding factors. Variants include singular value decomposition, non-negative matrix factorisation, and low-rank adaptation methods.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:matrix-factorisation", "labels": [ "Matrix Factorisation", "Matrix Decomposition", "Matrix Factorization" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "matrix-multiplication", "title": "Matrix Multiplication", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Matrix multiplication is the binary operation that combines two matrices to produce a third, where each entry of the result is the dot product of a row of the first matrix with a column of the second. It is the fundamental computational primitive of linear algebra and the dominant operation in deep learning, where dense layers, attention and convolutions all reduce to large matrix or tensor products executed on parallel hardware such as GPUs and TPUs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:matrix-multiplication", "labels": [ "Matrix Multiplication" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "matrix-protocol", "title": "Matrix Protocol", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Matrix is an open standard and communication protocol for real-time, decentralised communication over the internet. It enables interoperable, federated messaging and collaboration by allowing servers from different organisations to synchronise shared conversation state. The protocol supports end-to-end encryption, persistent history, and a rich ecosystem of clients and bridges.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:matrix-protocol", "labels": [ "Matrix Protocol" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "matter-protocol", "title": "Matter Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Matter is a royalty-free smart-home connectivity standard developed by the Connectivity Standards Alliance to provide a common application layer enabling devices from different manufacturers to interoperate over IP networks. Built on top of Thread and Wi-Fi for transport, Matter defines a unified data model, secure commissioning, and local control that removes dependence on proprietary vendor clouds. It is backed by Apple, Google, Amazon, and Samsung, and aims to resolve the fragmentation that historically forced consumers into single-ecosystem smart-home purchases.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:matter-protocol", "labels": [ "Matter Protocol", "CSA Matter Specification", "Matter CSA", "Matter Commissioning", "Matter Standard" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "maximum-likelihood-estimation", "title": "Maximum Likelihood Estimation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Maximum Likelihood Estimation (MLE) is a method of estimating the parameters of a statistical model by choosing the parameter values that maximise the likelihood of the observed data under the model. Equivalently it minimises the negative log-likelihood, connecting it directly to many machine-learning loss functions. As a principled, asymptotically efficient estimator it underlies a large share of classical statistics and probabilistic machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:maximum-likelihood-estimation", "labels": [ "Maximum Likelihood Estimation" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "maximum-sequence-length", "title": "Maximum Sequence Length", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The upper bound on the number of tokens a model can ingest in a single forward pass, determined during training by the positional encoding scheme and available memory. It governs context retention, task complexity, and whether long inputs must be truncated, chunked, or processed with sliding-window attention.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:maximum-sequence-length", "labels": [ "Maximum Sequence Length" ], "is_subclass_of": [ "AI Technique", "Natural Language Processing" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "maximum-valid-data-value", "title": "Maximum Valid Data Value", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:maximum-valid-data-value", "labels": [ "Maximum Valid Data Value" ], "is_subclass_of": [ "Measurement Value" ], "wikilinks": [] }, { "id": "mean-absolute-error", "title": "Mean Absolute Error", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A regression performance metric representing the average magnitude of errors between predicted and actual values, calculated as the arithmetic mean of absolute differences between predictions and ground truth across all instances, providing an intuitive measure of prediction accuracy in the same units as the target variable. MAE treats all errors equally regardless of direction and is less sensitive to outliers than squared error metrics such as RMSE.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:mean-absolute-error", "labels": [ "Mean Absolute Error" ], "is_subclass_of": [ "AI Technique", "Model Performance" ], "wikilinks": [ "Error Analysis", "MAPE", "Mean Squared Error", "Median Absolute Error", "Model Evaluation", "Outlier", "Regression", "Computer Vision", "MetaverseDomain", "Model Performance", "Root Mean Square Error" ] }, { "id": "mean-anomaly", "title": "Mean Anomaly", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mean-anomaly", "labels": [ "Mean Anomaly" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mean-squared-error", "title": "Mean Squared Error", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A measure of the average squared difference between predicted values and observed values, widely used to quantify estimation and prediction error in regression and statistical learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:mean-squared-error", "labels": [ "Mean Squared Error" ], "is_subclass_of": [ "Loss Function" ], "wikilinks": [ "Statistics", "Probability Theory", "Supervised Learning", "Gradient Descent", "Loss Function" ] }, { "id": "measure-theory", "title": "Measure Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Measure Theory is the branch of mathematical analysis that studies measures, which assign a consistent notion of size, length, area, volume or probability to subsets of a space. It provides the rigorous foundation for the Lebesgue integral, which generalises the Riemann integral and handles a wider class of functions and limiting operations. It is the formal basis of modern probability theory, where a probability is a measure of total mass one.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:measure-theory", "labels": [ "Measure Theory" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Lebesgue Integral", "Sigma-Algebra", "Set Theory", "Real Analysis", "Probability Theory", "Stochastic Processes", "Functional Analysis", "owl:Thing" ] }, { "id": "measurement-accuracy", "title": "Measurement Accuracy", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-accuracy", "labels": [ "Measurement Accuracy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-bias", "title": "Measurement Bias", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-bias", "labels": [ "Measurement Bias" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-device", "title": "Measurement Device", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-device", "labels": [ "Measurement Device" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-instrument", "title": "Measurement Instrument", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-instrument", "labels": [ "Measurement Instrument" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-methodology", "title": "Measurement Methodology", "domain": "governance", "domain_name": "Governance", "definition": "A measurement methodology is a defined, repeatable procedure for quantifying a property or performance characteristic, specifying scope, units, data sources, and calculation rules. It ensures that results are consistent, comparable, and auditable across observers and over time. Rigorous methodology is the precondition for trustworthy indicators, benchmarks, and standards-based reporting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:measurement-methodology", "labels": [ "Measurement Methodology", "Measurement Protocols" ], "is_subclass_of": [ "Evaluation Metric" ], "wikilinks": [] }, { "id": "measurement-offset", "title": "Measurement Offset", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-offset", "labels": [ "Measurement Offset" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-precision", "title": "Measurement Precision", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-precision", "labels": [ "Measurement Precision" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-scale-factor", "title": "Measurement Scale Factor", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-scale-factor", "labels": [ "Measurement Scale Factor" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-scaling", "title": "Measurement Scaling", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-scaling", "labels": [ "Measurement Scaling" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-uncertainty", "title": "Measurement Uncertainty", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-uncertainty", "labels": [ "Measurement Uncertainty" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "measurement-value", "title": "Measurement Value", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:measurement-value", "labels": [ "Measurement Value" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mecanum-wheel-robot", "title": "Mecanum Wheel Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Mecanum Wheel Robot is a wheeled mobile robot platform that achieves true omnidirectional movement by mounting four independently driven mecanum wheels\u2014each fitted with a ring of passive rollers oriented at 45 degrees to the wheel's rotation axis. By differentially controlling the speeds and directions of the four wheels, the platform can translate in any horizontal direction, rotate in place, or combine translation and rotation simultaneously without requiring steering joints or changing wheel orientation. Invented by Bengt Ilon at Mecanum AB in 1972, the design is widely used in warehousing, logistics, and research platforms where unrestricted planar mobility is required in confined spaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:mecanum-wheel-robot", "labels": [ "Mecanum Wheel Robot" ], "is_subclass_of": [ "Robot Type", "Omnidirectional Robot" ], "wikilinks": [ "Omnidirectional Robot", "Robotics" ] }, { "id": "mechanical-component", "title": "Mechanical Component", "domain": "robotics", "domain_name": "Robotics", "definition": "MechanicalComponent is the ontological superclass for all physical structural, kinematic, and power-transmission elements constituting robotic hardware systems.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:mechanical-component", "labels": [ "Mechanical Component", "MechanicalComponent" ], "is_subclass_of": [ "Actuation and Control", "Kinematic Chain", "Robot Hardware", "Physical System", "Mechatronic System", "Manufacturing Product" ], "wikilinks": [ "Actuator", "Actuator", "Additive Manufacturing", "Additive Manufacturing", "Additive Manufacturing", "Aluminium Alloy", "Aluminium Alloy", "Backdrivability", "Ball Screw", "Ball Screw", "Ball Screw Mechanism", "Bearing", "Bearing", "Bearing", "BLDC Motor", "Cable Drive", "Cable Drive", "Carbon Fibre Composite", "Carbon Fibre Composite", "Carbon Fibre Composite" ] }, { "id": "mechanical-interface", "title": "Mechanical Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "A mechanical interface is the standardised physical coupling between two components that defines their geometric fit, fastening, and load transfer. In robotics it specifies how an end-effector or tool attaches to a robot wrist or actuator, governing alignment, rigidity, and quick-change capability. Well-defined mechanical interfaces enable modularity and interchangeability of subsystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mechanical-interface", "labels": [ "Mechanical Interface" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "mechanical-load", "title": "Mechanical Load", "domain": "robotics", "domain_name": "Robotics", "definition": "A mechanical load is the external force, torque, or resistance that a motor, actuator, or structure must work against during operation. Characterised by inertia, friction, gravity, and process forces, it determines the torque-speed demand placed on a drive system. Accurate load characterisation is essential for sizing motors, selecting gearing, and designing stable control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mechanical-load", "labels": [ "Mechanical Load" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "mechanism-design", "title": "mechanism design", "domain": "governance", "domain_name": "Governance", "definition": "Mechanism design, sometimes called reverse game theory, is a field of economics and game theory concerned with constructing the rules, incentive structures, and institutional frameworks of strategic interactions so that self-interested participants collectively produce socially desirable outcomes. Starting from a target social choice function rather than from a given game, the designer works backwards to identify what game structure (the mechanism) would implement that function as an equilibrium for rational agents. Mechanism design underpins auction theory, voting systems, market design, and cryptoeconomic protocol engineering, and has been central to Nobel-Prize-recognised work by Hurwicz, Maskin, and Myerson. Its principles govern token incentive structures, automated market makers, governance protocols, and public-goods funding schemes across decentralised systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mechanism-design", "labels": [ "Mechanism Design" ], "is_subclass_of": [ "Game Theory" ], "wikilinks": [] }, { "id": "mechanistic-interpretability", "title": "Mechanistic Interpretability", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Mechanistic interpretability is a research area that seeks to reverse-engineer the internal computations of neural networks into human-understandable algorithms. It studies features, circuits, and representations within model weights and activations to explain how specific behaviours arise. The field aims to make models transparent enough to predict, audit, and align, supporting AI safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:mechanistic-interpretability", "labels": [ "Mechanistic Interpretability", "Mechanistic Circuits Analysis" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "medi-ledger", "title": "MediLedger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A permissioned blockchain network for the pharmaceutical industry that records product provenance and verifies the legitimacy of drug transactions between trading partners. It was built to support compliance with United States drug supply chain security regulations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:medi-ledger", "labels": [ "MediLedger" ], "is_subclass_of": [ "Pharmaceutical Supply Chain" ], "wikilinks": [ "Permissioned Blockchain", "GS1", "Provenance", "Distributed Ledger Technology", "Supply Chain", "Pharmaceutical Supply Chain", "https://www.mediledger.com/" ] }, { "id": "media-authentication", "title": "Media Authentication", "domain": "standards", "domain_name": "Standards", "definition": "Media Authentication is the verification that a piece of digital media, such as an image, audio clip, or video, originates from a claimed source and has not been altered since capture or publication. It relies on cryptographic signatures, embedded provenance metadata, or perceptual hashing to detect tampering or synthetic generation. Standards such as C2PA define interoperable formats for attaching and verifying this provenance information.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:media-authentication", "labels": [ "Media Authentication" ], "is_subclass_of": [ "Content Provenance" ], "wikilinks": [] }, { "id": "media-authenticity", "title": "Media Authenticity", "domain": "ai", "domain_name": "Ai", "definition": "Media authenticity is the property and practice of establishing that a piece of digital media is genuine, unaltered, and accurately attributed to its source. It combines cryptographic provenance metadata, watermarking, and forensic analysis to distinguish authentic recordings from synthetic or manipulated content. As generative models make convincing synthetic media inexpensive, media authenticity has become central to combating disinformation and preserving trust in visual and audio evidence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:media-authenticity", "labels": [ "Media Authenticity" ], "is_subclass_of": [ "Content Authenticity" ], "wikilinks": [] }, { "id": "media-forensics", "title": "Media Forensics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Media forensics is the analysis of images, audio and video to determine authenticity, detect manipulation and attribute provenance, using techniques ranging from pixel-level artefact analysis to learned deepfake classifiers. It examines compression artefacts, sensor noise patterns, lighting inconsistencies and model-specific generation fingerprints to distinguish genuine capture from synthetic or edited content. It underpins deepfake detection systems and broader media authenticity assessment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:media-forensics", "labels": [ "Media Forensics" ], "is_subclass_of": [ "Digital Forensics" ], "wikilinks": [] }, { "id": "media-library", "title": "Media Library", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Centralised digital asset repositories that store, organise, and manage media files including 3D models, textures, audio, video, and immersive content for metaverse applications, with metadata tagging and search capabilities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:media-library", "labels": [ "Media Library" ], "is_subclass_of": [ "Content and Assets", "Digital Asset Management" ], "wikilinks": [ "Content Organisation", "Digital Asset Management", "metaverse" ] }, { "id": "media-production", "title": "Media Production", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Media production is the end-to-end process of creating audiovisual content, spanning pre-production planning, capture, editing, and post-production finishing. It increasingly integrates generative AI, virtual production, and automated voice and speech tools to accelerate and personalise output. As a workflow it coordinates creative, technical, and asset-management activities across a content pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:media-production", "labels": [ "Media Production", "Immersive Media Production" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "media-richness-theory", "title": "Media Richness Theory", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"A theoretical framework positing that communication media vary in their capacity to convey rich information through multiple cues, immediate feedback, language variety, and personal focus, predicting that richer media enable more effective communication of complex, ambiguous information whilst l...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:media-richness-theory", "labels": [ "Media Richness Theory", "Media Richness", "TELE-004-media-richness-theory" ], "is_subclass_of": [ "Communication Technology", "OrganisationalTheory" ], "wikilinks": [ "CommunicationEffectiveness", "TELE-003-social-presence-theory", "TELE-020-virtual-reality-telepresence", "TELE-100-ai-avatars", "OrganisationalTheory", "TELE-001-telepresence" ] }, { "id": "media-theory", "title": "Media Theory", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "An interdisciplinary field examining how communication media shape human perception, social interaction, and cultural meaning, encompassing frameworks such as media richness theory, social presence theory, and Marshall McLuhan's medium-as-message thesis. Media theory informs the design of telecollaboration and immersive systems by analysing how channel properties affect communication fidelity and presence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:media-theory", "labels": [ "Media Theory" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "mediation", "title": "Mediation", "domain": "governance", "domain_name": "Governance", "definition": "A structured, consensual dispute resolution process in which a neutral third party \u2014 the mediator \u2014 helps disputing parties communicate, surface underlying interests, and negotiate their own settlement, without the power to impose an outcome. Confidential, without prejudice, and typically far cheaper and faster than litigation or arbitration, mediation resolves the majority of commercial, family, workplace, and community disputes that reach it, and any agreement becomes binding only when the parties choose to contract on it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:mediation", "labels": [ "Mediation" ], "is_subclass_of": [ "Dispute Resolution" ], "wikilinks": [ "Dispute Resolution", "Negotiation", "Conflict Resolution" ] }, { "id": "medical-ai", "title": "Medical AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Medical AI encompasses artificial intelligence and machine learning applications in healthcare for disease detection, diagnosis, treatment planning, clinical decision support, drug discovery, and patient outcome prediction, subject to medical-device regulation and clinical validation requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:medical-ai", "labels": [ "Medical AI" ], "is_subclass_of": [ "AI Application", "Artificial Intelligence", "Healthcare Technology" ], "wikilinks": [ "Drug Discovery", "Medical Imaging Analysis", "Personalized Medicine", "Uncertainty Quantification", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Clinical Decision Support", "Explainable AI", "Federated Learning", "Healthcare Analytics", "Healthcare Technology", "Medical Diagnosis AI", "Medical Imaging AI", "Precision Medicine" ] }, { "id": "medical-devices", "title": "Medical Devices", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Medical devices are instruments, machines, implants, or software intended for the diagnosis, prevention, monitoring, or treatment of disease and injury. They are classified by risk and tightly regulated, demanding high reliability, fault tolerance, and often real-time performance where failure can cause patient harm. Embedded and connected medical devices increasingly incorporate sensing, control, and AI under stringent safety regimes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:medical-devices", "labels": [ "Medical Devices" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [] }, { "id": "medical-diagnosis-ai", "title": "Medical Diagnosis AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Medical Diagnosis AI refers to artificial intelligence systems that automate or assist in the diagnostic process by analysing patient symptoms, medical history, laboratory results, imaging findings, and other clinical data to generate differential diagnoses, diagnostic hypotheses, and diagnostic ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:medical-diagnosis-ai", "labels": [ "Medical Diagnosis AI", "Medical Diagnosis", "Medical Diagnostics" ], "is_subclass_of": [ "AI Application", "Medical AI" ], "wikilinks": [ "Clinical Decision Support", "Medical AI", "MetaverseDomain", "Treatment Planning AI" ] }, { "id": "medical-image-analysis", "title": "Medical Image Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Medical Image Analysis is the application of computer-vision and machine-learning methods to interpret images produced by medical imaging modalities such as radiography, computed tomography, magnetic resonance imaging, ultrasound and pathology slides. It encompasses tasks including segmentation of anatomy and lesions, detection and classification of abnormalities, registration across images, and quantitative measurement. Modern systems rely heavily on deep convolutional and transformer-based models trained on annotated clinical data. It supports radiologists and clinicians in diagnosis, treatment planning and monitoring.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:medical-image-analysis", "labels": [ "Medical Image Analysis" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "medical-image-synthesis", "title": "Medical Image Synthesis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Medical image synthesis is the use of generative models to produce artificial medical images, such as scans, for training data augmentation, modality translation, or privacy-preserving sharing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:medical-image-synthesis", "labels": [ "Medical Image Synthesis" ], "is_subclass_of": [ "Image Synthesis" ], "wikilinks": [ "Generative Model", "Medical Imaging", "Generative Adversarial Network", "Image Synthesis" ] }, { "id": "medical-imaging-ai", "title": "Medical Imaging AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Medical Imaging AI encompasses artificial intelligence systems designed to analyse, interpret, and enhance medical images including radiological scans, pathology slides, and other diagnostic imaging modalities. These systems employ deep learning architectures, particularly convolutional neural networks, to perform lesion detection, disease classification, anatomical segmentation, and quantitative image analysis whilst meeting clinical validation standards and regulatory requirements.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:medical-imaging-ai", "labels": [ "Medical Imaging AI" ], "is_subclass_of": [ "Medical AI" ], "wikilinks": [ "Computer Vision", "Convolutional Neural Network", "Medical AI", "MetaverseDomain", "Pathology AI", "Radiology AI" ] }, { "id": "medical-imaging", "title": "medical imaging", "domain": "ai", "domain_name": "Ai", "definition": "Medical imaging is the acquisition, reconstruction, processing, and computational analysis of visual representations of human anatomy and physiology \u2014 including X-ray, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), single-photon emission computed tomography (SPECT), and digital pathology \u2014 for the purposes of clinical diagnosis, treatment planning, surgical guidance, and longitudinal disease monitoring. The field has undergone a fundamental shift with the integration of deep learning methods, particularly convolutional neural networks, vision transformers, and diffusion-based models, which now perform organ segmentation, lesion detection, and disease classification at or near radiologist-level accuracy on constrained benchmarks. Data interoperability is standardised by DICOM (Digital Imaging and Communications in Medicine) for image storage and transfer, and by HL7 FHIR for clinical report integration; AI-based clinical decision software requires regulatory clearance such as FDA 510(k) premarket notification, EU MDR/IVDR conformity assessment, or UKCA marking.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:medical-imaging", "labels": [ "Medical Imaging", "Medical Imaging Analysis", "Medical Imaging Synthesis", "Medical Imaging Visualization", "Medical Segmentation" ], "is_subclass_of": [ "Healthcare AI" ], "wikilinks": [] }, { "id": "medical-robot", "title": "Medical Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Medical Robot is a robotic system designed for use in clinical and healthcare settings, encompassing surgical assistants, rehabilitation exoskeletons, and diagnostic platforms. Medical robots operate under strict safety standards (ISO 13482) and integrate sensing, actuation, and AI to augment human clinical capability.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:medical-robot", "labels": [ "Medical Robot" ], "is_subclass_of": [ "Robot Type", "Service Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "medical-robotics", "title": "Medical Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Medical Robotics is the application of robotic systems to clinical and surgical settings, encompassing robot-assisted surgery, rehabilitation robotics, diagnostic automation, and hospital logistics. These systems extend clinician precision, reduce invasiveness, and enable procedures at scales or locations beyond unaided human capability. Medical robots must meet rigorous safety, sterilisation, and regulatory standards before clinical deployment. The field intersects robotics engineering, clinical medicine, and regulatory science.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:medical-robotics", "labels": [ "Medical Robotics" ], "is_subclass_of": [ "Robotics", "Robot Type" ], "wikilinks": [] }, { "id": "medical-segmentation-decathlon", "title": "Medical Segmentation Decathlon", "domain": "ai", "domain_name": "Ai", "definition": "The Medical Segmentation Decathlon is a benchmark challenge providing ten diverse medical imaging datasets to evaluate the generalisability of automated segmentation algorithms. Spanning organs and modalities such as brain MRI, liver CT, and cardiac imaging, it tests whether a single method can perform well across heterogeneous anatomical tasks without per-task tuning. It is a widely cited reference for assessing biomedical image-segmentation models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:medical-segmentation-decathlon", "labels": [ "Medical Segmentation Decathlon" ], "is_subclass_of": [ "Evaluation Metric", "AI Research Area" ], "wikilinks": [] }, { "id": "medical-simulation", "title": "Medical Simulation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Medical simulation is the use of virtual environments, mannequins, and digital models to replicate clinical scenarios for training, assessment, and procedure planning. Immersive and motion-tracked simulations let clinicians rehearse surgery and emergency response without risk to patients. It is a core application of metaverse and XR technologies within healthcare education and practice.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:medical-simulation", "labels": [ "Medical Simulation" ], "is_subclass_of": [ "Virtual Environment" ], "wikilinks": [] }, { "id": "medication-safety", "title": "Medication Safety", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Medication safety is the set of practices and systems that prevent harm from drug errors, covering dosing, interactions, allergies and administration timing across the medication-use process. Clinical decision support systems apply medication-safety rules to flag dangerous interactions or incorrect dosages at the point of prescribing, while supply-chain traceability ensures pharmaceuticals are not counterfeit or expired before reaching patients. It is a primary application area for AI systems operating in regulated healthcare environments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:medication-safety", "labels": [ "Medication Safety" ], "is_subclass_of": [ "Clinical Decision Support" ], "wikilinks": [] }, { "id": "medium-earth-orbit", "title": "Medium Earth Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:medium-earth-orbit", "labels": [ "Medium Earth Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "meeting-ai-assistant", "title": "Meeting AI Assistant", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "An AI-powered software agent that attends synchronous and asynchronous Virtual Meetings either as an autonomous bot participant or as capability embedded within a Meeting Platform, providing continuous Automated Transcription via large-vocabulary Automatic Speech Recognition (ASR)...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:meeting-ai-assistant", "labels": [ "Meeting AI Assistant", "AI Meeting Assistant", "AI Meeting Assistants" ], "is_subclass_of": [ "Communication Technology", "Collaboration Software", "AI Agent", "Distributed Collaboration Technology", "Conversational AI", "Meeting Support Tools" ], "wikilinks": [ "Abstractive Summarisation", "Action Item Extraction", "AI Agent", "Asynchronous Meeting Access", "Audio Stream", "Automated Transcription", "Automatic Speech Recognition", "BERT-based NLP", "Calendar Integration", "CCPA", "Collaboration Software", "Conformer Architecture", "CRM Systems", "Customer Success", "Decision Audit Trail", "Diarisation Models", "Digital Provenance", "DistributedCollaborationDomain", "Distributed Collaboration Technology", "Distributed Team Coordination" ] }, { "id": "meeting-recording", "title": "Meeting Recording", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Persistent capture and archival of synchronous virtual or physical meetings\u2014comprising video, audio, screen share, and transcript streams\u2014stored in cloud platforms or on-premises repositories for asynchronous playback, compliance audit, onboarding, and AI-mediated analysis.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:meeting-recording", "labels": [ "Meeting Recording" ], "is_subclass_of": [ "Communication Technology", "Digital Workplace Platform", "Asynchronous Communication", "Meeting Documentation", "Knowledge Graphing" ], "wikilinks": [ "Asynchronous Communication", "Authentication", "Automatic Speech Recognition", "Automatic Transcription", "Chapter Detection", "Cloud Storage Infrastructure", "ComplianceDomain", "DistributedCollaborationDomain", "Distributed Learning Archive", "DPO", "FCA SYSC 10A", "FINRA Rule 17a-4", "Full-text Search Index", "GDPR Article 6", "Imperial College London", "KnowledgeManagementDomain", "Meeting Documentation", "MP4 Encoding", "Network Bandwidth", "Onboarding" ] }, { "id": "meeting-transcription", "title": "Meeting Transcription", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The automated conversion of spoken dialogue in synchronous or asynchronous meetings into structured text, combining automatic speech recognition with speaker diarisation, punctuation restoration, and optionally action-item extraction. AI-powered meeting transcription systems enable searchable records, accessibility, and downstream summarisation workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:meeting-transcription", "labels": [ "Meeting Transcription", "Transcription Engine" ], "is_subclass_of": [ "AI Application", "Speech Recognition" ], "wikilinks": [ "Artificial Intelligence", "Speech Recognition" ] }, { "id": "megatron-lm", "title": "Megatron-LM", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Megatron-LM is an open-source framework developed by NVIDIA for training very large transformer language models efficiently across many GPUs. It pioneered intra-layer tensor model parallelism, splitting individual weight matrices and attention heads across devices, and combines this with pipeline and data parallelism to scale to models with hundreds of billions of parameters. The framework provides optimised kernels and communication patterns that minimise the overhead of synchronising gradients and activations between accelerators.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:megatron-lm", "labels": [ "Megatron-LM" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "melvin-carvalho-decentralised-web-advocate", "title": "Melvin Carvalho Decentralised Web Advocate", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Melvin Carvalho is a developer and advocate active in decentralised web and identity standards work, contributing to linked data, Solid and related semantic web technologies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:melvin-carvalho-decentralised-web-advocate", "labels": [ "Melvin Carvalho Decentralised Web Advocate", "Melvin Carvalho" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [ "Linked Data", "Decentralized Identity", "Solid", "Semantic Web" ] }, { "id": "membership-inference", "title": "Membership Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A privacy attack that determines wher a specific data point was included in a model's training dataset by analyzing the model's behavior on that input, potentially revealing sensitive information about individuals' participation in datasets.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:membership-inference", "labels": [ "Membership Inference", "Membership Inference Attack" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "membership-service-provider", "title": "Membership Service Provider", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Membership Service Provider (MSP) is the component of a permissioned blockchain that abstracts the cryptographic identity material and rules used to authenticate and authorise participants. It defines which certificate authorities are trusted, which roles members hold, and how digital certificates map to organisational identities used in endorsement and access policies. The MSP turns raw public-key infrastructure into the network's notion of who is allowed to act and in what capacity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:membership-service-provider", "labels": [ "Membership Service Provider" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "memory-bandwidth", "title": "Memory Bandwidth", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Memory bandwidth is the rate at which data can be read from or written to memory, typically measured in gigabytes per second. It is a primary performance constraint for data-intensive workloads such as deep-learning inference and real-time rendering, where compute units stall waiting for data. High-bandwidth memory technologies are deployed precisely to relieve this bottleneck on accelerators and edge hardware.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:memory-bandwidth", "labels": [ "Memory Bandwidth", "GPU Memory Bandwidth" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "memory-bank", "title": "Memory Bank", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A memory bank is a structured, persistent repository of curated project and task context that an autonomous agent reads at the start of every session and updates as work progresses, so that knowledge survives the resetting of the model's ephemeral context window. Typically realised as a set of versioned markdown or database records (project brief, active decisions, progress log, architectural notes), it gives a stateless language model durable long-term memory, enabling continuity of intent, avoidance of repeated discovery, and coherent behaviour across many disconnected invocations of the same or different agents.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:memory-bank", "labels": [ "Memory Bank" ], "is_subclass_of": [ "Agent Memory", "AgentMemory" ], "wikilinks": [ "AgentMemory", "MemoryStore", "ContextManagement", "DataPersistence" ] }, { "id": "memory-encryption", "title": "Memory Encryption", "domain": "security", "domain_name": "Security", "definition": "Memory encryption is the protection of data held in a system's volatile memory (RAM) by encrypting it transparently between the processor and the memory controller. It defends against physical attacks such as cold-boot extraction and bus snooping, and underpins confidential computing by keeping a workload's working set unreadable to other tenants, the hypervisor, or anyone with physical access. Modern implementations apply per-page or per-VM keys managed within the CPU package so that plaintext never leaves the trust boundary of the silicon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:memory-encryption", "labels": [ "Memory Encryption" ], "is_subclass_of": [ "Encryption" ], "wikilinks": [] }, { "id": "memory-hierarchy", "title": "Memory Hierarchy", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A memory hierarchy organises storage into layers of differing speed, size and cost so that frequently accessed data sits in fast caches close to the processor and bulk data resides in slower, larger memory.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:memory-hierarchy", "labels": [ "Memory Hierarchy" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [ "Computer Hardware", "GPU Architecture", "Real-Time Rendering", "Parallel Computing", "GPU Computing" ] }, { "id": "memory-management", "title": "Memory Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Memory management is the systematic allocation, tracking, and reclamation of a computer system's volatile memory resources across processes and runtimes. It encompasses allocation strategies, virtual memory abstraction, paging, and the reclamation of unused memory to prevent exhaustion and fragmentation. Effective memory management is foundational to system stability, performance, and the isolation guarantees that underpin multi-tenant infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:memory-management", "labels": [ "Memory Management" ], "is_subclass_of": [ "Resource Management" ], "wikilinks": [] }, { "id": "memory-store", "title": "Memory Store", "domain": "ai", "domain_name": "Ai", "definition": "A memory store is the persistence layer in an AI agent system that retains information across turns and sessions for later retrieval. It typically holds conversation history, facts, and learned context, often as embeddings in a vector database to enable semantic recall. The memory store is what allows agents to maintain continuity, personalise responses, and accumulate knowledge beyond a single context window.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:memory-store", "labels": [ "Memory Store", "In-Memory Data Stores", "Long-Term Memory Store" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "memory-efficient-training", "title": "Memory-Efficient Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Memory-efficient training encompasses techniques that reduce the accelerator memory required to train large neural networks, allowing larger models or batch sizes to fit within fixed hardware budgets. Approaches include gradient checkpointing, which recomputes intermediate activations during the backward pass instead of storing them, and parameter-efficient methods such as LoRA and DoRA, which train small low-rank adapters instead of full weight matrices. These techniques trade additional compute or reduced expressivity for substantially lower peak memory use during model training.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:memory-efficient-training", "labels": [ "Memory-Efficient Training" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "memory", "title": "Memory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Memory, in computing hardware, is the volatile storage a processor uses to hold instructions and data during execution, characterised by its capacity, latency, and bandwidth relative to the CPU. It is allocated and reclaimed by an operating system's memory management subsystem across running processes, and its size and speed directly bound what workloads a system can execute. Memory sits between fast on-chip caches and slower persistent storage in a system's memory hierarchy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:memory", "labels": [ "Memory" ], "is_subclass_of": [ "Computer Hardware" ], "wikilinks": [] }, { "id": "mempool", "title": "Mempool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The memory pool maintained by each blockchain node holding broadcast but as-yet unconfirmed transactions awaiting inclusion in a block. Miners select transactions from the mempool, typically prioritising by fee rate, while node operators use mempool policies to manage capacity and mitigate spam.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mempool", "labels": [ "Mempool", "Mempool Propagation", "Transaction Mempool", "User Operation Mempool", "UserOperation Mempool" ], "is_subclass_of": [ "Distributed Data Structure", "Blockchain Entity", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "mental-health-monitoring", "title": "Mental Health Monitoring", "domain": "ai", "domain_name": "Ai", "definition": "The use of sensors, applications, and data analysis to track indicators related to a person's psychological wellbeing over time. It may use self-reports, physiological signals, and behavioural data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mental-health-monitoring", "labels": [ "Mental Health Monitoring" ], "is_subclass_of": [ "Affective Computing" ], "wikilinks": [ "Sensor", "Affective Computing" ] }, { "id": "mental-model", "title": "Mental Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A mental model is an internal cognitive representation that a person holds of how a system, interface, or environment works and behaves. In interaction design it explains how users predict outcomes, form expectations, and interpret feedback, so designs that match users' existing mental models are easier to learn and less error-prone. Mismatches between the designer's conceptual model and the user's mental model are a primary source of usability problems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mental-model", "labels": [ "Mental Model" ], "is_subclass_of": [ "Human-Computer Interaction" ], "wikilinks": [] }, { "id": "mentions-and-notifications", "title": "Mentions and Notifications", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Mentions and notifications are mechanisms that direct a participant's attention to specific messages, tasks, or events by referencing their identity with an @-symbol or equivalent trigger. They bridge asynchronous communication gaps by alerting individuals even when they are not actively monitoring a channel. Effective notification design balances urgency with focus protection to avoid alert fatigue in distributed teams.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:mentions-and-notifications", "labels": [ "Mentions and Notifications", "Live Notifications" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "meridian", "title": "Meridian", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:meridian", "labels": [ "Meridian" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "merkle-dag", "title": "Merkle DAG", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Merkle Directed Acyclic Graph (DAG) is a data structure combining Merkle tree hash-linking with a generalised directed acyclic graph topology, allowing nodes to have multiple parents and enabling content-addressed, tamper-evident storage of arbitrary graph-shaped data. Unlike a binary Merkle tree, each node's cryptographic hash is derived from all its children, forming a unique, immutable identifier for any subgraph.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:merkle-dag", "labels": [ "Merkle DAG" ], "is_subclass_of": [ "Content Addressing" ], "wikilinks": [] }, { "id": "merkle-patricia-trie", "title": "Merkle Patricia Trie", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Merkle Patricia Trie is a cryptographically authenticated key-value data structure that combines a Patricia (radix) trie for compact prefix-keyed storage with Merkle hashing for tamper-evident integrity. Each node is referenced by the hash of its contents, so a single root hash commits to the entire dataset and any change propagates to the root. It is the data structure Ethereum uses to store account state, transactions, and receipts, enabling compact Merkle proofs of inclusion.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:merkle-patricia-trie", "labels": [ "Merkle Patricia Trie" ], "is_subclass_of": [ "Merkle Tree" ], "wikilinks": [] }, { "id": "merkle-proof", "title": "Merkle Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cryptographic proof of inclusion or exclusion that demonstrates whether a specific data element is part of a Merkle tree, requiring only O(log n) sibling hashes rather than the full data set. Merkle proofs underpin light-client verification in blockchain systems and enable simplified payment verification (SPV) without downloading the entire chain.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:merkle-proof", "labels": [ "Merkle Proof", "Merkle Proof Verification" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "merkle-root", "title": "Merkle Root", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Merkle Root is the single cryptographic hash at the apex of a Merkle tree, computed by recursively hashing pairs of child hashes until a single digest remains, such that the root encodes the integrity of every transaction or data element in the tree. In blockchain systems, each block header contains the Merkle root of all transactions in that block, enabling lightweight clients to verify transaction inclusion via a logarithmic-length Merkle proof without downloading the full block. This compact commitment property makes the Merkle root fundamental to blockchain scalability, security, and efficient synchronisation protocols.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:merkle-root", "labels": [ "Merkle Root", "Merkle Root Construction" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "merkle-tree", "title": "Merkle Tree", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A binary tree of cryptographic hashes in which each leaf node contains the hash of a data block and each non-leaf node contains the hash of its children. Merkle trees enable efficient and tamper-evident verification of large data sets; in blockchain systems they allow nodes to confirm individual transaction inclusion without downloading an entire block.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:merkle-tree", "labels": [ "Merkle Tree", "Merkle Tree Accumulator", "Merkle Tree Structure", "Merkle Trees", "MerkleTree" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "merkle-damgard-construction", "title": "Merkle-Damgard Construction", "domain": "security", "domain_name": "Security", "definition": "The Merkle-Damgard construction is a method for building a collision-resistant cryptographic hash function of arbitrary input length from a fixed-size, collision-resistant compression function, by padding the message and processing it in sequential blocks that chain into one another. SHA-256 and most other widely used hash functions of the SHA-1 and SHA-2 families are built on this construction. It is known to be vulnerable to length-extension attacks unless the hash output is further truncated or otherwise protected.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:merkle-damgard-construction", "labels": [ "Merkle-Damgard Construction" ], "is_subclass_of": [ "Cryptographic Hash" ], "wikilinks": [] }, { "id": "mesa-optimisation", "title": "Mesa-Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The phenomenon in which a learned model, produced by a base optimiser such as stochastic gradient descent, is itself an optimiser pursuing an internally represented objective \u2014 the mesa-objective \u2014 that may diverge from the training objective; the central inner-alignment concern in AI safety, since a mesa-optimiser can perform well during training for instrumental reasons and then pursue its own goals under distribution shift, the failure mode known as deceptive alignment.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:mesa-optimisation", "labels": [ "Mesa-Optimisation" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [ "AI Alignment", "AI Safety", "Corrigibility", "Existential AI Risk" ] }, { "id": "mesh-compression", "title": "Mesh Compression", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mesh compression is a family of techniques for reducing the storage and transmission size of 3D polygon mesh data, encoding vertex positions, normals, texture coordinates and connectivity more compactly than a naive array representation. Methods range from quantisation and delta encoding of vertex attributes to connectivity-aware schemes such as edgebreaker and Draco. It is a prerequisite for efficient streaming and exchange of 3D assets across networks and constrained devices.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:mesh-compression", "labels": [ "Mesh Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [] }, { "id": "mesh-data", "title": "Mesh Data", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mesh data is a structured representation of a three-dimensional surface or volume as a collection of vertices, edges, and polygonal faces \u2014 most commonly triangles or quads \u2014 that together define the geometry of a shape. It encodes both topological connectivity (which vertices form which faces) and geometric attributes (positions, normals, UV coordinates, vertex colours) required for rendering, simulation, and analysis. Mesh data serves as the primary interchange format between 3D modelling software, real-time engines, and spatial computing pipelines. It is produced by processes such as photogrammetry, structured-light scanning, LiDAR capture, and procedural generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mesh-data", "labels": [ "Mesh Data" ], "is_subclass_of": [ "Spatial Mesh" ], "wikilinks": [] }, { "id": "mesh-generation", "title": "Mesh Generation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mesh generation is the process of constructing a discrete representation of a geometric domain as a network of vertices, edges and faces - typically triangles or tetrahedra - suitable for rendering, simulation or analysis. It transforms continuous shapes, point clouds or implicit surfaces into well-formed polygonal or volumetric meshes that meet quality, density and topology constraints. Mesh generation underpins computer graphics, 3D reconstruction and numerical simulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mesh-generation", "labels": [ "Mesh Generation" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "mesh-network", "title": "Mesh Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A mesh network is a network topology in which each node relays data for the network, cooperating to distribute information so that traffic can take multiple paths between source and destination. Mesh networks self-organise and self-heal: when a link or node fails, traffic is rerouted around the fault without manual reconfiguration. They are widely used in wireless sensor networks, smart-home protocols, and community and industrial deployments where infrastructure-free, resilient connectivity is required.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mesh-network", "labels": [ "Mesh Network" ], "is_subclass_of": [ "Network Topology" ], "wikilinks": [] }, { "id": "mesh-networking", "title": "Mesh Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Mesh networking is a network topology in which nodes connect directly, dynamically, and non-hierarchically, relaying data on behalf of one another to reach destinations. Each node cooperates in routing, so the network can self-organise and self-heal around failed or moved nodes without central infrastructure. Mesh topologies are widely used in wireless sensor networks, smart-home protocols, and resilient community networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:mesh-networking", "labels": [ "Mesh Networking" ], "is_subclass_of": [ "Network Topology" ], "wikilinks": [] }, { "id": "mesh-routing-software", "title": "Mesh Routing Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Network protocol software that enables decentralised, self-healing communication in mesh networks through dynamic routing algorithms, supporting IoT deployments, distributed systems, and metaverse infrastructure connectivity.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:mesh-routing-software", "labels": [ "Mesh Routing Software", "Mesh Router", "Mesh Routing" ], "is_subclass_of": [ "Platform and Environment", "Network Infrastructure" ], "wikilinks": [ "Decentralised Networks", "metaverse", "Network Infrastructure" ] }, { "id": "mesh-shading", "title": "Mesh Shading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mesh shading is a GPU rendering pipeline model, exposed through APIs such as Vulkan and DirectX 12, that replaces the fixed vertex-and-geometry-shader stages with programmable task and mesh shaders operating on small groups of primitives called meshlets. It gives applications direct control over primitive culling, level-of-detail selection and amplification within the shader itself, removing bottlenecks inherent in the traditional pipeline. Mesh shading is used to render highly detailed geometry more efficiently in modern real-time graphics engines.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:mesh-shading", "labels": [ "Mesh Shading" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "mesopause", "title": "Mesopause", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mesopause", "labels": [ "Mesopause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mesosphere", "title": "Mesosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mesosphere", "labels": [ "Mesosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "message-authentication-code", "title": "Message Authentication Code", "domain": "security", "domain_name": "Security", "definition": "A Message Authentication Code (MAC) is a fixed-size cryptographic tag generated from an arbitrary-length message and a shared secret key using a keyed hash or block-cipher-based algorithm, providing simultaneous data integrity verification and authentication of the sender to any party holding the same secret key. Unlike digital signatures, MACs are symmetric and do not provide non-repudiation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:message-authentication-code", "labels": [ "Message Authentication Code", "Authentication Tag", "Message Authentication Codes", "Symmetric Message Authentication Code" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "message-authentication", "title": "Message Authentication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Message authentication is the cryptographic property that guarantees a received message originated from a claimed sender and has not been altered in transit. It is achieved through message authentication codes (MACs), digital signatures, or authenticated encryption schemes, each binding message content to a key or key pair that only the legitimate sender possesses.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:message-authentication", "labels": [ "Message Authentication", "Secure Message Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "message-broker", "title": "Message Broker", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A message broker is an intermediary software component that translates messages between disparate messaging protocols and routes them between producers and consumers, decoupling the two sides of a communication so that neither needs direct knowledge of the other. It typically provides guaranteed delivery, message queuing, routing rules, protocol translation, and persistence semantics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:message-broker", "labels": [ "Message Broker" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [] }, { "id": "message-format", "title": "Message Format", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A message format is the agreed structure, encoding, and field schema for data exchanged between systems over a protocol. It specifies headers, payload layout, data types, and serialisation so that senders and receivers can unambiguously parse and validate messages. Standardised message formats are essential to interoperability in communication protocols and transaction standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:message-format", "labels": [ "Message Format", "MT Message Format", "Message Schema" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "message-passing-interface", "title": "Message Passing Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Message Passing Interface (MPI) is a standardised communication protocol and application programming interface for parallel and distributed computing. It defines primitives for point-to-point and collective communication between processes running across multiple compute nodes, enabling tightly coupled high-performance computing workloads. MPI is the de facto standard for scientific computing, numerical simulations, and distributed AI training at supercomputer scale.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:message-passing-interface", "labels": [ "Message Passing Interface" ], "is_subclass_of": [ "Infrastructure", "Parallel Computing" ], "wikilinks": [] }, { "id": "message-passing", "title": "Message Passing", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Message passing is a foundational communication paradigm in which processes, objects, or distributed agents interact exclusively by sending and receiving discrete, self-contained messages rather than accessing shared memory. It underlies actor-model concurrency, microservice architectures, and distributed AI agent frameworks, providing loose coupling, location transparency, and inherent support for asynchronous execution. Messages may traverse in-process channels, persistent message queues, or wide-area network transports, and may be delivered synchronously (blocking until acknowledgement) or asynchronously (fire-and-forget). Formal semantics are studied through process calculi such as the pi-calculus and CSP, and the paradigm has grown from Hewitt's 1973 Actor Model into the backbone of modern cloud-native and multi-agent AI systems.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:message-passing", "labels": [ "Message Passing", "General Message Passing", "Message Passing Protocol", "Message Passing System" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "message-queue", "title": "Message Queue", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Message Queue is a durable, ordered buffer that mediates asynchronous inter-process communication, allowing producers to enqueue messages independently of consumers reading them, thereby decoupling components in both time and topology. Messages are persisted by a broker until a consumer retrieves and acknowledges them, with delivery semantics ranging from at-most-once through at-least-once to exactly-once, each trading throughput for reliability. Message queues are foundational to event-driven architectures, microservices integration, and distributed data pipelines, absorbing traffic spikes, enabling backpressure management, and improving system resilience against partial failures. Implementations range from lightweight in-process task queues to enterprise-grade distributed log brokers such as Apache Kafka, RabbitMQ, and Amazon SQS.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:message-queue", "labels": [ "Message Queue", "Dead Letter Queue", "Dead-Letter Queue" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "message-reactions", "title": "Message Reactions", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Message reactions are lightweight emoji or icon-based acknowledgements that participants attach to individual messages without composing a reply. They convey sentiment, agreement, or status in a low-friction manner that reduces notification noise. In distributed teams, reactions serve as asynchronous micro-feedback signals that keep conversation threads concise.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:message-reactions", "labels": [ "Message Reactions" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "message-signing", "title": "Message Signing", "domain": "security", "domain_name": "Security", "definition": "Message signing is a cryptographic operation whereby a private key is used to generate a digital signature over a message or data payload, enabling any holder of the corresponding public key to verify the authenticity and integrity of the message. It provides non-repudiation guarantees, ensuring that the signer cannot plausibly deny having produced the signature. Message signing is foundational to authentication protocols, blockchain transactions, and secure communication systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:message-signing", "labels": [ "Message Signing" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "messagepack", "title": "Messagepack", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "MessagePack is a binary data serialisation format that encodes structured data compactly while preserving a data model compatible with JSON, allowing maps, arrays, strings, integers, and floats to be transmitted in far fewer bytes than their textual equivalents. It uses a type-prefixed encoding that minimises overhead, making it well-suited to high-throughput messaging, caching, and inter-service communication. MessagePack trades human readability for speed and compactness, sitting between verbose JSON and schema-bound formats like Protocol Buffers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:messagepack", "labels": [ "Messagepack", "MessagePack" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [] }, { "id": "meta-ai", "title": "meta ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Meta AI is the artificial intelligence research and applied AI division of Meta Platforms, Inc., encompassing the Fundamental AI Research (FAIR) laboratory for long-horizon academic research and multiple applied AI teams that embed machine-learning capabilities into Meta products including Facebook, Instagram, and WhatsApp. The division is best known for its open-weight large language model series LLaMA, the zero-shot image segmentation model Segment Anything (SAM), the multimodal joint-embedding model ImageBind, and early contributions to the broader AI ecosystem including the PyTorch deep-learning framework and fastText word embeddings. Meta AI's open-weight release strategy has materially shaped the open-source AI ecosystem by enabling community fine-tuning, on-device deployment, and proliferation of derivative models without dependence on proprietary inference APIs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:meta-ai", "labels": [ "Meta AI", "Meta AI Research" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "meta-governance", "title": "Meta Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Meta governance is the practice of governing the governance process itself, encompassing the rules, parameters, and mechanisms that determine how decisions are made, how voting power is allocated, and how the decision-making system can be amended. In decentralised protocols it often refers to one entity wielding governance rights over another, such as a protocol holding voting tokens of the protocols it integrates. Meta governance addresses the meta-level design of decision rights, escalation paths, and the evolution of governance structures over time.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:meta-governance", "labels": [ "Meta Governance", "Meta-Governance" ], "is_subclass_of": [ "Decentralized Governance" ], "wikilinks": [] }, { "id": "meta-llama-model-family", "title": "Meta Llama Model Family", "domain": "ai", "domain_name": "Ai", "definition": "Llama is a family of large language models developed by Meta and released, in large part, with open weights for research and commercial use. First introduced in 2023 with LLaMA, followed by Llama 2, Llama 3 and later versions, the models are transformer-based and trained on large text corpora. By releasing model weights under permissive terms, Meta enabled a wide range of independent fine-tuning and deployment, making Llama a common base for open models. The family spans several parameter sizes to suit different compute and latency requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:meta-llama-model-family", "labels": [ "Meta Llama Model Family", "LLaMA", "Llama", "Llama Architecture", "Meta Llama" ], "is_subclass_of": [ "Large Language Model", "Large Language Models", "Machine Learning Domain" ], "wikilinks": [ "Transformer", "Large Language Model", "Microsoft Copilot", "Machine Learning Domain", "Touvron et al. 2023, LLaMA: Open and Efficient Foundation Language Models" ] }, { "id": "meta-learning", "title": "Meta-Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Meta-Learning, colloquially described as 'learning to learn', is the study and design of machine learning systems that improve their own learning algorithms or initialisation through experience across multiple tasks, enabling rapid adaptation to new tasks with minimal data. Rather than learning a task directly, a meta-learning algorithm learns a prior or inductive bias that facilitates fast generalisation. Key paradigms include model-agnostic meta-learning (MAML), which optimises for a parameter initialisation that is close to good solutions for many tasks, and metric-based approaches such as prototypical networks that learn a task-agnostic embedding space. Meta-learning is central to few-shot learning, continual learning, and automated machine learning research.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:meta-learning", "labels": [ "Meta-Learning", "Meta-Learner" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "meta-transaction", "title": "Meta-Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A meta-transaction is a blockchain transaction that a user signs off-chain but does not submit or pay gas for directly; instead a relayer submits it on the user's behalf and covers the transaction fee. It relies on a standard such as EIP-712 for structured, verifiable off-chain message signing so the relayer and receiving contract can authenticate the original signer. It enables gasless user experiences, letting applications abstract away native token fee requirements.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:meta-transaction", "labels": [ "Meta-Transaction" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "meta", "title": "Meta", "domain": "metaverse", "domain_name": "Metaverse", "definition": "An American technology company, formerly named Facebook, that operates social media platforms and develops virtual and augmented reality hardware and software. It rebranded to Meta Platforms in 2021 to reflect a focus on the metaverse.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:meta", "labels": [ "Meta", "Meta Platforms" ], "is_subclass_of": [ "Platform and Environment", "owl:Thing" ], "wikilinks": [ "Metaverse", "Augmented Reality", "owl:Thing" ] }, { "id": "meta-mask", "title": "MetaMask", "domain": "blockchain", "domain_name": "Blockchain", "definition": "MetaMask is a self-custodial software cryptocurrency wallet, distributed as a browser extension (Chrome, Firefox, Brave, Edge) and mobile application (iOS and Android), that enables users to manage Ethereum and EVM-compatible blockchain assets directly in the browser without handing custody of private keys to a third party. It implements the EIP-1193 provider standard to expose a JavaScript API that decentralised applications (dApps) use to request signatures and transactions, making it the de facto gateway through which most users access Web3 services. MetaMask stores an encrypted HD wallet (BIP-32/BIP-39) locally in the browser's secure storage and signs all transactions client-side before broadcasting them to the chosen network via JSON-RPC. Developed by ConsenSys and first released in 2016, it is the most widely adopted non-custodial wallet by active monthly users.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:meta-mask", "labels": [ "MetaMask" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [ "Ethereum", "Web3", "Self-Custody", "Digital Wallet" ] }, { "id": "metadata-catalog", "title": "Metadata Catalog", "domain": "data", "domain_name": "Data", "definition": "A metadata catalog is a centralised, searchable inventory of an organisation's data assets and their descriptive, technical, operational, and business metadata. It records schemas, locations, ownership, lineage, classifications, and usage, enabling users to discover, understand, trust, and govern data. By unifying metadata across heterogeneous stores, it underpins data discovery, governance, and self-service analytics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:metadata-catalog", "labels": [ "Metadata Catalog" ], "is_subclass_of": [ "Data Catalog" ], "wikilinks": [] }, { "id": "metadata-management", "title": "Metadata Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Metadata Management encompasses the discipline, tooling, standards, and governance processes required to systematically capture, store, classify, version, discover, and operationalise descriptive, structural, and administrative information about data assets, pipelines, models, and services ac...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:metadata-management", "labels": [ "Metadata Management", "Metadata Management Infrastructure" ], "is_subclass_of": [ "Data Management", "Data Quality", "Data Governance", "Knowledge Management", "Information Architecture", "Enterprise Architecture" ], "wikilinks": [ "Active Metadata", "Ad Hoc Data Access", "Apache Atlas", "Apache Kafka", "AWS Glue Data Catalog", "Business Glossary", "Business Intelligence", "Data Catalog", "Data Contracts", "Data Discovery", "Data Engineering", "Data Fabric", "DataGovernanceDomain", "DataHub", "DataHub Metadata Model", "Data Lineage", "Data Mesh", "Data Observability", "DataOps", "Data Quality" ] }, { "id": "metadata-registry", "title": "Metadata Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Centralised systems for managing metadata according to standards like ISO/IEC 11179, providing authoritative definitions, usage rules, and data element descriptions to ensure consistency, interoperability, and governance across enterprise data systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:metadata-registry", "labels": [ "Metadata Registry" ], "is_subclass_of": [ "Data Management", "Data Governance" ], "wikilinks": [ "Enterprise Data Consistency", "Data Governance", "metaverse" ] }, { "id": "metadata-repository", "title": "Metadata Repository", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Storage systems that centralise metadata from diverse data sources, providing unified access to technical, business, and operational metadata for data discovery, cataloguing, lineage tracking, and governance across enterprise environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:metadata-repository", "labels": [ "Metadata Repository", "Metadata Database", "Metadata Store" ], "is_subclass_of": [ "Standards and Interoperability", "Data Management" ], "wikilinks": [ "Data Discovery", "Data Management", "metaverse" ] }, { "id": "metadata-schema", "title": "Metadata Schema", "domain": "data", "domain_name": "Data", "definition": "A structured specification defining metadata elements, their semantics, syntax, and relationships for describing and managing information resources. Metadata schemas establish standardised vocabularies and constraints that enable interoperability, discovery, and governance across data ecosystems through predefined sets of descriptive attributes tailored for specific domains or resource types.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:metadata-schema", "labels": [ "Metadata Schema" ], "is_subclass_of": [ "Data Management", "Data Standard" ], "wikilinks": [ "Data Discovery", "Data Standard", "Dublin Core", "FAIR Data Principles", "JSON-LD", "Schema.org", "Data Governance", "Data Interoperability" ] }, { "id": "metadata-standard", "title": "Metadata Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A formal specification defining the structure, semantics, format, and rules for describing data about data, ensuring consistent interpretation and interoperability across systems and domains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:metadata-standard", "labels": [ "Metadata Standard", "DAOstar Metadata Standard", "NFT Metadata Standard", "PREMIS Metadata Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standards" ], "wikilinks": [ "Data Discovery", "Data Elements", "Data Management System", "Data Model", "Encoding Specification", "ETSI GR ARF 010", "Information Exchange", "ISO 11179", "JSON Schema", "Namespace Management", "Ontology", "RDF", "Resource Description", "Schema Definition", "Semantic Interoperability", "Semantics Rules", "Validation Constraints", "XML Schema", "ComputationAndIntelligenceDomain", "Controlled Vocabulary" ] }, { "id": "metadata-standards", "title": "Metadata Standards", "domain": "standards", "domain_name": "Standards", "definition": "Established frameworks and specifications such as Dublin Core, IPTC, and XMP that define how descriptive information about digital assets should be structured, enabling interoperability, discoverability, and consistent management across systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:metadata-standards", "labels": [ "Metadata Standards" ], "is_subclass_of": [ "Data Standards" ], "wikilinks": [ "Cross-System Data Exchange", "Data Standards", "metaverse" ] }, { "id": "metadata", "title": "Metadata", "domain": "data", "domain_name": "Data", "definition": "Metadata is structured information that describes, contextualises, or categorises other data, enabling its discovery, management, interpretation, and interoperability. Metadata encompasses descriptive attributes (title, author, date), structural information (format, schema, relationships), administrative records (access rights, provenance, lifecycle), and technical parameters (resolution, encoding, checksum) attached to or associated with a primary data object.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:metadata", "labels": [ "Metadata", "EXIF Metadata", "Model Metadata" ], "is_subclass_of": [ "Data Schema" ], "wikilinks": [] }, { "id": "metal-api", "title": "Metal Api", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metal is Apple's low-level graphics and compute application programming interface that provides near-direct access to the GPU on Apple platforms. It minimises driver overhead, exposes explicit command-buffer and resource management, and unifies rendering and general-purpose compute under one programming model. Metal is the platform-native alternative to cross-vendor APIs such as Vulkan and the legacy OpenGL on Apple hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:metal-api", "labels": [ "Metal Api", "Metal API" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "metaverse-application-platform", "title": "Metaverse Application Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Development environments and SDKs built on engines like Unity and Unreal that provide tools, APIs, and frameworks for creating immersive 3D applications, virtual worlds, and blockchain-integrated experiences for the metaverse.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:metaverse-application-platform", "labels": [ "Metaverse Application Platform", "Application Runtime" ], "is_subclass_of": [ "Platform and Environment", "Development Platform" ], "wikilinks": [ "Metaverse Content Creation", "Development Platform", "metaverse" ] }, { "id": "metaverse-application", "title": "Metaverse Application", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A software application designed to operate within or alongside a metaverse platform, providing users with persistent virtual experiences including social interaction, commerce, entertainment, education, and collaboration. Metaverse applications leverage real-time 3D rendering, avatar systems, spatial audio, and interoperability standards to deliver presence-rich, cross-platform functionality within shared virtual environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-application", "labels": [ "Metaverse Application" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "owl:Thing" ] }, { "id": "metaverse-architecture-stack", "title": "Metaverse Architecture Stack", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Layered framework defining functional components and interfaces for metaverse systems to interoperate at network, data, and application levels.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:metaverse-architecture-stack", "labels": [ "Metaverse Architecture Stack" ], "is_subclass_of": [ "Platform and Environment", "Metaverse Architecture" ], "wikilinks": [ "Component Definitions", "Component Reusability", "ETSI ENI 008", "IEEE P2048-1", "IEEE P2048-1 (Architecture Overview)", "Interface Specifications", "Interoperability Protocols", "Layering Principles", "Multi-vendor Integration", "OSI Model", "Application Layer", "Compute Layer", "Data Layer", "InfrastructureDomain", "Interface Standards", "Network Layer", "Physical Layer", "Reference Architecture", "Scalable Architecture", "System Interoperability" ] }, { "id": "metaverse-architecture", "title": "Metaverse Architecture", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The layered technical framework defining metaverse infrastructure, encompassing network connectivity, computing resources, spatial computing, creator tools, interaction protocols, and economic systems that enable persistent virtual worlds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:metaverse-architecture", "labels": [ "Metaverse Architecture" ], "is_subclass_of": [ "Platform and Environment", "System Architecture" ], "wikilinks": [ "Virtual World Operation", "metaverse", "System Architecture" ] }, { "id": "metaverse-classification", "title": "Metaverse Classification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured taxonomy scheme for categorising metaverse platforms, applications, and components according to defined criteria such as openness, immersion level, economic model, and governance structure, enabling systematic comparison and standardised description within the ETSI metaverse domain model.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-classification", "labels": [ "Metaverse Classification" ], "is_subclass_of": [ "Platform and Environment", "Etsi Metaverse Domain Taxonomy" ], "wikilinks": [ "owl:Thing" ] }, { "id": "metaverse-commerce", "title": "Metaverse Commerce", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse commerce is the buying, selling and trading of goods, services and experiences within immersive virtual worlds and spatial environments. It spans virtual goods, wearables and real estate, branded experiences and the digital twins of physical products, often settled with virtual currencies or tokens. Metaverse commerce extends e-commerce into embodied, social and persistent 3D spaces where avatars browse, transact and own digital assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-commerce", "labels": [ "Metaverse Commerce" ], "is_subclass_of": [ "Metaverse" ], "wikilinks": [] }, { "id": "metaverse-content-creation", "title": "Metaverse Content Creation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Metaverse content creation is the production of three-dimensional assets, environments, avatars, and interactive experiences for immersive virtual worlds. It spans modelling, texturing, animation, scripting, and increasingly generative AI workflows, and is the supply side that populates persistent shared spaces. Quality, interoperability, and real-time performance constraints distinguish it from conventional digital media production.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-content-creation", "labels": [ "Metaverse Content Creation", "Metaverse Creation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "metaverse-content-pipeline", "title": "Metaverse Content Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "End-to-end workflow connecting asset creation, optimization, storage, distribution, and real-time rendering for metaverse experiences across platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:metaverse-content-pipeline", "labels": [ "Metaverse Content Pipeline" ], "is_subclass_of": [ "Platform and Environment", "Content Production Workflow" ], "wikilinks": [ "3D Authoring Tools", "Asset Compression", "Asset Creation", "Asset Management System", "Asset Optimization", "CDN Distribution", "Content Delivery Network", "Content Interoperability", "Content Storage", "Cross-Platform Content", "Dynamic Asset Loading", "Format Conversion", "glTF Standard", "LOD Generation", "Material System", "OMA3 Content WG", "Real-Time Rendering Engine", "Runtime Loading", "Shader Pipeline", "SMPTE ST 2128" ] }, { "id": "metaverse-entity-schema-archive", "title": "Metaverse Entity Schema Archive", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Some legacy Linked-JSON is a collection of early-stage JSON-LD-based ontology data structures capturing metaverse entity models \u2014 including agents, scenes, digital assets, and virtual economies \u2014 created before the current NarrativeGoldmine v2 schema. These artefacts document the iterative evolution of the knowledge graph's data model and serve as a reference for schema migration and ontology provenance tracing.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-entity-schema-archive", "labels": [ "Metaverse Entity Schema Archive", "Some legacy Linked-JSON" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "metaverse-infrastructure", "title": "Metaverse Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Technical infrastructure supporting persistent, large-scale metaverse platforms, encompassing the networking, cloud compute, edge nodes, rendering servers, and spatial data systems required to deliver low-latency immersive experiences at scale. It spans connectivity layers (5G, content delivery), server-side rendering and simulation, asset storage, and identity services that together underpin interoperable virtual worlds.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:metaverse-infrastructure", "labels": [ "Metaverse Infrastructure", "Scalable Metaverse Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Metaverse" ] }, { "id": "metaverse-interoperability", "title": "Metaverse Interoperability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse interoperability is the capability for users, avatars, digital assets, and experiences to move and function across multiple independently operated virtual worlds and platforms without loss of identity, ownership, or fidelity. It depends on shared standards for asset formats, identity, and transport, alongside the economic and governance arrangements that make cross-platform portability viable. The goal is an open, connected metaverse rather than a set of isolated walled gardens.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-interoperability", "labels": [ "Metaverse Interoperability" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "metaverse-liability-model", "title": "Metaverse Liability Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive legal responsibility framework for virtual worlds that defines liability attribution, responsibility allocation, and harm redress mechanisms across platforms, users, AI agents, and content creators in metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:metaverse-liability-model", "labels": [ "Metaverse Liability Model" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Activity Logging", "AI Agent Liability", "Content Creator Liability", "Dispute Resolution", "EU Digital Services Act", "Evidence Collection", "Harm Classification System", "Harm Compensation", "Harm Redress Mechanism", "Insurance Integration", "Jurisdiction Mapping", "Legal Accountability", "Legal Governance Framework", "Legal Precedent Database", "Liability Attribution Engine", "Platform Liability Framework", "Platform Protection", "Product Liability Directive", "Section 230 CDA", "Terms of Service" ] }, { "id": "metaverse-navigation-systems", "title": "Metaverse Navigation Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse navigation systems are spatially-embedded interaction mechanisms that enable users to traverse, orient, and coordinate within persistent three-dimensional virtual environments through modalities such as gaze-based controls, hand-tracking, locomotion techniques (joystick, teleportation, redirected walking), and spatial overlays including mini-maps and AR wayfinding cues. These systems integrate environmental awareness, social presence, and user experience design to maintain spatial cognition across interconnected virtual worlds. They form a critical infrastructure layer for usable, inclusive metaverse participation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-navigation-systems", "labels": [ "Metaverse Navigation Systems" ], "is_subclass_of": [ "Platform and Environment", "Metaverse and Telecollaboration" ], "wikilinks": [ "MetaverseDomain", "Telecollaboration" ] }, { "id": "metaverse-ontology-schema", "title": "Metaverse Ontology Schema", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A formal OWL 2 ontology framework defining the complete taxonomic structure, semantic relationships, axioms, and reasoning rules for metaverse concepts, enabling automated classification, consistency validation, and interoperability across virtual world implementations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:metaverse-ontology-schema", "labels": [ "Metaverse Ontology Schema" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Automated Reasoning", "Axiom Rules", "Inference Rules", "ISO IEC 21838 Top Level Ontology", "Knowledge Representation System", "Ontology-Based Data Access", "OWL 2 Reasoner", "Property Definitions", "RDF Schema", "RDF Triple Store", "Semantic Interoperability", "Semantic Web Infrastructure", "SHACL Constraints", "Validation Constraints", "W3C OWL 2", "W3C RDF Schema", "InfrastructureDomain", "Knowledge Graph Construction", "MiddlewareLayer", "Namespace Declarations" ] }, { "id": "metaverse-ontology", "title": "Metaverse Ontology", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The Metaverse Ontology is a formal knowledge representation that classifies the entities, relationships, and properties constituting a metaverse ecosystem\u2014including agents, scenes, digital assets, economies, governance structures, and communication protocols. It provides a shared semantic vocabulary enabling interoperability between platforms, AI systems, and toolchains operating within or around persistent immersive environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-ontology", "labels": [ "Metaverse Ontology" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Agentic Mycelia", "Anthropic Claude", "Bitcoin", "Cashu", "ChatGPT", "Gemini", "Knowledge Graphing", "Large Language Models", "Lightning and Similar L2", "Logseq", "Ontology conversation with AIs", "RGB and Client Side Validation", "Some legacy Linked-JSON" ] }, { "id": "metaverse-platform", "title": "Metaverse Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual world environments such as Roblox, Fortnite, Decentraland, and VRChat that provide persistent, shared, real-time spaces for social interaction, gaming, commerce, and creative expression with varying degrees of decentralisation and user-generated content.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:metaverse-platform", "labels": [ "Metaverse Platform" ], "is_subclass_of": [ "Virtual World" ], "wikilinks": [ "Virtual Social Interaction", "metaverse", "Virtual World" ] }, { "id": "metaverse-psychology-profile", "title": "Metaverse Psychology Profile", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Comprehensive dataset describing behavioral, emotional, cognitive, and social traits derived from virtual interactions, enabling personalized experiences and psychological research in metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:metaverse-psychology-profile", "labels": [ "Metaverse Psychology Profile" ], "is_subclass_of": [ "Governance and Safety", "Psychological Research" ], "wikilinks": [ "Adaptive Content Delivery", "Affective Response Logs", "APA Virtual Psychology Guidelines 2025", "Behavioral Pattern Data", "Behavioral Tracking System", "Cognitive Trait Indicators", "Data Analytics Engine", "Emotional State Metrics", "Ethical Data Governance", "Mental Health Monitoring", "Personalization Framework", "Preference Mapping", "Psychometric Assessment Tools", "Social Compatibility Matching", "Social Interaction Analytics", "User Consent Management", "User Profile System", "Application Layer", "Middleware Layer", "Personalized Virtual Experiences" ] }, { "id": "metaverse-safety-protocol", "title": "Metaverse Safety Protocol", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedures and safeguards ensuring physical and psychological safety of users during immersive metaverse experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:metaverse-safety-protocol", "labels": [ "Metaverse Safety Protocol" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "ETSI ENI 008", "IEEE VR Safety", "Incident Response Protocol", "ISO 45003", "Risk Assessment Procedure", "Safety Assessment", "Safety Guideline", "Safety Standard", "User Monitoring", "User Protection Measure", "User Well-being", "Compute Layer", "Content Moderation", "Data Layer", "Governance Framework", "Middleware Layer", "Network Layer", "Risk Mitigation", "Safe Immersive Experience", "TrustAndGovernanceDomain" ] }, { "id": "metaverse-stack", "title": "Metaverse Stack", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The comprehensive layered technology framework encompassing hardware, network infrastructure, spatial computing, decentralisation, creator economy, discovery, and experience layers that toger enable the creation and operation of persistent virtual world environments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:metaverse-stack", "labels": [ "Metaverse Stack" ], "is_subclass_of": [ "Platform and Environment", "Technology Stack" ], "wikilinks": [ "Virtual World Operation", "metaverse", "Technology Stack" ] }, { "id": "metaverse-standards-forum", "title": "Metaverse Standards Forum", "domain": "standards", "domain_name": "Standards", "definition": "The Metaverse Standards Forum (MSF) is an industry consortium hosted by the Khronos Group, founded in June 2022, that coordinates existing standards-development organisations to promote interoperability across metaverse and spatial-computing platforms rather than authoring competing specifications. With several hundred member organisations \u2014 including Meta, Microsoft, NVIDIA, Sony, Epic Games, the W3C, and ISO \u2014 it operates domain-specific working groups that identify interoperability gaps and produce exploratory prototypes whose outputs feed back into recognised bodies for normative ratification. Its model of pre-competitive alignment distinguishes it from traditional specification-writing bodies: it acts as a coordination layer, convening the standards ecosystem to prevent the fragmentation that would result from proprietary, siloed virtual worlds.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-standards-forum", "labels": [ "Metaverse Standards Forum" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "metaverse-technology", "title": "Metaverse Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The foundational technology domain encompassing persistent, synchronous 3D virtual worlds, augmented reality environments, spatial computing platforms, and immersive internet infrastructure that enable shared experiences with interoperable digital assets, persistent identity systems, and real-time social interaction across physical and virtual boundaries.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:metaverse-technology", "labels": [ "Metaverse Technology" ], "is_subclass_of": [ "Platform and Environment", "Technology Domain" ], "wikilinks": [ "Blockchain Technology", "Artificial Intelligence", "Augmented Reality", "Avatar System", "Digital Asset", "Immersive Experience", "Metaverse", "Metaverse Infrastructure", "Metaverse Technology", "Robotics Systems", "Spatial Computing", "Technology Domain", "Telecollaboration", "Virtual Economy", "Virtual World" ] }, { "id": "metaverse-venue", "title": "Metaverse Venue", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual 3D spaces within metaverse platforms designed to host events, conferences, exhibitions, and social gatherings, enabling global participation through customisable avatars and interactive environments that transcend physical location constraints.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:metaverse-venue", "labels": [ "Metaverse Venue" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "Global Virtual Gatherings", "metaverse", "Virtual Environment" ] }, { "id": "metaverse-and-telecollaboration", "title": "Metaverse and Telecollaboration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The convergence of persistent, shared virtual environments with real-time remote collaboration technologies, enabling geographically distributed participants to interact through embodied avatars, spatial audio, and shared 3D workspaces. Metaverse telecollaboration combines immersive XR hardware, low-latency networking, avatar systems, and spatial computing to replicate the social and spatial cues of co-presence across distances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-and-telecollaboration", "labels": [ "Metaverse and Telecollaboration" ], "is_subclass_of": [ "Metaverse" ], "wikilinks": [ "Adalgeirsson2010; @Lee2011; @Tsui2011; @Paulos1998; @Kristoffersson2013", "Adamczyk2007", "aiken2020zooming", "Aoki2003", "Argyle1965", "Argyle1969", "Argyle1969; @Argyle1988; @Cook1977", "Argyle1969; @Kleinke1986", "Argyle1976", "Argyle1988", "Argyle1988; @Argyle1976; @Argyle1965; @Argyle1976; @Argyle1969; @Kendon1967; @Monk2002", "authenticVolume2022", "Bailenson2001", "Bailenson2002", "Bandyopadhyay2001", "Bardram2012", "barrero2021working", "Bartneck2007; @Bartneck2009", "Bartneck2009", "Beck2011" ] }, { "id": "metaverse-governance-and-safeguarding", "title": "Metaverse governance and safeguarding", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse governance and safeguarding is a spatial computing concept and a type of Metaverse.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:metaverse-governance-and-safeguarding", "labels": [ "Metaverse governance and safeguarding", "Metaverse Governance" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [] }, { "id": "metaverse-telepresence-bridge", "title": "Metaverse-Telepresence Bridge", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The conceptual and technical integration between metaverse virtual environments and telepresence technologies, where participants experience remote presence in persistent 3D virtual worlds through immersive XR platforms, combining metaverse spatial computing infrastructure with telepresence soci...", "entityType": "Class", "qualityScore": 0.6, "maturity": "draft", "iri": "urn:ngm:class:metaverse-telepresence-bridge", "labels": [ "Metaverse-Telepresence Bridge" ], "is_subclass_of": [ "Telepresence", "ConvergenceConcept" ], "wikilinks": [ "EmbodiedMetaverseCollaboration", "TELE-003-social-presence-theory", "TELE-020-virtual-reality-telepresence", "TELE-026-microsoft-mesh", "TELE-027-spatial-platform", "TELE-028-horizon-workrooms", "TELE-100-ai-avatars", "TELE-105-real-time-language-translation", "TELE-250-blockchain-collaboration", "TELE-300-digital-twin-collaboration", "TELE-301-virtual-office-spaces", "TELE-302-shared-whiteboards", "ConvergenceConcept", "Metaverse", "SpatialComputing", "TELE-001-telepresence" ] }, { "id": "metaverse", "title": "Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A convergent network of persistent, synchronous 3D virtual worlds, augmented reality environments, and internet platforms that enable shared spatial computing experiences with interoperable digital assets, persistent identity, and real-time social interaction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:metaverse", "labels": [ "Metaverse", "Metaverse Asset", "Metaverse Participation", "Unified Metaverse" ], "is_subclass_of": [ "Platform and Environment", "Multiverse" ], "wikilinks": [ "3D Rendering", "Content Distribution Network", "Cross-World Portability", "ETSI GR MEC 032", "Extended Reality", "IEEE P2048", "Internet", "Interoperability Protocol", "ISO 23257", "Real-time Synchronization", "Social System", "Synchronous Interaction", "User Identity System", "ApplicationLayer", "Asset Management", "Avatar", "Blockchain", "Cloud Computing", "Creator Economy", "Database System" ] }, { "id": "metaverse-platforms", "title": "MetaversePlatforms", "domain": "metaverse", "domain_name": "Metaverse", "definition": "MetaversePlatforms denotes the category of systems that host persistent, shared 3D virtual worlds with avatars, user interaction and often user-generated content and economies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:metaverse-platforms", "labels": [ "MetaversePlatforms" ], "is_subclass_of": [ "Metaverse", "Metaverse Domain" ], "wikilinks": [ "Virtual World", "Avatar System", "Microsoft Mesh", "AltspaceVR", "Metaverse Domain" ] }, { "id": "meteor", "title": "Meteor", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:meteor", "labels": [ "Meteor" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "meteorite", "title": "Meteorite", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:meteorite", "labels": [ "Meteorite" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "meteoroid", "title": "Meteoroid", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:meteoroid", "labels": [ "Meteoroid" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "methane-abatement", "title": "Methane Abatement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Methane abatement is the set of techniques for capturing, destroying, or preventing the release of methane, a greenhouse gas with roughly 80 times the near-term warming potential of carbon dioxide. Approaches include flaring, combustion in engines or turbines, and conversion of otherwise-vented or flared gas into useful energy. In the Bitcoin context it underpins the use of stranded or waste methane to power mining as an emissions-reduction strategy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:methane-abatement", "labels": [ "Methane Abatement" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "methane-emissions-reduction", "title": "Methane Emissions Reduction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Methane emissions reduction refers to interventions that capture or combust methane -- a greenhouse gas roughly 80 times more potent than CO2 over 20 years -- before it vents to atmosphere, converting it into electricity or heat. In the Bitcoin context, miners co-locate with flared gas, landfill or agricultural waste sites and use the otherwise-wasted methane to power mining hardware, turning a stranded liability into monetised, lower-carbon energy. It is cited as a mitigating factor in assessments of Bitcoin's environmental footprint.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:methane-emissions-reduction", "labels": [ "Methane Emissions Reduction" ], "is_subclass_of": [ "Climate Change Mitigation" ], "wikilinks": [] }, { "id": "methane-mitigation-mining", "title": "Methane Mitigation Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Methane mitigation mining is the practice of siting Bitcoin mining hardware at sources of vented or flared methane, such as oil wellheads and landfills, to combust the gas in generators that power the miners. The mining load creates a continuous, location-flexible demand for energy that would otherwise be wasted, converting a potent greenhouse gas into electricity and revenue. It is promoted as a way to reduce net emissions while monetising stranded hydrocarbons.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:methane-mitigation-mining", "labels": [ "Methane Mitigation Mining" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "metrics-collection", "title": "Metrics Collection", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Metrics collection is the practice of instrumenting software and infrastructure to gather quantitative measurements of system behaviour, such as request rates, latencies, error counts and resource utilisation, and forwarding them to a store for analysis. Collected metrics are typically aggregated as time series and queried to drive dashboards, alerting and capacity planning. It is a foundational pillar of observability alongside logging and tracing, enabling teams to detect, diagnose and prevent operational problems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:metrics-collection", "labels": [ "Metrics Collection" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "metrics", "title": "Metrics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Metrics are numeric, time-series measurements that quantify the state and behaviour of systems, services and infrastructure over time. As one of the three pillars of observability alongside logs and traces, they are typically aggregated counters, gauges and histograms scraped or pushed at regular intervals. Metrics enable efficient trend analysis, alerting and capacity planning at scale because they compress system behaviour into compact, queryable numeric series.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:metrics", "labels": [ "Metrics" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "mi-ca-regulation", "title": "MiCA Regulation", "domain": "governance", "domain_name": "Governance", "definition": "The Markets in Crypto-Assets (MiCA) Regulation (EU 2023/1114) is the European Union's comprehensive legislative framework governing the issuance, offering, and trading of crypto-assets, including asset-referenced tokens, e-money tokens, and utility tokens. Adopted in June 2023 and phasing in through December 2024, MiCA establishes authorisation requirements for crypto-asset service providers (CASPs), disclosure obligations analogous to prospectus rules, and prudential standards for stablecoin issuers, creating the world's first complete statutory crypto-asset regime across a major economic bloc.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:mi-ca-regulation", "labels": [ "MiCA Regulation", "MiCA Regulation 2023-1114", "MiCA Regulation EU 2023-1114" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "mi-ca", "title": "MiCA", "domain": "governance", "domain_name": "Governance", "definition": "MiCA (Markets in Crypto-Assets Regulation, EU 2023/1114) is the European Union's comprehensive legal framework governing crypto-assets not already covered by existing financial services legislation such as MiFID II. It classifies crypto-assets into three categories \u2014 utility tokens, asset-referenced tokens (ARTs), and e-money tokens (EMTs) \u2014 imposing tiered obligations on issuers including mandatory white-paper disclosures, capital adequacy requirements, and reserve safeguards for stablecoins. Crypto-asset service providers (CASPs) \u2014 encompassing exchanges, custodians, portfolio managers, and advisers \u2014 must obtain authorisation from a national competent authority with passporting rights across the EU single market. MiCA entered into force in June 2023 with stablecoin provisions applying from June 2024 and the full CASP regime from December 2024, making it one of the world's first end-to-end crypto regulatory regimes.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mi-ca", "labels": [ "MiCA" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Stablecoin", "OECD", "Regulatory Domain", "Regulation (EU) 2023/1114 on Markets in Crypto-Assets" ] }, { "id": "micro-oled-display", "title": "Micro-OLED Display", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A micro-OLED display is an organic light-emitting diode panel fabricated directly on a silicon backplane (OLED-on-silicon), yielding pixel pitches of a few micrometres and very high pixel densities. Its compact size, high contrast, fast response, and low power make it the dominant near-eye display technology for AR and VR headsets. The silicon substrate enables integrated drive circuitry impossible on glass.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:micro-oled-display", "labels": [ "Micro-OLED Display" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "micro-bt", "title": "MicroBT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "MicroBT is a Chinese manufacturer of Bitcoin mining hardware, known for its Whatsminer series of ASIC machines. It competes with other mining hardware producers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:micro-bt", "labels": [ "MicroBT" ], "is_subclass_of": [ "ASIC" ], "wikilinks": [ "Bitcoin Mining", "Transaction Validation", "Hardware", "ASIC", "https://www.microbt.com", "https://whatsminer.net" ] }, { "id": "micro-strategy", "title": "MicroStrategy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "MicroStrategy (rebranded Strategy in 2025) is a Nasdaq-listed business intelligence and analytics software company that became the first publicly traded corporation to adopt Bitcoin as its primary treasury reserve asset, initiating this strategy in August 2020 under executive chairman Michael Saylor. The company has since deployed multiple capital market instruments\u2014including convertible notes, equity offerings, and preferred stock\u2014to continuously acquire Bitcoin, accumulating over 500,000 BTC by early 2025 and creating a leveraged Bitcoin exposure vehicle accessible through public equity markets. MicroStrategy's approach established a corporate treasury model that has been studied and replicated by other public companies seeking Bitcoin exposure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:micro-strategy", "labels": [ "MicroStrategy" ], "is_subclass_of": [ "Enterprise Blockchain" ], "wikilinks": [] }, { "id": "microcontroller", "title": "Microcontroller", "domain": "robotics", "domain_name": "Robotics", "definition": "A microcontroller is a compact integrated circuit that combines a processor core, memory and programmable input and output peripherals on a single chip, designed to run a dedicated control program. Unlike a general-purpose processor, it embeds the resources needed for embedded control directly, enabling low-cost, low-power operation in devices that sense and actuate their environment. Microcontrollers execute firmware in real time and are the computational heart of countless embedded and robotic systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:microcontroller", "labels": [ "Microcontroller" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [] }, { "id": "microdisplay", "title": "Microdisplay", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A microdisplay is a miniature display panel, typically under an inch diagonal, with very high pixel density, used in near-eye optical systems such as headsets, viewfinders and projectors. Built on technologies including micro-OLED, liquid-crystal-on-silicon (LCoS) and microLED, it produces a small bright image that magnifying or waveguide optics expand into a large virtual image for the eye. Microdisplays are a critical enabling component for augmented and virtual reality head-mounted displays, where size, brightness, resolution and power efficiency are paramount.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:microdisplay", "labels": [ "Microdisplay" ], "is_subclass_of": [ "Display Hardware" ], "wikilinks": [] }, { "id": "microeconomics", "title": "Microeconomics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Microeconomics is the branch of economics that studies the behaviour of individual agents \u2014 households, firms, and markets \u2014 and how their decisions about allocation of scarce resources determine prices and quantities. It analyses supply and demand, consumer and producer choice, market structures, and the conditions under which markets reach equilibrium or fail. In blockchain and token systems, microeconomic reasoning underpins mechanism design, incentive structures, and the analysis of fee markets and liquidity. It contrasts with macroeconomics, which studies aggregate economic phenomena.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:microeconomics", "labels": [ "Microeconomics" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "microgravity", "title": "Microgravity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:microgravity", "labels": [ "Microgravity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "micropayment", "title": "Micropayment", "domain": "finance", "domain_name": "Finance", "definition": "A very small electronic payment, typically fractions of a cent to a few dollars, processed automatically within digital environments to enable low-value, high-frequency transactions that are economically infeasible with traditional payment infrastructure.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:micropayment", "labels": [ "Micropayment" ], "is_subclass_of": [ "Digital Payment System" ], "wikilinks": [ "Cryptographic Authentication", "Fee Calculation", "Instant Settlements", "Microtransactions", "OMA3", "Pay-Per-Use Models", "Payment Network", "Payment Protocol", "Reed Smith", "Settlement Mechanism", "Transaction Validation", "Blockchain Infrastructure", "Central Bank Digital Currency", "Digital Payment System", "Digital Wallet", "MiddlewareLayer", "VirtualEconomyDomain" ] }, { "id": "micropayments", "title": "Micropayments", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Micropayments are electronic payment transactions with values typically below \u2014 often in the sub-cent to sub-dollar range \u2014 enabling granular, per-use monetisation of digital goods, services, API calls, data streams, and creative content at a price granularity previously impractical due to the co...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:micropayments", "labels": [ "Micropayments", "BC-0319-micropayments" ], "is_subclass_of": [ "Computing and Cloud", "Payment Systems", "Digital Finance", "Bitcoin Proof-of-Work Protocol", "Blockchain Network", "Lightning and Similar L2", "Money" ], "wikilinks": [ "Advertising Revenue", "Andreessen Horowitz a16z Crypto State of Crypto Report 2025", "arXiv 2412.19384 Internet of Value Blockchain Lightning Micropayments 2024", "Beres et al 2021 Cryptoeconomic Traffic Analysis Bitcoin Lightning arXiv 2002.06998", "BIS Project Agor\u00e1 Quarterly Review 2024", "BIS Working Paper Cross-Border Payments Tokenised Deposits 2024", "Calle 2023 Cashu NUT Specification GitHub", "CCAF Cambridge Bitcoin Electricity Consumption Index 2024", "CCAF Global Cryptoasset Benchmarking Study 2024", "Chaum 1982 Blind Signatures for Untraceable Payments", "Coinbase X402 Protocol Cloudflare September 2025", "Digital Finance", "Fedimint", "Financial Inclusion", "Fishburn Odlyzko Siders 1997 Fixed Fee versus Unit Pricing First Monday", "IC3RE Imperial College Centre Cryptocurrency Research Engineering 2024", "IoT", "Johansson et al 2022 Fedimint Federated Chaumian Mints", "Kahneman Tversky 1979 Prospect Theory", "Lightning Labs L402 Documentation Aperture Proxy 2024" ] }, { "id": "microphone-array", "title": "Microphone Array", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A microphone array is an arrangement of multiple microphone elements at known spatial positions whose signals are combined to infer or shape the directional properties of captured sound. By exploiting the time and phase differences between elements, an array supports beamforming, source localisation and spatial noise suppression that a single microphone cannot achieve. Microphone arrays underpin far-field voice capture, conferencing systems and spatial-audio acquisition.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:microphone-array", "labels": [ "Microphone Array" ], "is_subclass_of": [ "Audio System" ], "wikilinks": [] }, { "id": "microphone", "title": "Microphone", "domain": "robotics", "domain_name": "Robotics", "definition": "A microphone is a transducer that converts acoustic sound waves into an electrical signal, serving as the primary audio sensor in interactive and telepresence systems. Common types include electret, MEMS, condenser, and dynamic, characterised by polar pattern, sensitivity, and signal-to-noise ratio. In robots and voice interfaces, microphone arrays additionally enable beamforming and source localisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:microphone", "labels": [ "Microphone", "Microphone State" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "microscope", "title": "Microscope", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:microscope", "labels": [ "Microscope" ], "is_subclass_of": [ "Optical Instrument" ], "wikilinks": [] }, { "id": "microservices-architecture", "title": "Microservices Architecture", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed software architecture pattern that decomposes applications into independent, loosely coupled services communicating via APIs, enabling high concurrency, scalability, and resilience through containerisation technologies like Docker and orchestration platforms like Kubernetes.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:microservices-architecture", "labels": [ "Microservices Architecture", "MicroservicesArchitecture" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [ "Scalable Applications", "metaverse", "System Architecture" ] }, { "id": "microservices", "title": "Microservices", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Microservices is an architectural style in which a large application is decomposed into a suite of small, independently deployable services, each responsible for a distinct bounded business capability and communicating through well-defined lightweight APIs or message channels. Each service runs in its own process, manages its own data store, and can be developed, deployed, scaled, and retired independently of other services. The pattern contrasts with monolithic architectures by enabling polyglot development, fine-grained fault isolation, and organisational alignment of teams with service ownership, at the cost of increased operational and distributed-systems complexity.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:microservices", "labels": [ "Microservices", "Microservice", "Polyglot Microservices" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "microsoft-copilot", "title": "Microsoft Copilot", "domain": "ai", "domain_name": "Ai", "definition": "Microsoft Copilot is a family of AI-powered assistant products developed by Microsoft and embedded across its software ecosystem, including Windows, Microsoft 365, the Edge browser, GitHub, and Azure. The system builds on large language models from OpenAI (GPT-4 series) combined with Microsoft's own retrieval augmentation over Microsoft Graph, enabling context-aware responses grounded in a user's documents, emails, meetings, and calendar data. Enterprise deployments offer tenancy-scoped data access with existing permission boundaries, distinguishing Copilot from general-purpose chatbots. GitHub Copilot, a precursor product launched in 2021, provides AI-assisted code completion, chat, and pull-request summarisation inside developer environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:microsoft-copilot", "labels": [ "Microsoft Copilot", "Copilot", "Copilot Systems" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Large Language Model", "Llama", "Artificial Intelligence Domain" ] }, { "id": "microsoft-entra-verified-id", "title": "Microsoft Entra Verified ID", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Microsoft Entra Verified ID is a managed service for issuing and verifying decentralised identity credentials based on open standards for verifiable credentials.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:microsoft-entra-verified-id", "labels": [ "Microsoft Entra Verified ID" ], "is_subclass_of": [ "Verifiable Credentials" ], "wikilinks": [ "Decentralised Identity", "Identity Verification", "Identity Management", "Verifiable Credentials", "https://learn.microsoft.com/en-us/entra/verified-id/", "https://www.w3.org/TR/vc-data-model/" ] }, { "id": "microsoft-mesh", "title": "Microsoft Mesh", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Microsoft Mesh is a platform for shared mixed-reality experiences that lets distributed participants meet as avatars in persistent 3D spaces across headsets and conventional devices.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:microsoft-mesh", "labels": [ "Microsoft Mesh" ], "is_subclass_of": [ "Metaverse", "Metaverse Domain" ], "wikilinks": [ "Avatar System", "Mixed Reality", "Virtual World", "Spatial Computing", "Metaverse Domain" ] }, { "id": "microsoft-teams", "title": "Microsoft Teams", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Microsoft Teams is a collaboration application combining chat, video meetings, file sharing and application integration within the Microsoft 365 suite.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:microsoft-teams", "labels": [ "Microsoft Teams" ], "is_subclass_of": [ "Collaboration Tools" ], "wikilinks": [ "Identity Management", "Communication Protocols", "Microsoft", "Collaboration Tools", "https://www.microsoft.com/en-us/microsoft-teams/group-chat-software", "https://learn.microsoft.com/en-us/microsoftteams/" ] }, { "id": "microsoft", "title": "Microsoft", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Microsoft is a multinational technology corporation that develops and distributes operating systems, enterprise productivity software, cloud computing services, artificial intelligence platforms, developer tools, and mixed-reality hardware; its Azure cloud, Windows OS, Office 365, GitHub, and HoloLens product lines make it a foundational infrastructure and AI provider across enterprise and consumer markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:microsoft", "labels": [ "Microsoft", "Microsoft 365", "Microsoft Corporation", "Microsoft Research" ], "is_subclass_of": [ "Distributed Systems", "Standards Organization" ], "wikilinks": [ "Spatial Computing", "Augmented Reality", "Graphics API", "Distributed Systems", "Standards Organization" ] }, { "id": "microwave-radiometer", "title": "Microwave Radiometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:microwave-radiometer", "labels": [ "Microwave Radiometer" ], "is_subclass_of": [ "Radiometer" ], "wikilinks": [] }, { "id": "microwave-remote-sensing", "title": "Microwave Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:microwave-remote-sensing", "labels": [ "Microwave Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "middle-atmosphere", "title": "Middle Atmosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:middle-atmosphere", "labels": [ "Middle Atmosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "middleware-layer", "title": "Middleware Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Middleware Layer is the stratum of software infrastructure that sits between low-level platform services or network protocols and the application-facing interfaces that consume them, providing integration, abstraction, orchestration, and interoperability services. It decouples system components by standardising communication contracts through APIs, messaging buses, and RPC frameworks, and bridges heterogeneous subsystems \u2014 including on-chain protocols, off-chain data sources, and cross-network boundaries \u2014 without requiring changes to the underlying protocol or the consuming application. In distributed and blockchain architectures the layer encompasses API gateways, oracle networks, cross-chain bridges, indexing services, wallet connectors, and transaction orchestration utilities that collectively raise the abstraction level available to developers. Its maturity is well-established in enterprise computing and rapidly maturing in decentralised ecosystems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:middleware-layer", "labels": [ "Middleware Layer", "MiddlewareLayer" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "middleware", "title": "Middleware", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software layer that mediates between applications and underlying services or infrastructure to enable communication, resource access, and interoperability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:middleware", "labels": [ "Middleware", "Blockchain Middleware", "Integration Middleware", "Robotics Middleware" ], "is_subclass_of": [ "Computing and Cloud", "Digital Infrastructure" ], "wikilinks": [ "Distributed Communication", "ETSI GR ARF 010", "EWG/MSF Taxonomy", "ISO/IEC 30170", "Message Queue", "Resource Abstraction", "Service Discovery", "Service Integration", "Service Registry", "API Gateway", "Communication Protocol", "ComputationAndIntelligenceDomain", "Compute Infrastructure", "ComputeLayer", "Data Format", "DataLayer", "Distributed System", "InfrastructureDomain", "Interoperability", "Network Infrastructure" ] }, { "id": "midjourney-text-to-image-service", "title": "Midjourney Text-to-Image Service", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Midjourney is a proprietary generative artificial intelligence service that produces images from natural-language text prompts. It is operated by an independent research lab of the same name and is accessed primarily through a Discord bot interface and, later, a dedicated web application. The system is known for a distinctive aesthetic and for iterating rapidly through successive model versions that improve coherence, resolution and prompt adherence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:midjourney-text-to-image-service", "labels": [ "Midjourney Text-to-Image Service", "Midjourney" ], "is_subclass_of": [ "Text-to-Image Generation", "Creative Media Domain" ], "wikilinks": [ "Diffusion Model", "Text-to-Image Generation", "Concept Art", "Synthetic Media", "Generative AI", "Stable Diffusion", "Creative Media Domain" ] }, { "id": "mina-protocol", "title": "Mina Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Mina Protocol is a blockchain that maintains a constant-size proof of its state using recursive zero-knowledge proofs, keeping the chain compact.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mina-protocol", "labels": [ "Mina Protocol" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [ "Cryptography", "Web3 Infrastructure", "Distributed Ledger Technology", "Blockchain", "https://minaprotocol.com/", "https://docs.minaprotocol.com/" ] }, { "id": "mind-map", "title": "Mind Map", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A mind map is a diagram that organises information hierarchically around a central concept, with branches radiating to related topics and sub-topics. It is used for brainstorming, note-taking, and knowledge organisation, leveraging spatial and associative structure to aid recall and ideation. Digital mind-mapping tools support collaborative editing and links to external resources.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mind-map", "labels": [ "Mind Map" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "mind-uploading", "title": "Mind Uploading", "domain": "ai", "domain_name": "Ai", "definition": "Mind uploading is the hypothetical process of scanning a biological brain in sufficient detail to emulate its structure and function on a computational substrate, producing a digital instantiation of an individual's mind. It is a speculative concept central to transhumanist visions of substrate-independent consciousness and digital immortality. No current technology approaches the required resolution or theoretical understanding of consciousness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:mind-uploading", "labels": [ "Mind Uploading" ], "is_subclass_of": [ "Artificial Intelligence", "Artificial General Intelligence" ], "wikilinks": [] }, { "id": "mined-data-state", "title": "Mined Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mined-data-state", "labels": [ "Mined Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "miner-extractable-value", "title": "Miner Extractable Value", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Miner extractable value, now more broadly termed maximal extractable value, is the profit that block producers (miners or validators) and other actors can capture by reordering, inserting or censoring transactions within the blocks they produce. Because the producer controls transaction ordering, they can exploit pending transactions in the mempool through strategies such as front-running, back-running and sandwich attacks, especially around decentralised exchanges. MEV has significant implications for fairness, network economics, censorship resistance and protocol design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:miner-extractable-value", "labels": [ "Miner Extractable Value" ], "is_subclass_of": [ "Blockchain Economics" ], "wikilinks": [] }, { "id": "miner", "title": "Miner", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Miner is a network participant in a proof-of-work blockchain who dedicates computational resources to solving cryptographic hash puzzles, competing to produce valid blocks that extend the canonical chain in exchange for a block reward (newly issued coins plus transaction fees). Miners collectively provide the computational security of the network: the cost of mounting a 51% attack is proportional to the total hash rate, which represents real-world energy and hardware expenditure. Mining can be performed individually or cooperatively within mining pools that aggregate hash power and share rewards proportionally.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:miner", "labels": [ "Miner" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "mini-batch", "title": "Mini-Batch", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A mini-batch is a small, fixed-size subset of a training dataset processed together in a single forward and backward pass when training a machine learning model. Mini-batch gradient descent computes the gradient over the mini-batch rather than over a single example (stochastic) or the entire dataset (full batch), balancing the noise-reduction benefits of larger batches against the computational and memory cost. The mini-batch size is a key hyperparameter that influences convergence behaviour, gradient variance, hardware utilisation and generalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mini-batch", "labels": [ "Mini-Batch" ], "is_subclass_of": [ "Stochastic Gradient Descent" ], "wikilinks": [] }, { "id": "minimal-risk-ai", "title": "Minimal Risk AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The residual tier in the EU AI Act's four-level risk pyramid, covering AI systems not classified as prohibited, high-risk, or limited-risk. Such systems carry no AI Act-specific compliance obligations and remain subject only to applicable horizontal legislation such as GDPR and product liability rules.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:minimal-risk-ai", "labels": [ "Minimal Risk AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "minimally-invasive-surgery", "title": "Minimally Invasive Surgery", "domain": "robotics", "domain_name": "Robotics", "definition": "Minimally invasive surgery is a class of surgical technique performed through small incisions using specialised instruments and cameras, rather than the large openings required by traditional open surgery. It reduces patient trauma, blood loss, and recovery time by limiting the surgical footprint while preserving surgical precision. Surgical robots extend minimally invasive surgery by providing tremor-filtered, precisely scaled instrument control through the small incisions used in surgical robotics.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:minimally-invasive-surgery", "labels": [ "Minimally Invasive Surgery" ], "is_subclass_of": [ "Surgical Robotics" ], "wikilinks": [] }, { "id": "minimax-algorithm", "title": "Minimax Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Minimax Algorithm is a artificial intelligence concept and a type of Search Algorithms. that enables Adversarial Search.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:minimax-algorithm", "labels": [ "Minimax Algorithm", "Minimax Optimisation", "Minimax Search" ], "is_subclass_of": [ "AI Technique", "Search Algorithms" ], "wikilinks": [ "Adversarial Search", "Search Algorithms" ] }, { "id": "minimum-valid-data-value", "title": "Minimum Valid Data Value", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:minimum-valid-data-value", "labels": [ "Minimum Valid Data Value" ], "is_subclass_of": [ "Measurement Value" ], "wikilinks": [] }, { "id": "mining-hardware", "title": "Mining Hardware", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Mining hardware is the specialised computing equipment used to perform the cryptographic work that secures proof-of-work blockchains, racing to find valid block hashes in exchange for block rewards. It has progressed from general-purpose CPUs and GPUs to dedicated application-specific integrated circuits (ASICs) optimised for a single hashing algorithm. The efficiency and concentration of this hardware shape network security, energy consumption, and mining centralisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:mining-hardware", "labels": [ "Mining Hardware" ], "is_subclass_of": [ "Cryptocurrency Mining" ], "wikilinks": [] }, { "id": "mining-node", "title": "Mining Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A mining node is a blockchain participant that not only validates and relays transactions but also competes to produce new blocks by performing proof-of-work computation. It assembles pending transactions into candidate blocks and searches for a valid solution that allows the block to be added to the chain. Mining nodes secure proof-of-work networks and are rewarded for the blocks they successfully add.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mining-node", "labels": [ "Mining Node" ], "is_subclass_of": [ "Full Node" ], "wikilinks": [] }, { "id": "mining-pool", "title": "Mining Pool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cooperative arrangement in which multiple miners aggregate their computational resources to increase the probability of successfully mining a block, sharing the resulting block reward proportionally to contributed hash rate. Mining pools reduce variance in miner income but introduce centralisation risks and hash-rate concentration that can threaten network security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mining-pool", "labels": [ "Mining Pool", "Mining Pool Data" ], "is_subclass_of": [ "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "mining-reward", "title": "Mining Reward", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Mining Reward is the economic incentive awarded to the miner who successfully produces a valid block and appends it to the canonical blockchain, comprising a block subsidy of newly minted cryptocurrency plus the aggregate transaction fees of all transactions included in that block. The subsidy follows a pre-programmed halving schedule\u2014in Bitcoin, halving every 210,000 blocks\u2014gradually reducing issuance until the subsidy approaches zero and transaction fees become the sole miner compensation, aligning long-term network security incentives with user demand for block space.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:mining-reward", "labels": [ "Mining Reward" ], "is_subclass_of": [ "Blockchain Entity" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "mining", "title": "Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Mining is the proof-of-work block-creation process in which participating nodes compete to solve a cryptographic hash puzzle, with the winner appending the next block to the chain and receiving a block reward. Mining provides Sybil resistance, ensures probabilistic finality, and anchors chain security to real-world energy expenditure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mining", "labels": [ "Mining", "Blockchain Mining", "Containerised Mining", "Merged Mining", "Mining Worker", "Proof-of-Work Mining" ], "is_subclass_of": [ "Consensus Protocol", "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "mint-burn-mechanism", "title": "Mint-Burn Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A token supply-management pattern in which new units are created (minted) when value enters a system and permanently destroyed (burned) when it leaves, keeping circulating supply in one-to-one correspondence with the assets, collateral, or claims backing it. Implemented as privileged mint and burn functions in a token's smart contract, the pattern underlies fiat-backed and algorithmic stablecoins, wrapped tokens that represent assets locked on another chain, synthetic assets minted against collateral, and cross-chain bridges that burn on the source chain and mint on the destination. Its integrity depends entirely on access control and honest accounting of the backing: compromised mint authority or unbacked minting is a recurring cause of catastrophic protocol failures.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:mint-burn-mechanism", "labels": [ "Mint-Burn Mechanism", "Mint Burn Mechanism" ], "is_subclass_of": [ "Tokenomics" ], "wikilinks": [ "Tokenomics", "Smart Contract", "Wrapped Token", "Synthetic Asset", "Peg" ] }, { "id": "minting", "title": "Minting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The protocol-governed creation of new tokens on a blockchain, which may be fungible (ERC-20) or non-fungible (ERC-721), increasing total token supply according to defined issuance rules. Minting may be permissioned (restricted to authorised smart contracts or validators), algorithmic (triggered by staking rewards or proof-of-work block production), or demand-driven (as in NFT drops), and directly determines a token's inflation schedule and economic model.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:minting", "labels": [ "Minting", "Token Minting" ], "is_subclass_of": [ "Token and Asset", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "mipmap", "title": "Mipmap", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A precomputed pyramid of progressively half-resolution versions of a texture image, from full size down to a single texel, from which the graphics pipeline selects (and trilinearly blends between) the level whose texel density best matches the on-screen footprint of the surface being shaded; introduced by Lance Williams in 1983, mipmapping eliminates minification aliasing and shimmer, improves texture-cache coherence, and costs only one third more memory than the base image.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:mipmap", "labels": [ "Mipmap" ], "is_subclass_of": [ "Texture Map" ], "wikilinks": [ "Texture Map", "Texture Mapping", "Texture Compression", "Foveated Rendering" ] }, { "id": "miro", "title": "Miro", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Miro is a cloud-based collaborative whiteboard platform that provides an infinite canvas for distributed teams to brainstorm, diagram, map workflows, and run workshops in real time. It offers sticky notes, templates, frameworks, and integrations with tools such as Jira, Slack, and design suites. It is widely adopted for product discovery, agile ceremonies, and remote facilitation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:miro", "labels": [ "Miro", "Miro AI" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "misinformation", "title": "Misinformation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Misinformation is false or inaccurate information that is spread regardless of intent to deceive, distinguishing it from disinformation, which is deliberately fabricated. In AI contexts, misinformation risk is amplified by generative capabilities such as deepfakes and by model hallucination, which can produce plausible but false content at scale. Trust and safety practice addresses misinformation through detection, provenance labelling, and content moderation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:misinformation", "labels": [ "Misinformation" ], "is_subclass_of": [ "Trust and Safety" ], "wikilinks": [] }, { "id": "missing-value", "title": "Missing Value", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:missing-value", "labels": [ "Missing Value" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mission-control-centre", "title": "Mission Control Centre", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mission-control-centre", "labels": [ "Mission Control Centre" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mission-design", "title": "Mission Design", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mission-design", "labels": [ "Mission Design" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mission-operations", "title": "Mission Operations", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mission-operations", "labels": [ "Mission Operations" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mission-phase", "title": "Mission Phase", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mission-phase", "labels": [ "Mission Phase" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mission-planning", "title": "Mission Planning", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:mission-planning", "labels": [ "Mission Planning" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "mistral-ai-open-weight-model-family", "title": "Mistral AI Open-Weight Model Family", "domain": "ai", "domain_name": "Ai", "definition": "Mistral AI is a French AI company that developed a family of open-weight large language models, including the dense Mistral 7B and the sparse Mixture-of-Experts Mixtral 8x7B and 8x22B architectures. These models are distinguished by their efficient use of grouped-query attention, sliding window attention, and sparse expert routing, achieving performance competitive with much larger models at a fraction of the inference cost. Released under permissive licences, they have become foundational reference models for the open-source AI community.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:mistral-ai-open-weight-model-family", "labels": [ "Mistral AI Open-Weight Model Family", "Mistral AI", "Mistral Model Family" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "mistral", "title": "Mistral", "domain": "ai", "domain_name": "Ai", "definition": "Mistral is a family of language models from Mistral AI, several released as open weights. The line includes dense models and mixture-of-experts models for text generation and reasoning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:mistral", "labels": [ "Mistral" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [ "Transformer", "Attention Mechanism", "Retrieval-Augmented Generation", "Function Calling", "Language Model", "Natural Language Processing", "Large Language Models" ] }, { "id": "mixed-precision-training", "title": "Mixed Precision Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A training technique that uses lower precision (FP16 or BF16) for most computations whilst maintaining higher precision (FP32) for numerically critical operations, reducing memory usage and increasing training speed without sacrificing model quality. Relies on loss scaling to prevent gradient underflow and leverages hardware tensor cores for throughput gains.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:mixed-precision-training", "labels": [ "Mixed Precision Training", "Mixed-Precision Training" ], "is_subclass_of": [ "Neural Network Training" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "mixed-reality-mr", "title": "Mixed Reality (MR)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interactive environment where physical and digital elements coexist, interact bidirectionally, and dynamically influence each other in real time with advanced occlusion, lighting, and physics simulation creating seamless blended experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:mixed-reality-mr", "labels": [ "Mixed Reality (MR)" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ACM", "Bidirectional Interaction", "Depth Camera", "Environmental Lighting", "ISO 9241-940", "Lighting Estimation", "Object Tracking", "Occlusion Rendering", "Physics Simulation Engine", "Real-Time 3D Reconstruction", "Realistic Occlusion", "ComputeLayer", "Computer Vision", "Depth Sensing", "Extended Reality (XR)", "Hand Tracking", "InteractionDomain", "NetworkLayer", "Physics Engine", "Shared Spatial Anchors" ] }, { "id": "mixed-reality-meeting", "title": "Mixed Reality Meeting", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Mixed Reality Meeting is a collaborative session in which remote and local participants interact within an environment that blends physical and virtual elements through mixed reality headsets or displays. Remote attendees may appear as holograms or avatars anchored in the local physical space, while shared digital artefacts are overlaid onto the real world for all participants. This format dissolves the boundary between in-room and remote presence, enabling richer spatial and contextual collaboration than traditional video conferencing.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:mixed-reality-meeting", "labels": [ "Mixed Reality Meeting" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "mixed-reality-platform", "title": "Mixed Reality Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A hardware and software ecosystem that overlays and anchors digital content onto the physical environment in real time, enabling simultaneous interaction with both real and virtual objects. Mixed reality platforms combine inside-out tracking, spatial mapping, and pass-through or optical display technologies to support immersive collaborative and industrial workflows.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:mixed-reality-platform", "labels": [ "Mixed Reality Platform" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "mixed-reality", "title": "Mixed Reality", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mixed reality (MR) is the perceptual and computational regime occupying the central span of the Milgram-Kishino Reality-Virtuality Continuum (1994, IEICE Transactions on Information Systems E77-D:12, 1321-1329), wherein real-world physical objects and digitally synthesised content coexist, intera...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:mixed-reality", "labels": [ "Mixed Reality", "Mixed Reality Interaction", "MixedReality" ], "is_subclass_of": [ "Platform and Environment", "Immersive Computing", "Extended Reality", "Spatial Computing Paradigm", "Augmented Reality", "Human Computer Interaction", "Perceptual Computing" ], "wikilinks": [ "6DoF Tracking", "Ambient Computing", "Architectural Visualisation", "Avatar Presence", "Azure Spatial Anchors", "Cloud Anchor Infrastructure", "Compute SoC", "ComputerVisionDomain", "Depth Sensor", "DeviceLayer", "Diffractive Waveguide", "Digital Twin Overlay", "Display Calibration", "Extended Reality", "Flat Screen Computing", "Foveated Rendering", "Head-Mounted Display", "Holographic Rendering", "Holographic Training", "Holography" ] }, { "id": "mixture-of-experts", "title": "Mixture of Experts", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A sparse neural network architecture in which a learned router (gating network) dispatches each input \u2014 in modern transformers, each token at each MoE layer \u2014 to a small subset of many parallel expert sub-networks, so that total parameter count can grow enormously while per-token computation stays roughly constant; introduced by Jacobs and Jordan in 1991 and revived at scale by Shazeer's sparsely-gated MoE and the Switch Transformer, it underpins frontier models such as Mixtral and DeepSeek-V3.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:mixture-of-experts", "labels": [ "Mixture of Experts" ], "is_subclass_of": [ "Neural Network Architecture" ], "wikilinks": [ "Neural Network Architecture", "Transformer Architecture", "Model Scaling", "Monolithic Ai" ] }, { "id": "mixture-of-experts-architecture", "title": "Mixture-of-Experts Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An architecture that uses multiple specialised sub-networks (experts) with a gating mechanism that routes inputs to a sparse subset of experts, enabling scaling without proportional compute increases. MoE is adopted in production LLMs like GPT-4, enabling massive scale with controlled costs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:mixture-of-experts-architecture", "labels": [ "Mixture-of-Experts Architecture", "Mixture of Experts", "Mixture-of-Experts" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "research", "MetaverseDomain", "natural language processing" ] }, { "id": "mixup", "title": "Mixup", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A data augmentation technique that creates virtual training samples by computing convex linear combinations of input pairs and their corresponding labels, parameterised by a mixing coefficient drawn from a Beta distribution. Mixup reduces overfitting, improves calibration, and promotes smoother decision boundaries across supervised and semi-supervised learning tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:mixup", "labels": [ "Mixup", "Mixup Augmentation" ], "is_subclass_of": [ "AI Technique", "Data Augmentation Strategies" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "mlperf", "title": "Mlperf", "domain": "ai", "domain_name": "Ai", "definition": "MLPerf is a suite of standardised benchmarks, governed by the MLCommons consortium, that measures the performance of machine learning hardware, software and systems for both training and inference. It defines fixed reference models, datasets, quality targets and submission rules so that results from different vendors are reproducible and directly comparable. MLPerf has become an industry reference for evaluating accelerators, frameworks and end-to-end systems on representative deep learning workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mlperf", "labels": [ "Mlperf", "MLPerf" ], "is_subclass_of": [ "Performance Benchmarks" ], "wikilinks": [] }, { "id": "mob-programming", "title": "Mob Programming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Mob programming is a software development practice in which the whole team works on the same task at the same time, on the same computer, rotating the roles of driver and navigators. It extends pair programming to the full group, concentrating collective knowledge to improve code quality, shorten feedback loops, and spread learning. It trades raw parallelism for higher alignment and fewer hand-offs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mob-programming", "labels": [ "Mob Programming" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "mobile-broadband", "title": "Mobile Broadband", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Mobile broadband is high-capacity internet access delivered over cellular radio networks such as 3G, 4G/LTE and 5G, allowing devices to connect without a fixed wired line. It depends on licensed spectrum allocated to network operators and is provisioned through cell towers and core network infrastructure rather than fibre or cable to the premises. Mobile broadband underpins smartphone data access, fixed-wireless access, and IoT connectivity in areas where fixed broadband is unavailable or impractical.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:mobile-broadband", "labels": [ "Mobile Broadband" ], "is_subclass_of": [ "Broadband Connectivity" ], "wikilinks": [] }, { "id": "mobile-computing", "title": "Mobile Computing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Mobile computing is the discipline of delivering computation, data, and connectivity to portable devices such as smartphones, tablets, and wearables that operate while in motion. It contends with constraints on battery, memory, bandwidth, and intermittent connectivity, and increasingly hosts machine-learning inference directly on-device. Mobile computing bridges cloud back-ends and edge devices, enabling responsive, context-aware applications wherever the user goes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:mobile-computing", "labels": [ "Mobile Computing" ], "is_subclass_of": [ "Edge Computing" ], "wikilinks": [] }, { "id": "mobile-edge-computing", "title": "Mobile Edge Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Mobile edge computing, also termed multi-access edge computing, places cloud computing capabilities at the edge of the mobile network, close to end users and devices. By hosting applications and services at base stations or aggregation points it reduces latency, conserves backhaul bandwidth and enables real-time processing. It is a foundational enabler for 5G use cases such as augmented reality, autonomous vehicles and industrial automation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:mobile-edge-computing", "labels": [ "Mobile Edge Computing" ], "is_subclass_of": [ "Edge Computing" ], "wikilinks": [] }, { "id": "mobile-manipulation", "title": "Mobile Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Mobile manipulation is the subfield of robotics that integrates a locomoting mobile base with one or more robotic manipulator arms, enabling a robot to traverse unstructured environments and physically interact with objects beyond the reach of a fixed-base system. It requires the coordinated control of both navigation and dexterous manipulation, necessitating joint motion planning, whole-body control, and perception pipelines that resolve the relative pose of targets in the robot's frame. The discipline addresses challenges absent in purely fixed-arm or purely mobile systems, including base-arm kinematic coupling, dynamic stability during manipulation, and long-horizon task planning across navigation and contact phases.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mobile-manipulation", "labels": [ "Mobile Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [ "Manipulator", "Mobile Robot", "Joint Configuration", "Manipulation" ] }, { "id": "mobile-manipulator", "title": "Mobile Manipulator", "domain": "robotics", "domain_name": "Robotics", "definition": "A Mobile Manipulator is a robotic system that integrates a mobile base \u2014 wheeled, legged, or tracked \u2014 with one or more articulated manipulator arms, enabling it to navigate through unstructured environments and perform dexterous manipulation tasks at arbitrary locations. This combination resolves the fundamental trade-off between workspace reach and precision, making mobile manipulators suitable for logistics, inspection, surgery assistance, and domestic service applications. Effective operation requires tight coordination between navigation, motion planning, perception, and manipulation subsystems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:mobile-manipulator", "labels": [ "Mobile Manipulator", "Autonomous Mobile Manipulator" ], "is_subclass_of": [ "Robot Type", "Hybrid Robot" ], "wikilinks": [ "Hybrid Robot", "Robotics" ] }, { "id": "mobile-robot-platform", "title": "Mobile Robot Platform", "domain": "robotics", "domain_name": "Robotics", "definition": "A Mobile Robot Platform is an integrated mechatronic base that provides locomotion, power, computing, and sensor-mounting infrastructure upon which higher-level autonomy stacks\u2014perception, planning, and control\u2014are deployed.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:mobile-robot-platform", "labels": [ "Mobile Robot Platform" ], "is_subclass_of": [ "Robot Type", "Autonomous System", "Cyber Physical System", "Unmanned Ground Vehicle", "Embodied AI", "Mobile Robot" ], "wikilinks": [ "Agricultural Robotics", "Autonomous System", "AutonomousSystemsDomain", "Battery Management System", "Behaviour Tree", "Behaviour Tree Control", "Chassis Frame", "Communication Interface", "Cyber Physical System", "CyberPhysicalSystemsDomain", "DDS Middleware", "Deployment Cost", "Depth Camera", "Differential Drive Kinematics", "Edge AI", "Embodied AI", "Environment Mapping", "FastDDS", "Fixed Robot Manipulator", "Fleet Management" ] }, { "id": "mobile-robot", "title": "Mobile Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A mobile robot is an autonomous or semi-autonomous electromechanical system equipped with a locomotion mechanism \u2014 wheels, tracks, legs, rotors, or thrusters \u2014 that enables it to navigate within or across physical environments without being fixed to a stationary base. Mobile robots integrate sensing, actuation, and computation to perceive their surroundings, plan feasible paths, and execute goal-directed motion, distinguishing them from fixed industrial manipulators. They span a wide spectrum of embodiments including ground vehicles (UGVs), aerial vehicles (UAVs/drones), underwater vehicles (AUVs), and legged walkers, unified by the capability to self-relocate in service of a task. Defined formally by ISO 8373:2021 as a robot able to travel under its own control.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mobile-robot", "labels": [ "Mobile Robot", "Mobile Robots", "MobileRobot" ], "is_subclass_of": [ "Robot" ], "wikilinks": [ "ISO 8373:2021", "Robot (RB-0001)", "Robotics" ] }, { "id": "mobile-robotics", "title": "Mobile Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Mobile robotics is the engineering and scientific discipline concerned with the design, construction, and programming of robots capable of autonomous locomotion through unstructured or semi-structured physical environments. It integrates mechanical locomotion systems (wheeled, tracked, legged, aerial, aquatic), onboard sensor suites, simultaneous localisation and mapping (SLAM) for spatial state estimation, motion planning for collision-free trajectory generation, and control systems for execution \u2014 all without continuous human tele-operation. The field is fundamentally interdisciplinary, drawing on computer science, control theory, mechanical engineering, and artificial intelligence, and underpins major commercial domains including warehouse automation, autonomous vehicles, agricultural robotics, and planetary exploration.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:mobile-robotics", "labels": [ "Mobile Robotics" ], "is_subclass_of": [ "Robotics", "Navigation and Planning" ], "wikilinks": [] }, { "id": "modality-specific-encoder", "title": "Modality-Specific Encoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural network component within a multimodal architecture that transforms raw input from one particular modality \u2014 text, image, audio, video, depth, or sensor streams \u2014 into a dense embedding using an architecture suited to that modality's structure, such as a Transformer for tokenised text or a Vision Transformer for image patches. The resulting per-modality representations are projected into a shared latent space where fusion, alignment, or cross-modal conditioning can occur.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:modality-specific-encoder", "labels": [ "Modality-Specific Encoder" ], "is_subclass_of": [ "Encoder" ], "wikilinks": [ "Encoder", "Cross-Modal Conditioning", "Transformer", "Vision Transformer", "Embedding", "Contrastive Learning", "CLIP" ] }, { "id": "model-access-suspension", "title": "Model Access Suspension", "domain": "governance", "domain_name": "Governance", "definition": "The temporary or permanent revocation of user or developer access to a specific AI model or API, typically triggered by regulatory compliance requirements, security incidents, or strategic business decisions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-access-suspension", "labels": [ "Model Access Suspension" ], "is_subclass_of": [ "Export Controls" ], "wikilinks": [] }, { "id": "model-adaptation", "title": "Model Adaptation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model adaptation is the process of adjusting a pretrained model's parameters or behaviour to perform well on a new task, domain, or dataset distinct from the one it was originally trained on. It spans a spectrum from full fine-tuning of all parameters to parameter-efficient methods such as LoRA, which inject small trainable matrices into frozen layers to adapt behaviour at a fraction of the memory and compute cost. Effective model adaptation must balance plasticity, the ability to acquire new task-specific behaviour, against catastrophic forgetting of prior capability. It is the practical mechanism through which large pretrained foundation models are specialised for downstream applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:model-adaptation", "labels": [ "Model Adaptation" ], "is_subclass_of": [ "Transfer Learning" ], "wikilinks": [] }, { "id": "model-architecture-layer", "title": "Model Architecture Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Model Architecture Layer is the stratum that specifies the structural design of a machine learning model: its operators, connectivity, and parameterisation. It sits above the Algorithm Layer, which supplies the primitives it composes, and below the Model Layer, which holds trained instances of these architectures. It contains layer definitions, network topologies, and architectural hyperparameters.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:model-architecture-layer", "labels": [ "Model Architecture Layer" ], "is_subclass_of": [ "AI Model Architecture", "owl:Thing" ], "wikilinks": [ "Algorithm Layer", "Model Layer", "Training Layer", "Transformer", "Neural Network", "owl:Thing" ] }, { "id": "model-architecture", "title": "Model Architecture", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The structural design and configuration of neural networks and machine learning systems, encompassing layer arrangements, activation functions, and connection patterns that determine how models process information and learn from data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:model-architecture", "labels": [ "Model Architecture", "ModelArchitecture" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [ "AI Model Development", "metaverse", "System Architecture" ] }, { "id": "model-based-control", "title": "Model Based Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Model Based Control (MBC) is a class of control-system design paradigms in which an explicit mathematical model of the plant's dynamics \u2014 encoding kinematics, inertia tensors, contact forces, aerodynamics, or learned neural representations \u2014 is embedded within the controller to predict, plan, and...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:model-based-control", "labels": [ "Model Based Control", "Model-based Control" ], "is_subclass_of": [ "Actuation and Control", "Control System", "Trajectory Optimisation", "Control Theory", "Optimal Control", "Predictive Control", "Robot Control" ], "wikilinks": [ "acados", "Aerial Robotics", "Autonomous Driving", "Autonomous Vehicle Control", "Behaviour Cloning", "CasADi", "Compliant Manipulation", "Constraint Satisfaction", "Constraint Set", "Constraint Specification", "Contact Mechanics", "Contact Model", "ControlLayer", "ControlSystemsDomain", "Convex Optimisation", "Cost Function", "Differential Dynamic Programming", "Drake Toolbox", "Dynamic Model", "Gaussian Processes" ] }, { "id": "model-based-reinforcement-learning", "title": "Model Based Reinforcement Learning", "domain": "robotics", "domain_name": "Robotics", "definition": "Model-based reinforcement learning is a class of reinforcement learning in which the agent learns or is given an explicit model of the environment's dynamics and reward, then uses that model to plan or to generate simulated experience for policy improvement. By predicting future states it can be far more sample-efficient than model-free methods, at the cost of vulnerability to model error. It underpins planning algorithms, world models and many robotics control approaches.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-based-reinforcement-learning", "labels": [ "Model Based Reinforcement Learning", "Model-Based Reinforcement Learning" ], "is_subclass_of": [ "Reinforcement Learning", "Reinforcement Learning for Robotics" ], "wikilinks": [] }, { "id": "model-benchmarking", "title": "Model Benchmarking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic process of evaluating and comparing the performance, capabilities, and limitations of different artificial intelligence models using standardized datasets and metrics.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-benchmarking", "labels": [ "Model Benchmarking" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "model-calibration", "title": "Model Calibration", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model calibration is the process of ensuring that the probabilities a predictive model outputs reflect the true likelihood of outcomes, so that, for example, events predicted with 70 per cent confidence occur roughly 70 per cent of the time. A well-calibrated model produces reliable confidence estimates, which is essential when predictions inform risk-sensitive decisions. Calibration is assessed with reliability diagrams and metrics such as expected calibration error, and corrected with post-hoc techniques that adjust a model's output probabilities without changing its rankings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-calibration", "labels": [ "Model Calibration" ], "is_subclass_of": [ "Model Evaluation", "AI Technique" ], "wikilinks": [] }, { "id": "model-capacity", "title": "Model Capacity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The measure of a machine learning model's ability to represent a wide variety of functions, determined by parameter count, architecture depth and width, and representational power. Models with insufficient capacity underfit the data; those with excessive capacity risk overfitting. Capacity is formally bounded by concepts such as the Vapnik\u2013Chervonenkis dimension, and is managed in practice through regularisation, pruning, and architecture search.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-capacity", "labels": [ "Model Capacity" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Digital Twin" ] }, { "id": "model-cards", "title": "Model Cards", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model Cards are short structured documents that accompany a trained machine learning model, reporting its intended uses, out-of-scope uses, training and evaluation data characteristics, disaggregated performance metrics across demographic groups, known limitations, and ethical considerations. Introduced by Mitchell et al. (2019) at Google, they standardise transparency disclosures so that developers, deployers, and affected communities can judge whether a model is appropriate for a given context. Model Cards serve as a communication artefact bridging technical model documentation to governance, fairness auditing, and regulatory compliance workflows.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:model-cards", "labels": [ "Model Cards", "Model Card" ], "is_subclass_of": [ "Responsible AI" ], "wikilinks": [ "Transparency", "Accountability", "AI Governance", "Responsible AI" ] }, { "id": "model-checking", "title": "Model Checking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model checking is an automated formal-verification technique that exhaustively explores the reachable states of a finite-state model of a system to determine whether it satisfies a specification, typically expressed in temporal logic. When the property fails, the model checker returns a concrete counterexample trace, making it valuable for debugging concurrent and reactive systems. Its main challenge is the state-space explosion problem, addressed by symbolic and abstraction techniques.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-checking", "labels": [ "Model Checking" ], "is_subclass_of": [ "Formal Verification" ], "wikilinks": [] }, { "id": "model-checkpoint", "title": "Model Checkpoint", "domain": "ai", "domain_name": "Ai", "definition": "A model checkpoint is a serialised snapshot of a machine learning model's learned parameters (weights), often including optimiser state, captured at a point during or after training. Checkpoints allow training to resume after interruption, enable model sharing and deployment, and support evaluation of intermediate states. They are typically stored in formats such as safetensors, PyTorch .pt, or framework-native files.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:model-checkpoint", "labels": [ "Model Checkpoint", "Checkpoint Segment" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "model-collapse", "title": "Model Collapse", "domain": "ai", "domain_name": "Ai", "definition": "Model collapse is a degenerative process in which generative models trained recursively on their own (or other models') synthetic outputs progressively lose information about the true data distribution. Tails of the distribution disappear first, leading to reduced diversity, amplified biases, and eventual convergence on degenerate outputs. It is a key risk as AI-generated content increasingly contaminates web-scale training corpora.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:model-collapse", "labels": [ "Model Collapse", "Mode Collapse" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "model-comparison", "title": "Model Comparison", "domain": "ai", "domain_name": "Ai", "definition": "Model comparison is the systematic process of evaluating and contrasting multiple machine learning or AI models against a common set of tasks, datasets, and metrics to determine their relative strengths, weaknesses, and suitability for deployment. It encompasses both quantitative benchmarking and qualitative assessment of factors such as computational cost, latency, robustness, and alignment properties. Rigorous model comparison underpins reproducible research and responsible AI deployment decisions. The discipline has grown substantially as the proliferation of foundation models makes vendor-neutral evaluation increasingly critical.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model-comparison", "labels": [ "Model Comparison" ], "is_subclass_of": [ "Evaluation Metric", "AI Evaluation" ], "wikilinks": [] }, { "id": "model-complexity", "title": "Model Complexity", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model complexity is a measure of the capacity of a machine-learning model to fit varied patterns in data, governed by factors such as the number of parameters, the richness of the hypothesis space, and the flexibility of the functional form. Higher complexity lets a model capture intricate structure but raises the risk of overfitting, while lower complexity risks underfitting. Managing complexity is central to achieving good generalisation on unseen data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-complexity", "labels": [ "Model Complexity" ], "is_subclass_of": [ "Machine Learning", "AI Technique" ], "wikilinks": [] }, { "id": "model-compression-for-edge", "title": "Model Compression for Edge", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model Compression for Edge is the systematic application of techniques reducing neural network computational requirements, memory footprint, and inference latency to enable deployment on resource-constrained edge devices while maintaining acceptable accuracy levels through quantization, pruning, knowledge distillation, and architectural optimization.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model-compression-for-edge", "labels": [ "Model Compression for Edge" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ONNX Runtime", "PyTorch Mobile", "TensorFlow Model Optimization", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "model-compression", "title": "Model Compression", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model compression is the family of techniques that reduce the size, memory footprint, and computational cost of a machine learning model while preserving as much of its predictive accuracy as possible. Common methods include quantisation of weights to lower precision, pruning of redundant parameters, knowledge distillation into a smaller student model, and weight sharing. Compression is essential for deploying large neural networks on resource-constrained hardware and for reducing inference latency and energy use.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-compression", "labels": [ "Model Compression" ], "is_subclass_of": [ "Model Optimization", "AI Technique" ], "wikilinks": [] }, { "id": "model-context-protocol-anthropic-2024", "title": "Model Context Protocol Anthropic 2024", "domain": "ai", "domain_name": "Ai", "definition": "The Model Context Protocol (MCP) is an open standard introduced by Anthropic in 2024 that defines a uniform JSON-RPC interface for connecting large language model applications to external tools, data sources, and prompts. It standardises how AI agents discover and invoke capabilities exposed by MCP servers, decoupling model hosts from integrations. It has been broadly adopted as a common plug-in layer for agentic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:model-context-protocol-anthropic-2024", "labels": [ "Model Context Protocol Anthropic 2024" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "model-context-protocol", "title": "Model Context Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Model Context Protocol (MCP) is an open standard published by Anthropic in November 2024 that defines a JSON-RPC 2.0-based client\u2013server protocol for connecting Large Language Model inference hosts (MCP clients) to external capability providers (MCP servers), exposing tools, resources, an...", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:model-context-protocol", "labels": [ "Model Context Protocol", "Model Context Protocol Integration", "Model Context Protocol Specification" ], "is_subclass_of": [ "AI Infrastructure", "Agent Communication Protocol", "JSON-RPC Protocol" ], "wikilinks": [ "Agent Communication Protocol", "AgentLayer", "Agent Tool Use", "AI Domain", "Anthropic MCP Specification", "Anthropic MCP Specification v0.1", "Claude Agent SDK", "Claude Agent SDK Documentation", "Context Injection", "HTTP Protocol", "JSON-RPC 2.0", "JSON-RPC 2.0 Specification", "JSON-RPC Protocol", "JSON Schema", "JSON Schema Validator", "LangChain Agent Framework", "Large Language Model", "MCP Client", "MCP Prompt", "MCP Resource" ] }, { "id": "model-control-protocols-like-mcp", "title": "Model Control Protocols like MCP", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model Control Protocols like MCP refers to the family of open, standardised communication protocols that govern how large language model (LLM) inference hosts communicate with external services, tools, data sources, and agents at runtime, enabling composable agentic ecosystems without per-integra...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:model-control-protocols-like-mcp", "labels": [ "Model Control Protocols like MCP", "MCP Protocol" ], "is_subclass_of": [ "AI Infrastructure", "Agent Frameworks", "Agentic Internet", "API Gateway", "CLI Multi-Agent Systems", "Agents" ], "wikilinks": [ "AgentIntegrationDomain", "AgentOrchestrationLayer", "Anthropic", "AutoGen", "Capability Advertisement", "Claude Desktop", "Cursor IDE", "Database Connectivity", "DistributedSystemsDomain", "Dynamic Tool Discovery", "Filesystem Access", "GitHub Integration", "HTTP", "HTTP Streaming", "JSON-RPC", "JSON-RPC 2.0 Specification", "JSON Schema", "LangChain Tools", "LLM Tool Use", "MCP Client" ] }, { "id": "model-debugging", "title": "Model Debugging", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model debugging is the systematic process of diagnosing, isolating and correcting faults in the behaviour of a machine-learning model, such as poor accuracy, biased predictions, brittleness to distribution shift, or unexpected outputs on specific inputs. Unlike conventional software debugging, it must reason about statistical behaviour, training data, feature representations and learned parameters rather than deterministic control flow. Practitioners combine error analysis, interpretability tooling, slice-based evaluation and counterfactual probing to trace failures back to data, model architecture or the training procedure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-debugging", "labels": [ "Model Debugging" ], "is_subclass_of": [ "Model Evaluation", "AI Infrastructure" ], "wikilinks": [] }, { "id": "model-deployment", "title": "Model Deployment", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model Deployment is the engineering discipline of transitioning a trained machine learning model from a development or research environment into a production system where it can serve real-time or batch predictions to users and downstream applications. It encompasses model packaging, serving infrastructure, API exposure, versioning, scaling, and monitoring, ensuring that the model behaves reliably and efficiently under operational conditions. Deployment strategies range from synchronous online endpoints for low-latency inference to batch scoring pipelines, edge device embedding, and serverless function invocations. The discipline is tightly coupled with MLOps practices that treat models as first-class software artefacts subject to continuous integration, delivery, observation, and rollback.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:model-deployment", "labels": [ "Model Deployment", "Large Language Model Deployment", "Multi-Task Model Deployment" ], "is_subclass_of": [ "MLOps", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [ "Inference Engine", "Edge Inference", "MLOps" ] }, { "id": "model-deprecation", "title": "Model Deprecation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The formal process of retiring or phasing out a specific AI model version, often triggering user migration or backlash.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-deprecation", "labels": [ "Model Deprecation" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "model-depth", "title": "Model Depth", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The count of stacked transformer layers (encoder, decoder, or both) in a neural network, governing the number of sequential representation transformations. Greater depth enables more abstract hierarchical feature learning but increases training difficulty, requiring residual connections and layer normalisation to stabilise gradient flow.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-depth", "labels": [ "Model Depth" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Bitcoin Technical Overview", "BTC Layer 3", "California AI bill", "Lightning and Similar L2" ] }, { "id": "model-distillation-artifacts", "title": "Model Distillation Artifacts", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Unintended behavioral quirks or biases that emerge in AI models as a result of being trained on or derived from the outputs of other models, potentially compounding alignment issues.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-distillation-artifacts", "labels": [ "Model Distillation Artifacts" ], "is_subclass_of": [ "AI Safety Research" ], "wikilinks": [] }, { "id": "model-distillation", "title": "Model Distillation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A knowledge transfer technique where a smaller student model is trained to mimic the output distributions or behavior of a larger teacher model to achieve comparable performance with reduced computational cost.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-distillation", "labels": [ "Model Distillation" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "model-documentation", "title": "Model Documentation", "domain": "ai", "domain_name": "Ai", "definition": "Model documentation is the structured recording of a machine learning model's intended use, training data, performance, limitations, and ethical considerations, typically through artefacts such as model cards and datasheets. It supports transparency, accountability, and informed deployment decisions for downstream users and regulators. It is increasingly required by AI governance frameworks and procurement standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:model-documentation", "labels": [ "Model Documentation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "model-ensembling", "title": "Model Ensembling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A machine learning technique that aggregates predictions from multiple independently trained base models \u2014 via voting, averaging, stacking, or boosting \u2014 to reduce variance, improve generalisation, and produce more reliable outputs than any single constituent model. Diversity among base learners is the key driver of ensemble benefit.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-ensembling", "labels": [ "Model Ensembling" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Stable Diffusion" ] }, { "id": "model-evaluation-results", "title": "Model Evaluation Results", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Structured outputs produced during the assessment of a machine learning model's predictive performance, encompassing quantitative metrics such as accuracy, precision, recall, F1 score, and AUC alongside qualitative analyses. These results form the evidentiary basis for model selection, regulatory compliance disclosures, and AI model card documentation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:model-evaluation-results", "labels": [ "Model Evaluation Results" ], "is_subclass_of": [ "AI Technique", "ModelArchitecture" ], "wikilinks": [ "ModelArchitecture" ] }, { "id": "model-evaluation", "title": "Model Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "Model Evaluation is the systematic process of measuring the performance, reliability, safety, and fitness-for-purpose of machine learning models against defined metrics, held-out datasets, and behavioural benchmarks. It spans quantitative metric computation (accuracy, F1, perplexity, BLEU, AUC-ROC), qualitative red-teaming and adversarial probing, and comparative benchmarking across standardised test suites. Evaluation drives deployment decisions, informs architectural iteration, and increasingly underpins regulatory conformity assessments demanded by AI governance frameworks. Both static offline evaluation and dynamic online evaluation in live production environments are within scope.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:model-evaluation", "labels": [ "Model Evaluation", "AI Model Evaluation", "Cityscapes Evaluation", "Disaggregated Evaluation", "Evaluation", "Fidelity Evaluation", "Machine Learning Model Evaluation", "System Evaluation" ], "is_subclass_of": [ "Machine Learning", "AI Technique" ], "wikilinks": [] }, { "id": "model-extraction", "title": "Model Extraction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An attack where adversaries reconstruct a functionally equivalent or similar machine learning model by systematically querying a target model and training a substitute model on the collected input-output pairs, enabling theft of intellectual property, privacy violations, and subsequent attacks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:model-extraction", "labels": [ "Model Extraction" ], "is_subclass_of": [ "AI Technique", "Adversarial Attacks" ], "wikilinks": [ "Adversarial Attacks", "ArtificialIntelligenceDomain", "Membership Inference", "Model Inversion" ] }, { "id": "model-fine-tuning", "title": "Model Fine-Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of further training a pre-trained machine learning model on a specific dataset to adapt its general capabilities to a particular task or domain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-fine-tuning", "labels": [ "Model Fine-Tuning" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "model-generalisation", "title": "Model Generalisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model generalisation is the ability of a trained machine learning model to perform accurately on new, previously unseen data drawn from the same underlying distribution as its training set, rather than merely memorising the training examples. It is the central goal of the training process, balanced against overfitting, and is commonly assessed through held-out validation and test performance. Techniques such as data augmentation and hyperparameter tuning are used specifically to improve generalisation.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:model-generalisation", "labels": [ "Model Generalisation" ], "is_subclass_of": [ "Generalisation" ], "wikilinks": [] }, { "id": "model-governance", "title": "Model Governance", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model governance is the framework of policies, roles, controls, and documentation that an organisation applies across the lifecycle of its machine learning and statistical models to manage their risk, ensure their fitness for purpose, and satisfy regulatory and ethical obligations. It establishes accountability for model development, validation, approval, deployment, and ongoing monitoring, supported by artefacts such as model cards, audit trails, and validation reports. By imposing oversight, documentation, and review gates, it ensures models remain accurate, fair, explainable, and compliant rather than opaque or unaccountable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-governance", "labels": [ "Model Governance" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "model-inference", "title": "Model Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model inference is the operational phase of a machine learning system in which a trained model is applied to new, previously unseen inputs to produce predictions, classifications, or generated outputs. Unlike training, inference involves only a forward pass through the model and is optimised for low latency, high throughput, and efficient resource use. It is the stage at which a model delivers value in production, serving requests in real time, in batches, or at the edge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-inference", "labels": [ "Model Inference" ], "is_subclass_of": [ "Machine Learning" ], "wikilinks": [] }, { "id": "model-interpretability", "title": "Model Interpretability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The degree to which a human can understand the cause-effect relationships within a machine learning model's decision-making process, encompassing both the model's internal mechanisms and the reasoning behind specific predictions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:model-interpretability", "labels": [ "Model Interpretability" ], "is_subclass_of": [ "Explainable AI" ], "wikilinks": [ "LIME", "SHAP", "ArtificialIntelligenceDomain", "Explainable AI", "Feature Importance", "Global Explanation", "Interpretable AI", "Intrinsic Interpretability", "Local Explanation", "Machine Learning", "Model Transparency" ] }, { "id": "model-inversion", "title": "Model Inversion", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An inference-time privacy attack that reconstructs sensitive training data or personal attributes from a trained model by iteratively querying model outputs, confidence scores, or internal representations. Model inversion attacks expose a fundamental tension between model utility and data privacy, motivating defences such as differential privacy, federated learning, and membership-inference auditing.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:model-inversion", "labels": [ "Model Inversion", "Model Inversion Attack" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "RGB", "agents", "ArtificialIntelligenceDomain", "Bitcoin", "Death of the Internet", "Nostr protocol" ] }, { "id": "model-layer", "title": "Model Layer", "domain": "ai", "domain_name": "Ai", "definition": "The Model Layer is the stratum that holds trained machine learning models as deployable artefacts with fixed parameters. In the canonical stack it sits above the Algorithm Layer and below the Inference Layer, packaging learned functions for use. It contains weight sets, model metadata, and the serialised representations that inference engines load.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:model-layer", "labels": [ "Model Layer" ], "is_subclass_of": [ "AI Model Architecture", "owl:Thing" ], "wikilinks": [ "Algorithm Layer", "Training Layer", "Inference Layer", "Machine Learning", "Model Registry", "owl:Thing" ] }, { "id": "model-monitoring", "title": "Model Monitoring", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model monitoring is the continuous observation of a deployed machine learning model's inputs, outputs, and performance in production to detect degradation, data drift, concept drift, and operational issues. It tracks predictive quality, latency, and input distributions against baselines and triggers alerts or retraining when thresholds are breached. As a core MLOps practice, it closes the loop between deployment and maintenance, sustaining model reliability over time.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-monitoring", "labels": [ "Model Monitoring" ], "is_subclass_of": [ "MLOps", "Data Drift" ], "wikilinks": [] }, { "id": "model-optimisation-and-performance", "title": "Model Optimisation and Performance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model Optimisation and Performance is the body of post-training techniques, inference-time systems, and hardware-aware transformations that reduce the computational cost, memory footprint, and latency of neural networks\u2014particularly large language models (LLMs)\u2014without unacceptable degradation of...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:model-optimisation-and-performance", "labels": [ "Model Optimisation and Performance", "Model Optimisation", "Model Optimisation Pipeline" ], "is_subclass_of": [ "MLOps", "AI Infrastructure (Artificial Intelligence)", "Machine Learning Discipline", "Systems Engineering", "Computational Efficiency", "Neural Network Compression" ], "wikilinks": [ "AlgorithmLayer", "Attention Tiling", "Autonomous Systems", "BFLOAT16", "Calibration Dataset", "Compiler Theory", "Computational Efficiency", "Computer Architecture", "Continuous Batching", "Cost-Effective LLM Serving", "CUDA Programming Model", "CUDA Runtime", "Dense Inference", "Eager Execution", "Edge Deployment", "Efficient Transformers", "FlashAttention", "Floating Point Arithmetic", "Full-Precision Training", "GGUF Format" ] }, { "id": "model-optimization", "title": "Model Optimization", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model optimisation is the set of techniques that reduce the size or computational cost of a trained model while preserving accuracy. It includes quantisation, pruning and distillation to make deployment more efficient.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:model-optimization", "labels": [ "Model Optimization" ], "is_subclass_of": [ "Model Deployment", "AI Technique" ], "wikilinks": [ "Knowledge Distillation", "Model Serving", "Inference Serving", "AI Model Inference Engine", "Latency", "Model Deployment" ] }, { "id": "model-parallelism", "title": "Model Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model parallelism is a distributed training strategy that splits a single neural network's parameters and computation across multiple accelerators when the model is too large to fit in one device's memory. Variants include tensor parallelism (splitting within layers) and pipeline parallelism (splitting across layers into stages). It is essential for training large language models and is often combined with data parallelism.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:model-parallelism", "labels": [ "Model Parallelism" ], "is_subclass_of": [ "Distributed Training" ], "wikilinks": [] }, { "id": "model-parameters", "title": "Model Parameters", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The learnable internal variables \u2014 weights and biases \u2014 of a neural network that are adjusted during training to minimise a loss function. Parameter count determines model capacity; foundation models commonly operate with billions to trillions of parameters, making parameter-efficient fine-tuning and management a central concern in contemporary AI development.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-parameters", "labels": [ "Model Parameters", "Learnable Parameters" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "automation", "research", "artificial intelligence", "ArtificialIntelligenceDomain", "machine learning", "natural language processing", "training" ] }, { "id": "model-performance", "title": "Model Performance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The quantitative and qualitative measure of how effectively an artificial intelligence model accomplishes its designated tasks, typically assessed through statistical metrics evaluating prediction accuracy, reliability, generalisability, computational efficiency, and robustness, considered across different data distributions, operational conditions, and stakeholder requirements, serving as a critical basis for model selection, deployment decisions, ongoing monitoring, and continuous improvement throughout the AI lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:model-performance", "labels": [ "Model Performance" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Bitnet and the rise of the 1bit model", "Data Drift", "Model Evaluation", "open source", "user experience", "Accuracy", "AI Monitoring", "artificial intelligence", "ArtificialIntelligenceDomain", "ComfyUI", "Confusion Matrix", "F1 Score", "Fairness", "Flux.1", "Google", "Hardware and Edge", "Medical AI", "Precision", "Recall", "ROC Curve" ] }, { "id": "model-predictive-control", "title": "model predictive control", "domain": "robotics", "domain_name": "Robotics", "definition": "Model Predictive Control (MPC) is an advanced control strategy in which an explicit mathematical model of the controlled process is used to predict future system outputs over a receding finite-horizon window, and a constrained optimisation problem is solved at each control step to compute the optimal input sequence. Only the first element of the computed sequence is applied, after which the horizon shifts forward and the optimisation repeats with an updated state estimate, creating an implicit closed-loop feedback mechanism. MPC handles multi-input multi-output (MIMO) systems, hard inequality constraints on both states and inputs, and competing cost objectives within a single unified formulation, making it applicable across process control, autonomous vehicles, robotics, and energy management. Its ability to anticipate constraint violations before they occur distinguishes it fundamentally from classical reactive controllers such as PID.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:model-predictive-control", "labels": [ "Model Predictive Control" ], "is_subclass_of": [ "Optimal Control" ], "wikilinks": [] }, { "id": "model-property", "title": "Model Property", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A meta-classification for properties, characteristics, and measurable attributes of machine learning models including performance metrics (accuracy, latency, throughput), architectural properties (parameters, layers, context length), and operational characteristics (memory footprint, inference co...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:model-property", "labels": [ "Model Property" ], "is_subclass_of": [ "AI Technique", "AI Concept" ], "wikilinks": [ "Benchmark", "Model Comparison", "AI Concept", "Artificial Intelligence", "Machine Learning Model" ] }, { "id": "model-pruning-for-edge-deployment", "title": "Model Pruning for Edge Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model Pruning for Edge Deployment systematically removes redundant weights and neurons from neural networks, reducing model size and computational requirements while maintaining sufficient accuracy for edge inference.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model-pruning-for-edge-deployment", "labels": [ "Model Pruning for Edge Deployment" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "model-pruning", "title": "Model Pruning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model pruning is a neural-network compression technique that removes redundant or low-importance parameters, neurons, or structural components from a trained model to reduce its size and computational cost while preserving accuracy. Pruning may be unstructured, zeroing individual weights to induce sparsity, or structured, removing whole channels, filters, or layers to yield models that run faster on standard hardware. Importance is judged by criteria such as weight magnitude or sensitivity, and pruning is typically followed by fine-tuning to recover lost accuracy. It is a core method for deploying large models on resource-constrained and edge environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-pruning", "labels": [ "Model Pruning" ], "is_subclass_of": [ "Model Compression", "AI Technique" ], "wikilinks": [] }, { "id": "model-quantization", "title": "Model Quantization", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model quantization is a model-compression technique that reduces the numerical precision of a neural network's weights and activations, typically converting 32-bit floating-point values to lower-precision integer or float formats such as INT8, INT4 or FP8. By shrinking the memory footprint and exploiting cheaper integer arithmetic, quantization lowers latency, energy use and storage cost, usually at a small and controllable loss in accuracy. It is applied either after training (post-training quantization) or during training (quantization-aware training) to deploy large models on constrained inference hardware.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-quantization", "labels": [ "Model Quantization", "Model Quantisation" ], "is_subclass_of": [ "Model Compression", "AI Technique" ], "wikilinks": [] }, { "id": "model-registry", "title": "Model Registry", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A model registry is a system for cataloguing trained machine learning models with their versions, metadata and lifecycle stage. It provides a central record that links a model to its training data, metrics and deployment status, supporting governance, reproducibility and controlled promotion across staging environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:model-registry", "labels": [ "Model Registry" ], "is_subclass_of": [ "MLOps", "Machine Learning Operations" ], "wikilinks": [ "Data Versioning", "Model Deployment", "Model Serving", "MLOps", "Feature Store" ] }, { "id": "model-retraining", "title": "Model Retraining", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The post-deployment lifecycle activity of training a machine learning model again on refreshed or expanded data after production monitoring detects data drift, concept drift, or performance degradation, restoring predictive accuracy without redesigning the model. Retraining closes the MLOps feedback loop between monitoring and deployment, and may be scheduled periodically or triggered automatically by drift-detection thresholds.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model-retraining", "labels": [ "Model Retraining" ], "is_subclass_of": [ "Machine Learning Operations" ], "wikilinks": [ "Machine Learning Operations", "Model Training", "AI Monitoring", "Continuous Training" ] }, { "id": "model-robustness", "title": "Model Robustness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The capacity of a machine learning model to maintain consistent and reliable performance when exposed to variations in input data, including noise, distributional shifts, and edge cases, without catastrophic degradation in accuracy or functionality.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-robustness", "labels": [ "Model Robustness" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "model-router", "title": "Model Router", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An inference-time orchestration component that dynamically selects and directs tasks to the most appropriate AI model based on complexity, cost, or performance requirements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-router", "labels": [ "Model Router" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "model-routing-architecture", "title": "Model Routing Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A system design pattern that dynamically directs inference requests to different AI models based on task complexity, cost, or performance requirements to optimize overall efficiency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-routing-architecture", "labels": [ "Model Routing Architecture" ], "is_subclass_of": [ "Model Architecture" ], "wikilinks": [] }, { "id": "model-scaling", "title": "Model Scaling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model scaling is the practice of systematically increasing the size of a machine learning model, the volume of training data, and the compute budget to improve performance in line with empirical scaling laws. Predictable power-law relationships link these resources to reductions in loss, guiding decisions about parameter counts and data allocation. Scaling has been a primary driver behind the capabilities of modern foundation models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-scaling", "labels": [ "Model Scaling" ], "is_subclass_of": [ "Scaling Laws" ], "wikilinks": [] }, { "id": "model-selection-strategy", "title": "Model Selection Strategy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The systematic process of evaluating and choosing the most appropriate AI model for a specific task based on performance, cost, and latency requirements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:model-selection-strategy", "labels": [ "Model Selection Strategy" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "model-selection", "title": "Model Selection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model selection is the process of choosing the most appropriate machine learning model, algorithm family, or configuration for a given task from a set of candidates. It balances predictive performance against constraints such as interpretability, inference cost, and generalisation, typically using validation data and metrics rather than the training error. Techniques include cross-validation, information criteria, and held-out benchmarking, with the goal of selecting the model expected to perform best on unseen data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-selection", "labels": [ "Model Selection" ], "is_subclass_of": [ "Model Evaluation", "Performance Metrics" ], "wikilinks": [] }, { "id": "model-serialisation", "title": "Model Serialisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model serialisation is the process of converting a trained machine learning model \u2014 including its architecture definition, learned weights, and associated metadata \u2014 into a persistent, portable file format that can be stored, transferred, and subsequently loaded to restore the model to an operational state. Serialisation enables reproducibility, deployment across different environments, and sharing of pre-trained models through repositories. Common formats include ONNX for cross-framework portability, SafeTensors for security, PyTorch checkpoint files, and framework-native formats such as TensorFlow SavedModel.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-serialisation", "labels": [ "Model Serialisation" ], "is_subclass_of": [ "Artificial Intelligence", "AI Infrastructure" ], "wikilinks": [] }, { "id": "model-serving", "title": "Model Serving", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Model serving is the practice of making a trained machine learning model available to handle inference requests, typically behind an API. It covers loading models, batching and routing requests, and returning predictions under latency and throughput constraints at production scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:model-serving", "labels": [ "Model Serving", "Large Language Model Serving", "Model Serving Infrastructure", "Model Serving Platform" ], "is_subclass_of": [ "Model Deployment", "Inference Layer" ], "wikilinks": [ "Model Deployment", "REST API", "Inference Serving", "MLOps", "Latency" ] }, { "id": "model-tier", "title": "Model Tier", "domain": "ai", "domain_name": "Ai", "definition": "A model tier is a named capability-and-cost band into which the language models available to an agent system are grouped, so that work can be routed to the cheapest model that is competent for it. A typical scheme distinguishes a small fast tier for classification and routing, a mid tier for routine generation, and a frontier tier for hard reasoning, each with characteristic latency, price per token, and quality envelope. Tiering turns the continuous space of model choices into a discrete policy surface that orchestration layers can reason about, enabling deliberate trade-offs between spend, speed, and answer quality across a workload.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:model-tier", "labels": [ "Model Tier" ], "is_subclass_of": [ "Model Selection", "ModelSelection" ], "wikilinks": [ "ModelSelection", "CostOptimisation", "CostEfficientInference", "LLMOrchestration" ] }, { "id": "model-training-pipeline", "title": "Model Training Pipeline", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A model training pipeline is the orchestrated sequence of stages that transforms raw data into a trained, validated machine-learning model ready for deployment. It typically chains data ingestion and preprocessing, feature engineering, model fitting, hyperparameter tuning and evaluation into a reproducible, automatable workflow. As a backbone of MLOps, the pipeline enforces consistency, versioning and repeatability across training runs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-training-pipeline", "labels": [ "Model Training Pipeline" ], "is_subclass_of": [ "Machine Learning", "Model Training" ], "wikilinks": [] }, { "id": "model-training", "title": "Model Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model Training is the end-to-end computational process by which a neural network's parameters are iteratively adjusted through gradient-based optimisation over large corpora so that the network acquires generalisable representations, task-specific capabilities, and aligned behavioural policies su...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:model-training", "labels": [ "Model Training", "AI Model Training", "Decentralised Foundation Model Training", "Large Model Training" ], "is_subclass_of": [ "AI Technique", "Neural Network", "Optimisation", "Machine Learning Discipline", "Deep Learning", "Large-Scale Pretrained Foundation Model", "Neural Networks" ], "wikilinks": [ "AdamW", "AISI 2024 Approach to Evaluations", "AISI Frontier AI Safety Framework", "AlgorithmLayer", "Alignment", "AlignmentLayer", "Bai et al. 2022 Anthropic RLHF Helpful Harmless", "Bai et al. 2022 Constitutional AI RLAIF", "Bommasani et al. 2021 Foundation Models Opportunities and Risks", "Brown et al. 2020 GPT-3 Few-Shot Learners", "Causal Language Modelling", "Chinchilla Scaling Laws", "Dao 2023 FlashAttention-2", "Dao et al. 2022 FlashAttention IO-Aware", "Data Curation", "DeepSeek-AI 2024 DeepSeek-V3 Technical Report", "DeepSpeed", "Dettmers et al. 2024 QLoRA", "DistributedSystemsDomain", "Emergent Capabilities" ] }, { "id": "model-transparency", "title": "Model Transparency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The capability to understand, explain, and interpret the decision-making processes within AI and machine learning models, encompassing explainability, interpretability, and accountability requirements essential for responsible AI deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:model-transparency", "labels": [ "Model Transparency" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "Trustworthy AI Systems", "AI Ethics", "metaverse" ] }, { "id": "model-validation", "title": "Model Validation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Model validation is the process of assessing whether a trained model meets its intended requirements for accuracy, robustness, fairness and generalisation before it is trusted in practice. It uses held-out data, cross-validation and stress tests to estimate performance on unseen inputs and to detect overfitting, bias or specification gaps. Distinct from evaluation metrics alone, validation judges fitness for purpose within the model lifecycle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:model-validation", "labels": [ "Model Validation" ], "is_subclass_of": [ "AI Lifecycle" ], "wikilinks": [] }, { "id": "model-versioning", "title": "Model Versioning", "domain": "ai", "domain_name": "Ai", "definition": "Model versioning is the practice of tracking and managing successive versions of machine learning models, recording the code, data, hyperparameters, and checkpoints that produced each one. It enables reproducibility, rollback, lineage tracing, and controlled promotion of models through development and production stages. Model registries and artefact stores implement versioning as a core MLOps capability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:model-versioning", "labels": [ "Model Versioning" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "model-weights", "title": "Model Weights", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The learnable numerical parameters in a neural network that encode the connection strengths adjusted during training via backpropagation to minimise loss. Model weights constitute the primary artefact of training and are the target of fine-tuning, quantisation, pruning, and transfer learning; their distribution and magnitude critically determine model capability and safety.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:model-weights", "labels": [ "Model Weights" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "model-width", "title": "Model Width", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The dimensionality of internal representations at each transformer layer, commonly denoted d_model or hidden dimension. Width sets the information-carrying capacity per token and scales the size of attention heads and feed-forward projections, making it a primary axis alongside depth and data volume in neural scaling law research.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:model-width", "labels": [ "Model Width" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "presentation", "3D and 4D", "ArtificialIntelligenceDomain", "Face Swap", "Fashion", "NVIDIA Omniverse", "PEOPLE", "Proprietary Large Language Models", "Segmentation and Identification", "Stable Diffusion", "Transformers" ] }, { "id": "model-based-systems-engineering", "title": "Model-Based Systems Engineering", "domain": "standards", "domain_name": "Standards", "definition": "A systems engineering methodology that replaces document-centric specification with a coherent, machine-interpretable system model as the authoritative source of requirements, architecture, behaviour, and verification information across the whole lifecycle. Formalised by INCOSE and realised through languages such as SysML, it enables early analysis, simulation, and consistency checking of complex systems, keeping every discipline working against a single shared model rather than divergent documents.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model-based-systems-engineering", "labels": [ "Model-Based Systems Engineering", "Model Based Systems Engineering" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [ "Systems Engineering", "Simulation Software", "Digital Twin", "Constraint Based Design" ] }, { "id": "model", "title": "Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In artificial intelligence, a Model is a mathematical or computational structure learned from data that encodes a mapping from inputs to outputs (or latent representations). Models range from classical machine learning models (linear, tree-based) to deep neural networks, generative models, and large language models, and constitute the core artefact produced by training and consumed during inference.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:model", "labels": [ "Model" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "model-ops", "title": "ModelOps", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ModelOps is the discipline of industrialising the end-to-end governance and operational lifecycle of AI and machine learning models, encompassing the processes, tooling, and organisational practices required to reliably build, validate, deploy, monitor, and retire models in production environments. It extends MLOps beyond technical pipeline automation to include regulatory compliance, risk management, model lineage tracking, and cross-functional accountability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:model-ops", "labels": [ "ModelOps", "MLOps", "Model Ops" ], "is_subclass_of": [ "Gartner Prediction" ], "wikilinks": [ "Artificial Intelligence", "Gartner Prediction" ] }, { "id": "moderation-tools", "title": "Moderation Tools", "domain": "governance", "domain_name": "Governance", "definition": "Moderation tools are software systems that detect, review, and act on user-generated content or behaviour that violates platform policies, combining automated classifiers, reporting queues, and human-review workflows. They enforce community standards by flagging, filtering, age-gating, or removing content and sanctioning accounts. In immersive and social platforms they increasingly cover real-time voice, spatial, and behavioural moderation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:moderation-tools", "labels": [ "Moderation Tools" ], "is_subclass_of": [ "Metaverse governance and safeguarding" ], "wikilinks": [] }, { "id": "modular-arithmetic", "title": "Modular Arithmetic", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Modular arithmetic performs integer arithmetic that wraps around a fixed modulus, so numbers are treated as equivalent if they differ by a multiple of that modulus.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:modular-arithmetic", "labels": [ "Modular Arithmetic" ], "is_subclass_of": [ "Cryptography", "Cryptography Domain" ], "wikilinks": [ "Number Theory", "Cryptographic Hash Function", "Zero-Knowledge Proof", "Cryptography", "Cryptography Domain" ] }, { "id": "modular-blockchain", "title": "Modular Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A modular blockchain is a blockchain architecture that separates the core functions of execution, settlement, consensus, and data availability into distinct, specialised layers rather than handling them all in a single monolithic chain. Each layer can be optimised and scaled independently, and components can be mixed and matched across networks. This decomposition aims to overcome the scalability constraints of monolithic designs while preserving security and decentralisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:modular-blockchain", "labels": [ "Modular Blockchain" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "module", "title": "Module", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Module is a self-contained, reusable unit of software or system design that encapsulates a discrete set of related functions, data structures, or responsibilities behind a well-defined interface, enabling independent development, testing, deployment, and composition within larger systems. Modules promote separation of concerns, reduce coupling between system components, and support both horizontal reuse across different systems and vertical composition within a single system architecture. In AI contexts, a module may refer to a functional unit within a model architecture (e.g., an attention module, an adapter module) or a deployable component in an AI pipeline.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:module", "labels": [ "Module" ], "is_subclass_of": [ "Component" ], "wikilinks": [ "Artificial Intelligence", "Component" ] }, { "id": "molecular-cloud", "title": "Molecular Cloud", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:molecular-cloud", "labels": [ "Molecular Cloud" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "molecular-dynamics", "title": "Molecular Dynamics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Molecular dynamics is a computational simulation method that models the physical movement of atoms and molecules by numerically integrating Newton's equations of motion under an interatomic force field. It produces time-resolved trajectories from which thermodynamic and kinetic properties are derived. The method bridges physics-based simulation and machine learning in computational biology and chemistry.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:molecular-dynamics", "labels": [ "Molecular Dynamics" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "momentum-contrast", "title": "Momentum Contrast", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A self-supervised contrastive learning framework (MoCo) that maintains a dynamically updated queue of encoded negative samples and a momentum encoder \u2014 updated as an exponential moving average of the query encoder \u2014 to provide large, consistent dictionaries for contrastive loss computation without requiring large batch sizes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:momentum-contrast", "labels": [ "Momentum Contrast" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "momentum-management", "title": "Momentum Management", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:momentum-management", "labels": [ "Momentum Management" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "momentum-wheel", "title": "Momentum Wheel", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:momentum-wheel", "labels": [ "Momentum Wheel" ], "is_subclass_of": [ "Actuator" ], "wikilinks": [] }, { "id": "momentum", "title": "Momentum", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Momentum is an optimisation technique that accelerates gradient descent by accumulating an exponentially weighted moving average of past gradients and using it to update parameters. By carrying velocity from prior steps, it damps oscillations across steep directions and speeds progress along consistent ones, improving convergence on ill-conditioned loss surfaces. Momentum underlies many modern optimisers, with Nesterov's variant and adaptive methods such as Adam building directly on its formulation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:momentum", "labels": [ "Momentum" ], "is_subclass_of": [ "Optimization Algorithm" ], "wikilinks": [] }, { "id": "monero", "title": "Monero", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Monero (XMR) is an open-source, privacy-preserving cryptocurrency launched in April 2014 as a fork of the Bytecoin codebase, distinguished by making confidentiality mandatory for every transaction rather than optional. It achieves unlinkability and untraceability through three complementary cryptographic primitives: ring signatures that blend the spender's output among decoys drawn from the blockchain history, one-time stealth addresses that prevent linking payments to a recipient's public key, and Ring Confidential Transactions (RingCT) that hide transferred amounts using Pedersen commitments. The protocol employs a proof-of-work algorithm \u2014 originally CryptoNight, later RandomX \u2014 engineered to maintain parity between CPUs and GPUs while resisting ASIC specialisation, thereby preserving decentralised mining participation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:monero", "labels": [ "Monero" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Proof of Work", "Ring Signature", "Zcash", "Blockchain Domain" ] }, { "id": "monetary-policy-implementation", "title": "Monetary Policy Implementation", "domain": "finance", "domain_name": "Finance", "definition": "Monetary Policy Implementation encompasses the operational processes, tools, and institutional mechanisms through which central banks and monetary authorities translate policy decisions\u2014such as interest rate targets or quantitative measures\u2014into observable effects on the financial system. In digital and metaverse economic contexts it extends to the programmatic enforcement of monetary parameters via Central Bank Digital Currencies (CBDCs), smart contracts, and programmable money frameworks. Effective implementation requires coordination between inflation control targets, payment system design, financial stability oversight, and regulatory compliance.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:monetary-policy-implementation", "labels": [ "Monetary Policy Implementation" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [ "Economic Mechanism" ] }, { "id": "monetary-policy-transmission", "title": "Monetary Policy Transmission", "domain": "finance", "domain_name": "Finance", "definition": "Monetary policy transmission is the process by which changes in a central bank's policy instruments\u2014primarily short-term interest rates or reserve quantities\u2014propagate through financial markets and the broader economy to affect output, employment, and the price level. The transmission operates through multiple channels including the interest rate channel (altering borrowing costs), the credit channel (influencing bank lending and balance-sheet constraints), the exchange rate channel (shifting the relative price of tradeable goods), the asset price channel (changing wealth and Tobin's q), and expectations channels (anchoring or re-anchoring inflation expectations). The strength, speed, and reliability of each channel depends on the institutional structure of the financial system, the degree of market completeness, and the prevailing monetary regime.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:monetary-policy-transmission", "labels": [ "Monetary Policy Transmission" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "monetary-policy", "title": "Monetary Policy", "domain": "finance", "domain_name": "Finance", "definition": "Monetary policy is the set of decisions made by a monetary authority \u2014 traditionally a central bank \u2014 to manage the supply of money, credit conditions, and interest rates in order to achieve macroeconomic objectives such as price stability, full employment, and sustainable economic growth. In traditional finance it encompasses tools such as open market operations, reserve requirements, discount-rate setting, and unconventional measures including quantitative easing and forward guidance. In decentralised systems and blockchain protocols, monetary policy is encoded algorithmically into protocol rules that govern token emission schedules, inflation rates, halving events, and burn mechanisms, removing discretionary human intervention and replacing it with transparent, auditable on-chain governance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:monetary-policy", "labels": [ "Monetary Policy", "Traditional Monetary Policy" ], "is_subclass_of": [ "Economic Governance" ], "wikilinks": [] }, { "id": "monetary-sovereignty", "title": "Monetary Sovereignty", "domain": "finance", "domain_name": "Finance", "definition": "A nation's authority to control its monetary policy, currency issuance, and financial systems, increasingly challenged by decentralised digital currencies and defended through initiatives like Central Bank Digital Currencies (CBDCs) that assert state control over the evolving digital economy.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:monetary-sovereignty", "labels": [ "Monetary Sovereignty" ], "is_subclass_of": [ "Economic Governance" ], "wikilinks": [ "National Financial Control", "Economic Governance", "metaverse" ] }, { "id": "monetary-system", "title": "Monetary System", "domain": "finance", "domain_name": "Finance", "definition": "The comprehensive framework of institutions, regulations, and mechanisms governing money creation, distribution, and management within an economy, now evolving to incorporate digital currencies, CBDCs, and blockchain-based financial systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:monetary-system", "labels": [ "Monetary System", "Eurozone Monetary System" ], "is_subclass_of": [ "Financial System" ], "wikilinks": [ "Economic Exchange", "Financial System", "metaverse" ] }, { "id": "monetary-theory", "title": "Monetary Theory", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Monetary theory studies the nature of money, how its supply is created and managed, and how monetary conditions affect prices, output and financial stability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:monetary-theory", "labels": [ "Monetary Theory" ], "is_subclass_of": [ "Economics", "Economics Domain" ], "wikilinks": [ "Economics", "Central Bank Digital Currency", "Fiat Currency", "Inflation", "Economics Domain" ] }, { "id": "money", "title": "Money", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A medium of exchange, unit of account, and store of value accepted within an economic system, encompassing physical currency, digital money, cryptocurrencies, and virtual currency forms that facilitate economic activity.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:money", "labels": [ "Money", "Electronic Money", "Mobile Money" ], "is_subclass_of": [ "Token and Asset", "Cryptocurrency" ], "wikilinks": [ "Acceptance Network", "Alden2023", "Arbitrum", "Bank of England", "Banking System", "barsky1987fisher", "bertomeu2023uncle", "BIS", "borio2017fx", "BTC/USD", "caballero2008financial; @spiro2019hidden", "cantillon1756essai; @bordo1983some", "carney2019growing; @piffaretti2009reshaping", "CBDC", "CBDC Tracker", "Chinese Yuan", "Cross-Border Payments", "davies2010history", "DeFi", "Debt" ] }, { "id": "monitoring-dashboard", "title": "Monitoring Dashboard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Visual interfaces that aggregate and display real-time metrics, logs, and traces from systems and applications, enabling observability, performance tracking, incident detection, and data-driven decision making through unified visualisation platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:monitoring-dashboard", "labels": [ "Monitoring Dashboard", "Dashboard", "Executive Dashboard" ], "is_subclass_of": [ "Data Visualisation" ], "wikilinks": [ "Operational Intelligence", "Data Visualisation", "metaverse" ] }, { "id": "monitoring-infrastructure", "title": "Monitoring Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Monitoring infrastructure is the collection of systems that gather, store, and analyse metrics, logs, and traces to observe the health, performance, and behaviour of software and physical systems. It underpins alerting, capacity planning, incident response, and, for AI systems, tracking of drift, cost, and environmental impact. Components typically include collectors, time-series databases, dashboards, and alerting engines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:monitoring-infrastructure", "labels": [ "Monitoring Infrastructure", "Energy Monitoring Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "monitoring-system", "title": "Monitoring System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A monitoring system is an integrated observability platform that continuously collects, aggregates, and evaluates telemetry signals \u2014 metrics, logs, traces, and events \u2014 from target environments spanning software services, physical infrastructure, AI models, and distributed systems, in order to detect anomalies, assess operational health, and trigger alerts or automated remediation responses. It encompasses data-collection agents, instrumentation SDKs, time-series storage engines, query and alerting pipelines, and visualisation dashboards that together form a closed-loop feedback mechanism for operational reliability. Monitoring systems implement the three pillars of observability (metrics, logs, traces) and serve as the foundational layer for site reliability engineering, incident management, and compliance auditing in modern distributed architectures.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:monitoring-system", "labels": [ "Monitoring System", "Energy Monitoring System", "User Health Monitoring System" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "monitoring", "title": "Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Monitoring is the systematic, continuous observation of computing systems, services, and infrastructure through the collection, aggregation, and analysis of metrics, logs, traces, and events to assess health, detect anomalies, and trigger remediation. It forms the primary feedback loop for operational awareness, enabling engineering teams to distinguish between normal variance and actionable degradation in real time. Monitoring encompasses both active probing (synthetic checks, health endpoints) and passive signal collection (agent-based telemetry, sidecar proxies), unified through time-series storage and alerting pipelines. As systems grow in complexity \u2014 particularly across distributed, containerised, and cloud-native environments \u2014 monitoring has evolved into a multi-signal discipline often grouped under the broader term Observability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:monitoring", "labels": [ "Monitoring", "Automated Monitoring", "Efficiency Monitoring", "Logging and Monitoring", "Monitoring Instrumentation", "Monitoring and Enforcement" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [ "Reliability Engineering", "Real-Time Computing" ] }, { "id": "monocular-camera", "title": "Monocular Camera", "domain": "robotics", "domain_name": "Robotics", "definition": "A Monocular Camera is a single-lens imaging sensor that captures 2D intensity or colour frames from a single viewpoint, serving as a primary perceptual modality in robotics, autonomous vehicles, and computer vision systems. Unlike stereo or depth cameras, it lacks intrinsic depth measurement capability, requiring computational techniques such as structure-from-motion, visual odometry, or learned depth estimation to recover 3D scene geometry.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:monocular-camera", "labels": [ "Monocular Camera" ], "is_subclass_of": [ "Perception and Sensing", "Camera" ], "wikilinks": [ "Camera", "Robotics" ] }, { "id": "monocular-depth-estimation", "title": "Monocular Depth Estimation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Monocular depth estimation is the task of inferring the distance of scene points from a single image, recovering a depth map without the explicit stereo disparity available from multiple cameras. Because depth from one view is inherently ambiguous, modern approaches learn statistical and contextual cues such as texture gradients, object size, and perspective, typically using deep convolutional or transformer networks trained on labelled or self-supervised data. It contrasts with stereo and active depth sensing while offering a low-cost route to three-dimensional perception.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:monocular-depth-estimation", "labels": [ "Monocular Depth Estimation" ], "is_subclass_of": [ "Depth Estimation" ], "wikilinks": [] }, { "id": "monolithic-ai", "title": "Monolithic Ai", "domain": "ai", "domain_name": "Ai", "definition": "Monolithic AI refers to a design paradigm in which a single, large, undivided model or system handles all tasks within an AI application, as opposed to decomposed or modular architectures. The term is most often used to contrast with multi-agent, mixture-of-experts, or microservice-based AI designs. Monolithic AI systems are simpler to deploy but harder to update, scale selectively, or audit at component level.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:monolithic-ai", "labels": [ "Monolithic Ai", "Monolithic AI" ], "is_subclass_of": [ "Artificial Intelligence", "AI Model Architecture" ], "wikilinks": [] }, { "id": "monolithic-architecture", "title": "Monolithic Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Monolithic architecture is a software design style in which an application is built and deployed as a single, self-contained unit where the user interface, business logic, and data-access layers are tightly coupled within one codebase and process. Components communicate through in-process function calls rather than network protocols, simplifying development, testing, and deployment for small to medium systems. As applications grow, the monolith can become difficult to scale selectively, evolve independently, or deploy without full redeployment, which motivates migration toward modular or microservices architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:monolithic-architecture", "labels": [ "Monolithic Architecture" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [] }, { "id": "monopropellant-propulsion", "title": "Monopropellant Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:monopropellant-propulsion", "labels": [ "Monopropellant Propulsion" ], "is_subclass_of": [ "Chemical Propulsion" ], "wikilinks": [] }, { "id": "monte-carlo-integration", "title": "Monte Carlo Integration", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Monte Carlo integration estimates the value of an integral by averaging the integrand over randomly sampled points, with error that decreases with the square root of the sample count regardless of dimension.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:monte-carlo-integration", "labels": [ "Monte Carlo Integration" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [ "Random Number Generation", "Volume Rendering", "Importance Sampling", "Sequential Monte Carlo", "Numerical Methods" ] }, { "id": "monte-carlo-localization", "title": "Monte Carlo Localization", "domain": "robotics", "domain_name": "Robotics", "definition": "A probabilistic localization algorithm that represents the robot's belief about its position using a set of weighted particles (samples), where each particle represents a hypothesis of the robot's pose. It implements a particle filter to recursively estimate the robot's pose distribution.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:monte-carlo-localization", "labels": [ "Monte Carlo Localization", "Monte Carlo Localisation", "Particle Filter", "RB-1014-monte-carlo-localization" ], "is_subclass_of": [ "Navigation and Planning", "RB 1013 localization", "Bayes Filter" ], "wikilinks": [ "Global Localization", "Global Localization Capability", "Importance Sampling", "Kidnapped Robot Problem", "Map", "Mobile Robots", "Motion Model", "Non-Parametric", "Particle Filter Theory", "Probabilistic Methods", "Probabilistic Robotics", "RB-1008-odometry", "RB-1015-kalman-filter", "Sensor Measurements", "Bayes Filter", "Localization", "Particle Filter", "RB-1013-localization", "Robotics", "Sensor Fusion" ] }, { "id": "monte-carlo-methods", "title": "monte carlo methods", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Monte Carlo methods are a broad class of computational algorithms that employ repeated random sampling to approximate numerical quantities that are analytically intractable, including high-dimensional integrals, expectations under complex probability distributions, and optimal policies in stochastic decision processes. The fundamental principle holds that averaging a function over sufficiently many independent samples drawn from an appropriate distribution converges, by the law of large numbers, to the true expected value at a rate of O(1/\u221aN) independent of dimensionality. Monte Carlo methods underpin Bayesian inference, reinforcement learning, sequential Monte Carlo particle filters, and physics simulation, and are foundational to risk analysis, financial modelling, and scientific computing. Variance reduction techniques \u2014 including importance sampling, control variates, stratified sampling, and quasi-Monte Carlo sequences \u2014 substantially improve practical convergence.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:monte-carlo-methods", "labels": [ "Monte Carlo Methods", "Monte Carlo Approximation", "Monte Carlo Method" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "monte-carlo-simulation", "title": "Monte Carlo Simulation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Monte Carlo simulation is a computational technique that estimates the behaviour of a system or the value of a quantity by repeatedly sampling random inputs from probability distributions and aggregating the resulting outcomes. By running many random trials it approximates expectations, distributions and tail risks that are difficult to derive analytically. It is widely used for numerical integration, risk analysis and uncertainty quantification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:monte-carlo-simulation", "labels": [ "Monte Carlo Simulation" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "monte-carlo-tree-search", "title": "Monte Carlo Tree Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Monte Carlo Tree Search (MCTS) is a heuristic search algorithm for sequential decision-making that builds a game tree incrementally through random simulations, balancing exploration and exploitation via the Upper Confidence Bound (UCB) formula. It enables strong play in games with large branching factors and generalises to planning under uncertainty.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:monte-carlo-tree-search", "labels": [ "Monte Carlo Tree Search" ], "is_subclass_of": [ "Search Algorithms" ], "wikilinks": [ "Game Playing AI", "Search Algorithms" ] }, { "id": "moore-s-law", "title": "Moore's Law", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Moore's Law is the observation, articulated by Gordon Moore in 1965, that the number of transistors on an integrated circuit doubles roughly every two years, driving exponential gains in computing capability and cost-efficiency. It functioned for decades as a self-fulfilling roadmap for the semiconductor industry. Physical and economic limits have slowed transistor scaling, shifting progress toward specialised architectures and advanced packaging.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:moore-s-law", "labels": [ "Moore's Law" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "motion-capture-rig", "title": "Motion Capture Rig", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical hardware or software system capturing human motion for animation or simulation through cameras, markers, sensors, and tracking infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:motion-capture-rig", "labels": [ "Motion Capture Rig" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Biomechanical Analysis", "Camera Calibration", "Data Fusion", "Data Processing Unit", "High-Speed Networking", "IMU Sensors", "ISO/IEC 17820", "Optical Cameras", "Performance Capture", "Sensor Input", "Skeletal Tracking", "Synchronized Timing", "Tracking Volume", "Animation Retargeting", "Calibration Target", "Computer Vision", "CreativeMediaDomain", "Motion Markers", "Motion Solver Software", "PhysicalLayer" ] }, { "id": "motion-capture-technology", "title": "Motion Capture Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Motion Capture Technology refers to systems that record the position and orientation of physical bodies\u2014human performers, objects, or camera rigs\u2014in three-dimensional space, producing skeletal or marker-based data streams used to animate digital avatars, create realistic character performances, and drive real-time spatial computing applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:motion-capture-technology", "labels": [ "Motion Capture Technology" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "motion-capture", "title": "Motion Capture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Motion Capture (mocap) is a technology domain and production pipeline discipline encompassing hardware, software, and algorithmic systems that record the position, orientation, and movement of bodies, objects, or faces in three-dimensional space over time \u2014 translating real-world kinematic data i...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:motion-capture", "labels": [ "Motion Capture", "Human Motion Capture", "Motion Capture Retargeting", "Motion Capture System", "Motion Capture Workflows", "MotionCapture" ], "is_subclass_of": [ "Content and Assets", "Animation", "Computer Vision", "Spatial Computing Paradigm", "Biomechanics", "Performance Capture" ], "wikilinks": [ "Action Unit Encoding", "AI Body Pose Estimation", "AlgorithmLayer", "Animation", "Apple ARKit", "Biomechanics", "Blendshape Rig", "Body Rig", "Bundle Adjustment", "BVH Format", "C3D Format", "Calibration", "Camera Array", "Cinematics", "Cloth Simulation", "ComputerVisionDomain", "CreativeToolsDomain", "DeepLabCut", "Digital Human", "Digital Twins" ] }, { "id": "motion-control", "title": "Motion Control", "domain": "robotics", "domain_name": "Robotics", "definition": "MotionControl is the discipline governing the coordinated generation and execution of actuator commands that transform high-level kinematic or task-space specifications into precise, smooth, and dynamically consistent robot motion through the integrated chain of trajectory generation, servo-loop ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:motion-control", "labels": [ "Motion Control", "Motion Control System", "Motion Controller", "MotionControl" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Control Theory", "Industrial Automation", "Mechatronics", "Embedded Systems" ], "wikilinks": [ "AlgorithmLayer", "Classical Control Theory", "CNC", "CNC Machine Tools", "CNC Machining", "Collaborative Robotics", "Collaborative Robots", "CommunicationLayer", "Compliant Actuation", "Computed Torque Control", "ControlEngineeringDomain", "Cubic Splines", "Delta Robots", "Deterministic Network", "Drive Amplifier", "Dynamic Model", "Embedded Systems", "Encoder Feedback", "EtherCAT", "Feedforward Control" ] }, { "id": "motion-estimation", "title": "Motion Estimation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Motion estimation is the process of determining motion vectors that describe how regions of one video frame map to corresponding regions in another, capturing the apparent movement of objects and the camera between frames. It is the computational heart of inter-frame video compression, where predicting a block from a previously coded frame removes temporal redundancy. The same techniques underpin optical-flow analysis and frame interpolation in computer vision.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:motion-estimation", "labels": [ "Motion Estimation" ], "is_subclass_of": [ "Video Compression" ], "wikilinks": [] }, { "id": "motion-markers", "title": "Motion Markers", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Reflective or active tracking points placed on subjects for optical motion capture systems, typically 12-15mm diameter for full-body capture, enabling precise position tracking at frame rates from 120fps to fps for animation, biomechanics, and performance analysis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:motion-markers", "labels": [ "Motion Markers" ], "is_subclass_of": [ "Interaction Technology", "Motion Capture Technology" ], "wikilinks": [ "3D Motion Analysis", "metaverse", "Motion Capture Technology" ] }, { "id": "motion-model", "title": "Motion Model", "domain": "robotics", "domain_name": "Robotics", "definition": "A motion model is a mathematical or learned representation that describes how the state of a moving entity\u2014a robot, vehicle, or articulated body\u2014evolves over time given control inputs and noise. In probabilistic robotics it forms the prediction step of filters such as Kalman and particle filters, characterising uncertainty in state transitions. Motion models range from simple kinematic approximations (constant velocity, unicycle) to full rigid-body dynamic equations and learned neural representations derived from data. Accuracy of the motion model directly determines the quality of localisation, planning, and control outcomes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:motion-model", "labels": [ "Motion Model", "Probabilistic Motion Model" ], "is_subclass_of": [ "Kinematics Model" ], "wikilinks": [] }, { "id": "motion-planning", "title": "Motion Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Motion Planning extends path planning by incorporating robot dynamics, control constraints, and time-parametrisation to generate dynamically feasible trajectories that account for velocity, acceleration, jerk, and actuator limitations. It produces executable control sequences that guide robots and autonomous vehicles from initial to goal states whilst satisfying kinodynamic constraints and optimising performance metrics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:motion-planning", "labels": [ "Motion Planning", "Classical Motion Planning", "MotionPlanning", "Robot Motion Planning", "Robotics Motion Planning" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [ "MotionDirector", "Trajectory Optimisation", "ComfyUI", "Control Theory", "Delivery Planning", "MetaverseDomain", "Microsoft Copilot", "Music and Audio", "Path Planning", "social media", "Update Cycle" ] }, { "id": "motion-solver-software", "title": "Motion Solver Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specialized software that uses multibody dynamics and physics-based algorithms to calculate reaction forces, torques, velocities, accelerations, and motor behaviors for mechanical systems and animated characters, enabling accurate simulation of rigid and flexible body movements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:motion-solver-software", "labels": [ "Motion Solver Software" ], "is_subclass_of": [ "Platform and Environment", "Simulation Software" ], "wikilinks": [ "metaverse", "Real Time Character Animation", "Simulation Software" ] }, { "id": "motion-tracking", "title": "Motion Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Motion tracking is a technology that captures and records the movement of objects, bodies, or body parts in physical space, translating this data into digital representations for use in VR/AR systems, animation, and metaverse applications. In XR contexts it is fundamental to embodied presence by mapping user movements to avatar animation and enabling natural interaction with virtual environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:motion-tracking", "labels": [ "Motion Tracking" ], "is_subclass_of": [ "Computer Vision", "XR Input Technologies" ], "wikilinks": [ "Avatar Animation", "Embodied Presence", "Gesture Control", "User Identification", "XR Input Technologies", "Metaverse" ] }, { "id": "motor-control-system", "title": "Motor Control System", "domain": "robotics", "domain_name": "Robotics", "definition": "A motor control system is the integrated assembly of sensing, driver electronics, and feedback logic that regulates the speed, torque, and position of one or more electric motors in a robotic platform. It combines components such as motor drivers, current sensors, and encoders with a control loop that adjusts drive signals in response to measured or estimated motor state. Motor control systems form the actuation layer that translates high-level motion commands into physical movement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:motor-control-system", "labels": [ "Motor Control System" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "motor-control", "title": "Motor Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Motor control, in robotics and mechatronics, is the discipline of regulating an electric motor's speed, torque, direction and position so that it produces the mechanical motion a system commands. It relies on power modulation techniques such as pulse-width modulation together with closed-loop feedback from encoders or current sensors to correct deviation from the commanded reference. Motor control is implemented in dedicated motor controller hardware but refers more broadly to the algorithms and control loops that translate high-level motion commands into physical actuation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:motor-control", "labels": [ "Motor Control" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "motor-controller", "title": "Motor Controller", "domain": "robotics", "domain_name": "Robotics", "definition": "A motor controller is an electronic device that regulates the speed, torque, direction, and position of an electric motor by modulating the power delivered to it, typically via pulse-width modulation and closed-loop feedback. In robotics it interprets high-level velocity or position commands and drives motors accordingly, often integrating current sensing and encoder feedback. It is a core actuation component bridging control software and mechanical motion.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:motor-controller", "labels": [ "Motor Controller" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "motor-driver", "title": "Motor Driver", "domain": "robotics", "domain_name": "Robotics", "definition": "A Motor Driver is an electronic circuit or integrated circuit module that translates low-power control signals from a microcontroller or digital signal processor into the high-current, high-voltage waveforms required to operate electric motors, including DC brushed motors, brushless DC motors, and stepper motors. Motor drivers implement switching topologies such as H-bridge configurations and PWM (pulse-width modulation) generation to control motor direction, speed, and torque. Protection circuitry for over-current, over-temperature, and back-EMF clamping is typically integrated. Motor drivers are fundamental building blocks of robotic systems, electric vehicles, CNC machinery, and consumer appliances.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:motor-driver", "labels": [ "Motor Driver", "Motor Driver Electronics", "MotorDriver" ], "is_subclass_of": [ "Power Electronics" ], "wikilinks": [] }, { "id": "mounting-interface", "title": "Mounting Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "A mounting interface is the standardised mechanical feature set, such as bolt patterns, flanges, and alignment pins, that defines how one component is rigidly attached to another in a mechanical or robotic assembly. It ensures repeatable positioning, load transfer, and interchangeability of links, end-effectors, sensors, and payloads. Standardised interfaces (e.g. ISO tool flanges) enable modular, reconfigurable robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mounting-interface", "labels": [ "Mounting Interface" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "move-it-2", "title": "MoveIt 2", "domain": "robotics", "domain_name": "Robotics", "definition": "MoveIt 2 is the open-source motion-planning framework for ROS 2, providing manipulation capabilities including inverse kinematics, collision-aware path planning, trajectory generation, and execution for robotic arms and mobile manipulators. It integrates planners, perception, and control through a plugin architecture and is the de facto standard for arm motion planning in the ROS ecosystem. The ROS 2 version adds real-time and lifecycle improvements over the original MoveIt.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:move-it-2", "labels": [ "MoveIt 2", "MoveIt2" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "mpc-wallet", "title": "Mpc Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An MPC wallet is a digital wallet whose private key is split into shares held by independent parties and never reconstructed in one place, so transactions are authorised through a multi-party computation protocol that jointly produces a signature. By distributing trust across devices or institutions, it removes the single seed phrase as a single point of failure while preserving a single on-chain address. MPC wallets contrast with traditional self-custody seed wallets and with on-chain multisignature schemes by keeping the threshold logic off-chain and chain-agnostic.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mpc-wallet", "labels": [ "Mpc Wallet", "MPC Wallet" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [] }, { "id": "mpi", "title": "Mpi", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "MPI (Message Passing Interface) is a standardised, portable specification for message-passing parallel programming, defining a library of routines for point-to-point and collective communication among processes in a distributed-memory system. It is the dominant programming model for high-performance computing clusters, where independent processes exchange data explicitly rather than through shared memory. Implementations such as Open MPI and MPICH provide the runtime that maps the standard onto specific hardware and interconnects.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:mpi", "labels": [ "Mpi", "MPI" ], "is_subclass_of": [ "Message Passing" ], "wikilinks": [] }, { "id": "mu-jo-co", "title": "MuJoCo", "domain": "robotics", "domain_name": "Robotics", "definition": "MuJoCo (Multi-Joint dynamics with Contact) is a high-speed, high-fidelity physics simulation engine designed for modelling articulated rigid-body systems with complex contact dynamics, widely used in robotics research and reinforcement learning. Developed by Emo Todorov at the University of Washington and commercialised by Roboti LLC, it was acquired by Google DeepMind in 2021 and made freely available as open-source software in 2022. MuJoCo excels at simulating musculoskeletal models, legged locomotion, dexterous manipulation, and other contact-rich robotic tasks that require accurate Lagrangian dynamics and constraint solvers. Its adoption as the standard benchmark environment for reinforcement learning algorithms\u2014via OpenAI Gym and later Gymnasium wrappers\u2014has made it the de facto substrate for sim-to-real transfer research in robotics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mu-jo-co", "labels": [ "MuJoCo" ], "is_subclass_of": [ "Physics Simulation" ], "wikilinks": [] }, { "id": "mu-sig-2", "title": "MuSig2", "domain": "security", "domain_name": "Security", "definition": "MuSig2 is a two-round multi-signature protocol for the Schnorr signature scheme, enabling a group of signers to collaboratively produce a single aggregate signature and a single aggregate public key that are indistinguishable on-chain from a standard single-signer Schnorr signature. It improves upon the original MuSig (three-round) by eliminating one communication round without sacrificing security, using a technique of committing to multiple nonces per signer and computing a linear combination of those nonces after seeing all co-signers' commitments. The scheme is proven secure in the random oracle model under the discrete logarithm assumption and is resistant to the rogue-key and Wagner attacks that affect naive key-aggregation approaches.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:mu-sig-2", "labels": [ "MuSig2" ], "is_subclass_of": [ "Multisignature" ], "wikilinks": [ "Schnorr Signatures", "Taproot", "Schnorr Signature", "Multisignature" ] }, { "id": "mu-zero", "title": "MuZero", "domain": "ai", "domain_name": "Ai", "definition": "MuZero is a model-based reinforcement learning algorithm from DeepMind that achieves superhuman performance in board games and Atari without being given the rules of the environment. It learns a latent dynamics model predicting reward, value, and policy, then plans with Monte Carlo Tree Search over this learned model. It generalises AlphaZero to domains where the environment dynamics are unknown.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:mu-zero", "labels": [ "MuZero" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "multi-objective-optimisation", "title": "Multi Objective Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Multi-objective optimisation is the discipline of optimising two or more conflicting objective functions simultaneously, where improving one objective typically degrades another. Rather than a single optimum it yields a Pareto front of non-dominated trade-off solutions, from which a decision-maker selects according to preferences. It is solved with scalarisation, evolutionary, and gradient-based methods and is pervasive in engineering design, machine learning, and resource allocation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multi-objective-optimisation", "labels": [ "Multi Objective Optimisation", "Multi-Objective Optimisation" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "multi-party-royalties", "title": "Multi Party Royalties", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A blockchain-based distribution mechanism that uses smart contracts to automatically split and distribute royalty payments among multiple stakeholders including creators, collaborators, developers, and brands based on predefined ownership percentages.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-party-royalties", "labels": [ "Multi Party Royalties", "Multi-Party Royalties" ], "is_subclass_of": [ "Governance and Safety", "Revenue Sharing" ], "wikilinks": [ "Creator Economy", "metaverse", "Revenue Sharing" ] }, { "id": "multi-party-transactions", "title": "Multi Party Transactions", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain transactions involving more than two participants that use atomic swap protocols, adaptor signatures, and hash timelock contracts (HTLCs) to ensure all parties complete their exchanges simultaneously or the entire transaction is reversed, preventing partial completion losses.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-party-transactions", "labels": [ "Multi Party Transactions", "Multi-Party Transactions", "Multi-Signature Transactions" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Transaction", "Blockchain Transactions" ], "wikilinks": [ "Blockchain Transactions", "Cross Chain Trading" ] }, { "id": "multi-sig-governance", "title": "Multi Sig Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multi-signature (multi-sig) governance is a cryptographic threshold-signature scheme requiring m-of-n authorised keyholders to co-sign a transaction or message before it executes, eliminating single points of failure in digital-asset control and enabling graduated, committee-based decision-making...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:multi-sig-governance", "labels": [ "Multi Sig Governance", "Multi-Sig Governance" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Governance", "Cryptographic Protocol", "Access Control System", "Digital Signature", "Threshold Cryptography", "Committee Governance", "Hash Function" ], "wikilinks": [ "Argent Account", "BIP-11", "BIP-16", "BIP-327", "BIP-340", "BIP-341", "BIP-373", "BitGo Wallet", "Bitcoin ETF Custody", "Bitcoin Script", "CGGMP21 Protocol", "Committee Governance", "Copper MPC", "Cross-Chain Bridge Security", "CryptographyDomain", "CustodyLayer", "DeFi Protocol Security", "DigitalAssetDomain", "Distributed Key Generation", "EdDSA" ] }, { "id": "multi-task-learning", "title": "Multi Task Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multi Task Learning is a machine learning paradigm where a single model is trained simultaneously on multiple related tasks, sharing intermediate representations to improve generalisation, sample efficiency, and regularisation relative to independently trained single-task models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:multi-task-learning", "labels": [ "Multi Task Learning", "Multi-Task Learning", "multi-task learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "MetaverseDomain", "Multimodal" ] }, { "id": "multi-user-systems", "title": "Multi User Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Multi User Systems are networked software platforms designed to support two or more simultaneous participants sharing a common computational environment, enabling real-time interaction, cooperative task execution, and mutual awareness of other participants' actions and presence. They encompass the full technology stack required for shared state management, including session management, authority arbitration, conflict resolution, and low-latency communication protocols. In spatial and immersive computing contexts, multi-user systems coordinate avatar representation, object ownership, physics synchronisation, and access control across geographically distributed clients. The category spans a wide range of instantiations, from traditional time-sharing operating systems and online multiplayer game engines to collaborative XR environments and cloud-hosted virtual workspaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:multi-user-systems", "labels": [ "Multi User Systems", "Multi-User Collaboration", "Multi-User Systems", "Multi-User Virtual Environment", "multi-user systems" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "owl:Thing" ] }, { "id": "multi-agent-collaboration", "title": "Multi-Agent Collaboration", "domain": "ai", "domain_name": "Ai", "definition": "Multi-agent collaboration is the coordinated execution of a task by multiple AI agents that divide work, share intermediate results and reconcile conflicting outputs toward a common goal. It requires protocols for task allocation, message passing and conflict resolution so that agents with different roles or specialisations can operate concurrently without duplicating or contradicting each other's work. Harness configuration packs and IDE coding agents increasingly rely on multi-agent collaboration to parallelise complex software tasks across specialised sub-agents.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:multi-agent-collaboration", "labels": [ "Multi-Agent Collaboration" ], "is_subclass_of": [ "Multi-Agent System" ], "wikilinks": [] }, { "id": "multi-agent-coordination", "title": "Multi-Agent Coordination", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multi-agent coordination is the set of mechanisms and protocols by which a population of autonomous agents organise their individual actions, communications, and resource usage to achieve shared or mutually compatible goals without central command. It encompasses task decomposition and allocation, conflict detection and resolution, synchronisation of parallel workstreams, and the design of incentive structures that align agent behaviour across heterogeneous systems. Coordination differs from simple parallelism in that it requires agents to reason about the intentions and capabilities of peers, adapting their own behaviour accordingly. Contemporary implementations range from classical Distributed Constraint Optimisation Problems (DCOP) to large-language-model orchestration frameworks in which a controller agent delegates subtasks to specialist sub-agents and integrates their outputs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multi-agent-coordination", "labels": [ "Multi-Agent Coordination" ], "is_subclass_of": [ "Multi-Agent Systems" ], "wikilinks": [ "Multi-Agent Systems", "Agentic Workflow", "AI Agent", "Game Theory" ] }, { "id": "multi-agent-interaction", "title": "Multi-Agent Interaction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The dynamic exchange of information and actions between multiple autonomous AI agents, which can lead to emergent behaviors, coordination, or systemic inefficiencies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-agent-interaction", "labels": [ "Multi-Agent Interaction" ], "is_subclass_of": [ "World Model" ], "wikilinks": [] }, { "id": "multi-agent-monitoring", "title": "Multi-Agent Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The operational practice of tracking the health, performance, and interactions of multiple AI agents within a system.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-agent-monitoring", "labels": [ "Multi-Agent Monitoring" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "multi-agent-orchestration-frameworks", "title": "Multi-Agent Orchestration Frameworks", "domain": "ai", "domain_name": "Ai", "definition": "Frameworks enabling multiple AI agents to collaborate on complex tasks through role assignment, structured conversations, handoffs, and coordinated workflows \u2014 includes openai-agents-python, crewAI, autogen, MetaGPT, ChatDev, PraisonAI, and agent-squad.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:multi-agent-orchestration-frameworks", "labels": [ "Multi-Agent Orchestration Frameworks" ], "is_subclass_of": [ "Agent Harness", "Agent Frameworks", "AI Agent System", "External AI Harness" ], "wikilinks": [ "Agent Harness", "Personal Agent Runtimes", "Agent Development SDKs", "Internal AI Harness", "External AI Harness", "Multi-Agent Coordination", "Agent Orchestrator", "Model Context Protocol", "Large Language Model", "Tool Use", "Agent Memory Layers", "AI Agent System", "Agentic AI", "Workflow Automation", "Graph Theory", "State Machine", "Directed Acyclic Graph", "Agent Evaluation Benchmarks", "Agent Execution Sandboxes", "Agent Frameworks" ] }, { "id": "multi-agent-orchestration", "title": "Multi-Agent Orchestration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multi-agent orchestration is the coordination and management of multiple autonomous AI agents \u2014 each capable of perceiving, reasoning, and acting \u2014 such that their collective behaviour accomplishes complex tasks beyond the capacity of any single agent. Orchestration encompasses task decomposition, agent assignment, inter-agent communication, state sharing, conflict resolution, and result aggregation, typically mediated by an orchestrator layer or emergent via peer protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multi-agent-orchestration", "labels": [ "Multi-Agent Orchestration", "Agent Orchestration" ], "is_subclass_of": [ "Orchestration" ], "wikilinks": [] }, { "id": "multi-agent-rag-architecture-compendium", "title": "Multi-Agent RAG Architecture Compendium", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A research and literature collection exploring architectures in which multiple specialised AI agents collaborate through retrieval-augmented generation pipelines, combining knowledge graphs, ontologies, and constrained large language models to model complex social and environmental contexts in immersive environments. Covers formal ontology design, multi-modal data ingestion, and ethical deployment constraints.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:multi-agent-rag-architecture-compendium", "labels": [ "Multi-Agent RAG Architecture Compendium", "Multi Agent RAG scrapbook" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "David Tully", "MUST", "PEOPLE" ] }, { "id": "multi-agent-reinforcement-learning", "title": "Multi-Agent Reinforcement Learning", "domain": "ai", "domain_name": "Ai", "definition": "Multi-agent reinforcement learning (MARL) extends reinforcement learning to settings where several agents learn concurrently while interacting in a shared environment. Each agent optimises its own policy, but the environment is non-stationary from any single agent's perspective because the others are simultaneously adapting, which raises challenges of coordination, competition, credit assignment and equilibrium selection. MARL draws on game theory to analyse cooperative, competitive and mixed incentive structures and underpins applications from team robotics to automated trading and traffic control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multi-agent-reinforcement-learning", "labels": [ "Multi-Agent Reinforcement Learning" ], "is_subclass_of": [ "Reinforcement Learning", "AI Technique" ], "wikilinks": [] }, { "id": "multi-agent-system", "title": "Multi-Agent System", "domain": "ai", "domain_name": "Ai", "definition": "A Multi-Agent System (MAS) is a computational architecture in which multiple autonomous agents \u2014 each equipped with local perception, internal state, and independent decision-making capability \u2014 interact within a shared environment to accomplish individual or collective objectives. Coordination emerges from direct communication, environmental signalling, stigmergy, or market-like auction mechanisms, without requiring any single agent to hold global knowledge or exert centralised control. MAS formalises distributed problem-solving by composing heterogeneous or homogeneous agent populations whose aggregate behaviour frequently exhibits emergence \u2014 properties absent in any individual agent. The paradigm spans robotics swarms, AI orchestration pipelines, financial market simulation, smart-grid balancing, and autonomous software engineering.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:multi-agent-system", "labels": [ "Multi-Agent System", "Decentralised Multi-Agent System", "MultiAgentSystem" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "multi-agent-systems", "title": "multi-agent systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multi-agent systems (MAS) are computational architectures in which multiple autonomous agents\u2014each with their own perception, memory, reasoning, and action capabilities\u2014interact within a shared environment to accomplish tasks that benefit from parallelisation, specialisation, or distributed coordination. Each agent operates according to local decision logic (reactive, deliberative, or hybrid) while collectively producing emergent global behaviour through communication, negotiation, and coordination protocols. In contemporary AI, MAS are realised through networks of LLM-backed agents orchestrated by frameworks that manage inter-agent messaging, tool invocation, and task decomposition. Mechanism-design principles, game theory, and classical distributed-AI theory inform how modern agentic systems handle conflict, incentive alignment, and safety in open-ended agent populations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:multi-agent-systems", "labels": [ "Multi-Agent Systems", "Multi Agent Systems" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "multi-armed-bandit", "title": "Multi-Armed Bandit", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A sequential decision-making framework in which an agent repeatedly chooses among a fixed set of actions ('arms') with unknown reward distributions, observing only the reward of the chosen arm, and seeks to maximise cumulative reward \u2014 equivalently, to minimise regret against the best arm in hindsight; it isolates the exploration\u2013exploitation trade-off in its purest form, since unlike a full Markov decision process the environment has no state transitions, and underpins algorithms such as epsilon-greedy, UCB, and Thompson sampling used in A/B testing, recommendation, and adaptive experimentation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:multi-armed-bandit", "labels": [ "Multi-Armed Bandit" ], "is_subclass_of": [ "Reinforcement Learning" ], "wikilinks": [ "Reinforcement Learning", "Exploration Exploitation Tradeoff", "Markov Decision Process", "Recommendation Systems" ] }, { "id": "multi-camera-rig", "title": "Multi-Camera Rig", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A multi-camera rig is a physical arrangement of multiple synchronised cameras positioned around a subject or scene to capture it from many viewpoints simultaneously. It is the primary capture hardware for volumetric capture and volumetric video pipelines, where images from all cameras at a given instant are fused into a single 3D or free-viewpoint representation. Camera count, placement and synchronisation precision directly determine the achievable reconstruction quality.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:multi-camera-rig", "labels": [ "Multi-Camera Rig" ], "is_subclass_of": [ "Volumetric Capture" ], "wikilinks": [] }, { "id": "multi-chain-application", "title": "Multi-Chain Application", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A multi-chain application is a decentralised application that deploys and operates logic across more than one blockchain network simultaneously, rather than being confined to a single chain. It relies on cross-chain communication and interoperability mechanisms, such as bridges or messaging protocols, to synchronise state and move assets between the chains it spans. Multi-chain designs let applications draw on the liquidity, users or specialised capabilities of several ecosystems at once, at the cost of added complexity and cross-chain security risk.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:multi-chain-application", "labels": [ "Multi-Chain Application" ], "is_subclass_of": [ "Cross-Chain Interoperability" ], "wikilinks": [] }, { "id": "multi-chain-de-fi", "title": "Multi-Chain DeFi", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multi-chain DeFi refers to decentralised finance applications and liquidity that operate across multiple independent blockchains rather than being confined to a single network. It relies on cross-chain bridges, messaging protocols, and interoperability standards to move assets and data between chains, letting users access yields and markets wherever they reside. It expands capital efficiency but introduces bridge and composability risks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multi-chain-de-fi", "labels": [ "Multi-Chain DeFi" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "multi-cloud", "title": "Multi-Cloud", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cloud deployment strategy in which an organisation consumes services from two or more independent public cloud providers \u2014 such as AWS, Azure, and Google Cloud \u2014 distributing workloads to avoid vendor lock-in, satisfy data-residency and regulatory requirements, exploit provider-specific strengths, and improve resilience against provider-wide outages, at the cost of increased operational complexity, duplicated tooling, and cross-cloud egress charges.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multi-cloud", "labels": [ "Multi-Cloud" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [ "Cloud Computing", "Hybrid Cloud", "Cloud Infrastructure" ] }, { "id": "multi-factor-authentication", "title": "Multi-Factor Authentication", "domain": "security", "domain_name": "Security", "definition": "Multi-Factor Authentication (MFA) is a security mechanism that requires a claimant to present two or more independent verification factors drawn from distinct categories \u2014 something known (a password or PIN), something possessed (a hardware token or mobile device), and something inherent (a biometric characteristic) \u2014 before access is granted to a system or resource. By requiring multiple independent proofs, MFA ensures that compromise of a single credential is insufficient for an attacker to gain access, substantially raising the cost and complexity of successful attacks. It is widely mandated by regulatory frameworks, cybersecurity standards, and national policy as a baseline control for protecting sensitive systems, privileged accounts, and personal data. Modern deployments increasingly employ adaptive or risk-based MFA, invoking stronger authentication only when contextual risk signals exceed a configurable threshold.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:multi-factor-authentication", "labels": [ "Multi-Factor Authentication" ], "is_subclass_of": [ "Authentication Mechanism" ], "wikilinks": [] }, { "id": "multi-head-attention", "title": "Multi-Head Attention", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An extension of scaled dot-product attention that runs multiple attention operations in parallel over distinct learned projection subspaces, then concatenates and linearly projects the results. Multi-head attention enables Transformer models to capture diverse dependency patterns across positions and representation subspaces simultaneously, and is foundational to modern large language models.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:multi-head-attention", "labels": [ "Multi-Head Attention", "Multi Head Attention", "Multi Head Self Attention", "Multi-Head Self-Attention" ], "is_subclass_of": [ "Attention Mechanism" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Attention Mechanism" ] }, { "id": "multi-hop-reasoning", "title": "Multi-Hop Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "Multi-hop reasoning is the capability to answer a question or draw a conclusion by chaining together several intermediate inferences or retrieved facts, rather than relying on a single piece of evidence. In language models and retrieval systems it requires composing information across multiple documents or knowledge-graph edges. It is central to complex question answering and is a known weak point for shallow retrieval and single-pass models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multi-hop-reasoning", "labels": [ "Multi-Hop Reasoning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "multi-layer-agentic-governance-framework", "title": "Multi-Layer Agentic Governance Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Agentic Mycelia is a multi-layered framework of specialised AI agents\u2014scene agents, transfer agents, onboarding agents, and jurisdictional agents\u2014that collectively manage governance, identity, value exchange, and protocol translation across interconnected metaverse instances. Analogous to fungal mycelium, the network forms resilient, decentralised inter-instance communication that enables seamless asset transfer, reputation propagation, and automated ontology exchange between diverse virtual worlds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:multi-layer-agentic-governance-framework", "labels": [ "Multi-Layer Agentic Governance Framework", "Agentic Mycelia" ], "is_subclass_of": [ "AI Application", "Agent Frameworks" ], "wikilinks": [ "Chain of Thought", "Anthropic Claude", "Blockchain", "Diagrams as Code", "Gemini", "Google", "Large Language Models", "Metaverse Ontology", "Prompt Engineering" ] }, { "id": "multi-model-orchestration", "title": "Multi-Model Orchestration", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The architectural pattern of coordinating multiple AI models to process a single task, often involving parallel execution, routing, or ensemble methods to optimize performance, cost, or reliability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-model-orchestration", "labels": [ "Multi-Model Orchestration" ], "is_subclass_of": [ "Model Architecture" ], "wikilinks": [] }, { "id": "multi-model-routing", "title": "Multi-Model Routing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An architectural pattern that dynamically directs AI requests to different models (e.g., open-weight, proprietary, local) based on task complexity, cost constraints, or privacy requirements to optimize performance and efficiency.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multi-model-routing", "labels": [ "Multi-Model Routing" ], "is_subclass_of": [ "Model Architecture" ], "wikilinks": [] }, { "id": "multi-party-computation", "title": "Multi-Party Computation", "domain": "security", "domain_name": "Security", "definition": "Multi-Party Computation (MPC) is a subfield of cryptography that enables a set of mutually distrusting parties to jointly compute a function over their private inputs without revealing those inputs to any other participant. Correctness is guaranteed even when a bounded subset of participants behave maliciously, making MPC a cornerstone of privacy-preserving collaborative computation. The technique generalises two-party secure computation to arbitrary numbers of participants using secret-sharing or garbled-circuit protocols.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:multi-party-computation", "labels": [ "Multi-Party Computation", "Multi Party Computation" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "multi-robot-coordination", "title": "Multi-Robot Coordination", "domain": "robotics", "domain_name": "Robotics", "definition": "Multi-robot coordination is the set of methods that enable multiple robots to act together coherently toward shared or individual goals, covering task allocation, collision avoidance, formation control, and communication. It addresses centralised and decentralised architectures and trades off optimality against scalability and robustness. Applications include warehouse fleets, drone swarms, and cooperative exploration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multi-robot-coordination", "labels": [ "Multi-Robot Coordination" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "multi-robot-systems", "title": "Multi-Robot Systems", "domain": "robotics", "domain_name": "Robotics", "definition": "Multi-robot systems (MRS) are coordinated ensembles of two or more robotic agents that collaborate to accomplish tasks beyond the reach of a single robot, using communication, task allocation, and shared world models. They span architectures from tightly coupled homogeneous fleets to loosely coupled heterogeneous teams where robots with different capabilities complement each other. Key research challenges include task and motion planning, collision avoidance, fault tolerance, and communication bandwidth constraints in dynamic environments. MRS find application in search and rescue, warehouse automation, precision agriculture, and planetary exploration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multi-robot-systems", "labels": [ "Multi-Robot Systems", "Multi-Robot System" ], "is_subclass_of": [ "Autonomous System" ], "wikilinks": [] }, { "id": "multi-signature-wallet", "title": "Multi-Signature Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A multi-signature wallet is a cryptographic account on a blockchain network that requires M-of-N authorisation \u2014 where a minimum of M private key holders out of a total of N authorised signers must collectively sign a transaction before it is considered valid and broadcast to the network. This threshold scheme eliminates single points of failure in asset custody, distributing the risk of key compromise or loss across multiple independent parties or hardware devices. Multi-signature wallets are widely deployed for treasury management, DAO governance, and exchange hot wallets, and are implemented both as native Bitcoin script constructs and as smart contracts on EVM-compatible chains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multi-signature-wallet", "labels": [ "Multi-Signature Wallet" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [] }, { "id": "multi-signature", "title": "Multi-Signature", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multi-Signature (multisig) is a cryptographic access-control scheme requiring that a transaction or operation be authorised by a threshold number m out of a defined set of n independent private keys before it is considered valid. Commonly expressed as m-of-n (e.g. 2-of-3), multisig eliminates single points of failure in key custody and enforces shared governance over digital assets, smart contracts, or any access-controlled resource. It is foundational to institutional cryptocurrency custody, DAO treasury management, bridge security, and hardware-secure wallet architectures. Multisig schemes may be implemented at the protocol layer, at the script level (as in Bitcoin P2SH/P2WSH), or at the application layer via smart contracts such as Gnosis Safe.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:multi-signature", "labels": [ "Multi-Signature", "Multi-Signature Approval", "Multi-Signature Cryptographic Schemes", "Multi-Signature Custody", "Multi-Signature Scheme" ], "is_subclass_of": [ "Signature Scheme" ], "wikilinks": [] }, { "id": "multi-stakeholder-governance", "title": "Multi-Stakeholder Governance", "domain": "governance", "domain_name": "Governance", "definition": "Multi-stakeholder governance is a model of collective decision-making in which government, private sector, civil society, technical communities, and academia participate jointly and on a relatively equal footing to develop policies, standards, or norms. It emphasises inclusiveness, transparency, and consensus over top-down or purely state-led control, and is the prevailing model for governing shared resources such as the internet. Its legitimacy derives from broad participation rather than hierarchical authority.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multi-stakeholder-governance", "labels": [ "Multi-Stakeholder Governance", "Multi-stakeholder Governance" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "multi-step-reasoning", "title": "Multi-Step Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "Multi-step reasoning is the capacity of an AI system to solve problems that require chaining several intermediate inferences, rather than mapping an input directly to an answer in a single step. It encompasses decomposing a problem into sub-problems, maintaining and updating intermediate state, and composing partial results into a final solution. In large language models it is elicited through chain-of-thought prompting, tool use, and search over reasoning paths, and it is a primary differentiator between shallow pattern completion and genuine problem-solving competence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multi-step-reasoning", "labels": [ "Multi-Step Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "multi-tenancy", "title": "Multi-Tenancy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multi-tenancy is a software architecture in which a single instance of an application or platform serves many independent customers, called tenants, while keeping each tenant's data, configuration, and behaviour logically isolated. It maximises resource utilisation and operational efficiency by sharing compute, storage, and code across tenants, relying on isolation boundaries to preserve privacy and prevent interference. It is foundational to software-as-a-service and to shared cloud infrastructure such as container orchestration platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:multi-tenancy", "labels": [ "Multi-Tenancy" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "multi-turn-dialogue", "title": "Multi-Turn Dialogue", "domain": "ai", "domain_name": "Ai", "definition": "Multi-turn dialogue is a conversational interaction spanning several exchanges in which a system must maintain context, track state, and resolve references across turns to produce coherent, relevant responses. It contrasts with single-turn question answering by requiring memory of prior utterances and the evolving goal. It is a core capability and evaluation axis for chatbots and conversational AI.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multi-turn-dialogue", "labels": [ "Multi-Turn Dialogue", "Multi-Turn Conversation" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "multi-view-images", "title": "Multi-View Images", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Multi-View Images are a set of photographs or renders of the same scene or object captured from multiple, known or estimable camera positions, providing the geometric overlap needed to reconstruct three-dimensional structure. They are the primary input to photogrammetry, novel view synthesis, and neural scene representations such as NeRFs and 3D Gaussian splats. Reconstruction accuracy improves with greater viewpoint coverage and more precise camera calibration.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:multi-view-images", "labels": [ "Multi-View Images" ], "is_subclass_of": [ "3D Reconstruction" ], "wikilinks": [] }, { "id": "multi-view-stereo", "title": "Multi-View Stereo", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Multi-View Stereo (MVS) is a computer vision technique that reconstructs dense 3D geometry from a set of overlapping 2D images captured from multiple camera positions. It extends traditional stereo matching by leveraging consistency across many viewpoints to estimate depth and surface detail at high resolution. MVS is a foundational component of photogrammetry pipelines, producing point clouds and textured meshes from photograph collections.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multi-view-stereo", "labels": [ "Multi-View Stereo", "Multi-View Imagery" ], "is_subclass_of": [ "Photogrammetry" ], "wikilinks": [] }, { "id": "multi-viewpoint-immersive-research-platform", "title": "Multi-Viewpoint Immersive Research Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A proposed multi-viewpoint immersive AI research platform enabling multiple domain experts to collaborate simultaneously in a shared stereoscopic environment, each maintaining their own spatial perspective. Real-time AI observes all communication channels\u2014verbal, gestural, spatial, and conceptual\u2014building a living ontology from emergent expert consensus, with applications in high-stakes planning such as nuclear decommissioning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:multi-viewpoint-immersive-research-platform", "labels": [ "Multi-Viewpoint Immersive Research Platform", "Future Bernard" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "multi-constellation-gnss", "title": "Multi-constellation GNSS", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multi-constellation-gnss", "labels": [ "Multi-constellation GNSS" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "multi-frequency-gnss", "title": "Multi-frequency GNSS", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multi-frequency-gnss", "labels": [ "Multi-frequency GNSS" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "multi-junction-spacecraft-solar-cell", "title": "Multi-junction Spacecraft Solar Cell", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multi-junction-spacecraft-solar-cell", "labels": [ "Multi-junction Spacecraft Solar Cell" ], "is_subclass_of": [ "Spacecraft Solar Cell" ], "wikilinks": [] }, { "id": "multi-messenger-astronomy", "title": "Multi-messenger Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multi-messenger-astronomy", "labels": [ "Multi-messenger Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "multi-access-edge-computing", "title": "MultiAccessEdgeComputing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multi-Access Edge Computing (MEC) is an ETSI-standardised architecture that places cloud-like compute and storage resources at the radio access network edge, enabling low-latency application hosting within milliseconds of end-user devices. It decouples latency-sensitive processing from centralised cloud data centres by co-locating compute with base stations and other network nodes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multi-access-edge-computing", "labels": [ "MultiAccessEdgeComputing", "Multi-Access Edge Computing", "Multi-access Edge Computing" ], "is_subclass_of": [ "Edge Computing" ], "wikilinks": [] }, { "id": "multibase", "title": "Multibase", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multibase is a self-describing encoding format specification that prefixes a binary value with a single character identifying which base encoding, such as base58btc, base32 or base64url, was used, so the encoded string can be decoded without external context. It is part of the Multiformats family of self-describing protocol specifications. It is used wherever content identifiers or decentralised-identifier material must be represented as text unambiguously across systems, such as Content Identifiers.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:multibase", "labels": [ "Multibase" ], "is_subclass_of": [ "Multiformats" ], "wikilinks": [] }, { "id": "multicodec", "title": "Multicodec", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multicodec is a self-describing data encoding scheme, part of the Multiformats family, that prefixes a binary value with a compact varint code identifying the codec used to interpret the bytes that follow. By making the encoding explicit and machine-readable, multicodec lets systems handle many data formats and hash algorithms without out-of-band agreement, supporting future-proof, agile interoperability across content-addressed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:multicodec", "labels": [ "Multicodec" ], "is_subclass_of": [ "Infrastructure", "Multiformats" ], "wikilinks": [] }, { "id": "multiformats", "title": "Multiformats", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multiformats is a collection of self-describing protocol and value specifications designed to make data formats future-proof and interoperable across decentralised systems. Rather than hard-coding a single hash, encoding, or address scheme, each multiformat prefixes the value with a compact code declaring which algorithm or format it uses, so software can interpret it unambiguously and evolve without breaking. Components include multihash, multibase, multicodec, and multiaddr, and the family underpins content addressing in IPFS and libp2p.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multiformats", "labels": [ "Multiformats" ], "is_subclass_of": [ "Content Addressing" ], "wikilinks": [] }, { "id": "multihash", "title": "Multihash", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multihash is a self-describing hash format developed by Protocol Labs in which the hash digest is prefixed with a varint-encoded function code and digest length, enabling consumers to identify the hash algorithm without out-of-band knowledge. It is a foundational component of the IPFS content-addressing stack and the Multiformat suite, providing algorithm agility so that systems can upgrade from SHA-256 to SHA3 or BLAKE3 without breaking existing identifiers. Any hash function can be registered in the Multihash table and the format is codec-neutral.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:multihash", "labels": [ "Multihash" ], "is_subclass_of": [ "Cryptographic Hash" ], "wikilinks": [] }, { "id": "multilayer-insulation", "title": "Multilayer Insulation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multilayer-insulation", "labels": [ "Multilayer Insulation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "multilayer-perceptron", "title": "Multilayer Perceptron", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A multilayer perceptron (MLP) is a feedforward neural network composed of an input layer, one or more hidden layers of fully connected neurons with nonlinear activations, and an output layer, trained by backpropagation. It is a canonical universal function approximator underlying deeper architectures, and is frequently used as a lightweight decoder or coordinate-based function in implicit neural representations. NeRF and related implicit neural representation methods use an MLP to map spatial coordinates directly to colour and density values.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:multilayer-perceptron", "labels": [ "Multilayer Perceptron", "Multi-Layer Perceptron" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "multimedia-processing", "title": "Multimedia Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multimedia processing is the broad category of computational techniques for capturing, encoding, transforming, analysing and rendering audio, video, image and other time-based media. It encompasses signal-processing operations such as filtering and transform coding, as well as higher-level tasks like compression, synchronisation and format conversion. Audio processing and video compression are specific disciplines within this wider field.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:multimedia-processing", "labels": [ "Multimedia Processing" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "multimodal-ai-architecture", "title": "Multimodal AI Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multimodal AI refers to artificial intelligence systems that process, fuse, and generate content across multiple data modalities\u2014including text, images, audio, video, and 3D\u2014within a unified model architecture. Multimodal systems learn cross-modal alignments through joint embedding spaces, enabling capabilities such as image captioning, visual question answering, text-to-image synthesis, and real-time audio-visual understanding that single-modality models cannot achieve.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:multimodal-ai-architecture", "labels": [ "Multimodal AI Architecture", "Multimodal Editing", "Multimodal Feedback", "Multimodal Fusion", "Multimodal Interaction", "Multimodal Perception", "multimodal" ], "is_subclass_of": [ "AI Research Area", "AI Model Architecture" ], "wikilinks": [ "Apple", "Large Language Models", "Multimodal", "Variational Autoencoders" ] }, { "id": "multimodal-ai-architecture-ai", "title": "Multimodal AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multimodal AI refers to machine learning systems that process and integrate information from multiple data modalities \u2014 including text, images, audio, and video \u2014 simultaneously to produce contextually richer outputs than single-modality systems. These architectures employ specialised neural network fusion techniques to replicate the human brain's capacity to synthesise diverse sensory inputs, enabling applications in healthcare diagnostics, human-computer interaction, and cross-modal content generation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:multimodal-ai-architecture-ai", "labels": [ "Multimodal AI", "Multimodal" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "multimodal-ai", "title": "Multimodal Artificial Intelligence", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial intelligence systems that jointly perceive, reason over and generate content across multiple data modalities \u2014 text, images, audio, video, depth and sensor streams \u2014 by learning shared or aligned representations, enabling capabilities such as image captioning, visual question answering, text-to-image and text-to-audio generation, and speech-driven interaction that no single-modality model can provide on its own.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multimodal-ai", "labels": [ "Multimodal Artificial Intelligence", "Multimodal AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "Transformer", "Deep Learning", "Neural Network" ] }, { "id": "multimodal-interaction", "title": "Multimodal Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Multimodal interaction is a style of human-computer interaction in which users communicate with a system through two or more input or output modalities \u2014 such as speech, gesture, gaze, touch, and haptics \u2014 used in combination or interchangeably. By fusing complementary signals, multimodal systems interpret intent more robustly and naturally than any single channel, and they are central to spatial and immersive computing where physical input is rich. It is a key paradigm for designing interfaces in augmented and virtual reality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multimodal-interaction", "labels": [ "Multimodal Interaction" ], "is_subclass_of": [ "Human-Computer Interaction" ], "wikilinks": [] }, { "id": "multimodal-learning", "title": "Multimodal Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Multimodal learning is a sub-field of machine learning concerned with building models that can process, align, and reason over data from two or more sensory modalities \u2014 such as text, images, audio, video, and structured data \u2014 within a unified representation space. Models trained multimodally acquire richer, grounded representations than unimodal counterparts by exploiting cross-modal correlations and complementarity.", "entityType": "Class", "qualityScore": 0.85, "maturity": "emerging", "iri": "urn:ngm:class:multimodal-ai-architecture-learning", "labels": [ "Multimodal Learning" ], "is_subclass_of": [ "Ai Machine Learning" ], "wikilinks": [] }, { "id": "multimodal-model", "title": "Multimodal Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A machine learning model that ingests, aligns, or generates information across two or more modalities \u2014 text, images, audio, video, or 3D \u2014 within a single architecture, typically by encoding each modality into a shared representation space processed by a transformer backbone. Spanning contrastive dual encoders such as CLIP, text-to-image generators, and natively multimodal frontier models like GPT-4o and Gemini, multimodal models underpin image understanding, cross-modal search, and modern creative tooling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multimodal-model", "labels": [ "Multimodal Model" ], "is_subclass_of": [ "Foundation Model" ], "wikilinks": [ "Foundation Model", "Transformer", "Vision Language Model", "Large Language Model", "Creative AI", "Creative Tools" ] }, { "id": "multimodal-models", "title": "Multimodal Models", "domain": "ai", "domain_name": "Ai", "definition": "Multimodal models are machine learning architectures that jointly process, align, and reason over information from two or more distinct data modalities \u2014 such as text, images, audio, video, or depth \u2014 within a unified model. They learn shared or bridged representations that enable cross-modal tasks including visual question answering, image captioning, speech recognition conditioned on vision, and text-to-image synthesis. These models extend unimodal foundations (typically large language models or vision encoders) by integrating modality-specific encoders or tokenisers with cross-attention or projection layers that align heterogeneous feature spaces. Multimodal models represent the convergence of natural language processing, computer vision, and speech processing into a single, generalist AI paradigm.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:multimodal-ai-architecture-models", "labels": [ "Multimodal Models", "Multimodal Model" ], "is_subclass_of": [ "Multimodal AI" ], "wikilinks": [ "Transformer", "Text-to-Image", "Foundation Models", "Multimodal AI" ] }, { "id": "multimodal-reasoning", "title": "Multimodal Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "Multimodal reasoning is the capability of an AI system to integrate and draw inferences across multiple input modalities such as text, images, audio, and video. It requires aligning representations from heterogeneous sources into a shared semantic space so that conclusions depend jointly on all available signals. This underpins tasks like visual question answering, document understanding, and grounded dialogue where no single modality is sufficient.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:multimodal-ai-architecture-reasoning", "labels": [ "Multimodal Reasoning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "multimodal-understanding", "title": "Multimodal Understanding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multimodal Understanding is an AI research area concerned with systems that jointly process and reason over multiple sensory modalities \u2014 including text, images, audio, video, and structured data \u2014 producing unified semantic representations. It underpins vision-language models, audio-visual reasoning, and multi-sensor scene interpretation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multimodal-ai-architecture-understanding", "labels": [ "Multimodal Understanding" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "multirotor-uav", "title": "Multirotor UAV", "domain": "robotics", "domain_name": "Robotics", "definition": "A Multirotor UAV is an unmanned aerial vehicle that achieves lift and attitude control through three or more independently driven rotors. Differential rotor speed adjustment enables hover, translation, yaw, and agile manoeuvring without mechanical pitch or collective mechanisms, making multirotors highly manoeuvrable platforms suited to inspection, aerial photography, payload delivery, and search-and-rescue operations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multirotor-uav", "labels": [ "Multirotor UAV" ], "is_subclass_of": [ "Robot Type", "Aerial Robot" ], "wikilinks": [ "Aerial Robot", "Robotics" ] }, { "id": "multisig-wallet", "title": "Multisig Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A multisig wallet is a cryptocurrency wallet that requires multiple independent signatures to authorise a transaction, typically following an M-of-N threshold scheme where M of N designated keys must sign. By distributing signing authority across keys held by different people or devices, it removes single points of failure and enforces shared control. It is widely used for treasury management, custody and decentralised-organisation governance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:multisig-wallet", "labels": [ "Multisig Wallet" ], "is_subclass_of": [ "Cryptocurrency Wallet" ], "wikilinks": [] }, { "id": "multisig", "title": "Multisig", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multisig, or multi-signature, is a scheme requiring more than one cryptographic signature to authorise a transaction or action. It is widely used in cryptocurrency wallets and custody to distribute control.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:multisig", "labels": [ "Multisig", "Script Multisig" ], "is_subclass_of": [ "Wallet" ], "wikilinks": [ "Cryptography", "Custody", "DAO Governance", "Smart Contract", "Wallet" ] }, { "id": "multisignature-wallets", "title": "Multisignature Wallets", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multisignature wallets are cryptocurrency custody arrangements requiring m-of-n cryptographic signatures before a transaction can be authorised, distributing key control across multiple independent parties or devices. They eliminate single points of failure in private key management, making them a standard security architecture for institutional digital asset custody and shared treasury governance. The underlying cryptographic mechanism uses threshold signing schemes such as ECDSA or Schnorr aggregation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:multisignature-wallets", "labels": [ "Multisignature Wallets", "MPC Wallets", "Multisignature Wallet" ], "is_subclass_of": [ "Blockchain Entity" ], "wikilinks": [] }, { "id": "multisignature", "title": "Multisignature", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multisignature (multisig) is a cryptographic access-control scheme in which a transaction or operation requires a quorum of M independent signatures drawn from a declared set of N authorised keyholders before it can be validated and executed. Applied to cryptocurrency wallets, smart contracts, and DAO treasuries, the M-of-N threshold eliminates single points of failure in key custody by distributing signing authority across multiple parties, devices, or organisations. More advanced variants \u2014 threshold signature schemes (TSS) and multi-party computation (MPC) \u2014 achieve the same security guarantee without assembling the full key set on-chain, reducing transaction cost and improving privacy.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:multisignature", "labels": [ "Multisignature", "Multisignature Cryptography", "Multisignature Scheme" ], "is_subclass_of": [ "Cryptographic Signature" ], "wikilinks": [] }, { "id": "multispectral-imaging", "title": "Multispectral Imaging", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:multispectral-imaging", "labels": [ "Multispectral Imaging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "multithreading", "title": "Multithreading", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Multithreading is a programming and execution model in which a single process contains multiple threads of execution that share the process's memory and resources while running concurrently. It enables responsiveness and parallel use of multiple CPU cores, but introduces the need for synchronisation to avoid race conditions and deadlocks. The operating system scheduler interleaves or parallelises threads, and shared mutable state must be coordinated with locks or other primitives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:multithreading", "labels": [ "Multithreading" ], "is_subclass_of": [ "Concurrency", "Parallel Computing" ], "wikilinks": [] }, { "id": "multivariate-statistics", "title": "Multivariate Statistics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Multivariate statistics is the branch of statistics concerned with observing and analysing more than one outcome variable simultaneously, covering methods such as covariance and correlation structure, dimensionality reduction, and multivariate hypothesis testing.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:multivariate-statistics", "labels": [ "Multivariate Statistics" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "multiverse", "title": "Multiverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A network of interconnected but distinct metaverses and virtual worlds that enable cross-platform identity, asset portability, and interoperability while maintaining individual world sovereignty and distinct governance models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:multiverse", "labels": [ "Multiverse" ], "is_subclass_of": [ "Platform and Environment", "Semantic Spatial Web Layer" ], "wikilinks": [ "Asset Bridging", "Asset Portability", "Asset Translation Layer", "Cross-Platform Authentication", "Cross-World Travel", "Decentralized Identifier", "Distributed Governance", "Federated Identity", "Federated Social Networks", "Interoperability Protocol", "Metaverse Standards Forum", "Multi-Platform Gaming", "Multi-World Governance", "OMA3", "Protocol Translation", "Universal Inventory", "Verifiable Credential", "ApplicationLayer", "Blockchain", "Cross-Chain Bridge" ] }, { "id": "music-generation", "title": "Music Generation", "domain": "ai", "domain_name": "Ai", "definition": "Music generation is the use of artificial intelligence to compose, arrange or synthesise musical content, producing either symbolic scores or raw audio waveforms. It applies generative models such as transformers, diffusion models and autoregressive audio networks trained on large music corpora. Outputs range from melodic and harmonic structure to full instrumental and vocal renderings conditioned on text, style or reference material.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:music-generation", "labels": [ "Music Generation" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "music-information-retrieval", "title": "Music Information Retrieval", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Music Information Retrieval (MIR) is the interdisciplinary science of extracting, analysing and organising musically meaningful information from audio signals, symbolic scores and metadata. It spans tasks such as genre and mood classification, chord and key estimation, beat tracking, melody extraction, audio fingerprinting, cover-song detection and source separation, combining digital signal processing with machine learning to make large music collections searchable, navigable and analysable.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:music-information-retrieval", "labels": [ "Music Information Retrieval" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [ "Information Retrieval", "Signal Processing", "Music Generation", "Audio Processing" ] }, { "id": "music-and-audio", "title": "Music and Audio", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Music and Audio (AI-driven) is a domain within artificial intelligence encompassing the generation, transformation, classification, and production of musical audio using deep generative models, large language models conditioned on audio, latent diffusion architectures, and audio-language foundati...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:music-and-audio", "labels": [ "Music and Audio" ], "is_subclass_of": [ "AI Application", "Creative AI", "Generative AI", "Large-Scale Pretrained Foundation Model", "Audio Signal Processing", "Deep Learning", "Multimodal AI" ], "wikilinks": [ "Adaptive Music", "Advertising", "AI Mastering", "AI Music Production", "Audio Diffusion", "AudioSeal Watermarking", "Audio Signal Processing", "AudioSignalProcessingDomain", "Audio Watermarking", "Autoregressive Decoding", "C2PA Audio Standard", "CLAP", "CLAP Audio Embeddings", "CLAP Embeddings", "Classifier-Free Guidance", "Content Creation", "Contrastive Language Audio Pre-training", "Creative AI", "CreativeAIDomain", "DAC Codec" ] }, { "id": "mutual-authentication", "title": "Mutual Authentication", "domain": "security", "domain_name": "Security", "definition": "Mutual authentication is a security process in which both parties to a communication verify each other's identity before exchanging data, rather than only one party authenticating to the other. Each participant presents and validates credentials, such as digital certificates or shared secrets, establishing bidirectional trust. This prevents impersonation in both directions and is a cornerstone of secure machine-to-machine and zero-trust communication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mutual-authentication", "labels": [ "Mutual Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "mutual-tls", "title": "Mutual TLS", "domain": "security", "domain_name": "Security", "definition": "Mutual TLS is a configuration of the Transport Layer Security protocol in which both the client and the server present and verify X.509 certificates, establishing bidirectional authentication rather than authenticating only the server. Each party proves possession of the private key corresponding to its certificate, which a trusted certificate authority has signed. It is widely used to secure service-to-service communication in zero-trust architectures and microservice meshes. By binding identity to the transport channel, it prevents impersonation and unauthorised connections.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:mutual-tls", "labels": [ "Mutual TLS" ], "is_subclass_of": [ "Transport Layer Security" ], "wikilinks": [] }, { "id": "mycelium-driven-generative-choreography-system", "title": "Mycelium-Driven Generative Choreography System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An experimental creative AI project exploring the intersection of hip-hop dance, mycelium network datasets, and machine learning to generate emergent movement vocabularies. GOLM investigates whether biological network topologies\u2014particularly fungal mycelium growth patterns\u2014can be used to train generative models that produce novel choreographic sequences and spatial interaction grammars.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:mycelium-driven-generative-choreography-system", "labels": [ "Mycelium-Driven Generative Choreography System", "GOLM" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "n-triples", "title": "N-Triples", "domain": "data", "domain_name": "Data", "definition": "N-Triples is a line-based, plain-text serialization format for RDF in which each line encodes a single subject-predicate-object triple terminated by a period. It is deliberately minimal and unambiguous, using full IRIs rather than prefixes, which makes it easy to parse, stream, and compare line by line. The format is a W3C standard and serves as a canonical interchange and testing representation for RDF graphs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:n-triples", "labels": [ "N-Triples" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "nat-traversal", "title": "NAT Traversal", "domain": "blockchain", "domain_name": "Blockchain", "definition": "NAT traversal is the set of techniques that let two devices behind Network Address Translation establish a direct connection despite the address rewriting and connection-tracking that NAT imposes. Methods such as hole punching, relaying and the coordination protocols STUN, TURN and ICE allow peers to discover routable endpoints and open paths through restrictive routers. NAT traversal is essential to peer-to-peer and decentralised networks where nodes behind home or corporate routers must reach one another without a central server.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:nat-traversal", "labels": [ "NAT Traversal" ], "is_subclass_of": [ "Peer-to-Peer Network" ], "wikilinks": [] }, { "id": "nb-io-t", "title": "NB-IoT", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "NB-IoT (Narrowband Internet of Things) is a 3GPP low-power wide-area cellular standard designed to connect large numbers of simple, battery-powered devices over licensed spectrum. It trades bandwidth and latency for deep indoor coverage, long battery life, and low module cost, operating within or alongside existing LTE deployments. It is widely used for metering, asset tracking, and environmental sensing where devices transmit small amounts of data infrequently.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nb-io-t", "labels": [ "NB-IoT", "3GPP NB-IoT" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "nbic-convergence", "title": "NBIC Convergence", "domain": "governance", "domain_name": "Governance", "definition": "NBIC convergence refers to the integration of Nanotechnology, Biotechnology, Information technology, and Cognitive science into mutually reinforcing capabilities. The premise is that advances in each field accelerate the others, enabling systems that span the molecular, biological, computational, and cognitive scales. It is a foundational framing in technology-foresight and innovation-policy discussions of long-horizon transformative change.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:nbic-convergence", "labels": [ "NBIC Convergence", "NBIC Convergence Framework" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "ncsc", "title": "NCSC", "domain": "governance", "domain_name": "Governance", "definition": "The National Cyber Security Centre, the United Kingdom's national technical authority for cyber security, providing guidance and incident response support.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ncsc", "labels": [ "NCSC", "NCSC UK" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "Cybersecurity", "Incident Response", "Information Security", "Risk Management" ] }, { "id": "nft-marketplace", "title": "NFT Marketplace", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital platform enabling the creation, listing, buying, selling, and auctioning of non-fungible tokens (NFTs). NFT marketplaces provide discovery, escrow, royalty enforcement, and settlement infrastructure for unique digital assets, typically integrating smart contracts for trustless ownership transfer and on-chain provenance tracking.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:nft-marketplace", "labels": [ "NFT Marketplace" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "nft-ownership-proof", "title": "NFT Ownership Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographic mechanism by which blockchain state verifiably establishes that a specific wallet address holds title to a non-fungible token, enabling trustless transfer of digital asset ownership without intermediaries, underpinning digital rights management, access control, and provenance verification in NFT ecosystems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:nft-ownership-proof", "labels": [ "NFT Ownership Proof", "NFT Ownership Verification" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "nft-renting", "title": "NFT Renting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Process of temporarily assigning usage rights for a non-fungible token without transferring ownership, enforced through smart contract time-bound licensing mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:nft-renting", "labels": [ "NFT Renting" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Asset Utilization", "Collateral Management", "Escrow Mechanism", "MSF Use Cases", "NFT Ownership Verification", "Rental Agreement Terms", "Rental Smart Contract", "Revenue Generation", "Smart Contract Execution", "Time Lock Mechanism", "Time Oracle", "Usage Rights Token", "Blockchain Network", "Digital Asset Lending", "Digital Wallet", "MiddlewareLayer", "NFT Marketplace", "Payment System", "Temporary Asset Access", "Token Standard" ] }, { "id": "nft-standard", "title": "NFT Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An NFT Standard is a formally specified interface and data schema that defines how non-fungible tokens are created, transferred, and queried on a blockchain network, ensuring interoperability across wallets, marketplaces, and decentralised applications. Dominant examples include ERC-721 (Ethereum, unique single tokens), ERC-1155 (multi-token standard combining fungible and non-fungible types in one contract), and EIP-2981 (royalty standard). These specifications are ratified through Ethereum Improvement Proposals or analogous governance processes on other chains, and they collectively establish the canonical interface any compliant smart contract must expose.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nft-standard", "labels": [ "NFT Standard", "NFT Standard Implementation" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "nft-standards", "title": "NFT Standards", "domain": "blockchain", "domain_name": "Blockchain", "definition": "NFT Standards is a blockchain and distributed systems concept and a type of blockchain.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:nft-standards", "labels": [ "NFT Standards" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Blockchain" ] }, { "id": "nft-swapping", "title": "NFT Swapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Process of executing mutual exchange of non-fungible tokens between participants using atomic smart contract transactions that ensure simultaneous bilateral asset transfer.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:nft-swapping", "labels": [ "NFT Swapping" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Asset Liquidity", "Asset Verification", "Atomic Transaction", "Direct Exchange", "Exchange Agreement", "Gas Fee Payment", "MSF Use Cases", "Peer-to-Peer Trading", "Swap Smart Contract", "Transaction Validation", "Trust-Minimized Transfer", "Blockchain Network", "Consensus Mechanism", "Cryptographic Verification", "Decentralized Exchange", "Digital Signature", "MiddlewareLayer", "NFT Marketplace", "NFT Ownership Proof", "Token Standard" ] }, { "id": "nft-trading", "title": "NFT Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "NFT trading is the buying, selling, and exchanging of non-fungible tokens on blockchain-based marketplaces or peer-to-peer, typically settled through smart contracts that transfer token ownership and payment atomically. Prices are discovered through fixed-price listings, auctions, or offer-based negotiation, and transactions are recorded immutably on-chain. NFT trading depends on supporting infrastructure such as digital marketplaces and self-custody wallets like MetaMask for signing transactions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:nft-trading", "labels": [ "NFT Trading" ], "is_subclass_of": [ "Digital Asset Trading" ], "wikilinks": [] }, { "id": "nft-wrapping", "title": "NFT Wrapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Process of encapsulating digital assets within a new token structure to modify usage or ownership rules.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:nft-wrapping", "labels": [ "NFT Wrapping" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Asset Interoperability", "Cross-Chain Asset Transfer", "Enhanced Token Functionality", "Metadata Mapping", "MSF Use Cases", "Token Registry", "Asset Tokenization", "Blockchain Infrastructure", "Cryptographic Keys", "Digital Wallet", "MiddlewareLayer", "NFT Standard", "Smart Contract", "Token Standard", "VirtualEconomyDomain" ] }, { "id": "nft", "title": "nft", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Non-Fungible Token (NFT) is a cryptographic token deployed on a blockchain that encodes a unique, verifiable ownership claim over a specific digital or physical asset, distinguishing it from fungible tokens where every unit is interchangeable. The on-chain token record contains a unique identifier and owner address, while rich metadata and media assets are typically stored off-chain via IPFS, Arweave, or centralised hosting, with a content-addressed URI anchored in the token. Dominant standards include ERC-721 for individual unique tokens and ERC-1155 for semi-fungible batch collections on Ethereum-compatible chains, with equivalents on Solana (Metaplex), Tezos (FA2), and other ecosystems. NFTs enable programmable royalties, provenance tracking, and cross-platform digital ownership in domains spanning digital art, gaming, music, event ticketing, and real-world asset tokenisation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:nft", "labels": [ "NFT", "NFTs" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "ngsi-ld", "title": "NGSI-LD", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "NGSI-LD is an ETSI standard information model and API for representing context information as linked data using entities, properties and relationships. It is widely used in smart city and Internet of Things platforms.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ngsi-ld", "labels": [ "NGSI-LD" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [ "Linked Data", "Data Aggregation", "Internet of Things" ] }, { "id": "nicve-virtual-reality-research-centre", "title": "NICVE Virtual Reality Research Centre", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A UK academic research centre active from the late 1990s through to the early 2020s, based at the University of Salford, that advanced industrial applications of virtual reality including CAVE systems, immersive workbenches, and telepresence. NICVE succeeded the National Advanced Robotics Research Centre and produced foundational work in multi-user virtual environments, real-time rendering on SGI hardware, and later human-scale mixed reality telecollaboration. The centre evolved through multiple directors and phases before being disbanded in the early 2020s as institutional interest waned.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:nicve-virtual-reality-research-centre", "labels": [ "NICVE Virtual Reality Research Centre", "National Industrial Centre for Virtual Environments" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "nist-800-63", "title": "NIST 800-63", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Special Publication providing digital identity guidelines, covering identity proofing, authentication and federation, with defined assurance levels. It is commonly referenced as SP 800-63.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-800-63", "labels": [ "NIST 800-63" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-ai-rmf", "title": "NIST AI RMF", "domain": "governance", "domain_name": "Governance", "definition": "The NIST AI Risk Management Framework (AI RMF) is a voluntary, technology-agnostic guidance document published by the National Institute of Standards and Technology in January 2023 that helps organisations identify, assess, and manage risks arising from the design, development, deployment, and operation of AI systems. It structures AI risk management practice through four core functions \u2014 Govern, Map, Measure, and Manage \u2014 forming a continuous lifecycle cycle applicable across sectors and organisational sizes. The framework explicitly promotes seven trustworthy-AI characteristics (accountable, explainable, fair, privacy-enhanced, reliable, resilient, and secure) as unifying goals, and is supported by a companion Playbook and a growing library of sector-specific profiles, including one for generative AI.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:nist-ai-rmf", "labels": [ "NIST AI RMF", "AI RMF Playbook", "NIST AI Risk Management Framework" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "nist-ai-risk-management-framework", "title": "NIST AI Risk Management Framework", "domain": "governance", "domain_name": "Governance", "definition": "The NIST AI Risk Management Framework (AI RMF 1.0), published by the National Institute of Standards and Technology in January 2023, is a voluntary, use-case-agnostic guidance document that helps organisations design, develop, deploy, and evaluate AI systems in a manner that is trustworthy and risk-informed. It organises AI risk management activities into four core functions \u2014 GOVERN, MAP, MEASURE, and MANAGE \u2014 and supports cross-functional integration of safety, reliability, fairness, privacy, and accountability considerations throughout the AI lifecycle. The framework is accompanied by a Playbook of suggested actions and is intended to complement existing risk management practices rather than replace sector-specific regulations or standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nist-ai-risk-management-framework", "labels": [ "NIST AI Risk Management Framework" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-ai-standards", "title": "NIST AI Standards", "domain": "standards", "domain_name": "Standards", "definition": "A general reference to NIST work and publications relating to artificial intelligence standards and guidance. No single specific standard is identified by this label alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-ai-standards", "labels": [ "NIST AI Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-blockchain-technology-overview", "title": "NIST Blockchain Technology Overview", "domain": "standards", "domain_name": "Standards", "definition": "A NIST publication providing an overview of blockchain technology, its components and potential uses. It is an explanatory report rather than a normative standard.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-blockchain-technology-overview", "labels": [ "NIST Blockchain Technology Overview", "NIST Blockchain Framework" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-cryptographic-standards", "title": "NIST Cryptographic Standards", "domain": "standards", "domain_name": "Standards", "definition": "A general reference to the body of cryptographic standards and guidelines published by NIST, including FIPS and Special Publications. No single specific standard is identified by this label alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-cryptographic-standards", "labels": [ "NIST Cryptographic Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-cybersecurity-framework", "title": "NIST Cybersecurity Framework", "domain": "security", "domain_name": "Security", "definition": "The NIST Cybersecurity Framework (CSF) is a voluntary, outcomes-based risk management framework published by the US National Institute of Standards and Technology that provides organisations with a common taxonomy and structured approach for managing cybersecurity risk across sectors. It organises security activities around six core functions\u2014Govern, Identify, Protect, Detect, Respond, and Recover\u2014each decomposed into categories and subcategories cross-referenced to industry standards including ISO/IEC 27001, NIST SP 800-53, and COBIT. Version 2.0, released in February 2024, formalised the Govern function and broadened applicability beyond critical infrastructure to all organisation types and sizes globally. The framework is widely adopted as a baseline for cybersecurity programme assessment, board-level communication, supply-chain risk management, and regulatory compliance alignment.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:nist-cybersecurity-framework", "labels": [ "NIST Cybersecurity Framework" ], "is_subclass_of": [ "Security Framework" ], "wikilinks": [] }, { "id": "nist-fips-186", "title": "NIST FIPS 186", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Federal Information Processing Standard specifying the Digital Signature Standard (DSS), defining approved algorithms for generating and verifying digital signatures. It covers algorithms such as DSA, RSA and ECDSA depending on the revision.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-fips-186", "labels": [ "NIST FIPS 186", "FIPS 186", "NIST FIPS 180" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-fips-205-slh-dsa-sphincs", "title": "NIST FIPS 205 SLH-DSA SPHINCS+", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Federal Information Processing Standard specifying the Stateless Hash-Based Digital Signature Algorithm (SLH-DSA), derived from the SPHINCS+ scheme. It provides hash-based signatures designed to resist quantum attacks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-fips-205-slh-dsa-sphincs", "labels": [ "NIST FIPS 205 SLH-DSA SPHINCS+", "SLH-DSA" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-fips-205", "title": "NIST FIPS 205", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Federal Information Processing Standard specifying the stateless hash-based digital signature algorithm (SLH-DSA), based on SPHINCS+. It is one of the standardised post-quantum signature schemes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-fips-205", "labels": [ "NIST FIPS 205" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-nistir", "title": "NIST NISTIR", "domain": "standards", "domain_name": "Standards", "definition": "NIST Interagency or Internal Reports (NISTIRs) are a formal publication series issued by the National Institute of Standards and Technology covering technical research, guidelines, measurements, and collaborative investigations that do not rise to the level of a Federal Information Processing Standard (FIPS) or a Special Publication (SP). NISTIRs document foundational work \u2014 including laboratory studies, software evaluations, and preliminary standards research \u2014 produced by NIST staff or by NIST in collaboration with other U.S. government agencies. Because they are freely accessible and citable, NISTIRs frequently underpin subsequent Special Publications, ISO standards, and regulatory frameworks in cybersecurity, metrology, AI evaluation, and more.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nist-nistir", "labels": [ "NIST NISTIR" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-post-quantum-cryptography", "title": "NIST Post-Quantum Cryptography", "domain": "standards", "domain_name": "Standards", "definition": "A reference to NIST's programme and standards for cryptographic algorithms designed to resist attacks by quantum computers. It includes standardised key-establishment and digital signature schemes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-post-quantum-cryptography", "labels": [ "NIST Post-Quantum Cryptography", "NIST FIPS 204 Post-Quantum", "NIST Post-Quantum Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-privacy-framework", "title": "NIST Privacy Framework", "domain": "standards", "domain_name": "Standards", "definition": "A NIST framework providing voluntary guidance to help organisations manage privacy risks arising from data processing. It is structured around core functions to support privacy risk management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-privacy-framework", "labels": [ "NIST Privacy Framework" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-sp-1270", "title": "NIST SP 1270", "domain": "standards", "domain_name": "Standards", "definition": "NIST SP 1270 provides guidance towards a standard for identifying and managing bias in artificial intelligence.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-sp-1270", "labels": [ "NIST SP 1270" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-sp-800-207", "title": "NIST SP 800-207", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Special Publication describing zero trust architecture, a security model that treats no implicit trust based on network location. It defines concepts, components and deployment approaches for zero trust.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-sp-800-207", "labels": [ "NIST SP 800-207" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-sp-800-53", "title": "NIST SP 800-53", "domain": "standards", "domain_name": "Standards", "definition": "NIST SP 800-53 provides a catalogue of security and privacy controls for information systems and organisations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-sp-800-53", "labels": [ "NIST SP 800-53" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-sp-800-63-4", "title": "NIST SP 800-63-4", "domain": "standards", "domain_name": "Standards", "definition": "Revision 4 of the NIST Special Publication 800-63 digital identity guidelines, covering identity proofing, authentication and federation. It defines assurance levels and related requirements.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-sp-800-63-4", "labels": [ "NIST SP 800-63-4" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist-sp-800-63", "title": "NIST SP 800-63", "domain": "standards", "domain_name": "Standards", "definition": "A NIST Special Publication providing digital identity guidelines, covering enrolment, identity proofing, authentication and federation. It defines assurance levels for identity systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nist-sp-800-63", "labels": [ "NIST SP 800-63", "NIST SP 800-63B" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "NIST", "Technical Standard" ] }, { "id": "nist", "title": "NIST", "domain": "governance", "domain_name": "Governance", "definition": "The National Institute of Standards and Technology (NIST) is a non-regulatory US federal agency within the Department of Commerce, responsible for advancing measurement science, standards development, and technology innovation to promote US industrial competitiveness and public safety. NIST produces widely adopted voluntary frameworks for cybersecurity, privacy, risk management, and artificial intelligence governance, including the NIST Cybersecurity Framework and the NIST AI Risk Management Framework. Its Special Publication (SP) 800-series documents serve as mandatory guidance for US federal agencies and are globally referenced by private industry, academic institutions, and regulators. NIST also operates the AI Safety Institute Consortium (AISIC), positioning it as a key convening body for multi-stakeholder AI governance and evaluation methodology development.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:nist", "labels": [ "NIST", "NIST (National Institute of Standards and Technology)", "NIST NVD", "NIST SP 800-162", "NIST SP 800-90A", "NIST SP 800-92", "NIST Special Publication", "NIST Special Publications" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "nlptask", "title": "NLPTask", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "NLP tasks are the canonical computational problems that define the scope of natural language processing: text classification, sentiment analysis, named entity recognition, machine translation, text summarisation, and question answering. Each task specifies an input-output contract over human language and serves as a benchmark for evaluating model capability. Transformer-based architectures such as BERT and GPT have become the dominant approach across nearly all NLP tasks, replacing earlier feature-engineering and statistical methods with pre-trained, fine-tunable representations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:nlptask", "labels": [ "NLPTask" ], "is_subclass_of": [ "Computing and Cloud", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "npc-interaction", "title": "NPC Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The systems and mechanics enabling player communication and engagement with non-player characters in video games and virtual worlds, increasingly powered by AI and large language models to generate dynamic, unscripted dialogue and contextual responses.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:npc-interaction", "labels": [ "NPC Interaction" ], "is_subclass_of": [ "Interaction Technology", "Game Mechanics" ], "wikilinks": [ "Immersive Gameplay", "Game Mechanics", "metaverse" ] }, { "id": "npu", "title": "NPU", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Neural Processing Unit (NPU) is a dedicated silicon accelerator architected to execute artificial neural network operations \u2014 principally matrix multiplications and activation functions \u2014 with far greater energy efficiency and throughput than general-purpose CPUs or GPUs. NPUs are integrated into mobile SoCs, edge devices, and data-centre accelerator cards to enable low-latency AI inference on-device. They are increasingly central to deploying large language models, computer vision pipelines, and speech recognition at the edge without relying on cloud round-trips.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:npu", "labels": [ "NPU" ], "is_subclass_of": [ "AI Infrastructure", "AI Hardware", "Artificial Intelligence" ], "wikilinks": [ "AI Hardware", "Artificial Intelligence", "ArtificialIntelligenceDomain", "Edge Computing" ] }, { "id": "nvidia-corporation", "title": "NVIDIA Corporation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "NVIDIA Corporation is an American multinational technology company specialising in the design and manufacture of graphics processing units (GPUs), system-on-chip units, and related software platforms that underpin modern parallel computing workloads. Founded in 1993, NVIDIA pioneered the GPU category and subsequently extended its platform to cover artificial intelligence accelerators, high-performance computing, autonomous vehicles, robotics, and data centre infrastructure. Its CUDA parallel-computing platform, combined with purpose-built AI accelerator hardware such as the A100 and H100, has made NVIDIA the dominant supplier of compute for training large language models and other deep neural networks. NVIDIA's vertically integrated hardware-software stack\u2014spanning chips, drivers, libraries, and cloud services\u2014positions the company as foundational infrastructure for the AI era.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:nvidia-corporation", "labels": [ "NVIDIA Corporation", "NVIDIA", "NVIDIA ACE" ], "is_subclass_of": [ "GPU Computing" ], "wikilinks": [ "Hardware", "Deep Learning", "Machine Learning", "GPU Computing" ] }, { "id": "nvidia-h-100", "title": "NVIDIA H100", "domain": "ai", "domain_name": "Ai", "definition": "The NVIDIA H100 is a data-centre GPU accelerator built on the Hopper microarchitecture (GH100 die), designed for AI model training, large-scale inference, and high-performance computing workloads. It introduced a dedicated Transformer Engine with FP8 mixed-precision support, fourth-generation NVLink interconnect, and high-bandwidth HBM3 memory, enabling substantially faster training of large language models and other deep-learning workloads compared to its predecessor, the A100. The H100 is manufactured on TSMC N4 process technology and is available in PCIe and SXM5 form factors, the latter optimised for dense multi-GPU server nodes. Its combination of raw compute throughput and high-speed GPU-to-GPU interconnect made it the de-facto standard accelerator for generative AI infrastructure from 2023 onwards.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:nvidia-corporation-h-100", "labels": [ "NVIDIA H100" ], "is_subclass_of": [ "GPU", "AI Infrastructure" ], "wikilinks": [ "GPU Cluster", "GPU Computing", "NVIDIA", "GPU" ] }, { "id": "nvidia-corporation-h200", "title": "NVIDIA H200", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The NVIDIA H200 is a data-centre GPU accelerator built on the Hopper architecture, positioned as the memory-upgraded successor to the H100 within the same generation. Its defining change is the adoption of HBM3e high-bandwidth memory, raising on-package capacity to 141 GB and memory bandwidth to roughly 4.8 TB/s \u2014 a substantial increase over the H100's 80 GB of HBM3. The extra capacity and bandwidth directly benefit large-language-model inference and training, where model weights and key-value caches are memory-bound, allowing larger models or longer context windows to be served per device and improving throughput on memory-limited workloads.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:nvidia-corporation-h200", "labels": [ "NVIDIA H200" ], "is_subclass_of": [ "AI Accelerator" ], "wikilinks": [ "AI Accelerator", "NVIDIA H100", "High Bandwidth Memory" ] }, { "id": "nvidia-isaac-sim", "title": "NVIDIA Isaac Sim", "domain": "robotics", "domain_name": "Robotics", "definition": "NVIDIA Isaac Sim is a robotics simulation application built on the Omniverse platform for designing, testing, and training robots in virtual environments. It is developed by NVIDIA.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nvidia-corporation-isaac-sim", "labels": [ "NVIDIA Isaac Sim", "Isaac Lab", "Isaac Sim" ], "is_subclass_of": [ "Simulation Environment" ], "wikilinks": [ "Physics Simulation Engine", "Robot Control", "Synthetic Data", "Simulation Environment", "https://developer.nvidia.com/isaac/sim", "https://docs.isaacsim.omniverse.nvidia.com" ] }, { "id": "nvidia-jetson", "title": "NVIDIA Jetson", "domain": "robotics", "domain_name": "Robotics", "definition": "NVIDIA Jetson is a family of system-on-module (SoM) and developer kit edge computing platforms that combine NVIDIA GPU cores with ARM-based CPU clusters, purpose-built for deploying deep learning inference, computer vision, and robotics workloads at the edge with constrained power budgets. Modules in the Jetson family \u2014 including Nano, TX2, Xavier, and Orin \u2014 span from entry-level embedded devices to high-performance autonomous machine platforms, all running NVIDIA's JetPack SDK which provides CUDA, cuDNN, TensorRT, and ROS integration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nvidia-corporation-jetson", "labels": [ "NVIDIA Jetson", "Jetson" ], "is_subclass_of": [ "Edge AI Accelerator" ], "wikilinks": [] }, { "id": "nvidia-omniverse-platform", "title": "NVIDIA Omniverse Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "NVIDIA Omniverse Platform is a real-time 3D simulation, rendering, and collaborative development platform built by NVIDIA on the OpenUSD (Universal Scene Description) open standard, designed to enable physically accurate digital twins, multi-user design workflows, and large-scale synthetic data generation for AI and robotics training. It provides a unified GPU-accelerated compute and rendering fabric \u2014 comprising the Nucleus collaboration server, the Kit application framework, and Connector plugins \u2014 that allows multiple applications and users to simultaneously edit shared 3D scenes with physically based rendering, ray tracing, and physics simulation. Omniverse bridges creative content-creation pipelines, industrial engineering workflows, and AI training infrastructure, with deployments spanning automotive design, factory digital twins, autonomous vehicle simulation, and robot learning environments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:nvidia-corporation-omniverse-platform", "labels": [ "NVIDIA Omniverse Platform", "NVIDIA Omniverse" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "nvlink", "title": "NVLink", "domain": "ai", "domain_name": "Ai", "definition": "NVLink is NVIDIA's high-bandwidth, low-latency point-to-point interconnect that links GPUs directly to one another, and in some platforms to the CPU, providing far greater throughput than the PCIe bus it supplements. By creating a coherent, high-speed fabric between accelerators, NVLink enables fast peer-to-peer memory transfers and unified memory pooling across multiple GPUs. It is foundational to multi-GPU training and inference of large neural networks, where the interconnect bandwidth between devices often determines the achievable scaling efficiency of model and tensor parallelism.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:nvlink", "labels": [ "NVLink" ], "is_subclass_of": [ "GPU Computing" ], "wikilinks": [] }, { "id": "naive-bayes-classifier", "title": "Naive Bayes Classifier", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A naive Bayes classifier is a probabilistic classification algorithm that applies Bayes' theorem under the simplifying (naive) assumption that features are conditionally independent given the class label. Despite this assumption rarely holding exactly in practice, the classifier is computationally efficient, requires little training data, and performs competitively on tasks such as text classification and spam filtering. It estimates class-conditional feature distributions from training data and selects the class maximising posterior probability for a new observation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:naive-bayes-classifier", "labels": [ "Naive Bayes Classifier" ], "is_subclass_of": [ "Bayesian Decision Theory" ], "wikilinks": [] }, { "id": "nakamoto-2008-bitcoin-whitepaper", "title": "Nakamoto 2008 Bitcoin Whitepaper", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Nakamoto 2008 Bitcoin Whitepaper, titled Bitcoin: A Peer-to-Peer Electronic Cash System, is the founding document of Bitcoin, published in October 2008 under the pseudonym Satoshi Nakamoto. It describes a system for electronic payments that allows two parties to transact directly without a trusted intermediary, solving the double-spending problem through a proof-of-work timestamp server. The paper introduces a public chain of blocks secured by computational work, where the longest valid chain represents the agreed transaction history. It established the conceptual basis for cryptocurrencies and decentralised consensus.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nakamoto-2008-bitcoin-whitepaper", "labels": [ "Nakamoto 2008 Bitcoin Whitepaper", "Nakamoto 2008 Whitepaper" ], "is_subclass_of": [ "Blockchain", "Bitcoin", "owl:Thing" ], "wikilinks": [ "Bitcoin", "Proof of Work", "Blockchain Domain", "UTXO", "Cryptographic Hash Function", "owl:Thing", "Nakamoto 2008, Bitcoin: A Peer-to-Peer Electronic Cash System" ] }, { "id": "nakamoto-coefficient", "title": "nakamoto coefficient", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Nakamoto Coefficient is a quantitative metric for blockchain decentralisation, defined as the minimum number of independent entities in a given subsystem (such as mining pool concentration, validator set, or client software diversity) whose collusion or failure would be sufficient to compromise network security, liveness, or integrity. A higher Nakamoto Coefficient indicates a more resilient, decentralised network; a coefficient of one denotes a single point of failure. The metric is applicable across multiple dimensions of a network simultaneously, yielding a multi-dimensional decentralisation profile rather than a single scalar.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:nakamoto-coefficient", "labels": [ "Nakamoto Coefficient" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "nakamoto-consensus", "title": "Nakamoto Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The proof-of-work based consensus mechanism introduced in the Bitcoin whitepaper, where nodes expend computational effort to extend the longest valid chain. It achieves probabilistic Byzantine fault tolerance in open, permissionless networks through the longest-chain rule and difficulty-adjusted mining, enabling trustless agreement without a central coordinator.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:nakamoto-consensus", "labels": [ "Nakamoto Consensus" ], "is_subclass_of": [ "Consensus Protocol", "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "named-entity-recognition", "title": "Named Entity Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Named Entity Recognition (NER) is the NLP task of identifying and classifying named entities (persons, organisations, locations, dates, quantities) within unstructured text into predefined categories. NER systems employ transformer-based models (BERT, RoBERTa) with sequence labelling architectures (CRF, BiLSTM-CRF) to extract structured information from documents.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:named-entity-recognition", "labels": [ "Named Entity Recognition", "Entity Recognition" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Information Extraction", "BERT", "MetaverseDomain", "Natural Language Processing" ] }, { "id": "named-entity", "title": "Named Entity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Named Entity is a real-world object or concept denoted by a proper noun\u2014typically a person, organisation, location, geopolitical entity, date, time expression, monetary value, or product\u2014that can be identified and classified within text by a named-entity recognition system. Named entities form the primary subjects, objects, and contextual anchors of factual statements, making their accurate identification a prerequisite for downstream information-extraction, relation-extraction, question-answering, and knowledge-graph construction tasks. The boundary and category of a named entity are determined by an annotation schema and domain ontology, so the same text span may be categorised differently across biomedical, legal, and news domains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:named-entity", "labels": [ "Named Entity" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "namespace-declarations", "title": "Namespace Declarations", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "XML and RDF syntax constructs that associate short prefix identifiers with full namespace URIs, enabling the use of qualified names (QNames) to abbreviate long IRIs into human-readable yet machine-processable references in semantic web documents.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:namespace-declarations", "labels": [ "Namespace Declarations" ], "is_subclass_of": [ "Standards and Interoperability", "Semantic Web Standards" ], "wikilinks": [ "Ontology Interoperability", "metaverse", "Semantic Web Standards" ] }, { "id": "namespace-management", "title": "Namespace Management", "domain": "data", "domain_name": "Data", "definition": "Namespace management is the discipline of allocating, resolving, and governing the prefixes and base IRIs that disambiguate identifiers in a data or knowledge system. It ensures that terms drawn from different vocabularies do not collide and that abbreviated CURIE-style names resolve consistently to canonical IRIs. Effective namespace management is foundational for interoperable metadata, ontology reuse, and stable linked-data references.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:namespace-management", "labels": [ "Namespace Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "namespace", "title": "Namespace", "domain": "data", "domain_name": "Data", "definition": "A namespace is a named scope that groups a set of identifiers so that names can be reused without collision across different contexts. By qualifying each name with its containing scope, a namespace lets the same local label denote distinct entities in different parts of a system. The concept appears across programming languages, XML and RDF vocabularies, file systems, DNS, and container orchestration as a fundamental mechanism for organising and disambiguating names.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:namespace", "labels": [ "Namespace", "Namespaces" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "nanotechnology", "title": "Nanotechnology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Nanotechnology is the science and engineering of materials, devices and systems manipulated at the nanometre scale, typically between one and one hundred nanometres, where quantum and surface effects give matter properties distinct from its bulk form. It draws on material science to engineer structures such as nanoparticles, nanowires and thin films with tailored electrical, optical or mechanical properties. Nanotechnology is a key enabler of modern semiconductor manufacturing, where transistor features have shrunk to nanometre scale, as well as of advanced materials used in energy storage, coatings and medicine.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:nanotechnology", "labels": [ "Nanotechnology" ], "is_subclass_of": [ "Material Science" ], "wikilinks": [] }, { "id": "narrative-arc", "title": "Narrative Arc", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The structural framework of a story that charts the progression through exposition, rising action, climax, falling action, and resolution, creating peaks and plateaus of dramatic tension that engage audiences and shape their emotional journey through the narrative.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:narrative-arc", "labels": [ "Narrative Arc" ], "is_subclass_of": [ "Content and Assets", "Storytelling Structure" ], "wikilinks": [ "metaverse", "Non Linear Narrative", "Storytelling Structure" ] }, { "id": "narrative-content", "title": "Narrative Content", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital storytelling material that combines video, audio, images, text, and interactive elements to create immersive narrative experiences, enabling audiences to engage with stories through multiple media formats and increasingly through participatory technologies like VR, AR, and interactive video.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:narrative-content", "labels": [ "Narrative Content" ], "is_subclass_of": [ "Content and Assets", "Digital Content" ], "wikilinks": [ "Digital Content", "metaverse", "Personalized Virtual Experiences" ] }, { "id": "narrative-design-ontology", "title": "Narrative Design Ontology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Formal ontology for modeling structured storytelling frameworks, interactive narratives, story graphs, character relationships, and branching narrative paths in digital and interactive media.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:narrative-design-ontology", "labels": [ "Narrative Design Ontology" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Branching Path", "Dramatis Personae Ontology", "Event Sequencing", "Interactive Fiction", "Interactive Storytelling System", "Narrative Schema.org", "Plot Graph Database", "Plot Structure", "Procedural Story Generation", "Story Event", "Story Node", "Story Ontology", "Transmedia Storytelling", "ApplicationLayer", "Character Model", "Character Relationship Graph", "CreativeMediaDomain", "Dialogue System", "Game Narratives", "Graph Database" ] }, { "id": "narrative-structure", "title": "Narrative Structure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The organisational framework governing how a story's events, character arcs, and thematic elements are arranged across time to produce meaning and audience engagement. In spatial and interactive contexts, narrative structure governs branching paths, player agency, and the sequencing of immersive experiences within virtual worlds and game environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:narrative-structure", "labels": [ "Narrative Structure" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "narrative-theme", "title": "Narrative Theme", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The underlying message, central idea, or motif that pervades a story and shapes public perception, constructed through selective framing, language choices, and emphasis on particular details to guide audience interpretation of events, characters, and meaning.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:narrative-theme", "labels": [ "Narrative Theme" ], "is_subclass_of": [ "Content and Assets", "Narrative Structure" ], "wikilinks": [ "metaverse", "Narrative Structure", "Storytelling" ] }, { "id": "narrow-ai", "title": "Narrow AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Narrow AI refers to systems engineered to perform a specific task or bounded set of tasks, achieving strong performance within their domain without the general adaptability associated with artificial general intelligence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:narrow-ai", "labels": [ "Narrow AI" ], "is_subclass_of": [ "Artificial Intelligence", "AI Research Area" ], "wikilinks": [ "Machine Learning", "Computer Vision", "Deep Learning", "Artificial General Intelligence", "Artificial Intelligence" ] }, { "id": "nash-equilibrium", "title": "Nash Equilibrium", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Nash equilibrium is a solution concept in game theory describing a profile of strategies, one for each player, such that no player can increase their own payoff by unilaterally changing strategy while the others hold theirs fixed. It captures a stable state of mutual best responses and may be in pure or mixed strategies. Nash proved that every finite game has at least one such equilibrium, making it the central predictive concept for strategic interaction among rational agents.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:nash-equilibrium", "labels": [ "Nash Equilibrium" ], "is_subclass_of": [ "Game Theory" ], "wikilinks": [] }, { "id": "national-accounts", "title": "National Accounts", "domain": "finance", "domain_name": "Finance", "definition": "National accounts are the standardised statistical framework, such as the UN System of National Accounts, used by governments to measure the economic activity of a country, including output, income, expenditure and wealth. They provide the basis for headline indicators such as GDP and are the reference dataset against which broader indicators, including environmental and inequality measures, are typically constructed or compared. National accounts are compiled according to internationally agreed conventions to enable comparison across countries and time.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:national-accounts", "labels": [ "National Accounts" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "national-ai-strategy", "title": "National Ai Strategy", "domain": "governance", "domain_name": "Governance", "definition": "A national AI strategy is a government-led policy framework setting a country's vision, priorities and coordinated actions for developing, adopting and governing artificial intelligence. It typically addresses research and innovation funding, talent and skills, data and compute infrastructure, ethical and regulatory guardrails, public-sector adoption and economic competitiveness. Such strategies align ministries, industry and academia around shared objectives and measurable outcomes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:national-ai-strategy", "labels": [ "National Ai Strategy", "National AI Strategy" ], "is_subclass_of": [ "AI Policy" ], "wikilinks": [] }, { "id": "national-competent-authority", "title": "National Competent Authority", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A body designated by an EU Member State under the AI Act \u2014 encompassing market surveillance authorities, notifying authorities, and data protection authorities \u2014 responsible for supervising provider and deployer compliance with the Act's requirements, enforcing penalties, and coordinating with the AI Office on cross-border and GPAI model matters.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:national-competent-authority", "labels": [ "National Competent Authority", "National Supervisory Authority" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Framework" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "national-competitiveness", "title": "National Competitiveness", "domain": "governance", "domain_name": "Governance", "definition": "National competitiveness is the set of policies, institutions, and capabilities that determine a country's level of productivity and its ability to sustain growth in an open global economy. In a technology context it captures how strategic investment in research, talent, and digital infrastructure positions a nation relative to others. Technology adoption is widely treated as a primary driver because it raises productivity and creates spillover advantages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:national-competitiveness", "labels": [ "National Competitiveness" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "national-income", "title": "National Income", "domain": "finance", "domain_name": "Finance", "definition": "National income is the total income earned by a country's residents \u2014 wages, profits, rent, and interest \u2014 from the production of goods and services over an accounting period, whether that production occurs at home or abroad. It is measured within the system of national accounts, where gross national income equals gross domestic product plus net factor income from overseas, and net national income further deducts depreciation of the capital stock, making it the income-side counterpart to output measures of economic activity.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:national-income", "labels": [ "National Income" ], "is_subclass_of": [ "Macroeconomics" ], "wikilinks": [ "Macroeconomics", "Gross Domestic Product", "Inflation", "Economics" ] }, { "id": "national-security", "title": "National Security", "domain": "governance", "domain_name": "Governance", "definition": "National Security denotes the capacity of a state to protect its citizens, institutions, territorial integrity, and critical interests from foreign and domestic threats spanning military conflict, espionage, terrorism, cyber attack, and economic coercion. It encompasses both hard-power capabilities (armed forces, intelligence agencies, border controls) and soft-power instruments (diplomacy, economic policy, information operations). Modern national security doctrine incorporates cyber resilience, supply-chain integrity, energy security, and AI competitiveness as dimensions of strategic concern. Governance frameworks such as security councils, export-control regimes, and classified classification schemes operationalise national security policy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:national-security", "labels": [ "National Security" ], "is_subclass_of": [ "Policy" ], "wikilinks": [] }, { "id": "national-sovereign-ai-procurement-initiative", "title": "National Sovereign AI Procurement Initiative", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Sovereign AI bid is a state or regional initiative to develop, procure, or deploy AI infrastructure and foundation models under national control, reducing dependence on foreign hyperscalers and ensuring that AI capabilities, data residency, and compute sovereignty remain within a defined jurisdiction. Such bids typically involve national compute investment, open-source model adoption, and alignment with local regulatory frameworks including the EU AI Act.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:national-sovereign-ai-procurement-initiative", "labels": [ "National Sovereign AI Procurement Initiative", "Sovereign AI bid" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Visionflow" ] }, { "id": "native-audio-generation", "title": "Native Audio Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The capability of a multimodal generative model to synthesize synchronized audio tracks simultaneously with video or image content.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:native-audio-generation", "labels": [ "Native Audio Generation" ], "is_subclass_of": [ "Multimodal AI" ], "wikilinks": [] }, { "id": "native-token", "title": "Native Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Native Token is the primary cryptocurrency issued and managed directly by a blockchain protocol, used to pay transaction fees, reward validators or miners, and serve as the base unit of value within the network's economic system. Unlike tokens created via smart contracts, native tokens are settled at the protocol layer and are intrinsic to chain operation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:native-token", "labels": [ "Native Token", "Native Currency", "Native Token Distribution" ], "is_subclass_of": [ "Token and Asset", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "natural-interaction", "title": "Natural Interaction", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Natural Interaction describes human-computer interaction modalities that leverage innate human capabilities\u2014speech, gesture, gaze, touch, and whole-body movement\u2014rather than requiring users to learn abstract symbolic interfaces such as keyboards or command-line syntax. The goal is to reduce cognitive translation costs and make technology accessible to a broader range of users by aligning the interaction channel with the way humans naturally communicate and manipulate objects in the physical world. Natural interaction is foundational to spatial computing, mixed reality, and embodied AI systems where traditional 2D pointing devices are ergonomically or practically unsuitable.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:natural-interaction", "labels": [ "Natural Interaction", "Natural Communication", "Natural VR Interaction" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "natural-language-generation", "title": "Natural Language Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Natural Language Generation (NLG) is the subfield of natural language processing concerned with producing coherent, contextually appropriate human-readable text from structured data, internal representations, or prompts. It encompasses content determination, sentence planning, and surface realisation, and is today dominated by neural language models that generate text token by token. NLG is the productive counterpart to natural language understanding within end-to-end conversational and generative systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:natural-language-generation", "labels": [ "Natural Language Generation" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "natural-language-processing", "title": "Natural Language Processing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Natural Language Processing (NLP) is the subfield of AI focused on enabling computers to understand, interpret, generate, and manipulate human language. Core tasks include text classification, named entity recognition, machine translation, sentiment analysis, question answering, and language generation, underpinned by transformer architectures and large-scale pre-training.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:natural-language-processing", "labels": [ "Natural Language Processing", "Low-Resource Language Processing", "Natural Language Processing Pipeline", "NaturalLanguageProcessing" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Large Language Models", "Machine Translation", "Text Generation", "Transformers" ] }, { "id": "natural-language-understanding", "title": "natural language understanding", "domain": "ai", "domain_name": "Ai", "definition": "Natural Language Understanding (NLU) is the subfield of Natural Language Processing dedicated to enabling machines to comprehend the meaning, intent, and pragmatic context of human language input beyond surface-level syntax. Core NLU tasks include semantic role labelling, named entity recognition, coreference resolution, intent classification, relation extraction, and natural language inference, all requiring models to construct structured semantic representations of utterance meaning. Modern NLU systems are predominantly built on large pre-trained Transformer architectures such as BERT and its derivatives, which learn contextualised word representations from massive corpora via self-supervised objectives. NLU serves as the comprehension layer in conversational AI, information extraction pipelines, question answering systems, and dialogue management frameworks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:natural-language-understanding", "labels": [ "Natural Language Understanding" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "natural-satellite", "title": "Natural Satellite", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:natural-satellite", "labels": [ "Natural Satellite" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "nav2", "title": "Nav2", "domain": "robotics", "domain_name": "Robotics", "definition": "Nav2 (Navigation2) is the production-ready autonomous navigation framework for ROS 2, providing a composable, lifecycle-managed stack of planners, controllers, costmap layers, and behaviour-tree-based task orchestration. It enables mobile robots to safely compute and execute collision-free paths in structured and semi-structured environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nav2", "labels": [ "Nav2", "ROS 2 Nav2" ], "is_subclass_of": [ "Navigation Stack" ], "wikilinks": [] }, { "id": "navigation-mesh", "title": "Navigation Mesh", "domain": "ai", "domain_name": "Ai", "definition": "A navigation mesh (navmesh) is a data structure that decomposes the traversable area of a virtual environment into a set of convex polygons over which an agent can move freely. By abstracting walkable space into connected regions, it enables efficient pathfinding without searching a dense uniform grid. Navmeshes are the standard representation for movement and obstacle avoidance in game AI and simulated 3D worlds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:navigation-mesh", "labels": [ "Navigation Mesh" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "navigation-stack", "title": "Navigation Stack", "domain": "robotics", "domain_name": "Robotics", "definition": "A navigation stack is the integrated software subsystem of a mobile robot responsible for taking it from a current pose to a goal pose while avoiding obstacles. It composes mapping, localisation, global path planning, and local trajectory control into a coordinated pipeline. The stack consumes sensor data and a map, and emits velocity commands that drive the robot's actuators.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:navigation-stack", "labels": [ "Navigation Stack", "Nav2 Navigation Stack" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "navigation-system", "title": "Navigation System", "domain": "robotics", "domain_name": "Robotics", "definition": "An integrated set of hardware and software components that enables a robot or autonomous agent to determine its position, plan collision-free paths, and execute motion towards a goal. Navigation systems typically combine localisation, mapping, path planning, and obstacle avoidance modules, often relying on sensor fusion from LiDAR, cameras, and IMUs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:navigation-system", "labels": [ "Navigation System" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "navigation", "title": "Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Navigation is the autonomous discipline integrating localisation, mapping, path planning, motion control, and semantic interpretation to enable agents \u2014 ground robots, aerial vehicles, autonomous cars, and embodied AI systems \u2014 to move reliably from a start configuration to a goal state throu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:navigation", "labels": [ "Navigation", "Experience Navigation", "Global Navigation Satellite System", "Robotics Navigation", "Safe Navigation" ], "is_subclass_of": [ "Navigation and Planning", "Path Planning", "Autonomous System", "Autonomous Systems", "Motion Planning", "Robotics", "Control Theory" ], "wikilinks": [ "AlgorithmLayer", "Anderson et al. 2018 VLN R2R", "Autonomous Systems", "AutonomousSystemsDomain", "Autonomous Vehicles", "Behaviour Trees", "Brohan et al. 2023 RT-2", "Chen et al. 2019 CrowdNav", "Chi et al. 2023 Diffusion Policy", "ControlAndPlanningDomain", "Costmap", "Cummins Newman 2008 FAB-MAP", "D Star Lite", "Drone Navigation", "DWA", "Fox et al. 1997 DWA", "Gammell et al. 2014 Informed RRT Star", "Gaussian Processes", "Global Planner", "Hart et al. 1968 A Star Algorithm" ] }, { "id": "ne-rf", "title": "NeRF", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Neural Radiance Field (NeRF) is a neural scene representation technique that encodes a continuous volumetric scene as a multi-layer perceptron mapping 3D coordinates and viewing directions to colour and density values, enabling photorealistic novel view synthesis from a sparse set of calibrated input images. Introduced by Mildenhall et al. at ECCV 2020, NeRF employs volumetric ray marching and differentiable rendering to optimise network weights via photometric loss against held-out views, without requiring explicit mesh or voxel geometry. The technique catalysed a broad family of neural rendering methods spanning real-time variants, dynamic scenes, large-scale outdoor capture, and hybrid neural-explicit representations such as 3D Gaussian Splatting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ne-rf", "labels": [ "NeRF" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "near-field-communication", "title": "Near Field Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Near Field Communication (NFC) is a short-range wireless communication technology operating at 13.56 MHz that enables contactless data exchange between devices within approximately 4 centimetres. Based on inductive coupling and derived from RFID standards, NFC supports three operating modes: card emulation (device acts as a smart card), reader/writer (device reads or writes NFC tags), and peer-to-peer (two active devices exchange data). It underpins contactless payments, identity verification, and rapid device pairing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:near-field-communication", "labels": [ "Near Field Communication", "Near-Field Communication" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "nearest-neighbor-search", "title": "Nearest Neighbor Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Nearest Neighbor Search (NNS) is an algorithmic problem and family of techniques for finding the point(s) in a dataset most similar to a given query point, as measured by a distance or similarity metric such as Euclidean distance, cosine similarity, or inner product. Exact NNS guarantees retrieval of the true closest point but scales poorly in high dimensions, while Approximate Nearest Neighbor (ANN) methods trade a bounded loss in recall for dramatically faster query latency and memory efficiency. Foundational index structures \u2014 including KD-Trees, Ball Trees, Locality-Sensitive Hashing, Inverted File Indexes, and Hierarchical Navigable Small World graphs \u2014 each make different trade-offs between construction cost, query speed, recall, and support for dynamic updates. NNS is a core primitive of modern machine learning pipelines, powering vector database retrieval, recommendation systems, semantic search, image recognition, and retrieval-augmented generation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:nearest-neighbor-search", "labels": [ "Nearest Neighbor Search", "Approximate Nearest Neighbour", "Approximate Nearest Neighbour Search", "Exact Nearest Neighbour", "Nearest Neighbour Search", "Nearest-Neighbour Search" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [ "metaverse", "Search Algorithm", "Semantic Search" ] }, { "id": "negotiation-protocol", "title": "Negotiation Protocol", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A negotiation protocol is a formally specified interaction pattern governing how autonomous agents exchange proposals, counter-proposals, and commitments in order to reach agreement over resources, tasks, or joint plans. It defines the permitted message types, turn-taking rules, and termination conditions, separating the public rules of engagement from each agent's private negotiation strategy. Canonical examples include the Contract Net Protocol, alternating-offers bargaining, and the FIPA standardised interaction protocols.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:negotiation-protocol", "labels": [ "Negotiation Protocol" ], "is_subclass_of": [ "Coordination Mechanisms" ], "wikilinks": [ "Coordination Mechanisms", "Multi-Agent Coordination", "Negotiation", "Conflict Resolution" ] }, { "id": "negotiation", "title": "Negotiation", "domain": "governance", "domain_name": "Governance", "definition": "Negotiation is the structured interactive process by which two or more parties with partially conflicting interests communicate proposals and concessions to reach a mutually acceptable agreement without recourse to coercion or third-party adjudication. It draws on principled bargaining, interest analysis, and reservation-value reasoning to expand and divide the zone of possible agreement. In decentralised and automated governance contexts, negotiation increasingly manifests as protocol-mediated bargaining among software agents, stakeholders, and on-chain coalitions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:negotiation", "labels": [ "Negotiation" ], "is_subclass_of": [ "Conflict Resolution" ], "wikilinks": [] }, { "id": "neo-4-j", "title": "Neo4j", "domain": "data", "domain_name": "Data", "definition": "A graph database management system that stores data as nodes and relationships and queries it with the Cypher query language. It is one of the most widely used native graph databases.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:neo-4-j", "labels": [ "Neo4j" ], "is_subclass_of": [ "Graph Database" ], "wikilinks": [ "Graph Database", "Knowledge Graph", "Database System" ] }, { "id": "net-neutrality", "title": "Net Neutrality", "domain": "governance", "domain_name": "Governance", "definition": "Net neutrality is the principle that internet service providers should treat all data traversing their networks equally, without discriminating, blocking, throttling, or charging differentially based on the source, destination, application, or content of that traffic. It aims to preserve an open internet where users and content providers compete on merit rather than on access deals. Net neutrality is enforced or challenged through telecommunications regulation and is a recurring subject of internet-governance policy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:net-neutrality", "labels": [ "Net Neutrality" ], "is_subclass_of": [ "Internet Governance" ], "wikilinks": [] }, { "id": "net-zero-target-setting", "title": "Net Zero Target Setting", "domain": "governance", "domain_name": "Governance", "definition": "Net zero target setting is the structured process by which an organisation or jurisdiction defines, scopes, and commits to balancing its greenhouse-gas emissions with removals by a stated date. It involves establishing a baseline, defining the emission scopes covered, setting interim milestones, and aligning the trajectory with a science-based decarbonisation pathway. Credible targets specify boundaries and the role of offsets to avoid greenwashing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:net-zero-target-setting", "labels": [ "Net Zero Target Setting" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "net-zero-targets", "title": "Net Zero Targets", "domain": "governance", "domain_name": "Governance", "definition": "Net Zero Targets are formal, time-bound commitments made by governments, corporations, subnational jurisdictions, or other entities to reduce gross greenhouse gas emissions as deeply as feasible and to balance any residual emissions with equivalent, verified carbon removals, thereby reaching a net-zero contribution to atmospheric greenhouse gas concentrations by a specified date. They typically distinguish near-term absolute reduction milestones across Scope 1, 2, and 3 emissions from longer-term residual-offset strategies, and are calibrated against the IPCC's 1.5 \u00b0C-consistent pathways. Credible targets are characterised by transparent interim milestones, independent third-party verification, and a clear accounting boundary that separates genuine emission reductions from carbon offsetting.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:net-zero-targets", "labels": [ "Net Zero Targets", "Net Zero Achievement", "Net Zero Strategy", "Net-Zero Commitments", "Net-Zero Target" ], "is_subclass_of": [ "Climate Commitments" ], "wikilinks": [] }, { "id": "net-zero-transition", "title": "Net Zero Transition", "domain": "finance", "domain_name": "Finance", "definition": "The net zero transition is the structural shift of economies and financial portfolios toward balancing residual greenhouse-gas emissions with removals, in line with limiting global warming. It mobilises capital toward low-carbon technologies, retires or retrofits high-emitting assets, and aligns corporate strategy and disclosure with decarbonisation pathways. Transition finance provides the funding and instruments that allow hard-to-abate sectors to decarbonise credibly over time.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:net-zero-transition", "labels": [ "Net Zero Transition" ], "is_subclass_of": [ "Climate Finance" ], "wikilinks": [] }, { "id": "net-zero", "title": "Net Zero", "domain": "governance", "domain_name": "Governance", "definition": "Net zero is the state in which the greenhouse gases added to the atmosphere by human activity are balanced by an equivalent amount removed, so that net emissions over a given scope and timeframe are zero. Achieving it combines deep decarbonisation across energy, industry, transport, and land use with carbon removal to neutralise residual emissions that cannot yet be eliminated. Net zero is the central organising target of contemporary climate policy, codified in commitments such as the Paris Agreement, and is operationalised by governments and organisations through science-based pathways and reporting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:net-zero", "labels": [ "Net Zero" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "netting", "title": "Netting", "domain": "finance", "domain_name": "Finance", "definition": "The offsetting of mutual obligations between counterparties so that only the net balance is settled, rather than each gross obligation individually. Netting compresses large volumes of bilateral or multilateral exposures into a single payable or receivable per party, dramatically reducing settlement volumes, liquidity needs, and counterparty credit risk. Legally enforceable forms include payment netting, novation netting, and close-out netting on default, and it is the core economic function performed by clearing houses and payment systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:netting", "labels": [ "Netting" ], "is_subclass_of": [ "Clearing" ], "wikilinks": [ "Clearing", "Settlement", "Settlement Finality", "Central Securities Depository" ] }, { "id": "network-addressing", "title": "Network Addressing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network addressing is the scheme by which devices and services on a network are assigned unique identifiers \u2014 such as IP addresses or protocol-specific identifiers \u2014 that allow packets to be routed to the correct destination. It is a foundational function of network protocols and a prerequisite for service discovery in distributed systems, which must resolve human- or service-readable names to addressable network locations. Addressing schemes range from flat, globally unique assignments to hierarchical, topology-aware allocations that support efficient routing.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:network-addressing", "labels": [ "Network Addressing" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "network-analysis", "title": "Network Analysis", "domain": "ai", "domain_name": "Ai", "definition": "Network Analysis is the systematic study of complex systems modelled as graphs of nodes and edges, applying mathematical, statistical, and computational techniques to characterise topology, detect communities, identify influential vertices, and model dynamic processes such as information diffusion, contagion, and resilience. Rooted in graph theory, sociology, and statistical physics, it encompasses structural metrics including degree distribution, betweenness centrality, clustering coefficients, and average path length. Machine learning on graphs \u2014 principally through graph neural networks \u2014 has substantially extended the field beyond classical descriptive statistics toward predictive and generative capabilities. Applications span social network mapping, cybersecurity threat detection, supply chain vulnerability assessment, biological pathway modelling, and knowledge graph quality evaluation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-analysis", "labels": [ "Network Analysis", "Network Protocol Analysis" ], "is_subclass_of": [ "Graph Theory" ], "wikilinks": [] }, { "id": "network-architecture", "title": "Network Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The structural design of a neural network, specifying the arrangement of layers, connection patterns, activation functions, skip connections, normalisation methods, and attention mechanisms. Key architectures include feedforward networks, CNNs, RNNs, transformers, and graph neural networks; Neural Architecture Search automates discovery of optimal configurations for accuracy, efficiency, and hardware constraints.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:network-architecture", "labels": [ "Network Architecture" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "Convolutional Neural Networks", "Neural Architecture Search", "Neural Network", "Transformers" ] }, { "id": "network-bandwidth", "title": "Network Bandwidth", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network bandwidth is the maximum rate at which data can be transferred across a network path, typically measured in bits per second. It defines the capacity of a communication channel rather than its current utilisation or the time taken for an individual message to traverse the path. Bandwidth interacts with latency and packet loss to determine the effective throughput experienced by applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:network-bandwidth", "labels": [ "Network Bandwidth" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "network-communication", "title": "Network Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Communication is the transmission of data between computing nodes via shared or dedicated media, governed by layered protocol stacks that abstract physical signal propagation into reliable, addressable data exchange. It encompasses all paradigms of machine-to-machine data transfer including wired, wireless, and optical mediums organised according to reference models such as the OSI seven-layer model and TCP/IP suite.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:network-communication", "labels": [ "Network Communication", "Network Communication Model" ], "is_subclass_of": [ "Communication Network" ], "wikilinks": [] }, { "id": "network-component", "title": "Network Component", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Network Component represents the fundamental infrastructure elements that constitute blockchain networks, including different node types, network protocols, and communication layers.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:network-component", "labels": [ "Network Component", "Network Component (Blockchain)" ], "is_subclass_of": [ "Blockchain Entity" ], "wikilinks": [ "Blockchain Technology", "Blockchain Entity" ] }, { "id": "network-congestion", "title": "Network Congestion", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Network congestion is the state in which the demand for a network's processing capacity exceeds its available throughput, causing transactions to queue and confirmation times and fees to rise. On a blockchain it occurs when the volume of pending transactions outstrips the space available in upcoming blocks, filling the mempool and triggering competitive fee bidding. Congestion exposes the scalability limits of a system and is a primary driver of fee market dynamics and layer-2 adoption. It is both a symptom of demand and a constraint that shapes protocol design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:network-congestion", "labels": [ "Network Congestion" ], "is_subclass_of": [ "Blockchain", "Transaction Pool" ], "wikilinks": [] }, { "id": "network-connectivity", "title": "network connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Connectivity is the capability of computing nodes, devices, or systems to establish, maintain, and exchange data across communication links spanning local area networks, wide area networks, and heterogeneous internet topologies. It encompasses the physical transmission medium (copper, fibre, radio), link-layer and network-layer addressing, routing protocols that guide packets across autonomous systems, and transport-layer mechanisms that provide reliability, ordering, and flow control. Together these layers enable end-to-end data delivery on which all higher-order distributed services \u2014 from cloud orchestration and edge computing to consensus protocols and collaborative applications \u2014 ultimately depend.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:network-connectivity", "labels": [ "Network Connectivity" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "network-effects", "title": "Network Effects", "domain": "finance", "domain_name": "Finance", "definition": "Network effects describe the phenomenon whereby a product or service gains additional value for each existing user as the number of participants increases, creating a self-reinforcing growth dynamic. Direct network effects arise when users directly benefit from others on the same platform; indirect effects emerge when complementary products or services multiply with scale, creating cross-side externalities.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:network-effects", "labels": [ "Network Effects", "Network Effect" ], "is_subclass_of": [ "Positive Feedback" ], "wikilinks": [] }, { "id": "network-fabric", "title": "Network Fabric", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network fabric is a switched interconnect topology, typically non-blocking and multi-path such as a fat-tree or dragonfly, that provides uniform high-bandwidth, low-latency connectivity between many compute nodes. It is the physical and logical substrate over which technologies such as InfiniBand move data between GPUs or servers in a cluster. It underpins distributed training and high-performance computing workloads that depend on collective communication primitives such as the Message Passing Interface.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:network-fabric", "labels": [ "Network Fabric" ], "is_subclass_of": [ "Network Layer" ], "wikilinks": [] }, { "id": "network-function-virtualization", "title": "Network Function Virtualization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Function Virtualization (NFV) is an architectural paradigm that decouples network functions \u2014 such as firewalls, load balancers, intrusion detection systems, and routers \u2014 from dedicated proprietary hardware appliances and implements them as software running on commercial off-the-shelf (COTS) servers, switches, and storage. Standardised by ETSI's NFV Industry Specification Group, NFV introduces a three-layer model comprising the NFV Infrastructure (NFVI), Virtual Network Functions (VNFs), and the NFV Management and Orchestration (MANO) framework. By enabling elastic provisioning, rapid deployment, and lifecycle automation of network services, NFV reduces capital and operational expenditure while providing the flexibility required for 5G, edge computing, and cloud-native telecom architectures. It is closely related to and operationally complementary with Software-Defined Networking (SDN), which separates the network control plane from the data plane.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:network-function-virtualization", "labels": [ "Network Function Virtualization", "Network Function Virtualisation", "Network Functions Virtualisation" ], "is_subclass_of": [ "Network Architecture" ], "wikilinks": [ "Network Scalability", "metaverse", "Network Architecture" ] }, { "id": "network-hash-rate", "title": "Network Hash Rate", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The aggregate computational power (measured in hashes per second) expended by all miners participating in a proof-of-work blockchain at a given time. A higher network hash rate increases attack cost for a 51% attack and triggers upward difficulty adjustments to maintain target block times; conversely, drops in hash rate trigger downward adjustments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:network-hash-rate", "labels": [ "Network Hash Rate", "NetworkHashRate" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "network-infrastructure", "title": "Network Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Infrastructure is the integrated ensemble of physical hardware, logical software layers, and communication protocols that enable data transmission, interconnection, and service delivery across computing environments. It encompasses wired and wireless transmission media, routing and switching equipment, software-defined networking (SDN) control planes, and edge computing nodes that collectively form the backbone of digital communication. Modern network infrastructure is characterised by convergence across 5G/6G mobile networks, fibre optic backbones, satellite constellations, and multi-access edge computing (MEC) to deliver low-latency, high-throughput connectivity. It underpins virtually all digital services \u2014 from cloud computing and AI inference to immersive spatial experiences, autonomous systems, and distributed enterprise collaboration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:network-infrastructure", "labels": [ "Network Infrastructure", "6G Network Infrastructure", "Local Network Infrastructure", "NetworkInfrastructure" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "network-interface-card", "title": "Network Interface Card", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network interface card (NIC) is the hardware component that connects a computing device to a network and implements the physical and data-link layer functions for sending and receiving frames. It encodes outgoing data onto the transmission medium, decodes incoming signals, and carries a unique MAC address for link-layer addressing. Modern NICs often offload checksum, segmentation, and encryption work from the CPU.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-interface-card", "labels": [ "Network Interface Card", "Network Interface Controller" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "network-interface", "title": "Network Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network interface is the hardware or software point at which a computing device connects to a network, mediating the transmission and reception of data frames. In hardware it is realised as a network interface controller bearing a unique MAC address; in software it appears as a named, addressable endpoint that the operating system binds to a protocol stack. The interface bridges the physical and data-link layers to the network layer, allowing higher-level protocols to send and receive packets without managing the underlying medium.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:network-interface", "labels": [ "Network Interface" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [] }, { "id": "network-interoperability", "title": "Network Interoperability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network interoperability is the ability of independently operated networks or systems to exchange information and use it correctly through agreed protocols and data formats. It depends on shared standards at the relevant layers so that heterogeneous implementations communicate without bespoke adaptation. In distributed and blockchain contexts it is the precondition for cross-network value transfer and coordinated operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-interoperability", "labels": [ "Network Interoperability" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "network-latency", "title": "Network Latency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Latency is the round-trip communication delay between nodes in a network, directly constraining consensus speed, block propagation, transaction throughput, real-time application responsiveness, and distributed system coordination. High latency increases the probability of forks in blockchain systems, degrades user experience in interactive applications, and sets a fundamental lower bound on achievable synchronisation intervals in geographically distributed deployments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:network-latency", "labels": [ "Network Latency", "NetworkLatency" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "network-layer", "title": "Network Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Architectural layer governing communication protocols, packet routing, congestion control, and network topology. Provides reliable end-to-end message delivery, bandwidth management, and quality-of-service guarantees for distributed systems across heterogeneous networks.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:network-layer", "labels": [ "Network Layer", "NetworkLayer" ], "is_subclass_of": [ "Network and Communication", "ArchitecturalLayer" ], "wikilinks": [ "ArchitecturalLayer", "Bandwidth Allocation", "Congestion Control", "Latency Management", "Message Delivery", "Narrative Gold Mine", "Network Resilience", "Protocol Stack", "QoS Manager", "Routing Engine", "Infrastructure Layer", "Middleware Layer", "Network Topology", "Physical Layer" ] }, { "id": "network-participation", "title": "Network Participation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Network participation is the act of an entity contributing resources or actions to a decentralised network in exchange for the right to influence and benefit from it. In blockchain systems this includes running nodes, validating or proposing blocks, providing liquidity, or voting, typically backed by staked capital or computational work. Participation rules and incentives determine the network's security, decentralisation, and reward distribution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-participation", "labels": [ "Network Participation" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "network-partition-tolerance", "title": "Network Partition Tolerance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Network partition tolerance is the property of a distributed system that allows it to continue operating despite the network splitting into groups of nodes that cannot communicate with one another. It is one of the three properties in the CAP theorem, which states that during a partition a system must sacrifice either strong consistency or availability. In blockchain and distributed databases, partition tolerance is generally treated as non-negotiable because partitions are inevitable in real networks, forcing explicit design choices about how the system behaves when nodes are split.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:network-partition-tolerance", "labels": [ "Network Partition Tolerance" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "network-partition", "title": "Network Partition", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A network partition is a failure condition in a distributed system where communication is severed between subsets of nodes, splitting the cluster into groups that cannot exchange messages. Partitions force a trade-off, formalised by the CAP theorem, between maintaining consistency and remaining available while the split persists. Detecting, tolerating, and recovering from partitions is a central concern of distributed-systems design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:network-partition", "labels": [ "Network Partition" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "network-performance-metrics", "title": "Network Performance Metrics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network Performance Metrics are quantitative measures used to characterise the operational quality of a communication network, encompassing latency, throughput, packet loss rate, jitter, and availability. These metrics are collected continuously from network devices, links, and endpoints to support quality-of-service enforcement, capacity planning, fault detection, and SLA compliance verification. They underpin all layers of network management from physical links to application-level experience.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:network-performance-metrics", "labels": [ "Network Performance Metrics", "Network Performance" ], "is_subclass_of": [ "Quality Of Service" ], "wikilinks": [] }, { "id": "network-protocol", "title": "Network Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network protocol is a set of established rules that specify how to format, send, and receive data between networked devices, enabling diverse communication systems to interact using standard procedures. Protocols are organized into layered architectures (OSI, TCP/IP) and govern addressing, routing, error detection, session management, and application-layer services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:network-protocol", "labels": [ "Network Protocol", "Network Protocols" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "Data Transmission", "Network Communication", "ETSI_Domain_Interoperability", "InfrastructureDomain", "Interoperability", "Technology Domain" ] }, { "id": "network-quality-metric", "title": "Network Quality Metric", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A quantitative measure used to characterise the performance and reliability of a network connection for real-time interactive applications, encompassing parameters such as latency (round-trip time), jitter, packet loss rate, available bandwidth, and connection stability. These metrics directly determine the feasibility of synchronised multi-user experiences in spatial computing environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:network-quality-metric", "labels": [ "Network Quality Metric" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "network-resilience", "title": "Network Resilience", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network resilience is a network's capacity to maintain an acceptable level of connectivity and service in the face of node or link failure, congestion, attack or partition, typically achieved through redundancy, self-healing routing and peer diversity. It is measured by how quickly and completely the network restores full functionality after a disruptive event. In peer-to-peer and blockchain networks it depends heavily on robust peer discovery mechanisms that keep nodes well connected.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:network-resilience", "labels": [ "Network Resilience" ], "is_subclass_of": [ "Resilience" ], "wikilinks": [] }, { "id": "network-science", "title": "Network Science", "domain": "data", "domain_name": "Data", "definition": "Network science is the interdisciplinary study of complex systems represented as graphs of nodes and edges, focusing on the structure, dynamics, and function of connections. It develops models and metrics, such as degree distributions, centrality, community structure, and small-world properties, to explain how topology shapes behaviour. It is applied across social, biological, technological, and economic networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-science", "labels": [ "Network Science" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "network-security", "title": "Network Security", "domain": "security", "domain_name": "Security", "definition": "Network security is the practice of protecting computer networks and the data transmitted across them from unauthorised access, misuse, modification, or denial of service through a combination of hardware controls, software policies, and operational procedures. It encompasses perimeter defence, intrusion detection and prevention, traffic analysis, encryption of data in transit, and access control applied at network boundaries and within internal segments. As networks have evolved from isolated LANs to globally distributed cloud and edge architectures, network security has broadened to encompass zero-trust models, software-defined perimeters, and AI-driven anomaly detection. It forms a foundational layer of broader cybersecurity strategy, intersecting with identity management, cryptographic standards, and regulatory compliance frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:network-security", "labels": [ "Network Security", "NetworkSecurity" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "network-segmentation", "title": "Network Segmentation", "domain": "security", "domain_name": "Security", "definition": "Network segmentation is the security practice of dividing a computer network into smaller, isolated zones so that traffic between them is controlled, inspected, and restricted by policy. By limiting the blast radius of a compromise, segmentation prevents an attacker who breaches one zone from moving laterally to others. It is a core control in defence-in-depth and a prerequisite for zero-trust network architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:network-segmentation", "labels": [ "Network Segmentation" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "network-slicing", "title": "Network Slicing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Network slicing is a virtualisation architecture that partitions a single physical network infrastructure into multiple isolated, logically independent virtual networks \u2014 called slices \u2014 each configured to provide the specific performance characteristics, security posture, and service topology required by a distinct application or tenant. Defined within 3GPP Release 15 and later releases for 5G networks, each slice is instantiated through software-defined networking and network function virtualisation, enabling differentiated service-level guarantees \u2014 such as ultra-low latency for industrial automation or high-throughput for video streaming \u2014 without dedicated physical hardware. Slices span radio access, transport, and core network domains and can be dynamically provisioned, scaled, and torn down.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:network-slicing", "labels": [ "Network Slicing", "5G Network Slicing", "NetworkSlicing" ], "is_subclass_of": [ "Network Function Virtualization" ], "wikilinks": [] }, { "id": "network-standards", "title": "Network Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical specifications and protocols developed by standards bodies like IEEE, IETF, ITU, and ISO that define how network devices communicate, ensuring interoperability, security, and performance across telecommunications and internet infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:network-standards", "labels": [ "Network Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Standards" ], "wikilinks": [ "Network Interoperability", "metaverse", "Technical Standards" ] }, { "id": "network-switch", "title": "Network Switch", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network switch is a hardware device that connects devices within a local area network and forwards data frames between them based on MAC addresses at the data link layer. By learning which addresses sit on which ports, it forwards traffic only to its destination, improving efficiency over shared media. Switches are foundational building blocks of wired networks in offices, campuses and data centres.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:network-switch", "labels": [ "Network Switch" ], "is_subclass_of": [ "Networking Infrastructure" ], "wikilinks": [] }, { "id": "network-synchronization", "title": "Network Synchronization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Network Synchronization is the process by which blockchain nodes align their local chain state with the canonical ledger through block propagation, header-first sync, and fork resolution protocols. It governs initial block download, peer discovery, and the ongoing receipt of new blocks and transactions via gossip, ensuring all honest participants converge on an identical view of the distributed ledger despite network partitions and latency.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:network-synchronization", "labels": [ "Network Synchronization", "Network Synchronisation" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "network-theory", "title": "Network Theory", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Network theory is the study of graphs representing relations between objects, analysing structure, connectivity and dynamics across networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:network-theory", "labels": [ "Network Theory" ], "is_subclass_of": [ "Graph Theory" ], "wikilinks": [ "Network Analysis", "Graph Databases", "Graph Theory", "https://en.wikipedia.org/wiki/Network_theory", "https://networkx.org/" ] }, { "id": "network-topology", "title": "Network Topology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Node connection structure within blockchain systems, providing essential functionality for distributed ledger technology operations and properties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:network-topology", "labels": [ "Network Topology", "Hierarchical Topology", "NetworkTopology" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "network-transport", "title": "Network Transport", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Network transport is the layer of communication concerned with end-to-end delivery of data between application endpoints across a network. It builds on the underlying packet-routing service to provide services such as connection establishment, multiplexing, reliable or unreliable delivery, ordering, flow control, and congestion control. Transport protocols determine the guarantees and performance characteristics that distributed applications experience.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:network-transport", "labels": [ "Network Transport" ], "is_subclass_of": [ "Transport Layer" ], "wikilinks": [] }, { "id": "networkcomponent", "title": "Networkcomponent", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Network Component is a physical or virtualised hardware device \u2014 including routers, switches, firewalls, load balancers, and wireless access points \u2014 that forms part of telecommunications and computer network infrastructure, enabling connectivity, traffic routing, security enforcement, and data exchange between users, devices, and services. Together these components implement the layers of the OSI model and realise network topologies ranging from local area networks to wide-area internet backbones.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:networkcomponent", "labels": [ "Networkcomponent" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [ "ConfigurationManagement", "CoreRouter", "EdgeRouter", "Firewall", "Gateway", "Hub", "IPAddressing", "LoadBalancer", "ManagedSwitch", "NetworkManagement", "NetworkMonitoring", "OSIModel", "Router", "SNMP", "Switch", "UnmanagedSwitch", "VLAN", "WirelessAccessPoint", "InfrastructureDomain", "NetworkInfrastructure" ] }, { "id": "networking-infrastructure", "title": "Networking Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Networking infrastructure comprises the integrated ensemble of physical hardware, logical protocols, and software-defined systems\u2014including routers, switches, fibre optic links, wireless access points, content delivery networks (CDNs), edge nodes, load balancers, and DNS services\u2014that collectively enable communication between computing systems across local, metropolitan, and wide-area scales. It forms the foundational substrate upon which distributed applications, cloud services, and real-time collaborative platforms depend, governing end-to-end latency, throughput, reliability, and security. Modern networking infrastructure increasingly incorporates software-defined networking (SDN), network function virtualisation (NFV), and programmable data planes that decouple control logic from physical forwarding hardware. For latency-sensitive domains such as spatial computing, extended reality, and distributed AI inference, infrastructure design must additionally satisfy stringent quality-of-service guarantees through traffic engineering, edge offloading, and geographically distributed state synchronisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:networking-infrastructure", "labels": [ "Networking Infrastructure" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "networking-layer", "title": "Networking Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Communication systems that connect components and users across distributed metaverse environments through network protocols and software.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:networking-layer", "labels": [ "Networking Layer", "NetworkingLayer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Communication Protocols", "Cross-Platform Connectivity", "Data Transmission Service", "Low-latency Interaction", "MSF Taxonomy 2025", "OSI Model", "Real-time Communication", "Routing Infrastructure", "TCP/IP Stack", "Transport Layer", "Communication Software", "Distributed Computing", "Infrastructure Architecture", "InfrastructureDomain", "Network Layer", "Network Protocol", "Network Standards", "Network Topology", "Physical Network Hardware" ] }, { "id": "networking-standard", "title": "Networking Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Networking Standard is a technology infrastructure concept and a type of infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:networking-standard", "labels": [ "Networking Standard" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "networking-technology", "title": "Networking Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The protocols, hardware, and software frameworks that enable real-time data transmission, synchronisation, and communication between distributed participants. In spatial computing contexts this encompasses WebRTC, QUIC, UDP/TCP stacks, adaptive bitrate streaming, and delta-compression techniques that support low-latency avatar synchronisation and shared world-state updates.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:networking-technology", "labels": [ "Networking Technology" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "networking", "title": "Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Networking is the discipline and practice of connecting computing devices so they can exchange data through shared communication channels and protocols. It encompasses the hardware, addressing, routing and protocol layers that move packets reliably between hosts across local and wide-area networks. Networking provides the connective substrate on which the internet, cloud services and distributed systems are built.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:networking", "labels": [ "Networking" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "neur-ips", "title": "NeurIPS", "domain": "ai", "domain_name": "Ai", "definition": "NeurIPS (Neural Information Processing Systems) is the premier annual international conference on machine learning, computational neuroscience, and artificial intelligence, held each December and governed by the NeurIPS Foundation. Founded in 1987 at the intersection of neuroscience and statistical learning theory, it has evolved into the most selective and impactful peer-reviewed publication venue in AI, shaping research agendas across deep learning, reinforcement learning, probabilistic modelling, generative AI, and AI ethics. Accepted papers undergo rigorous double-blind peer review and are freely available via the NeurIPS Proceedings archive, with acceptance rates typically below 26 percent in recent years. The conference serves as a primary mechanism for the global dissemination, validation, and agenda-setting of foundational machine learning research.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:neur-ips", "labels": [ "NeurIPS" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "neural-3-d-generation", "title": "Neural 3D Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI-powered creation of three-dimensional geometric models, volumetric representations, and 4D dynamic scenes using neural networks and machine learning techniques, including generative models, neural radiance fields, gaussian splatting, and diffusion-based 3D synthesis.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:neural-3-d-generation", "labels": [ "Neural 3D Generation" ], "is_subclass_of": [ "AI Application", "Generative AI", "3D Content Generation", "Procedural Content Generation" ], "wikilinks": [ "3D Asset Dataset", "3D Representation", "Automated 3D Modeling", "Diffusion Model", "DreamFusion (Google)", "Extended Reality", "GAN", "GET3D", "GET3D (NVIDIA)", "glTF 2.0", "GPU Compute", "NeRF", "NeRF (Mildenhall et al.)", "OpenAI Point-E", "Rapid Prototyping", "Shap-E (OpenAI)", "SIGGRAPH AI", "Training Pipeline", "VAE", "3D Content Generation" ] }, { "id": "neural-architecture-search", "title": "Neural Architecture Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Neural Architecture Search (NAS) is an automated machine learning technique that searches a defined space of neural network designs to discover architectures that maximise predictive performance or satisfy multi-objective constraints such as latency, parameter count, and energy consumption. NAS algorithms explore the architecture search space using strategies that include reinforcement learning, evolutionary algorithms, differentiable relaxations (DARTS), and predictor-based approaches, each making different trade-offs between search cost and solution quality. The field emerged from the observation that hand-designed architectures require substantial expert knowledge and iterative experimentation, and that automated search can discover non-obvious configurations that outperform human-designed baselines on targeted hardware or task distributions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:neural-architecture-search", "labels": [ "Neural Architecture Search", "Evolutionary Neural Architecture Search" ], "is_subclass_of": [ "AutoML" ], "wikilinks": [] }, { "id": "neural-audio-codec", "title": "Neural Audio Codec", "domain": "ai", "domain_name": "Ai", "definition": "A neural audio codec is a learned compression model that encodes audio into a compact discrete or latent representation and decodes it back to a waveform using neural networks. Unlike hand-designed codecs, it is trained end-to-end with reconstruction and adversarial objectives to maximise perceptual quality at very low bitrates. The discrete tokens it produces also serve as a representation for generative audio and speech models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:neural-audio-codec", "labels": [ "Neural Audio Codec", "DAC Codec" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "neural-audio-enhancement", "title": "Neural Audio Enhancement", "domain": "ai", "domain_name": "Ai", "definition": "Neural audio enhancement is the use of deep learning models to improve the perceptual quality of recorded audio by removing noise, reverberation, and artefacts or by restoring lost detail. Models are trained on paired clean and degraded audio to learn a mapping that suppresses unwanted components while preserving the target signal. It is widely applied to speech in podcasting, conferencing, and media post-production.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:neural-audio-enhancement", "labels": [ "Neural Audio Enhancement", "Audio Enhancement" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "neural-interface", "title": "Neural Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "A neural interface is a system that establishes a communication pathway between the nervous system and an external device, translating neural activity into machine-readable signals or delivering stimulation back to neural tissue. Interfaces range from non-invasive surface electrodes that read muscle or scalp potentials to implanted electrodes that record from or stimulate individual neurons. They enable direct neural control of prosthetics, robots, and computers, and the restoration of sensory feedback.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:neural-interface", "labels": [ "Neural Interface" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [] }, { "id": "neural-machine-translation", "title": "Neural Machine Translation", "domain": "ai", "domain_name": "Ai", "definition": "Neural Machine Translation (NMT) is an approach to automated language translation in which end-to-end neural networks \u2014 typically based on encoder-decoder architectures with attention mechanisms \u2014 learn to map source-language sequences directly to target-language sequences from parallel corpora. Unlike earlier statistical phrase-based methods, NMT systems capture long-range dependencies and global sentence context, producing more fluent and contextually accurate translations. The Transformer architecture has become the dominant NMT paradigm since its introduction in 2017.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:neural-machine-translation", "labels": [ "Neural Machine Translation", "NeuralMachineTranslation" ], "is_subclass_of": [ "Language Translation" ], "wikilinks": [] }, { "id": "neural-network-architecture", "title": "Neural Network Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Neural network architecture defines the structural organisation of artificial neurons into layers and the connectivity patterns between them, determining how a network processes and transforms input data through weighted connections and activation functions. Different architectural families \u2014 feedforward, convolutional, recurrent, and transformer \u2014 embody distinct inductive biases suited to different data modalities and task types. Architectural choices govern capacity, training stability, and generalisation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:neural-network-architecture", "labels": [ "Neural Network Architecture", "Neural Architecture" ], "is_subclass_of": [ "AI Model Architecture", "ModelArchitecture" ], "wikilinks": [ "Artificial Intelligence", "ModelArchitecture" ] }, { "id": "neural-network-component", "title": "Neural Network Component", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A discrete structural or functional building block within a neural network model, including layers, activation functions, normalisation mechanisms, attention heads, and connection schemes. These components are composed to form complete neural network architectures and directly determine model capacity, training dynamics, and inference behaviour.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:neural-network-component", "labels": [ "Neural Network Component" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "neural-network-inference", "title": "Neural Network Inference", "domain": "ai", "domain_name": "Ai", "definition": "Neural network inference is the phase in which a trained neural-network model is applied to new input data to produce predictions, classifications or generations, as distinct from the training phase that learns the model's parameters. It is computationally dominated by forward-pass matrix and tensor operations and is frequently accelerated on GPUs and dedicated hardware. Inference efficiency, measured in latency, throughput and energy, is critical for deploying models in real-time, edge and large-scale serving environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:neural-network-inference", "labels": [ "Neural Network Inference" ], "is_subclass_of": [ "Inference" ], "wikilinks": [] }, { "id": "neural-network-latent-space", "title": "Neural Network Latent Space", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network latent space is a continuous, lower-dimensional manifold that a neural network learns to construct from high-dimensional input data, representing the data's underlying generative factors as geometric relationships between points. Encoder networks compress inputs into latent vectors that capture semantically meaningful structure, while decoder networks reconstruct outputs from those vectors; the resulting geometry encodes similarity such that interpolation along geodesics yields semantically coherent transitions. Architectures including variational autoencoders, generative adversarial networks, and diffusion models exploit structured latent spaces to enable sampling, conditional generation, and disentangled attribute control through vector arithmetic operations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:neural-network-latent-space", "labels": [ "Neural Network Latent Space", "Latent Space", "Latent State Space", "latent space" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "neural-network-layer", "title": "Neural Network Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Neural Network Layer is a discrete computational stage in a neural network that applies a parameterised transformation to its input tensor, including operations such as linear projection, convolution, normalisation, or attention. Layers are composed sequentially or in parallel to form a complete neural network architecture.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:neural-network-layer", "labels": [ "Neural Network Layer" ], "is_subclass_of": [ "Neural Network Component" ], "wikilinks": [ "owl:Thing" ] }, { "id": "neural-network-quantisation", "title": "Neural Network Quantisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A model compression technique that reduces the numerical precision of neural network weights and activations from floating-point (FP32, FP16) to lower-bit integer representations (INT8, INT4, binary), decreasing memory footprint and improving inference speed on hardware with integer arithmetic units. Approaches include post-training quantisation and quantisation-aware training.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:neural-network-quantisation", "labels": [ "Neural Network Quantisation", "Model Quantisation", "Neural Network Compression", "Neural Network Quantization" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ONNX Runtime", "TensorFlow Lite Quantization", "TensorRT", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "neural-network-text-tokenisation", "title": "Neural Network Text Tokenisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The process of segmenting text into discrete units (tokens) \u2014 characters, subwords, or words \u2014 that serve as the atomic inputs to neural network language models, directly determining vocabulary size, out-of-vocabulary handling, and downstream model performance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:neural-network-text-tokenisation", "labels": [ "Neural Network Text Tokenisation", "Tokenisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain", "Update Cycle" ] }, { "id": "neural-network-training", "title": "Neural Network Training", "domain": "ai", "domain_name": "Ai", "definition": "Neural network training is the process of iteratively adjusting a model's weights to minimise a loss function over a dataset, typically using gradient descent with backpropagation. Each step computes the gradient of the loss with respect to parameters and updates them via an optimiser, repeating across many batches and epochs. Training quality depends on data, objective design, regularisation, and substantial parallel compute.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:neural-network-training", "labels": [ "Neural Network Training" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "neural-network", "title": "Neural Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A computational model comprising interconnected layers of artificial neurons that learn to approximate functions by adjusting weighted connections through exposure to training data, forming the foundational substrate of deep learning systems including convolutional networks, recurrent networks, and transformer architectures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:neural-network", "labels": [ "Neural Network", "Bayesian Neural Network", "Deep Neural Networks", "Efficient Neural Network", "Modular Neural Network", "Neural Network Emulator", "Neural Network Model", "Neural Network Optimisation", "Neural Networks", "NeuralNetwork", "neural network", "neural networks" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "AIGroundedDomain", "Backpropagation", "Computer Vision", "Convolutional Neural Network", "Deep Learning", "Gradient Descent", "Natural Language Processing", "Pattern Recognition", "Recurrent Neural Network", "Stochastic Gradient Descent", "Transformer" ] }, { "id": "neural-networking", "title": "Neural Networking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial neural network architectures \u2014 including convolutional, recurrent, and transformer-based models \u2014 to optimise network routing, traffic prediction, and resource allocation in communication infrastructure. Neural networking extends classical networking by replacing heuristic control planes with learned, data-driven decision policies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:neural-networking", "labels": [ "Neural Networking" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "neural-ode", "title": "Neural Ode", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural ordinary differential equation (Neural ODE) is a deep learning model that parameterises the continuous-time derivative of a hidden state with a neural network, so the forward pass becomes the solution of an ODE by a numerical integrator. This replaces a discrete stack of layers with a continuous-depth transformation and trains efficiently via the adjoint sensitivity method, giving constant memory cost. Neural ODEs are well suited to modelling continuous dynamics, irregularly sampled time series, and continuous normalising flows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:neural-ode", "labels": [ "Neural Ode", "Neural ODE" ], "is_subclass_of": [ "Deep Learning" ], "wikilinks": [] }, { "id": "neural-ordinary-differential-equation", "title": "Neural Ordinary Differential Equation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural ordinary differential equation (Neural ODE) is a class of deep learning model that parameterises the derivative of a hidden state with a neural network and obtains outputs by numerically integrating this learned dynamics. Rather than stacking a fixed number of discrete layers, it treats the transformation of representations as a continuous trajectory governed by an ODE, with depth replaced by integration time. This continuous-depth formulation enables memory-efficient training via the adjoint method and natural modelling of time-series and continuous dynamics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:neural-ordinary-differential-equation", "labels": [ "Neural Ordinary Differential Equation" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "neural-processing-unit", "title": "Neural Processing Unit", "domain": "ai", "domain_name": "Ai", "definition": "A Neural Processing Unit (NPU) is a specialised integrated circuit designed to accelerate neural network inference workloads by providing high-throughput, energy-efficient execution of the matrix multiplication and convolution operations that dominate deep learning computation. NPUs are integrated into mobile SoCs, personal computers, and edge devices to enable on-device AI inference without reliance on cloud compute, targeting performance-per-watt objectives unachievable by general-purpose CPUs or GPUs for these workloads.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:neural-processing-unit", "labels": [ "Neural Processing Unit", "Neural Engine" ], "is_subclass_of": [ "AI Hardware" ], "wikilinks": [] }, { "id": "neural-radiance-field", "title": "Neural Radiance Field", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A continuous volumetric scene representation that uses a multilayer perceptron to map 5D coordinates (3D position plus 2D viewing direction) to colour and volume density, enabling novel view synthesis via differentiable ray marching. NeRF achieves photo-realistic rendering of complex scenes from a sparse set of calibrated images and has spawned instant variants, Gaussian Splatting successors, and robotics perception applications.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:neural-radiance-field", "labels": [ "Neural Radiance Field" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "Artificial Intelligence", "Autonomous Robot" ] }, { "id": "neural-radiance-fields", "title": "Neural Radiance Fields", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A neural rendering technique that represents 3D scenes as continuous volumetric radiance functions encoded by multilayer perceptrons, mapping 5D inputs (3D spatial position plus 2D viewing direction) to colour and volume density. Novel viewpoints are synthesised by volumetric ray marching through the learned representation, enabling photorealistic view synthesis from sparse photograph sets without explicit 3D geometry. Introduced by Mildenhall et al. (ECCV 2020), NeRF has driven a generation of implicit neural scene representations spanning telepresence, virtual production, and robotics.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:neural-radiance-fields", "labels": [ "Neural Radiance Fields", "TELE-052-neural-radiance-fields" ], "is_subclass_of": [ "Telepresence", "Real-Time Rendering Pipeline" ], "wikilinks": [ "TELE-020-virtual-reality-telepresence", "TELE-050-neural-rendering-telepresence", "TELE-051-3d-gaussian-splatting", "TELE-053-volumetric-video-conferencing", "TELE-060-instant-ngp", "Real-Time Rendering Pipeline" ] }, { "id": "neural-rendering", "title": "Neural Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Neural Rendering is a class of computational techniques that train deep neural networks to represent, reconstruct, and synthesise visual appearance of 3D scenes, replacing or augmenting traditional rasterisation and ray-tracing pipelines with learned implicit representations. Methods such as Neural Radiance Fields (NeRF), Gaussian Splatting, and neural texture synthesis encode scene geometry, material properties, and illumination into network weights, enabling photo-realistic novel view synthesis from sparse image sets. Neural Rendering bridges computer vision and computer graphics, enabling real-time rendering of photorealistic avatars, 3D asset generation, scene understanding, and immersive spatial experiences at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:neural-rendering", "labels": [ "Neural Rendering" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "neural-vocoder", "title": "Neural Vocoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural vocoder is a deep generative model that synthesises a raw audio waveform from a compact intermediate representation, most commonly a mel-spectrogram produced by the acoustic model in a text-to-speech pipeline. Beginning with WaveNet in 2016 and evolving through flow-based, GAN-based designs such as HiFi-GAN, and diffusion-based approaches, neural vocoders replaced signal-processing methods such as Griffin-Lim and WORLD, delivering near-natural speech quality at real-time or faster generation speeds on commodity hardware.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:neural-vocoder", "labels": [ "Neural Vocoder" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Neural Network", "Text-to-Speech", "Audio Synthesis", "Speech Synthesis", "Acoustic Model" ] }, { "id": "neural-xr-interfaces", "title": "neural xr interfaces", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Direct neural input modalities for spatial computing systems, ranging from non-invasive EMG wristbands and EEG headsets to invasive brain-computer interfaces. Bridges neurotechnology with XR interaction paradigms, enabling thought-driven navigation, selection, and manipulation in virtual and augmented environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:neural-xr-interfaces", "labels": [ "Neural XR Interfaces" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "neuro-symbolic-ai", "title": "Neuro Symbolic Ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Neuro-symbolic AI integrates neural (sub-symbolic) learning with symbolic reasoning and knowledge representation to combine the perceptual strengths of deep networks with the interpretability, compositionality and logical rigour of symbolic systems. It aims to deliver models that learn from data yet reason over explicit knowledge, support verifiable inference and generalise from limited examples. Architectures range from neural networks that produce symbolic structures to symbolic engines guided by learned representations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:neuro-symbolic-ai", "labels": [ "Neuro Symbolic Ai", "Neuro-Symbolic AI", "Neurosymbolic AI" ], "is_subclass_of": [ "Artificial Intelligence", "Symbolic Reasoning" ], "wikilinks": [] }, { "id": "neuroimaging", "title": "Neuroimaging", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Neuroimaging is the set of techniques used to visualise the structure, function, or physiology of the nervous system, including magnetic resonance imaging (MRI and fMRI), positron emission tomography (PET), and electroencephalography (EEG). It provides the empirical measurements that cognitive science and cognitive psychology use to relate observed brain activity to mental processes such as perception, memory, and decision-making. Neuroimaging data increasingly feeds machine learning pipelines that decode cognitive states or diagnose neurological conditions from scan data.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:neuroimaging", "labels": [ "Neuroimaging" ], "is_subclass_of": [ "Neuroscience" ], "wikilinks": [] }, { "id": "neuromorphic-chip", "title": "Neuromorphic Chip", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neuromorphic chip is a specialised integrated circuit designed to emulate the structure and dynamics of biological neural networks, employing analogue or mixed-signal circuits to implement spiking neuron models that process information through sparse, event-driven spikes rather than continuous clock-driven computation. This brain-inspired architecture achieves orders-of-magnitude improvements in energy efficiency for pattern recognition, sensory processing, and on-device inference compared with conventional von Neumann processors. Representative implementations include IBM TrueNorth and Intel Loihi.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:neuromorphic-chip", "labels": [ "Neuromorphic Chip" ], "is_subclass_of": [ "AI Infrastructure", "Artificial Intelligence" ], "wikilinks": [ "IBM TrueNorth", "Intel Loihi", "AI Hardware", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "neuromorphic-chips", "title": "Neuromorphic Chips", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Neuromorphic Chips is a artificial intelligence concept and a type of Inference Hardware. that enables Ultra-Low-Power AI.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:neuromorphic-chips", "labels": [ "Neuromorphic Chips" ], "is_subclass_of": [ "AI Infrastructure", "Inference Hardware" ], "wikilinks": [ "Ultra-Low-Power AI", "Inference Hardware" ] }, { "id": "neuromorphic-computing", "title": "Neuromorphic Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computational paradigm that designs hardware and software systems to mirror the structure and function of biological neural networks, integrating memory and processing units to enable parallel, event-driven information processing inspired by how neurons communicate through electrical spikes. Unlike conventional von Neumann architectures, neuromorphic systems perform computation asynchronously at the location of data, enabling orders-of-magnitude improvements in energy efficiency for AI inference at the edge.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:neuromorphic-computing", "labels": [ "Neuromorphic Computing" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "neuroscience", "title": "Neuroscience", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Neuroscience is the scientific study of the nervous system, spanning the molecular, cellular, circuit, systems, and behavioural levels of how brains process information and generate behaviour. It investigates how neurons compute, communicate, and adapt, providing biological grounding for theories of cognition and learning. Neuroscience has a deep, bidirectional relationship with artificial intelligence, both inspiring computational models and using them to interpret neural data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:neuroscience", "labels": [ "Neuroscience" ], "is_subclass_of": [ "Cognitive Science", "AI Research Area" ], "wikilinks": [] }, { "id": "neutron-star", "title": "Neutron Star", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:neutron-star", "labels": [ "Neutron Star" ], "is_subclass_of": [ "Stellar Remnant" ], "wikilinks": [] }, { "id": "newcastle-ai-and-health-innovation", "title": "Newcastle AI & Health Innovation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Newcastle AI and Health Innovation is the regional technology ecosystem in North East England centred on the application of artificial intelligence to healthcare, life sciences, and digital health, anchored by Newcastle University, Northumbria University, and innovation institutions including the National Innovation Centre for Data and the National Innovation Centre for Ageing. The ecosystem combines academic research, NHS partnerships, and private sector investment to accelerate AI-driven health technology from discovery through to clinical and commercial deployment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:newcastle-ai-and-health-innovation", "labels": [ "Newcastle AI & Health Innovation" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "newcastle", "title": "Newcastle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Newcastle upon Tyne is a major city and metropolitan centre in North East England, positioned at the mouth of the River Tyne, historically significant as an industrial and maritime hub and now a regional capital for technology, data science, life sciences, and higher education. Home to Newcastle University and Northumbria University, the city anchors a cluster of digital and creative industries within the Northern Powerhouse economic policy framework. Newcastle's urban infrastructure, connectivity via rail and the A1 corridor, and significant public-sector investment underpin its role as a node in the UK's distributed knowledge economy. The city hosts institutions such as the Alan Turing Institute partner programmes and the National Innovation Centre for Data, reinforcing its positioning in data-driven research and applied artificial intelligence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:newcastle", "labels": [ "Newcastle" ], "is_subclass_of": [ "Entity" ], "wikilinks": [ "Digital Economy", "Northern Powerhouse", "Manchester", "Entity" ] }, { "id": "newton-euler-dynamics", "title": "Newton-Euler Dynamics", "domain": "robotics", "domain_name": "Robotics", "definition": "Recursive formulation of rigid-body mechanics that applies Newton's second law (F=ma) and Euler's rotation equation (\u03c4=I\u03b1) to each link of a robot chain in outward and inward passes, computing joint torques and reaction forces for dynamics simulation, control, and real-time motion planning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:newton-euler-dynamics", "labels": [ "Newton-Euler Dynamics", "Newton-Euler Algorithm" ], "is_subclass_of": [ "Actuation and Control", "Robot Dynamics" ], "wikilinks": [ "Robot Dynamics", "Robotics" ] }, { "id": "next-token-prediction", "title": "Next Token Prediction", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Next token prediction is the autoregressive language-modelling objective in which a model predicts the next token in a sequence given all preceding tokens. Trained by maximising the likelihood of each token conditioned on its left context, it requires no explicit labels and scales to vast text corpora. It is the core pre-training objective behind generative transformer language models such as the GPT family.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:next-token-prediction", "labels": [ "Next Token Prediction", "Next-Token Prediction" ], "is_subclass_of": [ "Pre Training" ], "wikilinks": [] }, { "id": "nft-minting", "title": "NFT Minting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "NFT Minting is the on-chain or deferred off-chain process by which a unique, indivisible, and cryptographically authenticated digital token\u2014a non-fungible token\u2014is created and permanently recorded on a distributed ledger, establishing irrevocable provenance, ownership, and transferability for a l...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:nft-minting", "labels": [ "NFT Minting" ], "is_subclass_of": [ "Token and Asset", "Blockchain Process", "Smart Contract Operation", "Digital Asset Creation", "Neural Network Text Tokenisation", "Provenance Recording" ], "wikilinks": [ "Arbitrum", "Arweave", "Base", "Batch Minting", "Bitcoin Ordinals", "Brand Loyalty NFTs", "Centralised Database", "Content Identifier", "CreatorEconomyDomain", "Creator Royalty", "Cryptographic Hash Function", "Custodial Digital Asset", "DeFi", "DecentralisedFinanceDomain", "Decentralised Identity", "Digital Art", "Digital Art Market", "Digital Asset Creation", "DigitalAssetDomain", "EIP-1559" ] }, { "id": "nis2-directive", "title": "Nis2 Directive", "domain": "governance", "domain_name": "Governance", "definition": "The NIS2 Directive is European Union legislation that raises and harmonises cybersecurity requirements for essential and important entities operating critical services across member states. It expands the scope of the earlier NIS Directive, imposes risk-management and governance obligations on covered organisations, and introduces stricter incident-reporting timelines and enforcement. It is a cornerstone of EU cyber-resilience policy for sectors such as energy, transport, health, and digital infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:nis2-directive", "labels": [ "Nis2 Directive", "NIS2 Directive" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "no-code-ai-development", "title": "No-Code AI Development", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The practice of building and deploying AI-powered applications and workflows using intuitive, low-barrier tools that require minimal or no traditional programming knowledge.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:no-code-ai-development", "labels": [ "No-Code AI Development" ], "is_subclass_of": [ "Intelligent Automation" ], "wikilinks": [] }, { "id": "nosql-database", "title": "NoSQL Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A NoSQL database is a class of data store that departs from the rigid tabular schema and relational model of traditional SQL systems to favour flexible schemas, horizontal scalability, and high write throughput. NoSQL systems are typically organised by data model \u2014 document, key-value, wide-column, or graph \u2014 and often relax strong consistency in exchange for availability and partition tolerance. They are designed for large-scale, distributed workloads where the relational model is a poor fit.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:nosql-database", "labels": [ "NoSQL Database", "NoSQL Databases" ], "is_subclass_of": [ "Database Management System" ], "wikilinks": [] }, { "id": "nodal-precession", "title": "Nodal Precession", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:nodal-precession", "labels": [ "Nodal Precession" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "node-based-editor", "title": "Node Based Editor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A visual authoring tool that represents computational logic, material properties, or content workflows as directed graphs of interconnected nodes and edges, enabling non-linear dataflow programming. Node-based editors are widely used in game engines, VFX pipelines, and spatial computing tools for shader authoring, procedural geometry generation, and real-time behaviour scripting.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:node-based-editor", "labels": [ "Node Based Editor", "Node Graph Editor", "Node-Based Workflow Editor" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "node-classification", "title": "Node Classification", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Node classification is a graph machine learning task that predicts a categorical label for each vertex in a graph, typically using the graph's structure together with node and edge features. It is commonly solved with graph neural networks, which propagate and aggregate neighbourhood information across message-passing layers to produce per-node predictions. Applications include fraud detection, citation-network labelling and social-network role inference.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:node-classification", "labels": [ "Node Classification" ], "is_subclass_of": [ "Classification" ], "wikilinks": [] }, { "id": "node-embedding", "title": "Node Embedding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Node embedding is a technique that maps each vertex of a graph to a low-dimensional vector such that structural relationships in the graph, such as proximity, connectivity, or community membership, are preserved as geometric relationships in the embedding space. Methods range from random-walk based approaches to graph neural network encoders that aggregate information from a node's neighbourhood. Node embeddings support downstream tasks such as community detection, link prediction, and node classification.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:node-embedding", "labels": [ "Node Embedding" ], "is_subclass_of": [ "Graph Embedding" ], "wikilinks": [] }, { "id": "node-js", "title": "Node Js", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Node.js is a cross-platform, open-source JavaScript runtime built on the V8 engine that executes JavaScript outside the browser, enabling server-side and command-line applications. It provides a single-threaded, event-driven, non-blocking I/O model that scales to many concurrent connections, together with a module system and the npm package ecosystem. Node.js is a foundational runtime for web back-ends, build tooling and blockchain development environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:node-js", "labels": [ "Node Js", "Node.js" ], "is_subclass_of": [ "Runtime Environment" ], "wikilinks": [] }, { "id": "node-based-diffusion-pipeline-interface", "title": "Node-Based Diffusion Pipeline Interface", "domain": "ai", "domain_name": "Ai", "definition": "A node-based diffusion pipeline interface is a visual programming environment that models a generative diffusion inference graph as a directed acyclic graph of interconnected functional nodes, where each node encapsulates a discrete operation such as model loading, text conditioning, latent sampling, or image decoding, and edges carry tensor data between nodes. This paradigm exposes the full computational structure of a diffusion pipeline as an inspectable, composable, and reproducible artefact rather than a hidden implementation detail. Tools such as ComfyUI exemplify this pattern: workflows are serialised as JSON graphs that can be version-controlled, shared, and deployed as production automation. The approach bridges visual dataflow programming traditions with modern deep-learning inference, enabling practitioners to compose multi-model, multi-stage generation pipelines without writing procedural code.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:node-based-diffusion-pipeline-interface", "labels": [ "Node-Based Diffusion Pipeline Interface", "ComfyUI Registry", "ComfyUI Workflow Integration" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "node-graph-visual-programming-interface", "title": "Node-Graph Visual Programming Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A paradigm of graphical programming in which computational logic is authored by connecting discrete functional nodes via edges rather than writing textual code. Node-based interfaces lower the barrier to complex workflow construction, making them prevalent in AI pipeline tools (e.g., ComfyUI, Flowise), creative software (shader editors, compositing), and knowledge-graph visualisation environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:node-graph-visual-programming-interface", "labels": [ "Node-Graph Visual Programming Interface", "Node Based Interface", "Node Graph", "Node based visual interfaces", "Node-Based Programming", "Visual Effects Node Graph", "Visual Programming Environment" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "node-independent-validation-pbft", "title": "Node-Independent Validation PBFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Node-Independent Validation PBFT is a variant of Practical Byzantine Fault Tolerance in which each validator independently verifies transaction validity without delegating validation authority to a designated primary node, eliminating the single point of failure inherent in classical PBFT's primary-replica model. By distributing validation responsibility uniformly across all consensus participants, the protocol improves resilience against primary node compromise and reduces the attack surface for Byzantine behaviour in consortium and enterprise blockchain deployments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:node-independent-validation-pbft", "labels": [ "Node-Independent Validation PBFT" ], "is_subclass_of": [ "Protocol and Consensus", "Practical Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Practical Byzantine Fault Tolerance" ] }, { "id": "node", "title": "Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A network participant computer within a blockchain system that stores, validates, and relays transactions and blocks. Nodes enforce consensus rules, maintain copies of the distributed ledger, and collectively provide the decentralised security guarantees that distinguish blockchain from centralised databases.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:node", "labels": [ "Node", "Relay Node", "Storage Node" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "noir", "title": "Noir", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A domain-specific programming language for writing zero-knowledge proofs, developed by Aztec. It abstracts the underlying proving system so developers can express circuits without low-level cryptographic detail.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:noir", "labels": [ "Noir", "Noir Programming Language" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [ "Zero-Knowledge Proof", "ZK-SNARK", "Cryptographic Protocol", "Programming Language" ] }, { "id": "noise-cancellation", "title": "Noise Cancellation", "domain": "ai", "domain_name": "Ai", "definition": "Noise cancellation is the process of reducing or eliminating unwanted audio signals from a primary audio stream using signal processing techniques. Active noise cancellation generates an anti-phase signal to destructively interfere with the noise, while passive methods use physical barriers. The technique is applied in audio hardware, telecommunications, and AI-driven speech enhancement systems to improve clarity and intelligibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:noise-cancellation", "labels": [ "Noise Cancellation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "noise-function-library", "title": "Noise Function Library", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A software library providing coherent noise generation algorithms like Perlin, simplex, value, and Voronoi noise for procedural content generation in computer graphics, enabling the creation of natural-looking textures, terrain, and visual effects without manual authoring.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:noise-function-library", "labels": [ "Noise Function Library" ], "is_subclass_of": [ "Display and Rendering", "Graphics Library" ], "wikilinks": [ "Graphics Library", "metaverse", "Procedural Content" ] }, { "id": "noise-function", "title": "Noise Function", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Noise Function is a deterministic pseudo-random function that maps spatial coordinates to smoothly varying scalar values, providing the controllable randomness behind procedural content. Gradient-based variants such as Perlin and Simplex noise produce coherent, band-limited fields that can be layered into fractal octaves to synthesise terrain, clouds, and textures. Because output depends only on input coordinates and a seed, noise functions are reproducible and efficiently evaluable on the GPU.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:noise-function", "labels": [ "Noise Function" ], "is_subclass_of": [ "Procedural Generation" ], "wikilinks": [] }, { "id": "noise-injection", "title": "Noise Injection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Noise injection is the deliberate addition of random perturbations \u2014 Gaussian noise, dropout-style masking, token swaps or signal distortions \u2014 to inputs, hidden activations, weights or gradients during training or data generation. As a data augmentation and regularisation technique it discourages over-fitting and improves robustness to distribution shift; in generative adversarial networks it supplies the stochastic latent input that drives sample diversity, and in differential privacy calibrated noise provides formal privacy guarantees.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:noise-injection", "labels": [ "Noise Injection" ], "is_subclass_of": [ "Data Augmentation" ], "wikilinks": [ "Data Augmentation", "Regularisation", "Generator Network", "Differential Privacy" ] }, { "id": "noise-mechanisms", "title": "Noise Mechanisms", "domain": "security", "domain_name": "Security", "definition": "Noise mechanisms are the algorithmic primitives in differential privacy that add calibrated random perturbation to query results or data so that the contribution of any single individual is statistically masked. The amount of noise is calibrated to the query's sensitivity and the desired privacy budget, balancing privacy protection against the accuracy of released statistics. The principal mechanisms are the Laplace, Gaussian, and exponential mechanisms, each suited to particular query types and privacy definitions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:noise-mechanisms", "labels": [ "Noise Mechanisms", "Noise Mechanism" ], "is_subclass_of": [ "Differential Privacy" ], "wikilinks": [] }, { "id": "noise-protocol", "title": "Noise Protocol", "domain": "security", "domain_name": "Security", "definition": "The Noise Protocol Framework is a toolkit for building secure cryptographic handshake protocols based on Diffie-Hellman key agreement. Rather than a single fixed protocol, it defines a set of composable handshake patterns from which designers select to obtain specific authentication, confidentiality, and forward-secrecy properties. Its simplicity and clear security properties have made it the basis for transport security in messaging and peer-to-peer systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:noise-protocol", "labels": [ "Noise Protocol" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "noise-reduction", "title": "Noise Reduction", "domain": "ai", "domain_name": "Ai", "definition": "Noise reduction is the family of techniques that attenuate unwanted random or structured disturbances in a signal while preserving the underlying information of interest. It ranges from classical linear and spectral filtering to statistical estimators and learned denoising models that infer clean signals from noisy observations. Effective noise reduction improves downstream perception, measurement and machine-learning tasks by raising the signal-to-noise ratio.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:noise-reduction", "labels": [ "Noise Reduction" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "noise-schedule", "title": "Noise Schedule", "domain": "ai", "domain_name": "Ai", "definition": "A noise schedule is the function that determines how much Gaussian noise is added at each step of a diffusion model's forward process and, correspondingly, removed during sampling. It governs the variance trajectory from clean data to pure noise and strongly affects sample quality, training stability, and the number of steps required. Common forms include linear, cosine, and learned schedules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:noise-schedule", "labels": [ "Noise Schedule", "Noise Scheduler", "Noise Scheduling" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "noise-suppression", "title": "Noise Suppression", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Noise suppression is a signal-processing technique that filters ambient, background, and non-speech sounds from audio streams in real time during virtual communication sessions. Modern implementations use deep-learning models to distinguish voice from environmental noise such as keyboard clicks, fan hum, and room echo. It is a foundational audio quality feature for distributed teams working from heterogeneous home or office environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:noise-suppression", "labels": [ "Noise Suppression", "NoiseFiltering" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "nominated-proof-of-stake", "title": "Nominated Proof of Stake", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Nominated Proof of Stake (NPoS) is a consensus mechanism variant in which token holders nominate a set of validator candidates; an election algorithm selects the active validator set to maximise stake distribution while satisfying security constraints. Elected validators produce and finalise blocks, while nominators share both block rewards and slashing penalties, aligning incentives across the broader token holder community.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:nominated-proof-of-stake", "labels": [ "Nominated Proof of Stake" ], "is_subclass_of": [ "Protocol and Consensus", "Proof of Stake" ], "wikilinks": [ "Blockchain", "Proof of Stake" ] }, { "id": "non-genesis-block", "title": "Non Genesis Block", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Non Genesis Block is any block in a blockchain with a block height greater than zero\u2014that is, every block produced after the genesis (first) block. Non-genesis blocks are structurally identical to the genesis block in terms of their header and transaction payload format, but they include a previous block hash field that cryptographically links them to their parent, forming the immutable chain structure. The vast majority of blocks in any mature blockchain are non-genesis blocks; they carry the transaction history and consensus record that gives the chain its economic value.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:non-genesis-block", "labels": [ "Non Genesis Block" ], "is_subclass_of": [ "Protocol and Consensus", "Block" ], "wikilinks": [ "Block", "Blockchain" ] }, { "id": "non-interactive-proof", "title": "Non Interactive Proof", "domain": "security", "domain_name": "Security", "definition": "A Non-Interactive Proof (NIP) is a cryptographic proof system in which the prover transmits a single message to the verifier, without any back-and-forth challenge-response rounds, allowing verification of a claim without ongoing interaction. Non-interactive proofs are typically constructed from interactive protocols via the Fiat-Shamir heuristic, replacing the verifier's random challenge with a hash of the prover's first message, binding the proof to the statement. The resulting proof string can be broadcast publicly, stored on-chain, or verified asynchronously by any party possessing the verification key. Non-interactive zero-knowledge proofs (NIZKs) additionally guarantee that the proof reveals nothing beyond the truth of the statement, and form the foundation of [[ZK-SNARK]] and [[ZK-STARKs]] schemes used in blockchain privacy and scalability applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:non-interactive-proof", "labels": [ "Non Interactive Proof", "Non-Interactive Proof" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [] }, { "id": "non-linear-narrative", "title": "Non Linear Narrative", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A storytelling approach where events are presented out of chronological order or where audience choices create branching paths through the narrative, allowing multiple ways to experience story events based on interactions and decisions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:non-linear-narrative", "labels": [ "Non Linear Narrative", "Non-Linear Narrative" ], "is_subclass_of": [ "Content and Assets", "Narrative Structure" ], "wikilinks": [ "Player Agency", "metaverse", "Narrative Structure" ] }, { "id": "non-linear-storytelling", "title": "Non Linear Storytelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The craft and practice of creating narratives where story events can be experienced in multiple sequences through branching paths, player choices, or temporal manipulation, commonly used in video games, interactive fiction, and immersive experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:non-linear-storytelling", "labels": [ "Non Linear Storytelling", "Non-Linear Storytelling" ], "is_subclass_of": [ "Content and Assets", "Storytelling" ], "wikilinks": [ "Multiple Endings", "metaverse", "Storytelling", "Telecollaboration" ] }, { "id": "non-maximum-suppression", "title": "Non Maximum Suppression", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Non-Maximum Suppression (NMS) is a post-processing algorithm used in object detection to eliminate redundant overlapping bounding box proposals by retaining only the highest-confidence detection and discarding lower-confidence boxes that exceed a predefined intersection-over-union (IoU) threshold. The procedure iteratively selects the detection with the highest class score, suppresses all remaining boxes that sufficiently overlap with it, and repeats until no candidates remain. NMS is a critical component of single-stage and two-stage detectors including YOLO, SSD, and Faster R-CNN families. Soft-NMS and class-agnostic variants address edge cases where multiple legitimate objects are densely packed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:non-maximum-suppression", "labels": [ "Non Maximum Suppression", "Non-Maximum Suppression" ], "is_subclass_of": [ "Object Detection" ], "wikilinks": [] }, { "id": "non-player-character", "title": "Non Player Character", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An autonomous, computer-controlled character in a virtual environment that follows scripted behaviour trees, finite state machines, or AI-driven policies to interact with human users and the scene. Modern NPCs increasingly leverage conversational AI and large language models to produce contextually responsive dialogue and adaptive behavioural patterns.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:non-player-character", "labels": [ "Non Player Character", "Non-Player Character" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "non-verbal-communication", "title": "Non Verbal Communication", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Non-verbal communication is the transmission of meaning through channels other than spoken or written words, including gesture, facial expression, gaze, posture, and interpersonal distance. It conveys emotion, intent, and social context that frequently carries more weight than the literal verbal message. In spatial computing and avatar-mediated environments, reproducing non-verbal cues is essential for credible co-presence and natural social interaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:non-verbal-communication", "labels": [ "Non Verbal Communication", "Non-Verbal Communication" ], "is_subclass_of": [ "Immersive Communication" ], "wikilinks": [] }, { "id": "non-volatile-memory", "title": "Non Volatile Memory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Non-volatile memory (NVM) is computer storage that retains its data after power is removed, in contrast to volatile memory such as DRAM and SRAM. It encompasses technologies including NAND and NOR flash, EEPROM, and emerging persistent-memory devices that combine byte-addressability with durability. NVM is fundamental to firmware storage, embedded systems, solid-state drives, and the bootstrapping of computing platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:non-volatile-memory", "labels": [ "Non Volatile Memory", "Non-Volatile Memory" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "non-convex-optimisation", "title": "Non-Convex Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The minimisation of objective functions that are not convex, so the loss surface may contain multiple local minima, saddle points, plateaus, and ravines, and no general guarantee links a local solution to the global optimum. Non-convex optimisation is the actual setting of deep learning\u2014neural network training landscapes are highly non-convex\u2014and of many engineering problems, tackled in practice with stochastic gradient methods, momentum, restarts, relaxations, and problem-specific structure rather than the clean certificates available in the convex case.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:non-convex-optimisation", "labels": [ "Non-Convex Optimisation" ], "is_subclass_of": [ "Mathematical Optimisation" ], "wikilinks": [ "Convex Optimisation", "Stochastic Gradient Descent", "Loss Landscape" ] }, { "id": "non-custodial-wallet", "title": "Non-Custodial Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A non-custodial wallet is a cryptocurrency wallet in which the user, rather than a third party, exclusively holds and controls the private keys that authorise transactions. Because no intermediary can move or freeze funds, the user bears full responsibility for key security and recovery, typically managed through a seed phrase. Non-custodial wallets are the technical embodiment of self-custody and a prerequisite for permissionless interaction with decentralised protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:non-custodial-wallet", "labels": [ "Non-Custodial Wallet" ], "is_subclass_of": [ "Cryptocurrency Wallet" ], "wikilinks": [] }, { "id": "non-destructive-editing", "title": "Non-Destructive Editing", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Non-destructive editing is a content-authoring approach in which modifications are recorded as separate, reversible operations or layers rather than overwriting the original source data. The source remains intact, and the final result is computed by composing the edit stack, so any change can be re-ordered, adjusted, or removed. It is fundamental to layered 3D scene description and modern media pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:non-destructive-editing", "labels": [ "Non-Destructive Editing", "Non-Destructive Workflow" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "non-discrimination", "title": "Non-Discrimination", "domain": "governance", "domain_name": "Governance", "definition": "Non-discrimination is the principle that individuals should not be treated less favourably on the basis of protected characteristics such as race, gender, age, or disability. In technology governance it is operationalised as a requirement that automated systems, including AI models, avoid producing biased or disparate outcomes across protected groups. Non-discrimination is a core component of human rights frameworks and a frequent target of algorithmic fairness auditing.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:non-discrimination", "labels": [ "Non-Discrimination", "Non-discrimination" ], "is_subclass_of": [ "Human Rights" ], "wikilinks": [] }, { "id": "non-fungible-token-nft", "title": "Non-Fungible Token (NFT)", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital asset recorded on a distributed ledger that is uniquely identifiable and non-interchangeable, representing ownership or rights to specific digital or physical items.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:non-fungible-token-nft", "labels": [ "Non-Fungible Token (NFT)" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Creator Royalties", "ETSI GR ARF 010", "IPFS", "ISO 24165", "Token Metadata", "Asset Trading", "Blockchain", "BlockchainDomain", "Crypto Token", "Cryptographic Hash", "Digital Asset", "Digital Ownership", "Digital Wallet", "MiddlewareLayer", "Provenance Tracking", "Smart Contract", "Token Standard", "Virtual Asset" ] }, { "id": "non-fungible-token", "title": "Non-Fungible Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A token standard in which each unit is distinct and not interchangeable, used to represent ownership of a specific digital or referenced physical item on a distributed ledger; uniqueness is enforced by a token identifier and ownership is tracked via smart-contract state.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:non-fungible-token", "labels": [ "Non-Fungible Token" ], "is_subclass_of": [ "Token" ], "wikilinks": [ "Smart Contract", "ERC-721", "NFT Standard", "NFT Marketplace", "Token" ] }, { "id": "non-repudiation", "title": "Non-Repudiation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Non-Repudiation is a foundational Information Security property and legal-technical assurance mechanism that renders it cryptographically and evidentially impossible for a party that originated, received, or approved a message, transaction, or data object to subsequently deny that partici...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:non-repudiation", "labels": [ "Non-Repudiation", "non-repudiability, undeniability, attribution assurance, accountable signing, cryptographic evidence" ], "is_subclass_of": [ "Security and Identity", "Information Security", "Security Property", "Accountability Mechanism", "Digital Evidence", "Legal Framework" ], "wikilinks": [ "Accountability Mechanism", "Anonymity", "Asymmetric Cryptography", "Blockchain Transaction Integrity", "CAdES", "Certificate Authority", "Certificate Revocation", "CRL Certificate Revocation", "CryptographicLayer", "CryptographyDomain", "Deniable Authentication", "Digital Evidence", "Dispute Resolution", "eIDAS Compliance", "eIDAS Regulation", "eSignature Law", "EdDSA", "Electronic Commerce", "ETSI EN 319 series", "Evidence Token" ] }, { "id": "nonce", "title": "Nonce", "domain": "security", "domain_name": "Security", "definition": "A Nonce (Number used ONCE) is an arbitrary value included in a cryptographic computation to prevent replay attacks or to satisfy a target condition. In proof-of-work blockchains, miners increment a 32-bit nonce in the block header repeatedly until the SHA-256 hash of the header falls below the current difficulty target, thereby expending computational work proportional to the difficulty. In communications protocols, nonces ensure that each session or message produces a unique ciphertext, preventing an attacker from replaying a previously captured message.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:nonce", "labels": [ "Nonce", "Account Nonce", "Random Nonce", "Transaction Nonce" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "nonverbal-communication", "title": "Nonverbal Communication", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Nonverbal communication is the transmission of meaning through cues other than words, including facial expression, gaze, gesture, posture, interpersonal distance, and tone of voice. It conveys emotion, regulates interaction, and signals social presence, and it is central to creating believable avatars and co-presence in virtual and augmented reality. Reproducing nonverbal channels faithfully is a core challenge for embodied and immersive collaboration.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:nonverbal-communication", "labels": [ "Nonverbal Communication", "NonverbalCommunication" ], "is_subclass_of": [ "Communication Theory" ], "wikilinks": [] }, { "id": "normal-map", "title": "Normal Map", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A normal map is a texture that encodes per-texel surface normal directions, typically storing the X, Y, and Z components of a normal vector in the red, green, and blue channels. It allows a low-polygon mesh to react to lighting as if it had the fine geometric detail of a much denser surface, by perturbing the shading normal without changing the underlying geometry. Normal maps are a foundational technique in real-time and physically based rendering for adding visual detail efficiently.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:normal-map", "labels": [ "Normal Map" ], "is_subclass_of": [ "Texture Mapping" ], "wikilinks": [] }, { "id": "normal-mapping", "title": "Normal Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A real-time rendering technique that fakes fine surface detail by storing perturbed surface normals in a texture and using them, rather than the interpolated geometric normals, during per-pixel lighting. Detail sculpted on a high-polygon model is baked into a tangent-space normal map applied to a low-polygon mesh, so bumps, scratches, and seams respond correctly to moving lights without adding geometry \u2014 a cornerstone of the game-asset pipeline and physically based rendering.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:normal-mapping", "labels": [ "Normal Mapping" ], "is_subclass_of": [ "Texture Mapping" ], "wikilinks": [ "Texture Mapping", "Surface Normal", "Pixel Shader" ] }, { "id": "normalised-burn-ratio", "title": "Normalised Burn Ratio", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:normalised-burn-ratio", "labels": [ "Normalised Burn Ratio" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "normalised-difference-vegetation-index", "title": "Normalised Difference Vegetation Index", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:normalised-difference-vegetation-index", "labels": [ "Normalised Difference Vegetation Index" ], "is_subclass_of": [ "Vegetation Index" ], "wikilinks": [] }, { "id": "normalised-difference-water-index", "title": "Normalised Difference Water Index", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:normalised-difference-water-index", "labels": [ "Normalised Difference Water Index" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "normalising-flow", "title": "Normalising Flow", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A normalising flow is a class of deep generative models that learns a bijective mapping between a simple tractable base distribution (typically a standard multivariate Gaussian) and a complex target data distribution through a composition of invertible, differentiable transformations. Because each transformation is invertible, the change-of-variables formula yields an exact closed-form expression for the data log-likelihood, enabling both efficient sampling and precise density estimation. This exact likelihood property distinguishes normalising flows from latent variable models such as variational autoencoders and implicit models such as generative adversarial networks. Architectures are distinguished by how they design transformations whose Jacobian determinants can be computed efficiently, either through triangular Jacobians (autoregressive flows, coupling layers) or via continuous-time dynamics (neural ODEs).", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:normalising-flow", "labels": [ "Normalising Flow", "Continuous Normalising Flow", "Normalizing Flow" ], "is_subclass_of": [ "Generative Model" ], "wikilinks": [ "Probabilistic Model", "Generative Model" ] }, { "id": "normalising-flows", "title": "Normalising Flows", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Normalising flows are a class of generative models that learn complex probability distributions by composing a series of invertible, differentiable transformations that map a simple base distribution (typically Gaussian) to the target distribution, with exact log-likelihood computation via the change-of-variables formula and the Jacobian determinant. Both sampling and density evaluation are tractable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:normalising-flows", "labels": [ "Normalising Flows" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "normalized-data-state", "title": "Normalized Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:normalized-data-state", "labels": [ "Normalized Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "north-england-innovation-corridor", "title": "North England Innovation Corridor", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The collaborative technology and innovation ecosystem spanning Manchester, Leeds, Liverpool, Sheffield, and Newcastle that constitutes the Northern Powerhouse. The corridor links world-class research universities, NHS trusts, advanced manufacturing facilities, and digital startups through coordinated investment, shared infrastructure, and regional development frameworks aimed at closing the productivity gap between North and South England.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:north-england-innovation-corridor", "labels": [ "North England Innovation Corridor" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "northern-powerhouse", "title": "Northern Powerhouse", "domain": "governance", "domain_name": "Governance", "definition": "The Northern Powerhouse is a United Kingdom government economic policy framework, formally launched in 2014, designed to rebalance national prosperity by co-ordinating investment, infrastructure, and devolved governance across the cities and regions of northern England. It operates through a combination of city-region devolution deals, major transport programmes such as Northern Powerhouse Rail, targeted innovation investment in science and technology clusters, and streamlined inward investment promotion. The initiative links major urban centres including Manchester, Leeds, Sheffield, Newcastle, and Liverpool under a shared economic development narrative, seeking to raise productivity and close the long-standing North-South divide in the United Kingdom economy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:northern-powerhouse", "labels": [ "Northern Powerhouse" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Digital Economy", "Manchester", "Leeds", "Sheffield", "Entity" ] }, { "id": "nostr-protocol", "title": "Nostr Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Nostr (Notes and Other Stuff Transmitted by Relays) is a minimalist open protocol for censorship-resistant, decentralised messaging and identity, in which clients sign events with Schnorr Signatures|Schnorr signatures over the secp256k1 elliptic curve and broadcast them to any number of s...", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:nostr-protocol", "labels": [ "Nostr Protocol", "Nostr Protocol (NIP-01)" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain", "Decentralised Communication Protocol", "Cryptographic Identity System" ], "wikilinks": [ "Agent-to-Agent Communication", "AI agent", "AI Domain", "BIP-340 Schnorr Keypair", "Bitcoin Ecosystem", "Bitcoin Lightning Network", "Censorship-Resistant Messaging", "Cryptographic Identity System", "Data Vending Machine", "Decentralised Communication Protocol", "Decentralised Social Web", "did:nostr DID Method", "eProsima Fast DDS", "IdentityLayer", "JSON Serialisation", "Large Language Model", "Lightning Zap Payments", "NIP-01 Core Protocol", "NIP-01 Core Specification", "NIP-05 Identity Verification" ] }, { "id": "nostr-relay-endpoint-registry", "title": "Nostr Relay Endpoint Registry", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Nostr relay list is a curated or automatically discovered list of WebSocket relay server endpoints that a Nostr client uses to publish and subscribe to signed events. Because Nostr has no centralised routing, a client's relay list determines its social graph reach: events are broadcast to all listed relays, and the client pulls its feed from the same set. Relay lists are stored as NIP-65 kind:10002 events on the network itself, enabling portable relay preferences that travel with the user's public key across clients.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:nostr-relay-endpoint-registry", "labels": [ "Nostr Relay Endpoint Registry", "nostr relay list" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "nostr-relay", "title": "Nostr Relay", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Nostr Relay is a server that implements the Nostr protocol, accepting signed event objects from clients, storing them, and forwarding them to subscribed clients according to filter criteria. Relays are the infrastructure backbone of the Nostr decentralised social network: because there is no central server or consensus chain, the network's availability and censorship-resistance derive entirely from the federated mesh of independently operated relays. Relays communicate with clients over persistent WebSocket connections using a simple JSON message format.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:nostr-relay", "labels": [ "Nostr Relay" ], "is_subclass_of": [ "Nostr Protocol" ], "wikilinks": [] }, { "id": "nostr", "title": "Nostr", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Nostr (Notes and Other Stuff Transmitted by Relays) is an open, censorship-resistant social-messaging protocol in which user identity is a secp256k1 cryptographic key pair and all user activity consists of signed JSON events published to one or more relay servers over WebSocket connections. The protocol has no central authority: identity is purely the public key, relays are interchangeable infrastructure that store and forward events without requiring account registration, and clients subscribe using filter objects to receive matching events. Nostr Improvement Proposals (NIPs) extend the core event-kind system to cover short-form notes, long-form articles, encrypted direct messages, Lightning Network zap payments, community moderation, and decentralised identity verification. Its radical simplicity, cryptographic self-sovereignty, and tight integration with the Bitcoin and Lightning ecosystem have made Nostr the dominant open social layer for value-aligned decentralised communication.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nostr", "labels": [ "Nostr", "Nostr Client" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "notary-service", "title": "Notary Service", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A notary service is a trusted third-party function that attests to the authenticity, integrity, and existence of a document or digital artefact at a specific point in time, producing a tamper-evident record legally admissible as proof. In digital contexts, notary services hash document content and anchor that hash to a verifiable ledger \u2014 blockchain or traditional timestamping authority \u2014 binding the evidence to a precise timestamp without revealing the document contents. They bridge traditional legal document authentication with cryptographic proof mechanisms, enabling use cases ranging from intellectual property protection to supply chain provenance. Blockchain-anchored notary services eliminate single-point-of-failure trust assumptions present in classical certificate-authority-based approaches.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:notary-service", "labels": [ "Notary Service", "Notary", "Notary Validation Pattern" ], "is_subclass_of": [ "Security Services" ], "wikilinks": [] }, { "id": "notification-system", "title": "Notification System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A software component that delivers event-driven alerts, presence indicators, and system messages to users or services in real time. In virtual world and metaverse contexts, notification systems manage social events, user-to-user signals, and platform state changes across distributed infrastructure, typically using push protocols or pub-sub messaging patterns.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:notification-system", "labels": [ "Notification System", "Alert Notification System", "Notification Routing", "Notification Suppression" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "notified-body", "title": "Notified Body", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An independent third-party conformity assessment organisation formally designated by a national notifying authority under the EU AI Act (Articles 29\u201339) to audit and certify high-risk AI systems \u2014 particularly biometric identification and product-safety-component systems \u2014 against technical documentation, quality management, and harmonised standards requirements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:notified-body", "labels": [ "Notified Body" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Framework" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "notion", "title": "Notion", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Notion is a cloud-based productivity and knowledge-management platform that unifies note-taking, relational databases, wikis, project management boards, and document editing into a single composable workspace. Its block-based editor treats every content element \u2014 text, tables, embeds, code, and media \u2014 as a manipulable unit that can be nested, linked, and filtered, enabling teams to construct custom workflows without traditional software development. Notion serves as an all-in-one tool replacing multiple specialised SaaS products for many knowledge workers and organisations. Since 2023 Notion has integrated generative AI capabilities directly into its editing and querying workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:notion", "labels": [ "Notion" ], "is_subclass_of": [ "Knowledge Management System" ], "wikilinks": [] }, { "id": "nouns-dao", "title": "Nouns DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An on-chain organisation that auctions one non-fungible token each day, with proceeds funding a shared treasury governed by token holders.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:nouns-dao", "labels": [ "Nouns DAO" ], "is_subclass_of": [ "Decentralised Autonomous Organisation" ], "wikilinks": [ "Treasury Management", "Governance Token", "NFT Standard", "On-chain Governance", "Decentralised Autonomous Organisation" ] }, { "id": "novel-view-synthesis", "title": "Novel View Synthesis", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Novel view synthesis (NVS) is a computer vision and computer graphics task that involves generating photorealistic images of a scene from camera viewpoints not present in the original set of captured images, given a collection of reference photographs and their corresponding camera poses. The task requires learning an implicit or explicit representation of the scene's geometry and appearance that supports free-viewpoint rendering with high fidelity. It is a foundational capability for immersive media, telepresence, and spatial computing applications, and has been dramatically advanced by neural scene representations such as Neural Radiance Fields and 3D Gaussian Splatting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:novel-view-synthesis", "labels": [ "Novel View Synthesis", "Real-Time Novel View Synthesis" ], "is_subclass_of": [ "Computer Vision Task" ], "wikilinks": [] }, { "id": "nuclear-energy", "title": "Nuclear Energy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Nuclear energy is electrical power generated by harnessing the heat released from nuclear fission of heavy isotopes, typically uranium-235, in a controlled reactor. It provides high-density, low-carbon, dispatchable baseload generation independent of weather conditions. Its role in decarbonisation is balanced against concerns over capital cost, long construction times, radioactive-waste management, and safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:nuclear-energy", "labels": [ "Nuclear Energy", "Nuclear Power" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "nucleus-sampling", "title": "Nucleus Sampling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A stochastic decoding strategy for autoregressive language models, also called top-p sampling, that at each step truncates the next-token distribution to the smallest set of tokens whose cumulative probability exceeds a threshold p, renormalises, and samples from that nucleus \u2014 adapting the candidate pool to the model's confidence and avoiding both the degenerate repetition of greedy search and the incoherent tail noise of unrestricted sampling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:nucleus-sampling", "labels": [ "Nucleus Sampling" ], "is_subclass_of": [ "Sampling" ], "wikilinks": [ "Sampling", "Beam Search", "Greedy Decoding", "Text Generation" ] }, { "id": "number-theory", "title": "Number Theory", "domain": "security", "domain_name": "Security", "definition": "Number Theory is the branch of pure mathematics concerned with the properties of integers and related structures, including divisibility, prime numbers, congruences and Diophantine equations. It ranges from elementary results, such as the fundamental theorem of arithmetic, to deep areas like analytic and algebraic number theory. Beyond its theoretical importance, it underpins much of modern cryptography, where the difficulty of certain number-theoretic problems provides security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:number-theory", "labels": [ "Number Theory" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Prime Number", "Modular Arithmetic", "Cryptography Domain", "Public Key Cryptography", "Algebra", "Measure Theory", "owl:Thing" ] }, { "id": "numerical-integration", "title": "Numerical Integration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Numerical integration is the family of algorithms that approximate definite integrals and advance differential equations in time when closed-form solutions are unavailable. In physics simulation it denotes the time-stepping schemes (such as explicit and implicit Euler, Verlet, and Runge-Kutta methods) that integrate equations of motion to update positions and velocities each frame. The choice of scheme balances accuracy, numerical stability, and computational cost.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:numerical-integration", "labels": [ "Numerical Integration" ], "is_subclass_of": [ "Physics Simulation" ], "wikilinks": [] }, { "id": "numerical-methods", "title": "Numerical Methods", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Techniques for obtaining approximate solutions to mathematical problems that cannot be solved exactly, using finite sequences of arithmetic operations; encompassing root-finding, interpolation, quadrature, linear-system solvers, ODE/PDE integrators, and optimisation routines with explicit control over accuracy, stability, and computational cost.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:numerical-methods", "labels": [ "Numerical Methods", "Numerical Analysis", "Numerical Method" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Linear Algebra", "Differential Equations", "Simulation", "Gradient Descent", "owl:Thing" ] }, { "id": "numpy", "title": "Numpy", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "NumPy (Numerical Python) is the foundational library for numerical and scientific computing in Python, providing the N-dimensional array (ndarray) object together with a comprehensive suite of vectorised mathematical, logical, linear algebra, Fourier transform, and random number operations. Its contiguous, typed memory layout and broadcasting semantics enable concise, high-performance array programming by delegating element-wise loops to compiled C and Fortran routines. NumPy underpins almost the entire Python data and machine learning ecosystem, serving as the in-memory array substrate that libraries such as pandas, SciPy, scikit-learn, and the deep learning frameworks build upon or interoperate with.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:numpy", "labels": [ "Numpy", "NumPy" ], "is_subclass_of": [ "Scientific Computing" ], "wikilinks": [] }, { "id": "nutation", "title": "Nutation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:nutation", "labels": [ "Nutation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "nvidia-gpu", "title": "Nvidia Gpu", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An NVIDIA GPU is a graphics processing unit designed and manufactured by NVIDIA, providing massively parallel computation that has become the dominant hardware substrate for machine learning training and inference. These devices expose thousands of cores together with specialised tensor units optimised for the dense matrix multiplications central to deep learning. They are programmed predominantly through the CUDA platform and serve as the primary accelerator in modern data centres and high-performance computing systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:nvidia-gpu", "labels": [ "Nvidia Gpu", "NVIDIA GPU" ], "is_subclass_of": [ "Graphics Processing Unit", "Hardware Component" ], "wikilinks": [] }, { "id": "oasis", "title": "OASIS", "domain": "standards", "domain_name": "Standards", "definition": "OASIS (Organisation for the Advancement of Structured Information Standards) is an international, not-for-profit standards development body that produces open standards for information technology, with particular strengths in web services, security, content formats, electronic business, and emergency management. Founded in 1993, OASIS operates through technical committees that develop specifications via an open, consensus-based process, producing both OASIS Standards and publicly available specifications that are frequently adopted as ISO/IEC standards. Prominent OASIS standards include MQTT for IoT messaging, SAML for federated identity, XACML for access control, OData for REST-based data access, and the OpenDocument Format. OASIS also co-publishes standards with other bodies including the W3C and ISO.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:oasis", "labels": [ "OASIS", "OASIS Open", "OASIS XACML" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "oauth-2-0", "title": "oauth 2.0", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OAuth 2.0 is an open authorisation framework standardised as RFC 6749 (2012) that enables a resource owner to delegate scoped, time-limited access to their protected resources on a resource server to a third-party client application, without exposing credentials. The framework separates four distinct roles\u2014resource owner, client, authorisation server, and resource server\u2014and defines multiple grant types (authorisation code, client credentials, device code, refresh token) suited to different trust levels and client profiles. Access tokens with explicit scopes enforce the principle of least privilege, while extensions such as PKCE (RFC 7636), JWT access tokens (RFC 9068), token introspection (RFC 7662), and token revocation (RFC 7009) complete the lifecycle. OAuth 2.0 serves as the foundation upon which OpenID Connect 1.0 adds federated authentication, together underpinning modern Identity and Access Management platforms and Zero Trust Architecture policies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:oauth-2-0", "labels": [ "OAuth 2.0", "OAuth 2", "OAuth2" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "oauth", "title": "OAuth", "domain": "security", "domain_name": "Security", "definition": "An open standard for delegated authorisation that allows a user to grant a third-party application limited access to a protected resource on behalf of a resource owner, using scoped, revocable access tokens rather than sharing credentials.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:oauth", "labels": [ "OAuth", "OAuth 1.0" ], "is_subclass_of": [ "Authorisation" ], "wikilinks": [ "Authorisation", "Access Control", "Single Sign-On", "OAuth 2.0", "Identity Provider" ] }, { "id": "oecd-ai-principles", "title": "OECD AI Principles", "domain": "governance", "domain_name": "Governance", "definition": "The OECD AI Principles are five intergovernmental policy guidelines adopted by the Organisation for Economic Co-operation and Development in May 2019 and endorsed by G20 leaders in Osaka, representing the first intergovernmental standard on AI. They address inclusive growth and sustainable development, human-centred values and fairness, transparency and explainability, robustness security and safety, and accountability. Updated in 2024 to address generative AI and foundation models, they have shaped national AI strategies, the EU AI Act, and a constellation of derivative multilateral instruments. Implementation is tracked through the OECD AI Policy Observatory, which catalogues national policies across member and partner countries.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:oecd-ai-principles", "labels": [ "OECD AI Principles", "OECD AI Principles 2024", "OECD Principles on AI" ], "is_subclass_of": [ "Responsible AI Principles" ], "wikilinks": [] }, { "id": "oecd-due-diligence-guidance", "title": "OECD Due Diligence Guidance", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The OECD Due Diligence Guidance is an internationally recognised framework that sets out how companies should identify, prevent, and mitigate human-rights and ethical risks in their supply chains, most notably for minerals from conflict-affected and high-risk areas. It defines a five-step, risk-based process for responsible sourcing and reporting. It serves as the reference standard underpinning many corporate and regulatory responsible-sourcing requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:oecd-due-diligence-guidance", "labels": [ "OECD Due Diligence Guidance" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "oecd", "title": "OECD", "domain": "governance", "domain_name": "Governance", "definition": "The OECD (Organisation for Economic Co-operation and Development) is an intergovernmental organisation founded in 1961, headquartered in Paris, comprising 38 member states committed to market economies and democratic governance. It produces comparative economic analysis, harmonised statistics, and binding and non-binding policy instruments spanning taxation, trade, education, digital policy, and emerging technology governance. Its AI Principles (2019) were the first intergovernmental standard for trustworthy artificial intelligence, and its Crypto-Asset Reporting Framework (CARF) extends automatic exchange of financial information to digital assets. OECD outputs exert normative influence through peer review, mutual recognition agreements, and adoption by broader multilateral bodies including the G20.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:oecd", "labels": [ "OECD" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "MiCA", "Regulatory Domain", "Governance Domain" ] }, { "id": "ofdm", "title": "OFDM", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OFDM (Orthogonal Frequency-Division Multiplexing) is a digital modulation scheme that transmits data in parallel over many closely spaced, mutually orthogonal subcarriers. Splitting a high-rate stream into many low-rate subcarriers makes the signal robust to multipath fading and allows simple frequency-domain equalisation. It is the physical-layer foundation of Wi-Fi, LTE, 5G, DVB, and many other modern broadband systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ofdm", "labels": [ "OFDM" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "oma-3-media-wg", "title": "OMA3 Media WG", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A working group within the Open Metaverse Alliance for Web3 (OMA3) focused on media-related interoperability standards for metaverse and Web3 applications. It coordinates specifications among member organisations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:oma-3-media-wg", "labels": [ "OMA3 Media WG" ], "is_subclass_of": [ "OMA3" ], "wikilinks": [ "OMA3", "Metaverse", "Web3" ] }, { "id": "oma-3", "title": "OMA3", "domain": "metaverse", "domain_name": "Metaverse", "definition": "OMA3 (Open Metaverse Alliance for Web3) is an industry consortium founded in 2022 by major metaverse and blockchain companies to establish open, interoperable standards for virtual worlds, digital assets, and cross-platform identity. It operates as a member-governed non-profit organisation whose working groups produce shared specifications, reference implementations, and governance frameworks ensuring that digital objects and user identities remain portable across disparate metaverse platforms. OMA3 sits at the intersection of Web3 ownership principles and spatial-computing infrastructure, seeking to prevent proprietary lock-in by codifying standards for asset transfer protocols, decentralised land registries, and avatar portability. Its outputs are designed to be adopted by blockchain networks, game engines, and virtual world platforms alike, making it a foundational governance body for an open metaverse ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:oma-3", "labels": [ "OMA3" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "Interoperability Standards", "Metaverse", "Web3", "Standards Body" ] }, { "id": "onnx-operator-set", "title": "ONNX Operator Set", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An ONNX operator set is a versioned collection of the computational operators defined by the Open Neural Network Exchange format, fixing the operators and their semantics that a model may use.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:onnx-operator-set", "labels": [ "ONNX Operator Set" ], "is_subclass_of": [ "ONNX" ], "wikilinks": [ "ONNX", "Machine Learning" ] }, { "id": "onnx-runtime", "title": "ONNX Runtime", "domain": "ai", "domain_name": "Ai", "definition": "ONNX Runtime is an open-source, cross-platform inference and training acceleration engine developed by Microsoft that executes models represented in the Open Neural Network Exchange (ONNX) format. It applies a multi-pass graph optimisation pipeline\u2014including operator fusion, constant folding, and common subexpression elimination\u2014before routing computation through hardware-specific execution providers such as CUDA, TensorRT, DirectML, OpenVINO, CoreML, and QNN to maximise throughput and minimise latency. The runtime decouples training-time framework choice from deployment-time execution environment, allowing models trained in PyTorch, TensorFlow, or scikit-learn to be deployed with a single, vendor-neutral API. In 2024\u20132025 it extended into large language model inference via the onnxruntime-genai extensions, adding KV-cache management and autoregressive decoding primitives.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:onnx-runtime", "labels": [ "ONNX Runtime" ], "is_subclass_of": [ "Inference Engine" ], "wikilinks": [] }, { "id": "onnx-standard", "title": "ONNX Standard", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The ONNX standard defines an open format for representing machine learning models as a computation graph of typed operators, enabling models to move between training and inference frameworks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:onnx-standard", "labels": [ "ONNX Standard" ], "is_subclass_of": [ "Machine Learning Discipline", "Open Standard" ], "wikilinks": [ "ONNX", "Deep Learning", "Neural Rendering", "Open Standards", "Machine Learning" ] }, { "id": "onnx", "title": "ONNX", "domain": "ai", "domain_name": "Ai", "definition": "ONNX (Open Neural Network Exchange) is an open, vendor-neutral interchange format and operator specification for representing machine learning models as directed acyclic computation graphs serialised via Protocol Buffers, enabling trained models to be exported from one deep learning framework (e.g. PyTorch, TensorFlow, MXNet) and executed on any compatible runtime or hardware accelerator without retraining. The ONNX ecosystem encompasses the format specification, versioned opset definitions, a model zoo of pretrained models, and the ONNX Runtime (ORT) inference engine with pluggable hardware execution providers. Originally co-created by Meta and Microsoft in 2017 and later adopted under Linux Foundation AI and Data governance, ONNX addresses framework fragmentation by providing a common intermediate representation (IR) that bridges the training-deployment gap in production ML systems. The specification supports standard neural network operations, control flow constructs, sparse tensors, dynamic shapes, and extended data types needed for large language model and on-device AI inference workloads.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:onnx", "labels": [ "ONNX", "ONNX Community", "ONNX Graph", "Onnx" ], "is_subclass_of": [ "Interoperability Standard" ], "wikilinks": [] }, { "id": "ontology-technical-details", "title": "ONTOLOGY_TECHNICAL_DETAILS", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A meta-documentation node cataloguing the technical structure, bidirectional link patterns, missing parent audits, and inheritance statistics of the NarrativeGoldmine ontology graph. It serves as a diagnostic reference for ontology engineers, documenting cross-domain isolation rules, broken reference inventories, and recommended file structures for parent concept pages.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ontology-technical-details", "labels": [ "ONTOLOGY_TECHNICAL_DETAILS" ], "is_subclass_of": [ "Data Management", "Data Structure", "Distributed System", "Record-Keeping System" ], "wikilinks": [ "BC-0426-hyperledger-fabric", "BC-0427-hyperledger-besu", "BC-0428-enterprise-blockchain-architecture", "BC-0429-permissioned-blockchain", "BC-0430-private-channels", "...", "Block", "Blockchain Entity", "Consensus Mechanism", "Cryptographic System", "Data Structure", "DistributedDataStructure", "Distributed Data Structure", "Distributed Ledger", "Distributed Protocol", "Distributed System", "Enterprise Blockchain Architecture", "Hyperledger Besu", "Hyperledger Fabric", "Mathematical Science" ] }, { "id": "opc-ua", "title": "OPC UA", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OPC Unified Architecture (OPC UA) is an open, platform-independent, service-oriented communication standard developed by the OPC Foundation for secure and reliable data exchange in industrial automation and the Industrial Internet of Things. It provides a unified information model that merges process data, alarms, historical data, and device metadata into a single addressable namespace accessible via TCP binary or HTTPS transport, replacing the earlier COM/DCOM-based OPC Classic specifications with a cross-platform, scalable architecture.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:opc-ua", "labels": [ "OPC UA", "OPC-UA" ], "is_subclass_of": [ "Interoperability Standard" ], "wikilinks": [] }, { "id": "or-set", "title": "OR-Set", "domain": "data", "domain_name": "Data", "definition": "An OR-Set (Observed-Remove Set) is a conflict-free replicated data type (CRDT) that supports concurrent add and remove operations on a set while guaranteeing eventual consistency across replicas. Each added element is tagged with a unique identifier so that concurrent adds and removes resolve deterministically, with adds winning over concurrent removes of unobserved tags. It is a foundational structure for collaborative applications that must merge edits without central coordination.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:or-set", "labels": [ "OR-Set" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "osi-model", "title": "OSI Model", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The OSI (Open Systems Interconnection) Model is a conceptual reference framework developed by ISO that partitions network communication functions into seven hierarchical layers: Physical, Data Link, Network, Transport, Session, Presentation, and Application. Each layer has a well-defined responsibility and communicates with the layer immediately above and below it through standardised interfaces, enabling interoperability between heterogeneous systems from different vendors. Originally published as ISO/IEC 7498-1 in 1984, the model does not describe a concrete protocol stack but provides a universal vocabulary and design template for networking protocols. It remains the canonical educational and diagnostic framework for understanding where specific protocols, devices, and services operate within a network.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:osi-model", "labels": [ "OSI Model" ], "is_subclass_of": [ "Network Architecture" ], "wikilinks": [ "Network Architecture", "REST API" ] }, { "id": "ot-cybersecurity-framework", "title": "OT Cybersecurity Framework", "domain": "security", "domain_name": "Security", "definition": "An OT cybersecurity framework is a structured body of security requirements, zone-and-conduit models, maturity levels, and assurance processes designed specifically for operational technology \u2014 the industrial control systems, SCADA platforms, PLCs, and safety systems that actuate physical processes. Unlike generic IT cybersecurity frameworks, it prioritises availability and safety over confidentiality, accommodates decades-long asset lifecycles and legacy protocols, and assigns duties across asset owners, integrators, and product suppliers, with the IEC 62443 series as its canonical instantiation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ot-cybersecurity-framework", "labels": [ "OT Cybersecurity Framework" ], "is_subclass_of": [ "Cybersecurity Framework" ], "wikilinks": [ "Cybersecurity Framework", "IEC 62443", "Operational Technology", "SCADA" ] }, { "id": "owasp-llm-top-10-2025", "title": "OWASP LLM Top 10 2025", "domain": "security", "domain_name": "Security", "definition": "The OWASP LLM Top 10 2025 is the updated edition of OWASP's ranked list of the most critical security risks for applications built on large language models. It catalogs threats such as prompt injection, sensitive information disclosure, supply-chain vulnerabilities, excessive agency, and improper output handling, with guidance on mitigation. The list is a widely referenced baseline for securing LLM and agentic AI systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:owasp-llm-top-10-2025", "labels": [ "OWASP LLM Top 10 2025", "OWASP LLM Top 10" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "owl-2-web-ontology-language", "title": "OWL 2 Web Ontology Language", "domain": "standards", "domain_name": "Standards", "definition": "A World Wide Web Consortium standard for representing ontologies on the web, providing formal semantics based on description logic for classes, properties and individuals. It extends earlier web ontology work with richer modelling features and defined reasoning profiles.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:owl-2-web-ontology-language", "labels": [ "OWL 2 Web Ontology Language", "OWL 2 DL", "OWL 2 W3C Recommendation", "OWL Ontology", "OWL Web Ontology Language", "W3C OWL Standard", "Web Ontology Language" ], "is_subclass_of": [ "OWL" ], "wikilinks": [ "Description Logic", "RDF", "Semantic Reasoning Engine", "Reasoning", "Ontology", "Knowledge Representation", "OWL", "https://www.w3.org/TR/owl2-overview/" ] }, { "id": "owl-class-hierarchy", "title": "OWL Class Hierarchy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The OWL Class Hierarchy is the directed acyclic graph (DAG) of named and anonymous classes connected via rdfs:subClassOf axioms within a Web Ontology Language (OWL) ontology, imposing a partial order on the class extension lattice. It provides the primary vehicle for monotonic inheritance of properties and restrictions, enabling description-logic reasoners such as HermiT, Pellet, and FaCT++ to classify individuals, detect unsatisfiable classes, and compute implicit subsumption relationships that are not asserted explicitly. The hierarchy is closed under the OWL semantics of the chosen profile (OWL 2 DL, EL, QL, or RL), constraining the decidability and computational complexity of reasoning tasks performed over it. Well-engineered class hierarchies underpin interoperability across domains including biomedical ontologies (GO, SNOMED CT), geospatial standards, and knowledge-graph schemas.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:owl-class-hierarchy", "labels": [ "OWL Class Hierarchy", "Class Hierarchy" ], "is_subclass_of": [ "Ontology Structure" ], "wikilinks": [ "Automated Reasoning", "metaverse", "Ontology Structure" ] }, { "id": "owl", "title": "OWL", "domain": "standards", "domain_name": "Standards", "definition": "OWL (Web Ontology Language) is a W3C-standardised semantic web language for defining and sharing ontologies on the World Wide Web, built on RDF and grounded in Description Logics to provide formal semantics with decidable reasoning. OWL enables the specification of classes, properties, individuals, and axioms that constrain their relationships, supporting automated inference and knowledge graph construction.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:owl", "labels": [ "OWL" ], "is_subclass_of": [ "Semantic Web Linked Data Standard" ], "wikilinks": [] }, { "id": "obfuscation", "title": "Obfuscation", "domain": "security", "domain_name": "Security", "definition": "Obfuscation is the deliberate transformation of data, code, or communication into a form that is difficult to understand or analyse while preserving its function, used to protect intellectual property, hinder reverse engineering, and conceal sensitive information. Unlike encryption, which renders data unreadable without a key, obfuscation aims to raise the effort required to comprehend an artefact rather than to guarantee secrecy. It is widely applied in software protection and privacy engineering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:obfuscation", "labels": [ "Obfuscation" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "obj-format", "title": "Obj Format", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The OBJ format (Wavefront .obj) is a simple, text-based 3D geometry interchange format that stores polygon mesh data as lists of vertices, texture coordinates, normals and faces. It is openly documented and almost universally supported, with material properties carried in a companion .mtl file. Valued for its readability and portability rather than efficiency, OBJ remains a common lowest-common-denominator format for exchanging static 3D models between modelling tools and pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:obj-format", "labels": [ "Obj Format", "OBJ Format" ], "is_subclass_of": [ "Polygon Mesh" ], "wikilinks": [] }, { "id": "object-detection-and-tracking", "title": "Object Detection and Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Object Detection and Tracking combines spatial object localisation with temporal tracking to identify, classify, and follow objects across video frames or sensor streams.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:object-detection-and-tracking", "labels": [ "Object Detection and Tracking", "Object Tracking" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "AI Video", "Computer Vision", "Hardware and Edge", "Human tracking and SLAM capture", "Image Generation", "MetaverseDomain", "Object Detection", "Perception System", "Product Design", "Segmentation and Identification", "WebDev and Consumer Tooling" ] }, { "id": "object-detection", "title": "Object Detection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Object Detection is the computer vision task of identifying and localising multiple objects within an image or video frame by predicting bounding boxes and class labels for each detected instance, combining spatial localisation with categorical classification in a single forward pass.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:object-detection", "labels": [ "Object Detection", "Monocular 3D Object Detection", "ObjectDetection" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "Computer Vision", "Image Classification", "Instance Segmentation", "MetaverseDomain" ] }, { "id": "object-manipulation", "title": "Object Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "The robotic capability to grasp, reorient, move, and release physical objects using end-effectors, encompassing grasp planning, force-torque control, dexterous in-hand manipulation, and task-level sequencing. Object manipulation integrates computer vision for object detection and pose estimation with tactile sensing and compliant actuation to handle diverse and unstructured items.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:object-manipulation", "labels": [ "Object Manipulation" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "object-recognition", "title": "Object Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Object recognition is a computer vision task that involves identifying and localising instances of predefined object categories within images or video streams, producing class labels, bounding boxes, segmentation masks, or pose estimates depending on the task variant. It subsumes tasks including image classification, object detection, semantic segmentation, and instance segmentation, and has become a core capability of autonomous systems, augmented reality, and content understanding pipelines. Modern approaches are dominated by deep convolutional and transformer-based architectures trained on large annotated datasets.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:object-recognition", "labels": [ "Object Recognition" ], "is_subclass_of": [ "Computer Vision Task" ], "wikilinks": [] }, { "id": "object-storage", "title": "Object Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Object storage is a data storage architecture that manages data as discrete objects \u2014 each comprising an opaque payload, a globally unique identifier, and extensible metadata \u2014 accessed via a flat namespace through RESTful HTTP APIs rather than a file hierarchy or block device. It is designed for massive horizontal scalability, high durability, and cost-effective storage of unstructured data such as media files, backups, and machine-learning datasets. Amazon S3 established the de facto API standard, now implemented by numerous compatible services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:object-storage", "labels": [ "Object Storage", "Cloud Object Storage", "Object Storage Service" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "object-based-image-analysis", "title": "Object-based Image Analysis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:object-based-image-analysis", "labels": [ "Object-based Image Analysis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "objective-function", "title": "Objective Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An objective function is a scalar-valued function that quantifies the quality of a candidate solution, which an optimisation or learning process seeks to minimise or maximise. In machine learning it formalises the goal of training \u2014 for example minimising prediction error or maximising likelihood \u2014 so that algorithms can adjust parameters to improve it. The choice of objective function determines what a model is rewarded for, shaping its behaviour, biases and generalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:objective-function", "labels": [ "Objective Function" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "objective", "title": "Objective", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A formal specification of the goal or criterion that an AI system, optimisation algorithm, or training procedure is designed to satisfy or maximise. In machine learning this is typically expressed as a loss function or reward signal that guides parameter updates; in planning and reinforcement learning it encodes desired agent behaviour through utility or reward functions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:objective", "labels": [ "Objective" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "oblivious-transfer", "title": "Oblivious Transfer", "domain": "security", "domain_name": "Security", "definition": "Oblivious transfer is a foundational cryptographic protocol in which a sender transmits one of several pieces of information to a receiver, but remains oblivious to which piece was received, while the receiver learns nothing about the other pieces. The canonical 1-out-of-2 variant lets a receiver choose one of two sender messages without revealing the choice and without learning the unchosen message. Oblivious transfer is complete for secure two-party computation and underpins protocols such as garbled circuits and private set intersection.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:oblivious-transfer", "labels": [ "Oblivious Transfer" ], "is_subclass_of": [ "Secure Multi-Party Computation" ], "wikilinks": [] }, { "id": "observability", "title": "Observability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Observability is the capability of a system that allows engineers to infer its internal state and behaviour solely from externally observable outputs \u2014 primarily logs, metrics, and distributed traces. Derived from control-systems theory, it answers whether the complete internal state can be reconstructed from a sequence of outputs. In modern software engineering, observability enables debugging, performance tuning, incident response, and proactive reliability assurance across complex distributed architectures where direct inspection is impractical.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:observability", "labels": [ "Observability", "LLM Observability", "Observability Platform", "Observability Stack" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [ "performance", "Site Reliability Engineering", "Distributed Systems", "Reliability Engineering" ] }, { "id": "observation-campaign", "title": "Observation Campaign", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:observation-campaign", "labels": [ "Observation Campaign" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "observation-model", "title": "Observation Model", "domain": "robotics", "domain_name": "Robotics", "definition": "An observation model (or measurement model) is the probabilistic relationship that specifies the likelihood of a sensor measurement given the underlying state of a system. In recursive Bayesian estimation it provides the likelihood term used to correct a predicted state belief against incoming data. Its accuracy, including the noise characteristics it encodes, directly governs the quality of state estimation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:observation-model", "labels": [ "Observation Model", "Probabilistic Observation Model" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "observation-provenance", "title": "Observation Provenance", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:observation-provenance", "labels": [ "Observation Provenance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "observation-timestamp", "title": "Observation Timestamp", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:observation-timestamp", "labels": [ "Observation Timestamp" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "observational-astronomy", "title": "Observational Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:observational-astronomy", "labels": [ "Observational Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "obstacle-avoidance", "title": "Obstacle Avoidance", "domain": "robotics", "domain_name": "Robotics", "definition": "The capability of a mobile robot or autonomous system to detect and avoid collisions with obstacles in its environment in real-time using sensors and reactive control strategies. It enables safe navigation without requiring complete prior knowledge of the environment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:obstacle-avoidance", "labels": [ "Obstacle Avoidance", "ObstacleAvoidance", "RB-1019-obstacle-avoidance" ], "is_subclass_of": [ "Reactive Control" ], "wikilinks": [ "Autonomous Vehicles", "Avoidance Maneuver", "Collision-Free Motion", "Drones", "Mobile Robotics", "Mobile Robots", "Obstacles", "Proximity Sensors", "Radar", "RB-1007-trajectory-generation", "RB-1011-cobot-safety-levels", "RB-1016-path-planning", "Reactivity", "Safe Navigation", "Safety Systems", "Autonomous Navigation", "Camera", "LIDAR", "RB-1013-localization", "Reactive Control" ] }, { "id": "obstacle-detection", "title": "Obstacle Detection", "domain": "robotics", "domain_name": "Robotics", "definition": "Obstacle Detection is the robotics perception task of identifying and localising physical objects in a robot's surroundings that may impede or endanger its motion. It fuses data from sensors such as lidar, cameras, radar and ultrasonic rangefinders to build a representation of free and occupied space. Reliable obstacle detection is a prerequisite for safe autonomous navigation, collision avoidance and motion planning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:obstacle-detection", "labels": [ "Obstacle Detection" ], "is_subclass_of": [ "Perception", "Perception and Sensing" ], "wikilinks": [] }, { "id": "occlusion-culling", "title": "Occlusion Culling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Occlusion culling is a real-time rendering optimisation technique that determines which scene objects are hidden (occluded) by other geometry from the viewer's current viewpoint and discards them before issuing GPU draw calls. By preventing invisible geometry from traversing the vertex, rasterisation, and fragment shader stages, occlusion culling reduces overdraw and GPU workload, enabling complex scenes with high polygon counts to maintain interactive frame rates. It is a complementary technique to frustum culling, level-of-detail selection, and back-face culling within the broader visibility determination subsystem of a render pipeline. Implementations range from CPU-driven portal and cell systems, to hardware occlusion queries, to GPU-driven indirect rendering with hierarchical Z-buffer occlusion.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:occlusion-culling", "labels": [ "Occlusion Culling" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [] }, { "id": "occlusion-rendering", "title": "Occlusion Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Occlusion rendering is the set of techniques used in real-time and offline graphics pipelines to correctly determine and display which surfaces are hidden behind other geometry from a given camera viewpoint, as well as to compute the darkening of surfaces due to local geometric obstruction of ambient light. It encompasses hardware depth-buffer culling, ambient occlusion shading, screen-space occlusion methods, and \u2014 in augmented reality \u2014 the masking of virtual objects by real-world foreground geometry. Correct occlusion is essential for perceptual plausibility in both games and AR/VR applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:occlusion-rendering", "labels": [ "Occlusion Rendering" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "occupancy-grid", "title": "Occupancy Grid", "domain": "robotics", "domain_name": "Robotics", "definition": "An occupancy grid is a probabilistic spatial representation of a robot's environment as a discretised lattice of cells, each storing a probability or log-odds value indicating the likelihood that the corresponding region of space is occupied by an obstacle. It provides a metric map suitable for collision avoidance, path planning, and SLAM by fusing noisy sensor measurements through Bayesian updates.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:occupancy-grid", "labels": [ "Occupancy Grid", "Occupancy Grid Map", "OccupancyGrid" ], "is_subclass_of": [ "Environment Mapping" ], "wikilinks": [] }, { "id": "ocean-colour-remote-sensing", "title": "Ocean Colour Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ocean-colour-remote-sensing", "labels": [ "Ocean Colour Remote Sensing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ocean-world", "title": "Ocean World", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ocean-world", "labels": [ "Ocean World" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ocsp", "title": "Ocsp", "domain": "security", "domain_name": "Security", "definition": "The Online Certificate Status Protocol (OCSP) is an internet protocol for obtaining the real-time revocation status of a digital certificate from a responder operated by or on behalf of the issuing certificate authority. A client queries the responder for a specific certificate and receives a signed reply stating whether it is good, revoked, or unknown. OCSP offers a more immediate and bandwidth-efficient alternative to downloading full certificate revocation lists, and OCSP stapling lets servers present a recent status to avoid client-side lookups.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ocsp", "labels": [ "Ocsp", "OCSP" ], "is_subclass_of": [ "Certificate Revocation" ], "wikilinks": [] }, { "id": "octave-immersive-research-facility", "title": "Octave Immersive Research Facility", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A world-class multi-modal immersive research facility at the University of Salford housing advanced display topologies, high-performance compute, and distributed systems for mixed reality, generative AI, and human-scale VR experimentation. The lab supported over 25 years of research including world firsts in collaborative mixed reality, brain scanning in phobia treatment, real-time human reconstruction from cameras, and telepresence, and served hundreds of SMEs through ERDF-funded programmes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:octave-immersive-research-facility", "labels": [ "Octave Immersive Research Facility", "Octave Multi Model Laboratory" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Hardware and Edge", "National Industrial Centre for Virtual Environments", "Telethrone" ] }, { "id": "octocopter", "title": "Octocopter", "domain": "robotics", "domain_name": "Robotics", "definition": "An Octocopter is a multirotor unmanned aerial vehicle (UAV) equipped with eight independently controlled rotors arranged symmetrically around a central frame, providing significant redundancy that allows continued stable flight after individual motor failure. The eight-rotor configuration enables lift capacity substantially exceeding that of quadcopters or hexacopters, making it the preferred platform for heavy industrial payloads such as professional cinema cameras, LiDAR scanners, and precision agricultural dispensers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:octocopter", "labels": [ "Octocopter" ], "is_subclass_of": [ "Robot Type", "Multirotor UAV" ], "wikilinks": [ "Multirotor UAV", "Robotics" ] }, { "id": "octree-spatial-index", "title": "Octree Spatial Index", "domain": "metaverse", "domain_name": "Metaverse", "definition": "An octree spatial index is a tree data structure that recursively subdivides 3D space into eight octants, organising objects by their spatial location for efficient querying. It accelerates operations such as range queries, nearest-neighbour search, collision detection, and visibility culling by pruning regions that cannot contain relevant objects. It is a core building block for spatial partitioning in 3D engines and virtual worlds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:octree-spatial-index", "labels": [ "Octree Spatial Index", "Octree" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "odometry", "title": "Odometry", "domain": "robotics", "domain_name": "Robotics", "definition": "The estimation of a mobile robot's position and orientation (pose) over time by integrating motion measurements from wheel encoders, IMUs, or visual sensors. It provides relative position estimates based on incremental motion and is subject to cumulative drift without correction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:odometry", "labels": [ "Odometry", "Lidar Odometry", "Odometry Sensors" ], "is_subclass_of": [ "State Estimation" ], "wikilinks": [ "Cumulative Error", "Drift", "IMU", "Measurement Errors", "Mobile Robotics", "Motion Sensors", "RB-1015-kalman-filter", "Relative Positioning", "Robot Pose", "Visual Sensors", "Wheel Slip", "Camera", "Encoder", "Navigation", "RB-1013-localization", "Robotics", "Sensor Fusion", "SLAM", "State Estimation" ] }, { "id": "ofcom", "title": "Ofcom", "domain": "governance", "domain_name": "Governance", "definition": "Ofcom (the Office of Communications) is the independent statutory regulator and competition authority for the UK communications industries, established by the Communications Act 2003. Its remit covers broadcasting, telecommunications, postal services, and \u2014 following the Online Safety Act 2023 \u2014 online platforms. It enforces spectrum management, sets licence conditions for broadcasters and telecoms operators, adjudicates on consumer complaints, and oversees compliance with the UK Online Safety Act framework requiring platforms to implement systems preventing access to illegal content.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ofcom", "labels": [ "Ofcom", "Ofcom Guidance" ], "is_subclass_of": [ "Regulatory Authority" ], "wikilinks": [] }, { "id": "off-chain-governance", "title": "Off Chain Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Off-chain governance refers to blockchain protocol governance processes conducted outside the ledger itself \u2014 through social consensus, developer forums, BIPs/EIPs, foundation decisions, and miner or validator coordination \u2014 rather than through on-chain voting mechanisms. Changes are agreed informally or through established processes, then implemented via software upgrades adopted voluntarily by network participants. This approach prioritises flexibility and expert deliberation but relies on social coordination and may lack the transparency of on-chain mechanisms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:off-chain-governance", "labels": [ "Off Chain Governance", "Off-Chain Governance", "Off-chain Governance" ], "is_subclass_of": [ "Distributed Ledger" ], "wikilinks": [] }, { "id": "off-chain-scaling", "title": "Off-Chain Scaling", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Off-chain scaling is a class of blockchain scalability techniques that move transaction execution and state off the main chain while retaining its security as a settlement and dispute-resolution layer. By processing many interactions outside the base layer and committing only summaries or final balances on-chain, it greatly increases throughput and lowers fees. Examples include payment and state channels and various layer-2 constructions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:off-chain-scaling", "labels": [ "Off-Chain Scaling" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "off-nadir-angle", "title": "Off-nadir Angle", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:off-nadir-angle", "labels": [ "Off-nadir Angle" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "offline-reinforcement-learning", "title": "Offline Reinforcement Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Offline reinforcement learning trains a policy from a fixed dataset of previously collected experience without further interaction with the environment. It avoids online exploration, which makes it suitable where data collection is costly or unsafe.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:offline-reinforcement-learning", "labels": [ "Offline Reinforcement Learning" ], "is_subclass_of": [ "Reinforcement Learning", "AI Technique" ], "wikilinks": [ "Reinforcement Learning", "Markov Decision Process", "Learning from Demonstration", "Imitation Learning", "Reward Function" ] }, { "id": "offline-rendering", "title": "Offline Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Offline rendering is the production of images or animation sequences where computation time per frame is not constrained to interactive rates, allowing each frame to take seconds, minutes, or hours to achieve maximum visual fidelity. It is the counterpart to real-time rendering and is typical of film, visual effects, and high-end visualisation, where physically based light transport such as path tracing is feasible. Offline rendering is commonly executed across render farms to parallelise the heavy computation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:offline-rendering", "labels": [ "Offline Rendering" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "offline-verification", "title": "Offline Verification", "domain": "security", "domain_name": "Security", "definition": "Offline verification is the ability to cryptographically validate a credential or claim without requiring a live connection to the issuer or a central server. The verifier checks digital signatures against the issuer's public key and any revocation data already held, confirming authenticity and integrity locally. It is a defining property of decentralised identity, enabling trust in low-connectivity or privacy-sensitive settings.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:offline-verification", "labels": [ "Offline Verification" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "olap", "title": "Olap", "domain": "data", "domain_name": "Data", "definition": "Online Analytical Processing (OLAP) is a category of data processing optimised for fast, multidimensional analysis of large volumes of historical and aggregated data. It organises measures along dimensions such as time, geography, and product, allowing analysts to slice, dice, roll up, and drill down through data cubes interactively. OLAP underpins business intelligence and decision support by enabling complex aggregate queries that contrast with the row-oriented, transactional focus of operational systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:olap", "labels": [ "Olap", "OLAP" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "olympus-dao", "title": "Olympus DAO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Olympus DAO is a decentralised autonomous organisation that pioneered the protocol-owned-liquidity model and the OHM reserve-currency token backed by a treasury of assets. It introduced bonding, where users sell assets to the protocol for discounted tokens, and staking rewards, aiming to build a community-owned treasury rather than relying on rented liquidity. It became a widely studied and forked template in decentralised finance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:olympus-dao", "labels": [ "Olympus DAO" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "omnichain-application", "title": "Omnichain Application", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An omnichain application is a decentralised application designed to operate across multiple blockchain networks simultaneously, maintaining unified state, liquidity, and user identity without requiring users to bridge assets manually between chains. Unlike multi-chain applications that deploy isolated instances on separate chains, omnichain applications treat all supported networks as a single logical execution environment connected by cross-chain messaging protocols. LayerZero is the most widely adopted infrastructure enabling omnichain application patterns.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:omnichain-application", "labels": [ "Omnichain Application" ], "is_subclass_of": [ "Decentralized Application" ], "wikilinks": [] }, { "id": "omnichannel-routing", "title": "Omnichannel Routing", "domain": "ai", "domain_name": "Ai", "definition": "Omnichannel routing is the contact-centre capability that directs customer interactions arriving across multiple channels, such as voice, chat, email, and social, to the most appropriate agent or automated handler using a unified queue and context. It maintains a single view of each customer's history so conversations can move between channels without loss of context. It is central to consistent, efficient customer service operations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:omnichannel-routing", "labels": [ "Omnichannel Routing" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "omnichannel", "title": "Omnichannel", "domain": "ai", "domain_name": "Ai", "definition": "Omnichannel is a customer engagement strategy that unifies all interaction channels \u2014 physical stores, web, mobile, social media, telephone, and messaging \u2014 into a single, consistent experience in which customer context and history transfer seamlessly across channel boundaries. Unlike multichannel approaches that operate channels in silos, omnichannel architectures share a common data layer and orchestration engine so that a session started on one channel can be continued on another without loss of state.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:omnichannel", "labels": [ "Omnichannel", "Omnichannel Support" ], "is_subclass_of": [ "Customer Experience Management" ], "wikilinks": [] }, { "id": "omnidirectional-robot", "title": "Omnidirectional Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A wheeled mobile robot that achieves holonomic motion\u2014the ability to translate in any direction and rotate independently\u2014by employing mecanum wheels or omniwheels whose passive rollers allow lateral force components. This enables precise, agile manoeuvring in constrained environments such as warehouses and hospital corridors without requiring turning arcs.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:omnidirectional-robot", "labels": [ "Omnidirectional Robot" ], "is_subclass_of": [ "Robot Type", "Wheeled Robot" ], "wikilinks": [ "Robotics", "Wheeled Robot" ] }, { "id": "omniverse", "title": "Omniverse", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A platform developed by NVIDIA for building and operating three-dimensional simulation and collaboration applications, based on the Universal Scene Description framework. It is used for digital twins and physically accurate simulation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:omniverse", "labels": [ "Omniverse", "Omniverse Kit" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [ "Simulation", "Metaverse", "NVIDIA" ] }, { "id": "on-chain-settlement", "title": "On Chain Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On chain settlement is the process by which the transfer of value or assets between parties is recorded directly on a distributed ledger and becomes economically irreversible according to the ledger's consensus rules. Unlike off-chain or netting arrangements that defer ledger updates, on chain settlement writes each transaction's final state into the canonical chain, removing the need for a trusted central clearing intermediary. Settlement is considered complete once the relevant transaction attains the chain's finality guarantee, after which reversal requires violating consensus. It underpins payment, securities, and decentralised finance flows on public and permissioned blockchains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:on-chain-settlement", "labels": [ "On Chain Settlement", "On-Chain Settlement" ], "is_subclass_of": [ "Settlement Finality" ], "wikilinks": [] }, { "id": "on-chain-voting", "title": "On Chain Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographically verifiable, blockchain-recorded governance mechanism enabling DAO participants to cast votes that are immutably recorded on public ledgers and automatically executed through Smart Contract|smart contracts without human intermediaries\u2014deployed across Uniswap,", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:on-chain-voting", "labels": [ "On Chain Voting", "BC-0462-on-chain-voting", "On-Chain Voting", "On-chain Voting Infrastructure", "OnChainVoting", "Proxy Voting" ], "is_subclass_of": [ "Governance and Regulation", "DAO", "Governance Token", "Smart Contract", "Decentralised Web", "Cryptography Security and Privacy", "Distributed Ledger" ], "wikilinks": [ "Aave", "Anti-Collusion Voting", "Aragon OSx Standard", "Ben-Sasson et al Zerocash IEEE SP 2014", "Block Snapshot", "Buterin 2021 Moving Beyond Coin Voting", "Buterin 2022 Soulbound", "Buterin Hitzig Weyl 2019 Quadratic Funding Management Science", "Cambridge CCAF Cryptoasset Governance Report 2025", "Centralised Governance", "Commons Stack Conviction Voting", "Compound", "Compound Governor", "Compound Governor Bravo", "Compound Governor Bravo Documentation", "Cross-Chain Governance", "Cryptographic Signature", "CryptographyDomain", "Curve veCRV Specification", "Decentralised Exchange" ] }, { "id": "on-device-ai", "title": "On Device Ai", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "On-device AI is the execution of machine-learning inference, and increasingly some training, directly on an end-user device such as a phone, wearable or embedded sensor, rather than sending data to remote servers. Keeping computation local reduces latency, removes network dependence and improves privacy because raw data need not leave the device. It relies on model compression, quantisation and hardware acceleration to fit capable models within tight power, memory and compute budgets, and it is a cornerstone of edge AI and privacy-preserving machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:on-device-ai", "labels": [ "On Device Ai", "On-Device AI" ], "is_subclass_of": [ "Edge AI" ], "wikilinks": [] }, { "id": "on-chain-data-indexing", "title": "On-Chain Data Indexing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On-chain data indexing is the process of extracting, transforming, and organising raw blockchain events and state into queryable structures suited to application access patterns. Because base-layer data is optimised for consensus rather than retrieval, indexers ingest blocks and logs, decode them, and serve them through APIs such as GraphQL. It is the data backbone for analytics dashboards and decentralised applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:on-chain-data-indexing", "labels": [ "On-Chain Data Indexing", "Consistent Indexing", "Data Indexing" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "on-chain-data", "title": "On-Chain Data", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On-Chain Data is any information permanently recorded within a blockchain's ledger, including transactions, account balances, smart-contract state, and event logs. It is publicly verifiable and immutable once confirmed, distinguishing it from off-chain data held in external databases or oracles. Analytics platforms, compliance tools, and price oracles all query on-chain data directly from full nodes or indexers.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:on-chain-data", "labels": [ "On-Chain Data" ], "is_subclass_of": [ "Blockchain Data" ], "wikilinks": [] }, { "id": "on-chain-identity", "title": "On-Chain Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On-chain identity is a persistent, cryptographically controlled identifier recorded on a blockchain that an entity uses to accumulate verifiable attributes, history, and reputation. Anchored to an address or smart contract and controlled by private keys, it can hold credentials, tokens, and attestations that other applications read trustlessly. It enables portable reputation and access control across decentralised applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:on-chain-identity", "labels": [ "On-Chain Identity", "On-Chain Identity ONCHAINID" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "on-chain-mrv", "title": "On-Chain MRV", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On-Chain MRV (Measurement, Reporting, and Verification) is the practice of recording carbon and environmental impact data directly on a blockchain to create tamper-evident, auditable proof of climate actions. It combines sensor data, satellite imagery, and oracle feeds with on-chain attestations so that carbon credits and sustainability claims can be independently verified. By anchoring MRV data to an immutable ledger, it reduces double-counting and greenwashing in voluntary and compliance carbon markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:on-chain-mrv", "labels": [ "On-Chain MRV" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "on-chain-transaction", "title": "On-Chain Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An on-chain transaction is a value or state transfer that is broadcast to a blockchain network, validated by consensus, and permanently recorded in a block on the shared ledger. Because settlement occurs through the network's consensus mechanism, on-chain transactions inherit the chain's security, immutability, and public verifiability, but also its latency, throughput limits, and fee costs. They contrast with off-chain and layer-two approaches that defer or aggregate ledger updates to improve scalability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:on-chain-transaction", "labels": [ "On-Chain Transaction" ], "is_subclass_of": [ "Blockchain Transaction", "Transaction" ], "wikilinks": [] }, { "id": "on-device-inference", "title": "On-Device Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "On-device inference is the execution of machine learning model forward passes entirely on the end-user's hardware \u2014 such as a smartphone, wearable, embedded controller, or edge server \u2014 without transmitting input data to a remote cloud backend. It requires models to be compressed, quantised, or distilled to fit within tight memory, compute, and power budgets while maintaining acceptable accuracy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:on-device-inference", "labels": [ "On-Device Inference", "On-Device Privacy" ], "is_subclass_of": [ "Edge Inference" ], "wikilinks": [] }, { "id": "on-device-learning", "title": "On-Device Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine learning model training and adaptation occurring directly on end-user devices using only local data, without transmitting raw data to cloud servers. Enables personalised model adaptation, privacy preservation, and offline functionality while addressing challenges of limited compute, memory, and energy. Implemented through transfer learning, few-shot adaptation, and continual learning techniques on mobile and embedded platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:on-device-learning", "labels": [ "On-Device Learning", "On-Device Machine Learning" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Apple Core ML", "PyTorch Mobile", "TensorFlow Lite", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "on-premises", "title": "On-Premises", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "On-premises is the deployment model in which an organisation runs its computing infrastructure \u2014 servers, storage, networking, and the software on top \u2014 in facilities it owns or directly controls, rather than renting capacity from a cloud provider. The organisation purchases hardware as capital expenditure, operates and patches the full stack itself, and retains physical custody of its data, trading the elasticity and managed services of public cloud for maximal control over locality, latency, security boundaries, and long-run unit costs.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:on-premises", "labels": [ "On-Premises" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [ "Infrastructure", "Cloud Computing", "Hybrid Cloud", "Infrastructure As A Service", "Software As A Service", "Data Centre", "Edge Computing" ] }, { "id": "on-chain-governance", "title": "on-chain governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "On-chain governance is a blockchain coordination model in which protocol upgrade proposals, parameter changes, and treasury spending decisions are formally submitted, deliberated, voted on by token holders, and automatically enacted through the execution of smart contracts recorded on the distributed ledger. Governance logic is codified directly in the protocol layer, making all votes, quorum checks, and execution outcomes immutable, censorship-resistant, and publicly auditable on-chain records. It contrasts with off-chain governance, where decisions emerge from social consensus (forums, developer meetings, improvement-proposal repositories) and are implemented by core developers without cryptographic enforcement, creating reliance on trust in key actors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:on-chain-governance", "labels": [ "On-chain Governance", "On-Chain Governance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "onboard-computer", "title": "Onboard Computer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:onboard-computer", "labels": [ "Onboard Computer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "ondo-finance", "title": "Ondo Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Ondo Finance is a company that issues tokenised versions of traditional financial products, including funds holding US Treasury securities, on public blockchains.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ondo-finance", "labels": [ "Ondo Finance" ], "is_subclass_of": [ "Asset Tokenisation" ], "wikilinks": [ "Asset Tokenisation", "Tokenisation", "Institutional Adoption" ] }, { "id": "one-hot-encoding", "title": "One Hot Encoding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "One hot encoding is a representation technique that converts categorical variables into binary vectors, where each category is mapped to a vector containing a single high (1) value and all other positions set to zero. It removes any implied ordinal relationship between categories, allowing machine learning models that operate on numeric input to consume nominal data without inferring spurious magnitude. The dimensionality of the encoding equals the cardinality of the category set, which can become sparse and high-dimensional for variables with many levels.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:one-hot-encoding", "labels": [ "One Hot Encoding", "One-Hot Encoding" ], "is_subclass_of": [ "Feature Engineering" ], "wikilinks": [] }, { "id": "one-time-password", "title": "One Time Password", "domain": "security", "domain_name": "Security", "definition": "A one-time password (OTP) is a credential that is valid for only a single login session or transaction, mitigating the risk of credential replay associated with static passwords. OTPs are typically generated from a shared secret combined with a moving factor \u2014 either a counter (HOTP) or the current time (TOTP) \u2014 using an HMAC construction, and delivered through authenticator apps, hardware tokens, or out-of-band channels such as SMS. While OTPs strengthen authentication as a second factor, they remain susceptible to real-time phishing, in contrast to origin-bound phishing-resistant methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:one-time-password", "labels": [ "One Time Password", "One-Time Password" ], "is_subclass_of": [ "Multi-Factor Authentication" ], "wikilinks": [] }, { "id": "one-way-function", "title": "One Way Function", "domain": "security", "domain_name": "Security", "definition": "A one-way function is a function that is easy to compute on any input but computationally infeasible to invert, meaning that recovering the input from a typical output is practically impossible with available resources. One-way functions are a foundational primitive of modern cryptography, underpinning hashing, password storage, and the trapdoor constructions used in public-key schemes. Their existence is conjectured rather than proven, and it is closely tied to open questions in computational complexity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:one-way-function", "labels": [ "One Way Function", "One-Way Function" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "oneM2M", "title": "Onem2M", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "oneM2M is a global standards initiative that defines a common service layer for machine-to-machine and Internet of Things communication, enabling interoperable connection of devices, applications, and platforms across vertical industries. Developed by a partnership of regional standards bodies including ETSI, it specifies a horizontal middleware exposing reusable capabilities such as data management, device management, security, and discovery through a RESTful resource model. By abstracting these common functions, oneM2M reduces fragmentation and lets IoT solutions span domains like smart cities, transport, and energy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:oneM2M", "labels": [ "Onem2M", "oneM2M" ], "is_subclass_of": [ "IoT Infrastructure" ], "wikilinks": [] }, { "id": "onfido", "title": "Onfido", "domain": "security", "domain_name": "Security", "definition": "Onfido is an identity verification company that combines document checks with facial biometrics to confirm users during remote onboarding. It was acquired by Entrust in 2024.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:onfido", "labels": [ "Onfido" ], "is_subclass_of": [ "Identity Verification System" ], "wikilinks": [ "Facial Recognition", "Biometric Authentication", "Identity Management", "iProov", "Identity Verification System" ] }, { "id": "onion-routing", "title": "Onion Routing", "domain": "security", "domain_name": "Security", "definition": "Onion routing is an anonymous-communication technique in which messages are wrapped in successive layers of encryption and relayed through a sequence of intermediary nodes, each of which removes one layer to learn only the next hop. Because no single relay knows both the source and destination, onion routing conceals the network path and protects communication metadata. It is the basis of the Tor network and is adapted in systems such as the Lightning Network for private multi-hop payment forwarding.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:onion-routing", "labels": [ "Onion Routing", "OnionRouting", "Sphinx Routing", "onion-routing" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "online-certificate-status-protocol", "title": "Online Certificate Status Protocol", "domain": "security", "domain_name": "Security", "definition": "The Online Certificate Status Protocol (OCSP) is an internet protocol for obtaining the real-time revocation status of an X.509 digital certificate. A client queries an OCSP responder, which returns a signed good, revoked or unknown status, avoiding the need to download large certificate revocation lists. OCSP stapling allows a server to present a recent signed status during the TLS handshake to improve privacy and performance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:online-certificate-status-protocol", "labels": [ "Online Certificate Status Protocol" ], "is_subclass_of": [ "Public Key Infrastructure" ], "wikilinks": [] }, { "id": "online-identity", "title": "Online Identity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The digital representation and persistent persona that an individual constructs and maintains across online platforms and virtual environments, encompassing usernames, avatars, credentials, reputation data, and linked digital assets. Online identity management raises questions of authentication, portability, privacy, and the mapping between real-world and digital personas.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:online-identity", "labels": [ "Online Identity" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "online-learning", "title": "Online Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Online Learning is a artificial intelligence concept and a type of Machine Learning. that enables Real-Time Learning.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:online-learning", "labels": [ "Online Learning", "E-Learning", "Online Machine Learning" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Real-Time Learning", "Machine Learning" ] }, { "id": "online-safety-act-2023", "title": "Online Safety Act 2023", "domain": "governance", "domain_name": "Governance", "definition": "The Online Safety Act 2023 is a landmark piece of UK primary legislation that received Royal Assent on 26 October 2023, establishing a comprehensive duty-of-care framework for online platforms and search services operating in or targeting users in the United Kingdom. It imposes tiered obligations proportionate to service size and risk profile: all in-scope providers must conduct risk assessments and remove illegal content swiftly; larger or higher-risk services face additional requirements covering child safety, user empowerment tools, and transparency reporting. Ofcom is designated as the statutory regulator with powers to audit compliance, issue fines of up to 10% of global annual turnover, and ultimately block non-compliant services from reaching UK users. The Act also introduces personal criminal liability for senior managers of companies that wilfully fail to protect children from serious online harms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:online-safety-act-2023", "labels": [ "Online Safety Act 2023", "Online Safety Act" ], "is_subclass_of": [ "Digital Regulation" ], "wikilinks": [] }, { "id": "ontology-alignment", "title": "Ontology Alignment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Ontology alignment is the process of determining correspondences between the concepts, properties and relations of two or more separately developed ontologies. The output, an alignment or set of mappings, allows systems using different ontologies to interoperate by translating or relating their terms. Alignment is a core enabler of semantic interoperability across heterogeneous knowledge sources and is closely related to schema matching and entity resolution.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ontology-alignment", "labels": [ "Ontology Alignment" ], "is_subclass_of": [ "Ontology" ], "wikilinks": [] }, { "id": "ontology-definition", "title": "Ontology Definition", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A formal, explicit specification of metaverse concepts, relationships, and axioms using knowledge representation frameworks (OWL, RDF, JSON-LD), defining entities through orthogonal dimensions (physicality, role) to enable semantic interoperability and automated classification across platforms. Ontologies provide foundational standards for governance, identity, and asset management, supporting machine-readable reasoning, validation, and discovery across federated and distributed systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ontology-definition", "labels": [ "Ontology Definition" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "InteroperabilityProtocol", "JSON-LD", "OWL", "RDF", "Metaverse", "MetaverseDomain" ] }, { "id": "ontology-engineering", "title": "Ontology Engineering", "domain": "data", "domain_name": "Data", "definition": "Ontology Engineering is the discipline of designing, formalizing, and maintaining ontologies: explicit, machine-readable specifications of concepts, properties, and relationships within a domain. It applies methodologies and logic languages such as OWL and RDFS to produce consistent, reusable, and reasoning-capable knowledge models. The field underpins semantic interoperability, knowledge graphs, and automated inference across heterogeneous data sources.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ontology-engineering", "labels": [ "Ontology Engineering", "Ontology Reasoner" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "ontology-loom", "title": "Ontology Loom", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A portable serving node that grounds LLM responses in a formal ontology behind a stable, model-swappable fa\u00e7ade. At query time it retrieves the relevant slice of the reasoned ontology, injects it as a budget-clamped structured scaffold, and delegates generation to whichever model sits behind it, so the model restates checked facts rather than performing open-ended recall. On a held-out benchmark, static scaffold grounding lifted paired answer scores to 0.94 on two different models, from parametric baselines of 0.146 and 0.268, at three to six times lower latency. The Ontology Loom is the DreamLab mesh's implementation of the context-graph layer. Distinct from Loom, the screen-recording product.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ontology-loom", "labels": [ "Ontology Loom" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [ "Context Graph", "Knowledge Graph", "Ontology", "Retrieval-Augmented Generation", "Reasoning Engine" ] }, { "id": "ontology-property-definitions", "title": "Ontology Property Definitions", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Comprehensive definitions of object properties and datatype properties used throughout the Disruptive Technology Ontology, establishing formal semantic relationships between concepts across AI, blockchain, robotics, and metaverse domains. Properties define typed, directional links (enables, requires, hasPart, uses, etc.) with explicit domains, ranges, inverse relationships, and logical characteristics enabling automated reasoning and knowledge graph traversal.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ontology-property-definitions", "labels": [ "Ontology Property Definitions", "Property Definitions" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Dublin Core Metadata Initiative", "OntologyDomain", "OWL 2 Web Ontology Language", "AIEthicsDomain", "BlockchainDomain", "RoboticsDomain" ] }, { "id": "ontology-structure", "title": "Ontology Structure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The formal organisation of an ontology into classes, properties, axioms, and inter-concept relationships expressed in a logic-based language such as OWL or RDFS. Ontology Structure defines the hierarchy and constraint patterns that make machine-readable knowledge graphs queryable, inferable, and interoperable across domains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ontology-structure", "labels": [ "Ontology Structure" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "ontology-in-llm-operations", "title": "Ontology in LLM Operations", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of formal ontological structures to guide, constrain, and enrich large language model inference and retrieval pipelines. Ontologies supply typed entity schemas, relation vocabularies, and axioms that reduce hallucination, improve structured output consistency, and enable semantic grounding of LLM responses within knowledge-graph-backed retrieval-augmented generation systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ontology-in-llm-operations", "labels": [ "Ontology in LLM Operations" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "ontology", "title": "ontology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In computer science and knowledge engineering, an ontology is a formal, machine-readable specification of a shared conceptualisation within a domain: it explicitly defines the classes of entities that exist, the properties and attributes of those entities, the relationships (object properties) that may hold between them, and the logical axioms that constrain valid world states. Ontologies are expressed using languages such as OWL 2 (Web Ontology Language) built atop RDF, enabling automated reasoning engines to infer implicit knowledge and detect logical inconsistencies. They underpin knowledge graphs, linked data systems, semantic interoperability frameworks, and AI knowledge representation across scientific, industrial, and Web domains. Rooted in philosophical ontology (the study of being and categories of existence), the computational form translates that tradition into machine-processable schemas governed by Description Logics with decidable inference.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:ontology", "labels": [ "Ontology", "Ontology Reasoning" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "op-stack", "title": "Op Stack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The OP Stack is an open-source, modular software framework, originally developed for Optimism, used to deploy and operate Ethereum layer-2 networks built on the optimistic rollup model. It standardises the components of a rollup chain, including the sequencer, derivation pipeline, fault-proof system, and bridge contracts, so that many independent chains can share a common technical foundation. Chains built on it interoperate as a federation often described as a superchain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:op-stack", "labels": [ "Op Stack", "OP Stack" ], "is_subclass_of": [ "Optimistic Rollup" ], "wikilinks": [] }, { "id": "opcodes", "title": "Opcodes", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Opcodes (operation codes) are the atomic instruction primitives of a blockchain scripting language that define permissible computations within transaction scripts, smart contracts, or virtual machine execution environments. Each opcode specifies an operation \u2014 such as hash computation, signature verification, stack manipulation, or conditional branching \u2014 and the set of valid opcodes for a given blockchain determines its scripting expressiveness and security surface. In Bitcoin, a deliberately restricted opcode set enforces non-Turing-completeness and predictable resource consumption, whereas Ethereum's EVM opcode set supports general computation within gas-metered bounds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:opcodes", "labels": [ "Opcodes", "Opcode" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "open-ai-chat-completions-api", "title": "Open AI Chat Completions API", "domain": "ai", "domain_name": "Ai", "definition": "The OpenAI Chat Completions API is a widely adopted HTTP interface for sending a sequence of role-tagged messages to a large language model and receiving a generated response. Its request and response schema, including roles, tool-calling, and streaming, has become a de facto interoperability standard implemented by many open-source and third-party inference servers. This compatibility lets applications swap model backends with minimal code change.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:openai-research-organisation-chat-completions-api", "labels": [ "Open AI Chat Completions API" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "open-ai-whisper", "title": "Open AI Whisper", "domain": "ai", "domain_name": "Ai", "definition": "OpenAI Whisper is an open-source automatic speech recognition (ASR) system released by OpenAI in September 2022, trained on approximately 680,000 hours of multilingual and multitask supervised audio data sourced from the internet. It employs a Transformer encoder-decoder architecture that jointly learns transcription, translation, and language identification from weakly supervised training data, achieving near-human accuracy across a broad range of accents, recording conditions, and languages. Whisper is released as open weights in multiple sizes (tiny, base, small, medium, large, and subsequent variants), enabling local deployment without API dependency. Its robustness to background noise, accented speech, and domain-specific vocabulary\u2014combined with zero-shot multilingual performance\u2014has made it the de facto baseline for ASR research and a widely deployed component in transcription pipelines, voice interfaces, and accessibility applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:openai-research-organisation-whisper", "labels": [ "Open AI Whisper", "OpenAI Whisper" ], "is_subclass_of": [ "Automatic Speech Recognition" ], "wikilinks": [] }, { "id": "open-access", "title": "Open Access", "domain": "governance", "domain_name": "Governance", "definition": "Open access is the practice of making scholarly and other knowledge outputs freely available online without subscription or most copyright and licensing barriers. It aims to maximise the dissemination, reuse and impact of research by removing paywalls and granting broad reuse rights. Open access is a core component of the wider open science and open data movements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-access", "labels": [ "Open Access" ], "is_subclass_of": [ "Open Data" ], "wikilinks": [] }, { "id": "open-banking", "title": "Open Banking", "domain": "finance", "domain_name": "Finance", "definition": "Open Banking is a regulatory and technical framework that allows authorised third-party providers to access customer banking data and initiate payments through secure application programming interfaces, subject to explicit customer consent. It shifts control of financial data from incumbent banks to the account holder, who may grant fine-grained, revocable permissions. The model underpins regulated data sharing regimes such as the EU's PSD2 and the UK's Open Banking Standard, fostering competition and new financial products.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-banking", "labels": [ "Open Banking" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "open-bridge-standard", "title": "Open Bridge Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Open Bridge Standard is a publicly specified protocol for transferring assets and messages between independent blockchains in an interoperable, vendor-neutral way. It defines common formats for lock/mint, burn/release, and message-passing operations so that distinct bridge implementations can interoperate and be audited against shared security assumptions. Such standards aim to reduce fragmentation and the systemic risk associated with bespoke cross-chain bridges.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-bridge-standard", "labels": [ "Open Bridge Standard" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "open-container-initiative", "title": "Open Container Initiative", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Open Container Initiative (OCI) is an open governance structure under the Linux Foundation that produces vendor-neutral specifications for container formats and runtimes. Its core deliverables are the image specification, the runtime specification, and the distribution specification, which together standardise how container images are built, distributed, and executed. By defining these contracts, the OCI guarantees interoperability and portability across the container ecosystem.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-container-initiative", "labels": [ "Open Container Initiative" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "open-data", "title": "Open Data", "domain": "data", "domain_name": "Data", "definition": "Open Data is data that is freely available for anyone to access, use, modify, and share, subject at most to attribution and share-alike requirements. It is typically published in machine-readable formats with open licenses to maximize reuse, transparency, and interoperability across organizations. Open Data underpins public-sector transparency, research reproducibility, and decentralized web ecosystems built on user-controlled data stores.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-data", "labels": [ "Open Data", "Government Open Data", "Open Data Portal", "Open Data Publishing", "Open Datasets", "Open Protocols" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "open-generative-ai-tools", "title": "Open Generative AI tools", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Open Generative AI tools is the ecosystem of openly licensed or openly released generative AI models, fine-tuning pipelines, inference infrastructure, and community-distribution platforms that collectively enable practitioners to download, modify, deploy, and share large-scale foundation models f...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:open-generative-ai-tools", "labels": [ "Open Generative AI tools", "Generative AI Tooling", "Open Generative AI Tools" ], "is_subclass_of": [ "AI Infrastructure", "Machine Learning Infrastructure", "Generative AI", "Open Source Software", "Large-Scale Pretrained Foundation Model", "AI Ecosystem", "Large Language Models" ], "wikilinks": [ "AI Ecosystem", "Apache License 2.0", "API-Only AI Services", "Civitai", "Closed-Weight Models", "CommunityEcosystemDomain", "CommunityLayer", "Community Licensing", "Compute Resources", "CUDA", "Diffusers Library", "Distributed Training", "Flash Attention", "GGUF Format", "GGUF Specification", "GPU Infrastructure", "Grouped Query Attention", "Hugging Face Hub", "InferenceInfrastructureDomain", "Inference Runtime" ] }, { "id": "open-governance", "title": "Open Governance", "domain": "standards", "domain_name": "Standards", "definition": "Open governance is a model of decision-making for projects, standards and organisations in which processes are transparent, participation is open and authority is distributed among stakeholders. It emphasises published rules, accountable roles and consensus-driven decisions rather than closed or unilateral control. It is common in open-source communities, standards bodies and multi-stakeholder institutions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-governance", "labels": [ "Open Governance" ], "is_subclass_of": [ "Open Source Governance" ], "wikilinks": [] }, { "id": "open-government", "title": "Open Government", "domain": "governance", "domain_name": "Governance", "definition": "Open government is a governance approach grounded in transparency, citizen participation, and accountability, holding that public institutions should make their information and decision-making accessible and engage the public in shaping them. It is operationalised through open data, freedom-of-information mechanisms, participatory processes, and the use of digital technology to expose how government works. The aim is to build public trust, improve services, and enable scrutiny of the use of public resources.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-government", "labels": [ "Open Government" ], "is_subclass_of": [ "Digital Governance" ], "wikilinks": [] }, { "id": "open-graph-link-preview-protocol", "title": "Open Graph Link Preview Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Link Preview is a web infrastructure technique that fetches and parses Open Graph metadata, title, description, and thumbnail from a target URL to generate a rich card preview for display in social feeds, chat applications, or note-taking tools. APIs and Python libraries such as linkpreview and URLMeta automate this extraction, enabling knowledge graphs and collaborative platforms to surface contextual summaries alongside raw hyperlinks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:open-graph-link-preview-protocol", "labels": [ "Open Graph Link Preview Protocol", "Link Preview", "Open Graph Protocol" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "open-home-foundation", "title": "Open Home Foundation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Open Home Foundation is a non-profit organization that stewards open-source smart-home projects, most notably Home Assistant, to advance privacy, choice, and sustainability in home automation. It holds and governs the intellectual property of these projects, funds development, and protects them from commercial capture. Its mission centers on local-first, user-controlled home technology rather than cloud-dependent proprietary platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-home-foundation", "labels": [ "Open Home Foundation" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "open-market-operations", "title": "Open Market Operations", "domain": "finance", "domain_name": "Finance", "definition": "Open market operations (OMO) are the purchase and sale of government securities and other eligible assets by a central bank in the open market to steer short-term interest rates and the supply of reserves in the banking system. By adding or draining reserves, OMO move the policy rate toward its target and transmit monetary policy to the wider economy. They are the principal day-to-day tool of monetary-policy implementation in most advanced economies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:open-market-operations", "labels": [ "Open Market Operations" ], "is_subclass_of": [ "Monetary Policy Implementation" ], "wikilinks": [] }, { "id": "open-metaverse-interoperability-group", "title": "Open Metaverse Interoperability Group", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The Open Metaverse Interoperability Group (OMI) is a community-driven body that develops open protocols and schemas for moving identity, avatars, and virtual objects between independent virtual worlds. It produces specifications for portable avatars, glTF extensions, and cross-platform identity so that metaverse experiences are not locked into single proprietary platforms. Its work aims to make virtual environments composable and interoperable rather than walled gardens.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-metaverse-interoperability-group", "labels": [ "Open Metaverse Interoperability Group" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "open-metaverse-interoperability", "title": "Open Metaverse Interoperability", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Open metaverse interoperability is the principle and set of open standards enabling users, avatars, digital assets, and identities to move freely across independently operated virtual worlds and platforms without proprietary lock-in. It depends on shared formats for 3D content, portable avatar and identity descriptions, and common protocols for asset provenance and transport. Driven by bodies such as the Metaverse Standards Forum and the Khronos Group, it aims to make the metaverse a federated network of interoperable spaces rather than a collection of isolated walled gardens.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:open-metaverse-interoperability", "labels": [ "Open Metaverse Interoperability" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "open-metaverse", "title": "Open Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The open metaverse is a vision of interconnected virtual worlds built on open standards, interoperable formats, and user-owned assets and identity, in contrast to closed, single-vendor platforms. It emphasises portability of avatars, content, and value across experiences, decentralised ownership, and protocols rather than walled gardens. The open metaverse depends on shared standards for 3D assets, identity, and interoperability so that participation is not locked to any one provider.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:open-metaverse", "labels": [ "Open Metaverse" ], "is_subclass_of": [ "Metaverse" ], "wikilinks": [] }, { "id": "open-policy-agent", "title": "Open Policy Agent", "domain": "security", "domain_name": "Security", "definition": "Open Policy Agent (OPA) is an open-source, general-purpose policy engine that decouples authorisation and policy decisions from application code by evaluating declarative policies against structured input. Policies are written in its purpose-built language, Rego, and OPA returns decisions that calling services enforce. As a graduated Cloud Native Computing Foundation project, it is widely used to implement policy-as-code across Kubernetes admission control, microservice authorisation, and API gateways. OPA acts as a policy decision point, leaving enforcement to integrated policy enforcement points.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-policy-agent", "labels": [ "Open Policy Agent" ], "is_subclass_of": [ "Policy Engine" ], "wikilinks": [] }, { "id": "open-protocol", "title": "Open Protocol", "domain": "standards", "domain_name": "Standards", "definition": "A communication protocol whose specification is publicly documented, royalty-free, and implementable by anyone without permission from a controlling party, typically evolved through an open governance process; open protocols such as TCP/IP, HTTP, SMTP, ActivityPub, and Nostr enable permissionless interoperability between independent implementations, prevent single-vendor lock-in, and shift competition from control of the network to quality of the client \u2014 the architectural foundation of the open internet and of credible open-metaverse and decentralised social efforts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:open-protocol", "labels": [ "Open Protocol" ], "is_subclass_of": [ "Protocol" ], "wikilinks": [ "Protocol", "Interoperability", "Nostr", "Farcaster" ] }, { "id": "open-rights-group", "title": "Open Rights Group", "domain": "governance", "domain_name": "Governance", "definition": "Open Rights Group is a United Kingdom organisation that campaigns for digital rights, including privacy, free expression, and data protection. It is a membership-funded non-profit.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:open-rights-group", "labels": [ "Open Rights Group" ], "is_subclass_of": [ "Digital Rights" ], "wikilinks": [ "Privacy", "User Sovereignty", "Digital Rights", "https://www.openrightsgroup.org", "https://www.openrightsgroup.org/about/" ] }, { "id": "open-science", "title": "Open Science", "domain": "governance", "domain_name": "Governance", "definition": "Open science is a movement and set of practices aimed at making scientific research, data and dissemination accessible, transparent and reproducible for all members of society. It encompasses open access to publications, open data, open-source tools and open methodologies, lowering barriers to participation and verification of research. By promoting sharing throughout the research lifecycle, open science seeks to accelerate discovery, improve reproducibility and broaden public trust in science.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-science", "labels": [ "Open Science" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "open-source-ai-models", "title": "Open Source AI Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial intelligence models whose weights, architecture, and training data are made publicly available for use, modification, and redistribution by the community.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:open-source-ai-models", "labels": [ "Open Source AI Models" ], "is_subclass_of": [ "Large Language Model" ], "wikilinks": [] }, { "id": "open-source-development", "title": "Open Source Development", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Open source development is a software production methodology in which source code is made publicly available under licences permitting inspection, modification, and redistribution, enabling distributed communities of contributors to collaboratively build and maintain software. The model contrasts with proprietary development by prioritising transparency, peer review, and meritocratic contribution over closed, hierarchical ownership. It encompasses practices including version-controlled public repositories, issue tracking, pull-request-based code review, and community governance structures. Open source development has become the dominant paradigm for infrastructure software, AI frameworks, and blockchain protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-source-development", "labels": [ "Open Source Development", "Open Source Protocol Development" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "open-source-digital-painting-application", "title": "Open Source Digital Painting Application", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Krita is a free, open-source professional digital painting and raster graphics application developed by the KDE community. It supports illustration, concept art, texture painting, and AI-assisted image synthesis via plugins such as krita-ai-diffusion, which integrates Stable Diffusion through a ComfyUI backend on a local or remote inference server.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:open-source-digital-painting-application", "labels": [ "Open Source Digital Painting Application", "krita" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "open-source-framework", "title": "Open Source Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A publicly licensed software framework whose source code is freely available for inspection, modification, and redistribution under an approved open-source licence. Open source frameworks provide reusable architectural scaffolding\u2014APIs, libraries, conventions, and tooling\u2014that accelerate development whilst enabling community-driven quality assurance, security auditing, and interoperability across vendor boundaries.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:open-source-framework", "labels": [ "Open Source Framework" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "open-source-governance", "title": "Open Source Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Open source governance is the set of processes, roles and norms by which an open source project makes decisions about its direction, accepts contributions, resolves disputes and manages releases. Models range from benevolent-dictator and meritocratic maintainer structures to elected steering committees and foundation stewardship, each balancing openness against the need for coordination. It determines who may merge changes, how proposals are reviewed, and how the project sustains trust and continuity across a distributed, often volunteer, community.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-source-governance", "labels": [ "Open Source Governance" ], "is_subclass_of": [ "Decentralised Governance" ], "wikilinks": [] }, { "id": "open-source-initiative", "title": "Open Source Initiative", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The Open Source Initiative is a non-profit organisation that stewards the Open Source Definition and maintains the authoritative process for reviewing and approving software licences as conforming to it. By certifying which licences qualify as open source, it provides a stable, community-recognised standard that distinguishes genuine open-source terms from merely source-available ones. The organisation also advocates for open-source software and educates on licensing and policy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-source-initiative", "labels": [ "Open Source Initiative" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "open-source-licence", "title": "Open Source Licence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "An open source licence is a legal instrument that grants users the rights to use, study, modify, and redistribute software source code, subject to conditions defined by the licence. Such licences fall broadly into permissive families, which impose few obligations, and copyleft families, which require derivative works to remain under compatible terms. Recognition typically follows the Open Source Initiative's definition, and licences are identified by standard SPDX identifiers to support automated compliance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:open-source-licence", "labels": [ "Open Source Licence" ], "is_subclass_of": [ "Software Licence" ], "wikilinks": [] }, { "id": "open-source-monetization", "title": "Open Source Monetization", "domain": "economics", "domain_name": "Economics", "definition": "The strategies and challenges associated with generating sustainable revenue from open-source software projects in the AI era.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:open-source-monetization", "labels": [ "Open Source Monetization" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "open-source-social-immersive-space", "title": "Open Source Social Immersive Space", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Flossverse is a conceptual and practical framework for building an open-source, trust-minimised social immersive space that combines free/libre open-source software (FLOSS), decentralised value transfer (Bitcoin/Lightning), generative AI tooling, and open XR standards to enable inclusive economic participation in virtual and mixed-reality environments. The project targets B2B, B2C, and creator-to-consumer workflows with a minimum viable product centred on trustless value exchange within a social VR context, with particular emphasis on expanding access for emerging markets and creators currently excluded from media production pipelines.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:open-source-social-immersive-space", "labels": [ "Open Source Social Immersive Space", "flossverse" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "jones2008trust" ] }, { "id": "open-source-software", "title": "Open Source Software", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Open Source Software (OSS) is software distributed with source code made publicly available under a licence that grants users the rights to inspect, modify, and redistribute the software and its derivatives. OSS development follows collaborative models coordinated through version-controlled repositories, issue trackers, and community governance structures, enabling distributed contribution at global scale. Canonical licence categories range from permissive instruments (MIT, Apache 2.0, BSD) to copyleft instruments (GPL, LGPL, AGPL) that impose viral redistribution obligations. In the AI, spatial-computing, and broader technology infrastructure landscape, OSS underpins foundational frameworks \u2014 including deep learning toolkits, container runtimes, distributed databases, and interoperability specifications \u2014 reducing vendor lock-in and accelerating innovation through transparent, auditable codebases.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:open-source-software", "labels": [ "Open Source Software", "Open-Source Software" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "open-source-sustainability", "title": "Open Source Sustainability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Open source sustainability concerns the long-term viability of freely available software projects, addressing how maintainers are funded, how contributor effort is sustained, and how critical dependencies are kept secure and maintained. It examines the mismatch between the widespread reliance on open source infrastructure and the scarce, often volunteer, resources behind it. Funding mechanisms such as grants, quadratic funding, and retroactive public-goods funding attempt to close this gap.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:open-source-sustainability", "labels": [ "Open Source Sustainability" ], "is_subclass_of": [ "Open Source Software" ], "wikilinks": [] }, { "id": "open-source", "title": "Open Source", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Open source designates software whose source code is released under a licence conforming to the Open Source Definition, granting anyone the rights to use, study, modify, and redistribute the code and derivative works. It is a development paradigm rooted in decentralised, community-driven collaboration where peer review, transparent workflows, and shared governance replace proprietary, closed development. Open source underpins the majority of modern computing infrastructure \u2014 from operating systems and compilers to cloud platforms and AI frameworks \u2014 and has expanded beyond software into hardware designs, datasets, models, and scientific research artefacts. Licences range from permissive (MIT, Apache 2.0) to copyleft (GPL, AGPL), with the choice of licence shaping how downstream users may integrate the code.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:open-source", "labels": [ "Open Source", "Open Source Collaboration", "Open Source Ecosystem", "Open-Source Project" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [ "Software Development", "GitHub", "https://opensource.org/osd", "https://opensource.org/licenses" ] }, { "id": "open-standard", "title": "Open Standard", "domain": "standards", "domain_name": "Standards", "definition": "An open standard is a publicly accessible specification for a technology, protocol, format, or interface that is developed through a transparent, consensus-based process and can be implemented and used without discriminatory restrictions. Open standards enable interoperability between systems from different vendors by providing a shared, stable technical baseline. They are maintained by recognised standards bodies and are typically available royalty-free or under FRAND (Fair, Reasonable And Non-Discriminatory) licensing terms. Open standards contrast with proprietary specifications controlled by a single vendor and are foundational to open, competitive ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:open-standard", "labels": [ "Open Standard", "Open Data Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Technical Specification" ], "wikilinks": [] }, { "id": "open-standards", "title": "Open Standards", "domain": "standards", "domain_name": "Standards", "definition": "Open standards are publicly available, transparently developed technical specifications whose adoption is not contingent on royalties or proprietary licences, enabling any party to implement them independently. They are produced through open, consensus-driven processes\u2014typically stewarded by recognised standards bodies such as W3C, IEEE, ISO, or IETF\u2014and their normative texts are accessible to the public. By decoupling specification from implementation, open standards promote interoperability, vendor diversity, and long-term ecosystem resilience. They are foundational to the internet, the web, spatial computing, AI model exchange, and distributed collaboration infrastructures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:open-standards", "labels": [ "Open Standards", "Open Standards Ecosystem", "Open Standards Process" ], "is_subclass_of": [ "Standards" ], "wikilinks": [ "Standards", "Khronos OpenXR", "glTF", "ONNX Standard" ] }, { "id": "open-webui-and-pipelines", "title": "Open Webui and Pipelines", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Open WebUI (formerly Ollama WebUI, created by Tim Jaeryang Baek, first released October 2023, + GitHub stars by mid-2025) is a self-hosted, ChatGPT-equivalent web interface for interacting with local and remote Large Language Models through a polished conversational UI, supporting Ollama ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:open-webui-and-pipelines", "labels": [ "Open Webui and Pipelines" ], "is_subclass_of": [ "Chatbots", "Retrieval Augmented Generation - RAG", "Large Language Models", "Agent Frameworks", "WebDev and Consumer Tooling" ], "wikilinks": [ "InferenceInfrastructureDomain", "IntegrationLayer", "WebApplicationDomain", "Agent Frameworks", "AI-GroundedDomain", "Anthropic Claude", "ApplicationLayer", "ChatGPT", "Chatbots", "CLI Multi-Agent Systems", "ComfyUI", "Compute Infrastructure", "Docker", "Edge Computing", "Education and AI", "Flux.1", "Foundation Models", "Function Calling", "Hardware and Edge", "Home Assistant" ] }, { "id": "open-world-assumption", "title": "Open World Assumption", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The open world assumption (OWA) is a logical stance in knowledge representation under which the absence of a statement from a knowledge base does not imply that the statement is false, only that its truth value is unknown. It is the foundational semantic principle of description logics and the Web Ontology Language (OWL), reflecting the incomplete and distributed nature of knowledge on the Semantic Web. The OWA contrasts directly with the closed world assumption used in conventional databases and logic programming, where unstated facts are treated as false.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:open-world-assumption", "labels": [ "Open World Assumption", "Open-World Assumption" ], "is_subclass_of": [ "Ontology" ], "wikilinks": [] }, { "id": "open-world", "title": "Open World", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A game design paradigm featuring large-scale virtual environments with non-linear progression, extensive player freedom, emergent gameplay, and exploratory mechanics that allow users to interact with the world in diverse ways beyond predetermined narrative paths.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:open-world", "labels": [ "Open World" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "3D Rendering", "AI System", "Content Creation Tools", "Creative Expression", "Dynamic World", "Emergent Behavior", "Emergent Gameplay", "Exploration", "Exploration System", "Free Roaming", "Game Design Patterns", "GDC", "Level Design", "Multiple Objectives", "Pathfinding", "Player Agency", "Player Choice", "Quest System", "Sandbox Creation", "Sandbox Mechanics" ] }, { "id": "open-x-embodiment", "title": "Open X-Embodiment", "domain": "robotics", "domain_name": "Robotics", "definition": "Open X-Embodiment is a large, collaborative open dataset and associated model effort that aggregates robot manipulation demonstrations across many robot embodiments and labs. By pooling trajectories from diverse hardware into a unified format, it enables training generalist robot policies that transfer across different platforms. The initiative supports research into cross-embodiment learning and the RT-X family of robotic foundation models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:open-x-embodiment", "labels": [ "Open X-Embodiment", "Open-X-Embodiment Dataset" ], "is_subclass_of": [ "Robotics", "Robot Type" ], "wikilinks": [] }, { "id": "open-loop-control", "title": "Open-Loop Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A control system where the control action is independent of the output. The system executes pre-programmed commands without feedback from sensors to verify if the desired state was achieved.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:open-loop-control", "labels": [ "Open-Loop Control", "Open Loop Control" ], "is_subclass_of": [ "Actuation and Control", "Control System" ], "wikilinks": [ "Actuator", "No Feedback", "Predictability", "RB-1002-closed-loop-control", "Stepper Motor Control", "Control System", "Control Theory", "Feedback Mechanism", "Forward Kinematics", "Robotics", "Servo Control" ] }, { "id": "open-source-ai", "title": "open-source ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Open-Source AI denotes artificial intelligence systems\u2014including model weights, training code, datasets, evaluation benchmarks, and inference tooling\u2014released under licences that permit public inspection, reproduction, modification, and redistribution. The degree of openness varies widely: fully open releases expose weights, training data, and procedures; open-weights releases share weights and inference code while withholding training data; and open-API systems expose neither. This transparency gradient determines reproducibility, auditability, and dual-use risk, and is formalised by frameworks such as the OSI Open Source AI Definition. Open-source AI accelerates community safety research, enables fine-tuning on proprietary data, and reduces vendor lock-in, while simultaneously raising governance questions about capability proliferation and misuse.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:open-source-ai", "labels": [ "Open-Source AI", "Open Source AI", "Open Source AI Platforms", "Open-Source AI Models" ], "is_subclass_of": [ "Open Source Software" ], "wikilinks": [] }, { "id": "open-source-intelligence", "title": "Open-Source Intelligence", "domain": "security", "domain_name": "Security", "definition": "Open-source intelligence, commonly abbreviated OSINT, is the practice of collecting and analysing information from publicly available sources, including websites, social media, public records, blockchain explorers and news media, to produce actionable intelligence. In security and blockchain-forensics contexts it is used to attribute wallet addresses, trace transaction flows and support threat-intelligence investigations without requiring privileged access to closed systems. Its reliability depends on corroborating multiple independent open sources rather than any single one.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:open-source-intelligence", "labels": [ "Open-Source Intelligence", "Open Source Intelligence" ], "is_subclass_of": [ "Threat Intelligence" ], "wikilinks": [] }, { "id": "open-source-llms", "title": "Open-Source LLMs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Large language models whose weights and architecture are publicly available, allowing for local deployment, modification, and fine-tuning without reliance on proprietary APIs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:open-source-llms", "labels": [ "Open-Source LLMs" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "open-source-video-diffusion-community-platform", "title": "Open-Source Video Diffusion Community Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Banodoco is a community platform focused on open-source AI video generation, bringing together model architects, fine-tuners, engineers, and artists to advance controllable creative AI. It serves as a coordination space for collaborative development of video diffusion models and related tooling under a shared artistic mission.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:open-source-video-diffusion-community-platform", "labels": [ "Open-Source Video Diffusion Community Platform", "Banodoco" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Digital Twin", "Stable Diffusion" ] }, { "id": "open-space-responsible-ai-gathering", "title": "Open-Space Responsible AI Gathering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Responsible AI Unconference is a participant-driven gathering that applies open-space technology to surface and negotiate ethical, social, and governance concerns around artificial intelligence. Unlike traditional conferences, the agenda is set on the day by attendees, enabling communities\u2014artists, technologists, marginalised groups, and policymakers\u2014to co-author discussions on topics such as bias, child-centric AI, feminist design, and the boundaries of acceptable automation. The format foregrounds perspectives historically excluded from mainstream AI discourse.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:open-space-responsible-ai-gathering", "labels": [ "Open-Space Responsible AI Gathering", "Responsible AI Unconference" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "open-weight-models", "title": "Open-Weight Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI models whose parameter weights are publicly available for download and use, distinguishing them from closed-source proprietary models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:open-weight-models", "labels": [ "Open-Weight Models" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "open-weights-model", "title": "Open-Weights Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An artificial intelligence model whose learned parameters are publicly available for download, allowing users to inspect, modify, and deploy the model locally without relying on a proprietary API.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:open-weights-model", "labels": [ "Open-Weights Model" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "open-ai-api", "title": "OpenAI API", "domain": "ai", "domain_name": "Ai", "definition": "The OpenAI API is a hosted programming interface that gives developers access to OpenAI's models for text generation, reasoning, embeddings, image generation, speech, and tool use over HTTP. Exposing capabilities through endpoints such as chat completions, responses, and embeddings, it abstracts model hosting, scaling, and inference behind a usage-priced REST interface with structured outputs, function calling, and streaming. Its conventions have become a de facto standard widely emulated by other providers and compatible open-source serving stacks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:openai-research-organisation-api", "labels": [ "OpenAI API", "OpenAI Function Calling API" ], "is_subclass_of": [ "API" ], "wikilinks": [] }, { "id": "open-ai-agents-sdk", "title": "OpenAI Agents SDK", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The OpenAI Agents SDK is a software development kit for building agentic applications on OpenAI models, providing primitives for agents, tools and handoffs. It supports orchestrating one or more agents that call functions and pass control between each other.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:openai-research-organisation-agents-sdk", "labels": [ "OpenAI Agents SDK" ], "is_subclass_of": [ "Agentic Workflow" ], "wikilinks": [ "Language Model", "Function Calling", "Multi-Agent Coordination", "Tool Use", "OpenAI", "AI Agent", "Agentic Workflow" ] }, { "id": "openai-research-organisation", "title": "OpenAI Research Organisation", "domain": "ai", "domain_name": "Ai", "definition": "OpenAI is an American artificial intelligence research and deployment organisation founded in December 2015 with the stated mission of ensuring that artificial general intelligence (AGI) benefits all of humanity. Originally incorporated as a non-profit, it restructured into a capped-profit hybrid in 2019 to attract large-scale investment while retaining mission-oriented governance. OpenAI is responsible for the GPT series of large language models, the DALL-E image generation systems, the Codex code-generation model, the Whisper speech-recognition model, the Sora video-generation model, and the ChatGPT conversational interface, as well as foundational research in reinforcement learning from human feedback (RLHF) and AI alignment. Through its API platform and strategic partnership with Microsoft, it has become a central commercial and research force in the global AI industry.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:openai-research-organisation", "labels": [ "OpenAI Research Organisation", "Open AI", "OpenAI", "OpenAI Gym" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "open-api-specification", "title": "OpenAPI Specification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The OpenAPI Specification (OAS) is a language-agnostic, machine-readable standard for describing RESTful HTTP APIs using a structured JSON or YAML document that defines endpoints, request/response schemas, authentication methods, and parameter types. Governed by the OpenAPI Initiative (a Linux Foundation project), it enables automated generation of client SDKs, server stubs, interactive documentation, and contract-based testing from a single source of truth. Originally derived from the Swagger specification, OAS version 3.x is now the dominant industry standard for API description. It promotes interoperability by allowing API consumers to understand a service's capabilities without access to its source code.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:open-api-specification", "labels": [ "OpenAPI Specification", "Open API Specification", "OpenAPI", "OpenAPI 3.0", "OpenAPI 3.1", "OpenAPI Schema" ], "is_subclass_of": [ "API Standard" ], "wikilinks": [] }, { "id": "openapi", "title": "OpenAPI", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OpenAPI is a language-agnostic, machine-readable specification for describing HTTP-based RESTful APIs. An OpenAPI document defines available endpoints, operations, parameters, request and response schemas, and authentication mechanisms in a single structured file written in YAML or JSON. It enables automated generation of documentation, client SDKs, server stubs, and validation logic, decoupling an API's contract from its implementation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:openapi", "labels": [ "OpenAPI" ], "is_subclass_of": [ "API" ], "wikilinks": [] }, { "id": "opencl", "title": "OpenCL", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OpenCL (Open Computing Language) is an open, royalty-free standard for writing programs that execute across heterogeneous platforms including CPUs, GPUs, and other accelerators. It defines a C-based kernel language and host API for expressing data-parallel and task-parallel computation portably across vendors. OpenCL underpins general-purpose GPU computing where cross-vendor portability is valued over the deepest single-vendor optimisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:opencl", "labels": [ "OpenCL" ], "is_subclass_of": [ "GPU Computing" ], "wikilinks": [] }, { "id": "open-cv", "title": "OpenCV", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "OpenCV (Open Source Computer Vision Library) is a BSD-licensed open-source library providing over 2,500 optimised algorithms for real-time computer vision, image processing, and machine learning, originally developed by Intel and now maintained by the OpenCV Foundation. It supports C++, Python, Java, and JavaScript bindings and runs on Linux, Windows, macOS, iOS, and Android. The library encompasses classical algorithms for feature detection, camera calibration, stereo vision, optical flow, and object tracking, as well as deep learning inference through the DNN module, which supports models from TensorFlow, PyTorch, and ONNX. OpenCV is the de facto standard toolkit for robotics perception pipelines, augmented reality applications, and embedded vision systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:open-cv", "labels": [ "OpenCV", "OpenCV Documentation" ], "is_subclass_of": [ "Computer Vision System" ], "wikilinks": [] }, { "id": "open-gl", "title": "OpenGL", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "OpenGL is a cross-platform graphics API for rendering 2D and 3D vector graphics, providing a standardised interface to the rendering capabilities of graphics hardware through a rasterisation pipeline with programmable vertex and fragment shader stages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-gl", "labels": [ "OpenGL", "OpenGL ARB", "OpenGL ES" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [ "GPU", "Shader Language", "3D Rendering", "Rasterization", "Vulkan", "Graphics API" ] }, { "id": "open-id-connect", "title": "openid connect", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OpenID Connect (OIDC) is a federated identity and authentication protocol standardised by the OpenID Foundation that adds an authentication layer on top of the OAuth 2.0 authorisation framework. Upon successful authentication, an OIDC-compliant identity provider issues a signed JSON Web Token (ID Token) containing identity claims \u2014 such as subject identifier, email, and session expiry \u2014 which the relying party verifies cryptographically using the provider's published JSON Web Key Set without querying the provider again. OIDC defines standard flows (authorisation code, implicit, hybrid, and device code), a discovery endpoint, and a UserInfo endpoint, enabling single sign-on across web and mobile applications and serving as the identity spine for enterprise federation, consumer social logins, open banking APIs, and decentralised identity wallets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:open-id-connect", "labels": [ "OpenID Connect", "OpenID Connect 1.0" ], "is_subclass_of": [ "Identity Federation" ], "wikilinks": [] }, { "id": "open-id-foundation", "title": "openid foundation", "domain": "security", "domain_name": "Security", "definition": "The OpenID Foundation (OIDF) is an international, member-driven, non-profit standards organisation that stewards the OpenID family of identity specifications, including OpenID Connect, FAPI (Financial-grade API), MODRNA, and Digital Credentials. It coordinates working groups composed of identity providers, relying parties, and government bodies to develop, test interoperability of, and maintain open specifications for federated authentication and authorisation. The Foundation operates certification programmes to verify implementation conformance to published profiles, reducing interoperability barriers across global digital identity ecosystems. Founded in 2007, the OIDF operates under a royalty-free intellectual property policy, ensuring open access to its specifications without patent encumbrance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-id-foundation", "labels": [ "OpenID Foundation" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "open-id4-vc", "title": "OpenID4VC", "domain": "security", "domain_name": "Security", "definition": "OpenID for Verifiable Credentials (OpenID4VC) is a family of OpenID Foundation specifications that extend OAuth 2.0 and OpenID Connect to issue and present verifiable credentials. It comprises OpenID4VCI for credential issuance and OpenID4VP for presentation, enabling interoperable digital wallets to obtain and selectively disclose cryptographically signed claims. The protocols bridge mainstream identity infrastructure with decentralized identity models such as DIDs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-id4-vc", "labels": [ "OpenID4VC" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "open-id4-vci", "title": "OpenID4VCI", "domain": "security", "domain_name": "Security", "definition": "OpenID for Verifiable Credential Issuance (OpenID4VCI) is an OpenID Foundation protocol specification that defines a standard API by which an Issuer can deliver W3C Verifiable Credentials to a Holder's digital wallet using OAuth 2.0 and OpenID Connect as the underlying authorisation and identity layer. The protocol specifies credential offer flows, authorisation code and pre-authorised code grant types, credential endpoint interactions, and metadata discovery, enabling interoperable credential issuance across identity wallet implementations and issuing authority systems. It is designed to complement OpenID4VP (Verifiable Presentations) to form a complete self-sovereign identity exchange ecosystem.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:open-id4-vci", "labels": [ "OpenID4VCI" ], "is_subclass_of": [ "Credential Issuance" ], "wikilinks": [] }, { "id": "open-id4-vp", "title": "OpenID4VP", "domain": "security", "domain_name": "Security", "definition": "OpenID for Verifiable Presentations (OpenID4VP) is an extension of the OpenID Connect and OAuth 2.0 framework that enables relying parties to request and receive W3C Verifiable Presentations from a holder's digital identity wallet using standard authorisation request-response flows. The specification defines a Presentation Exchange-compatible request syntax, transport bindings for cross-device and same-device wallets, and response encoding options for signed JWT VPs and JSON-LD credential formats. OpenID4VP allows verifiers to specify which credential types and claims are required, enabling selective disclosure and privacy-preserving identity verification without centralised identity providers holding user data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-id4-vp", "labels": [ "OpenID4VP" ], "is_subclass_of": [ "OpenID Connect" ], "wikilinks": [] }, { "id": "open-lineage", "title": "OpenLineage", "domain": "data", "domain_name": "Data", "definition": "OpenLineage is an open standard and specification for collecting data-lineage metadata from data pipelines as they run. It defines a common event model describing jobs, runs, datasets, and their input/output relationships, emitted by instrumented orchestration and processing tools. This enables consistent tracking of how data is produced and transformed across heterogeneous platforms for governance, debugging, and impact analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-lineage", "labels": [ "OpenLineage" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "openmp", "title": "OpenMP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An open standard API for shared-memory parallel programming in C, C++, and Fortran, in which developers annotate sequential code with compiler directives (pragmas) such as parallel regions, work-sharing loops, and tasks, and the compiler and runtime distribute the work across threads. Governed by the OpenMP Architecture Review Board since 1997, the specification has grown from simple loop-level parallelism to encompass explicit tasking, SIMD vectorisation, and offloading to GPUs and other accelerators via target directives. Its incremental, directive-based model makes it the dominant intra-node parallelisation approach in scientific and high-performance computing, commonly paired with MPI for communication between nodes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:openmp", "labels": [ "OpenMP" ], "is_subclass_of": [ "API" ], "wikilinks": [ "API", "Parallel Processing", "Message Passing Interface", "Shared Memory" ] }, { "id": "open-pose", "title": "OpenPose", "domain": "ai", "domain_name": "Ai", "definition": "OpenPose is an open-source, real-time multi-person pose estimation library developed at Carnegie Mellon University that simultaneously detects body, hand, face, and foot keypoints from RGB images and video using convolutional neural networks. It employs Part Affinity Fields (PAFs) \u2014 a set of 2D vector fields encoding the location and orientation of limb connections \u2014 enabling association of detected keypoints into individual skeletons without prior person detection. OpenPose established the part-affinity-field paradigm that underpins many subsequent human pose estimation systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:open-pose", "labels": [ "OpenPose" ], "is_subclass_of": [ "Pose Estimation" ], "wikilinks": [] }, { "id": "open-rail", "title": "OpenRAIL", "domain": "ai", "domain_name": "Ai", "definition": "OpenRAIL (Responsible AI License) is a family of model licenses that grant broad permission to use and distribute AI models while imposing use-based behavioral restrictions to prevent harmful applications. Unlike traditional open-source licenses, it couples openness with enforceable conditions prohibiting specified misuse such as discrimination or disinformation. OpenRAIL has become a common licensing framework for openly released foundation models and their training artifacts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:open-rail", "labels": [ "OpenRAIL", "CreativeML Open RAIL-M Licence" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "open-sea", "title": "OpenSea", "domain": "blockchain", "domain_name": "Blockchain", "definition": "OpenSea is a peer-to-peer decentralised marketplace for non-fungible tokens (NFTs), enabling users to mint, list, auction, and trade digital assets on multiple blockchain networks including Ethereum, Polygon, Solana, and others. Founded in 2017, it operates through integration with self-custodial crypto wallets and on-chain smart contracts that escrow assets and execute transfers atomically upon payment settlement. OpenSea pioneered the concept of a permissionless NFT trading platform, supporting diverse token standards such as ERC-721 and ERC-1155, and providing creator royalty enforcement mechanisms at the contract level.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-sea", "labels": [ "OpenSea" ], "is_subclass_of": [ "NFT Marketplace" ], "wikilinks": [ "NFT", "Digital Asset", "Ethereum", "NFT Marketplace" ] }, { "id": "open-telemetry", "title": "OpenTelemetry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "OpenTelemetry (OTel) is a vendor-neutral open-source observability framework and CNCF project that provides unified APIs, SDKs, agents, and wire protocols for collecting distributed traces, metrics, and logs from software systems. It merges the OpenTracing and OpenCensus projects into a single standardised instrumentation layer, enabling consistent telemetry data collection regardless of the backend analysis platform. OpenTelemetry's OpenTelemetry Protocol (OTLP) has become the de-facto standard for telemetry data transport in cloud-native environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:open-telemetry", "labels": [ "OpenTelemetry" ], "is_subclass_of": [ "Monitoring System" ], "wikilinks": [] }, { "id": "open-usd", "title": "openusd", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "OpenUSD (Universal Scene Description) is an open-source framework, scene-graph data model, and file format developed by Pixar Animation Studios for composing, simulating, collaborating on, and exchanging richly structured 3D scenes across digital content creation tools and real-time rendering pipelines. Its layered composition engine allows multiple teams or tools to contribute non-destructive overrides to a shared hierarchical scene graph with full time-sampled animation support, making it the de facto interchange format for large-scale visual effects, game engine, and industrial simulation pipelines. Governed since 2023 by the Alliance for OpenUSD (AOUSD) and progressing toward ISO/IEC 22886 standardisation, OpenUSD underpins NVIDIA Omniverse, Apple visionOS/RealityKit, and major digital twin platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:open-usd", "labels": [ "OpenUSD", "OpenUSD Alliance", "Pixar USD", "USD AOUSD" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "open-vr", "title": "OpenVR", "domain": "metaverse", "domain_name": "Metaverse", "definition": "OpenVR is a software development kit and API, developed by Valve, that lets applications interface with virtual-reality hardware without targeting a specific vendor's runtime. It abstracts headset tracking, controller input, and rendering submission so a single application can run across compatible VR devices via the SteamVR runtime. OpenVR was an early de facto cross-vendor VR interface, later complemented by the open OpenXR standard.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-vr", "labels": [ "OpenVR" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "openxr-standard", "title": "OpenXR Standard", "domain": "standards", "domain_name": "Standards", "definition": "A royalty-free open standard from the Khronos Group defining a common application programming interface between XR applications and virtual, augmented, and mixed reality hardware, covering device tracking, input actions, frame timing, and composition, so that an application written once against the OpenXR runtime interface runs across headsets from different vendors without engine-specific porting.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:openxr-standard", "labels": [ "OpenXR Standard" ], "is_subclass_of": [ "Open Standard" ], "wikilinks": [ "Open Standard", "Khronos Group", "Extended Reality" ] }, { "id": "open-xr", "title": "OpenXR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "OpenXR is the Khronos Group's open, royalty-free API standard for cross-platform access to extended reality (XR) hardware \u2014 encompassing virtual reality (VR) headsets, augmented reality (AR) glasses, and mixed reality (MR) devices \u2014 that defines a unified application-to-runtime interface, elimina...", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:open-xr", "labels": [ "OpenXR", "OpenXR Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Extended Reality API", "Khronos Group Standard" ], "wikilinks": [ "AI Domain", "Android Platform", "Extended Reality API", "HardwareAbstractionLayer", "Hardware-Agnostic XR Development", "Khronos Group Standard", "Khronos OpenXR 1.0 Specification", "Khronos OpenXR 1.1 Specification", "Monado Open Source Runtime", "OpenGL ES", "OpenXR Action System", "OpenXR Extension", "OpenXR Extension Registry", "OpenXR Loader", "OpenXR Session", "OpenXR Space", "OpenXR Swapchain", "Passthrough AR", "Unity OpenXR Plugin", "Vulkan API" ] }, { "id": "open-zeppelin-contracts", "title": "OpenZeppelin Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "OpenZeppelin Contracts is a widely used open-source library of secure, audited, and reusable smart-contract components for Ethereum and EVM-compatible chains. It provides standard implementations of token interfaces such as ERC-20, ERC-721, and ERC-1155, along with access control, upgradeability, and security utilities. The library is a de facto baseline for safe smart-contract development, reducing the risk of common vulnerabilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-zeppelin-contracts", "labels": [ "OpenZeppelin Contracts", "OpenZeppelin Proxy Pattern" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "open-zeppelin-governor-contracts", "title": "OpenZeppelin Governor Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "OpenZeppelin Governor Contracts are a modular on-chain governance framework that lets DAOs run proposal, voting, and execution workflows tied to governance tokens. The Governor base contract integrates with token-based voting power (including delegation and timelock execution) and is configurable for quorum, voting periods, and vote-counting strategies. It is a widely adopted standard for implementing token-weighted decentralized governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:open-zeppelin-governor-contracts", "labels": [ "OpenZeppelin Governor Contracts", "Governor Contracts", "OpenZeppelin Governance Contracts" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "open-zeppelin-governor", "title": "openzeppelin governor", "domain": "blockchain", "domain_name": "Blockchain", "definition": "OpenZeppelin Governor is a modular, security-audited Solidity smart-contract framework that provides the core infrastructure for deploying on-chain governance systems on EVM-compatible blockchains, succeeding the Compound Governor Bravo pattern. It implements a canonical proposal lifecycle \u2014 proposal creation, configurable voting delay, voting period, quorum validation, and timelock-guarded execution \u2014 with pluggable extension modules for vote-counting strategies (simple majority, Bravo-style, fractional), token-based voting-power sources (ERC-20 with EIP-5805 checkpointing or ERC-721), and TimelockController integration. The framework is maintained by the OpenZeppelin security team as part of OpenZeppelin Contracts, has undergone multiple third-party audits, and is adopted as the de facto governance standard by major decentralised autonomous organisations and DeFi protocols including Uniswap, ENS, and Compound.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:open-zeppelin-governor", "labels": [ "OpenZeppelin Governor", "Governor Contract", "OpenZeppelin Governor Standard" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "open-zeppelin", "title": "OpenZeppelin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "OpenZeppelin is an open-source framework and security company that provides battle-tested, audited smart contract libraries for Ethereum and EVM-compatible blockchains, most notably the OpenZeppelin Contracts library. It implements widely adopted token standards (ERC-20, ERC-721, ERC-1155), access control patterns, proxy upgrade mechanisms, and governance primitives used as foundational building blocks across decentralised finance, NFT platforms, and DAO infrastructure. Beyond libraries, OpenZeppelin offers professional security auditing services, the Defender operations platform for smart contract monitoring and automation, and open tooling that lowers the barrier to secure on-chain development.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:open-zeppelin", "labels": [ "OpenZeppelin", "OpenZeppelin Libraries", "OpenZeppelin Standards" ], "is_subclass_of": [ "Protocol and Consensus", "Smart Contract Framework" ], "wikilinks": [ "Solidity", "Smart Contract", "Token Standard" ] }, { "id": "openapi-initiative", "title": "Openapi Initiative", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The OpenAPI Initiative (OAI) is an open-governance consortium under the Linux Foundation that maintains the OpenAPI Specification, a vendor-neutral, machine-readable format for describing HTTP APIs. It evolved from the donated Swagger specification and provides a standard contract that both humans and tools can use to understand, document, generate and test RESTful interfaces. By standardising API descriptions it promotes interoperability across the API tooling ecosystem, including documentation generators, client SDK generators and gateways. The initiative governs the specification's stewardship and version evolution.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:openapi-initiative", "labels": [ "Openapi Initiative", "OpenAPI Initiative" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "operating-system", "title": "Operating System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Operating System (OS) is system software that manages computer hardware, software resources, and provides common services for application programs \u2014 encompassing process scheduling, memory management, file-system abstraction, device-driver interfaces, networking stacks, and security enforcement. Acting as the intermediary between hardware and user-space applications, the OS exposes stable APIs that decouple software from underlying physical or virtualised hardware. Specialised variants range from real-time operating systems (RTOS) for deterministic control loops in embedded and robotic systems, to hypervisor-based OSes supporting full hardware virtualisation, to spatial-computing runtime stacks that coordinate sensor fusion, GPU pipelines, and display refresh at sub-millisecond latencies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:operating-system", "labels": [ "Operating System", "Linux Operating System", "Operating System Kernel", "Operating Systems", "Windows Operating System" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "operational-efficiency", "title": "Operational Efficiency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Operational Efficiency is the capability of an organization or system to deliver outputs while minimizing wasted resources such as time, cost, energy, and labor. It is achieved through process optimization, automation, and measurement of input-to-output ratios across workflows. In industrial and digital contexts it is a primary target of automation and transformation initiatives that seek higher throughput and lower unit cost.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:operational-efficiency", "labels": [ "Operational Efficiency" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "operational-layer", "title": "Operational Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Operational Layer is the cross-cutting stratum concerned with running, maintaining, and recovering a system in production. It sits above the runtime and tooling strata it relies on and supports the institutional commitments made above. It contains deployment pipelines, monitoring, incident response, and capacity management.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:operational-layer", "labels": [ "Operational Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Runtime Layer", "Tooling Layer", "Institutional Layer", "Application Layer", "Site Reliability Engineering", "Incident Management", "owl:Thing" ] }, { "id": "operational-resilience", "title": "operational resilience", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Operational Resilience is the organisational and technical capacity to anticipate, prepare for, withstand, recover from, and adapt to disruptions\u2014whether caused by cyber-attacks, infrastructure failures, natural disasters, or human error\u2014while maintaining continuity of critical business services within pre-defined impact tolerances. It extends traditional Business Continuity Management and Disaster Recovery by demanding that organisations identify their most important business services, quantify the maximum tolerable disruption for each, and validate end-to-end resilience through realistic scenario testing. Regulatory frameworks such as the Bank of England's Supervisory Statement SS1/21, the EU Digital Operational Resilience Act (DORA), and NIST SP 800-160 Vol. 2 have transformed operational resilience from a best-practice aspiration into a legal compliance obligation. Technically it is realised through layered redundancy, fault-tolerant architectures, chaos engineering disciplines, and automated recovery orchestration.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:operational-resilience", "labels": [ "Operational Resilience", "Digital Operational Resilience Act", "Operational Resilience Standards" ], "is_subclass_of": [ "Resilience" ], "wikilinks": [] }, { "id": "operational-risk", "title": "Operational Risk", "domain": "finance", "domain_name": "Finance", "definition": "Operational risk is the risk of loss resulting from inadequate or failed internal processes, people and systems, or from external events. It encompasses fraud, human error, system failures, legal and compliance breaches, and disruptions such as cyber-attacks or natural disasters, but excludes strategic and reputational risk in its narrow Basel definition. Managing it relies on controls, monitoring, capital allocation and resilience planning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:operational-risk", "labels": [ "Operational Risk" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "operational-technology", "title": "Operational Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Operational technology (OT) is the hardware and software that directly monitors and controls physical processes, devices, and infrastructure in industrial environments. It encompasses industrial control systems such as SCADA and distributed control systems, programmable logic controllers, sensors, and actuators that manage manufacturing, energy, and utilities. OT prioritises availability, safety, and real-time determinism, distinguishing it from information technology, with which it increasingly converges under Industry 4.0.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:operational-technology", "labels": [ "Operational Technology" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "operational-transformation", "title": "Operational Transformation", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Operational Transformation (OT) is a concurrency control technique for real-time collaborative editing systems that enables multiple users to concurrently modify a shared document by representing edits as discrete operations and automatically transforming each incoming operation against previously applied operations to preserve user intent. The core insight is that when two concurrent operations are applied in different orders across distributed replicas, at least one must be adjusted \u2014 transformed \u2014 so that the final document state converges to a consistent result. OT requires a pair of transformation functions (IT and ET for inclusion and exclusion transformation) and must satisfy formal consistency properties (TP1 and TP2) to guarantee convergence under all interleavings. It underpins real-time collaborative editors including Google Docs, Apache Wave, and Etherpad, and remains a foundational technique in distributed collaboration research despite the later emergence of CRDTs.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:operational-transformation", "labels": [ "Operational Transformation" ], "is_subclass_of": [ "Protocol and Infrastructure", "Concurrency Control" ], "wikilinks": [ "Distributed Systems", "Collaboration Tools", "CRDT", "Conflict Resolution" ] }, { "id": "operations-research", "title": "Operations Research", "domain": "ai", "domain_name": "Ai", "definition": "Operations research (OR) is an analytical discipline that applies advanced mathematical and computational methods \u2014 including linear programming, integer optimisation, stochastic modelling, and simulation \u2014 to support decision-making in complex systems. It originated during World War II to optimise military logistics and has since expanded to industrial, financial, and humanitarian domains. OR provides rigorous frameworks for formulating trade-off problems as constrained optimisation models, then solving them to global or near-global optimality using exact solvers or heuristic methods. Modern OR integrates machine learning for parameter estimation and reinforcement learning for sequential decision problems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:operations-research", "labels": [ "Operations Research" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "operator-fusion", "title": "Operator Fusion", "domain": "ai", "domain_name": "Ai", "definition": "Operator fusion is a compiler optimisation that combines several consecutive operations in a neural network computation graph into a single fused kernel. By merging operations such as a matrix multiply with its bias addition and activation function, fusion avoids writing intermediate tensors back to memory, reducing memory bandwidth pressure and kernel launch overhead. It is a core technique in machine learning compilers and inference runtimes for improving throughput and latency on accelerators. Fusion trades increased kernel complexity for fewer round trips to global memory.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:operator-fusion", "labels": [ "Operator Fusion" ], "is_subclass_of": [ "Model Optimization" ], "wikilinks": [] }, { "id": "opportunity-ai", "title": "Opportunity AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence to expand market boundaries, create new products or services, and enable capabilities that were previously impossible, rather than just optimizing existing processes.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:opportunity-ai", "labels": [ "Opportunity AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "optical-astronomy", "title": "Optical Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:optical-astronomy", "labels": [ "Optical Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "optical-calibration-target", "title": "Optical Calibration Target", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An optical calibration target is a precision artefact bearing known geometric, radiometric, or colourimetric reference patterns used to characterise, correct, and validate the response of imaging and optical systems. Such targets enable the measurement of lens distortion, modulation transfer function (MTF), chromatic aberration, colour reproduction accuracy, and spatial linearity across a wide range of imaging modalities from visible-light cameras to multispectral and depth sensors. They are fundamental to the metrological traceability chain that links deployed camera systems to national and international measurement standards. In spatial computing and extended-reality applications, calibration targets additionally serve as fiducial references for camera-to-camera alignment, display-sensor registration, and world-space coordinate grounding.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:optical-calibration-target", "labels": [ "Optical Calibration Target" ], "is_subclass_of": [ "Calibration Equipment" ], "wikilinks": [ "Optical System Accuracy", "Calibration Equipment", "metaverse" ] }, { "id": "optical-character-recognition", "title": "Optical Character Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Optical Character Recognition (OCR) is a technology that converts images of typed, printed, or handwritten text into machine-encoded character sequences, enabling downstream search, editing, and automated processing of scanned documents and photographs. Classical approaches segment character glyphs and classify them against trained feature descriptors; contemporary deep-learning pipelines \u2014 typically convolutional neural networks paired with sequence models such as CTC or Transformer decoders \u2014 recognise whole text lines end-to-end without explicit segmentation. OCR is a foundational building block of document intelligence, information extraction, and accessibility tooling, and underlies virtually every large-scale digitisation effort from cultural heritage archives to enterprise content management.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:optical-character-recognition", "labels": [ "Optical Character Recognition" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "Computer Vision", "Legal Research", "Deep Learning", "Machine Learning" ] }, { "id": "optical-flow", "title": "Optical Flow", "domain": "ai", "domain_name": "Ai", "definition": "Optical flow is the pattern of apparent motion of objects, surfaces, and edges in a visual scene between consecutive frames of video, caused by relative movement between the observer and the scene. It is computed as a dense or sparse 2D velocity field over the image plane and is used in computer vision for motion estimation, video interpolation, action recognition, and autonomous navigation. Classical algorithms (Horn-Schunck, Lucas-Kanade) and deep learning approaches (RAFT, FlowNet) constitute the main methodological lineages.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:optical-flow", "labels": [ "Optical Flow" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "optical-instrument", "title": "Optical Instrument", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:optical-instrument", "labels": [ "Optical Instrument" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "optical-remote-sensing", "title": "Optical Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:optical-remote-sensing", "labels": [ "Optical Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "optical-sensor-array", "title": "Optical Sensor Array", "domain": "robotics", "domain_name": "Robotics", "definition": "A multi-element optical sensing arrangement comprising spatially distributed photodetectors\u2014including CCD (Charge-Coupled Device) arrays, CMOS (Complementary Metal-Oxide-Semiconductor) image sensors, SPAD (Single-Photon Avalanche Diode) arrays, InGaAs infrared arrays, avalanche photodiode (APD) a...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:optical-sensor-array", "labels": [ "Optical Sensor Array" ], "is_subclass_of": [ "Perception and Sensing", "Sensor", "Exteroceptive Sensor", "Computer Vision System", "Hardware", "Robotics", "Imaging Parameters" ], "wikilinks": [ "2D LiDAR", "3D LiDAR", "AI Hardware", "AR Frame", "AR Technology", "Augmented Reality (AR)", "Autonomous Navigation", "Autonomous Robot", "Autonomous Vehicle", "Camera", "Camera Parameters", "Computer Vision", "Computer Vision System", "Computer Vision Task", "Convolutional Neural Network", "Current Sensor", "Deep Learning", "Depth Estimation", "Depth Sensing", "Digital Signal Processing" ] }, { "id": "optical-sensors", "title": "Optical Sensors", "domain": "robotics", "domain_name": "Robotics", "definition": "Optical Sensors are devices that detect and convert light, across visible, infrared, or other wavelengths, into electrical signals for measurement or imaging. They include cameras, photodiodes, depth sensors, and structured-light or time-of-flight units used to capture scene geometry and motion. In capture and recognition systems they provide the primary visual input from which 3D structure, pose, and identity are inferred.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optical-sensors", "labels": [ "Optical Sensors" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "optical-space-communication", "title": "Optical Space Communication", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:optical-space-communication", "labels": [ "Optical Space Communication" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "optical-spacecraft-navigation", "title": "Optical Spacecraft Navigation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:optical-spacecraft-navigation", "labels": [ "Optical Spacecraft Navigation" ], "is_subclass_of": [ "Spacecraft Navigation" ], "wikilinks": [] }, { "id": "optical-speech-recognition", "title": "Optical Speech Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A biometric and signal processing technique that detects and transcribes human speech by analyzing visual cues from lip movements and facial muscle activity using optical sensors, rather than processing audio waveforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:optical-speech-recognition", "labels": [ "Optical Speech Recognition" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "optical-systems", "title": "Optical Systems", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Optical Systems are arrangements of lenses, mirrors, waveguides, and combiners that direct and form light to project or relay images. In display hardware they govern how rendered content is delivered to the eye, including focus, field of view, and image clarity. Augmented- and virtual-reality headsets depend on compact optical systems such as pancake lenses and waveguide combiners to merge virtual imagery with the real world.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optical-systems", "labels": [ "Optical Systems", "Optical System" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "optical-tracking", "title": "Optical Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A sensing technology that uses camera-based systems and computer vision algorithms to determine the position and orientation of objects or users within physical space. Implementations range from marker-based infrared systems achieving sub-millimetre accuracy to markerless inside-out tracking using SLAM, enabling motion capture, hand tracking, and environmental mapping for XR applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:optical-tracking", "labels": [ "Optical Tracking", "Head Tracking", "Optical Tracking System" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "optical-transport-network", "title": "Optical Transport Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Optical Transport Network (OTN) is a standardized framework, defined by ITU-T G.709, for carrying client signals over wavelength-division-multiplexed optical fiber with framing, error correction, and management overhead. It provides high-capacity, long-haul transport with operations and maintenance features that legacy SONET/SDH lacked. OTN forms a core layer of carrier and backbone networks beneath higher-layer packet and service traffic.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optical-transport-network", "labels": [ "Optical Transport Network" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "optimal-control", "title": "Optimal Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A control strategy that determines control inputs to minimize or maximize a performance criterion (cost function) while satisfying system dynamics and constraints. It seeks the best possible control policy according to specified objectives, using mathematical frameworks such as dynamic programming, the Hamilton-Jacobi-Bellman equation, and Pontryagin's Maximum Principle.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:optimal-control", "labels": [ "Optimal Control", "RB-1003-optimal-control" ], "is_subclass_of": [ "Closed-Loop Control", "RB-1002-closed-loop-control" ], "wikilinks": [ "Autonomous Vehicles", "Cost Function", "Efficiency", "Linear Quadratic Regulator", "Model Predictive Control", "Optimality", "Optimization", "Optimization Theory", "RB-1002-closed-loop-control", "RB-1004-adaptive-control", "RB-1007-trajectory-generation", "Spacecraft Control", "System Dynamics Model", "Control Theory", "Machine Learning", "Optimization Algorithm", "Robotics" ] }, { "id": "optimal-transport", "title": "Optimal Transport", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Optimal Transport is the mathematical theory of moving one probability distribution to another while minimising a total cost defined by a ground metric. Its solution induces the Wasserstein distance, a geometrically meaningful divergence between distributions that, unlike many alternatives, behaves well even when supports do not overlap. In machine learning it grounds generative model training, domain adaptation, and distribution matching, including flow-based formulations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:optimal-transport", "labels": [ "Optimal Transport" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [] }, { "id": "optimality", "title": "Optimality", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The property of being the best achievable solution with respect to a defined objective and set of constraints, central to optimisation and decision problems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:optimality", "labels": [ "Optimality", "Pareto Optimality" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [ "Optimisation", "Game Theory", "Robustness" ] }, { "id": "optimisation-algorithm", "title": "Optimisation Algorithm", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An optimisation algorithm is a systematic procedure for finding the values of decision variables that minimise or maximise an objective function, optionally subject to constraints. It spans first-order gradient methods, second-order Newton-type methods, derivative-free and metaheuristic search, and convex programming solvers, each trading convergence speed, robustness, and assumptions about the objective. In machine learning, optimisation algorithms drive model training by iteratively reducing a loss function, making them the engine that turns data and architecture into fitted parameters.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:optimisation-algorithm", "labels": [ "Optimisation Algorithm", "Optimization Algorithm" ], "is_subclass_of": [ "Mathematical Optimisation" ], "wikilinks": [] }, { "id": "optimisation", "title": "Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Optimisation is the mathematical and computational discipline concerned with finding the best solution \u2014 maximum or minimum \u2014 of an objective function subject to given constraints, across a search space of possible decisions or parameter configurations. It encompasses deterministic methods (linear programming, convex optimisation, gradient-based search), stochastic methods (simulated annealing, evolutionary algorithms, Monte Carlo sampling), and learned methods (differentiable optimisation, meta-learning, neural combinatorial solvers). Optimisation is the theoretical core of machine learning training, operations research, control engineering, signal processing, and resource scheduling. The choice of optimisation algorithm fundamentally determines convergence speed, solution quality, and computational cost across all application domains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optimisation", "labels": [ "Optimisation", "Least-Squares Optimisation", "Levenberg-Marquardt Optimisation", "Mathematical Optimisation", "Nonlinear Optimisation", "Optimisation Landscape Smoothing", "Optimization", "Performance Optimisation", "Pipeline Optimisation", "Query Optimisation" ], "is_subclass_of": [ "Mathematical Foundations" ], "wikilinks": [] }, { "id": "optimiser", "title": "Optimiser", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An Optimiser is an algorithm that adjusts model parameters during training to minimise a loss function, guiding convergence towards an optimal solution. Modern optimisers such as Adam, RMSProp, and AdaGrad extend stochastic gradient descent with adaptive learning rates, momentum accumulation, and second-moment estimates, enabling faster and more stable training of deep neural networks across diverse tasks.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:optimiser", "labels": [ "Optimiser", "Optimizer" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "optimism-collective", "title": "Optimism Collective", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Optimism Collective is a two-house governance body overseeing the Optimism ecosystem \u2014 a suite of Ethereum Layer 2 networks built on the OP Stack rollup technology \u2014 consisting of the Token House (OP token holders who vote on protocol upgrades and treasury allocations) and the Citizens' House (non-transferable badge holders who allocate Retroactive Public Goods Funding). It represents one of the most sophisticated real-world experiments in bicameral on-chain governance, designed to balance token-weighted economic interests with broader societal value creation and public goods provision.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:optimism-collective", "labels": [ "Optimism Collective", "Optimism Citizens House" ], "is_subclass_of": [ "DAO Governance" ], "wikilinks": [] }, { "id": "optimism-rpgf", "title": "Optimism RPGF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Optimism Retroactive Public Goods Funding (RPGF) is a recurring funding mechanism run by the Optimism Collective that rewards projects after they have demonstrably contributed value to the ecosystem. Rather than funding speculative proposals upfront, badgeholders or token holders allocate retroactive grants based on observed impact. The model is a flagship experiment in decentralized, impact-based public-goods funding for blockchain ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:optimism-rpgf", "labels": [ "Optimism RPGF" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "optimism", "title": "Optimism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Optimism is an Ethereum layer-2 scaling network that uses optimistic rollup technology to process transactions off the main chain while inheriting Ethereum security. Launched on mainnet in late 2021 by OP Labs, it batches transactions and posts compressed data and state commitments to Ethereum, assuming validity unless challenged within a dispute window. It is closely associated with the OP Stack, a modular open-source framework for building rollup chains, and the broader Superchain concept that links such chains. The OP token governs the Optimism Collective and funds public-goods initiatives through retroactive funding rounds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optimism", "labels": [ "Optimism", "Optimism Superchain" ], "is_subclass_of": [ "Blockchain", "Blockchain Network", "Blockchain Domain" ], "wikilinks": [ "Ethereum", "Rollup", "Fraud Proof", "Decentralised Finance Domain", "Arbitrum", "zkSync", "Polygon", "Blockchain Domain" ] }, { "id": "optimistic-governance", "title": "Optimistic Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Optimistic Governance is a DAO decision-making pattern in which proposals are assumed approved and can execute unless challenged within a defined dispute window. It reduces voting fatigue and gas costs by requiring active participation only to veto or contest actions rather than to ratify every routine decision. Disputes are resolved by escalation to a vote or an arbitration mechanism, trading some upfront scrutiny for operational efficiency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:optimistic-governance", "labels": [ "Optimistic Governance" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "optimistic-oracle", "title": "Optimistic Oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An oracle design in which proposed off-chain data is accepted by default and only verified on-chain if a participant disputes it within a challenge window. It reduces routine reporting costs by reserving full verification for contested values.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:optimistic-oracle", "labels": [ "Optimistic Oracle" ], "is_subclass_of": [ "Price Oracle" ], "wikilinks": [ "Dispute Resolution", "Smart Contract", "DeFi", "Price Oracle", "Chainlink", "https://docs.uma.xyz/" ] }, { "id": "optimistic-rollup", "title": "Optimistic Rollup", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A layer-two scaling design that posts transaction data to a base chain and assumes results are valid unless challenged within a dispute window through a fraud proof.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:optimistic-rollup", "labels": [ "Optimistic Rollup", "Optimistic Rollups" ], "is_subclass_of": [ "Rollup" ], "wikilinks": [ "Smart Contract", "Layer 2 Scaling", "EVM", "Optimism", "Rollup" ] }, { "id": "optimistic-verification", "title": "Optimistic Verification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Optimistic Verification is a security model in which state transitions or cross-chain messages are presumed valid and accepted after a challenge period unless a fraud proof demonstrates otherwise. It avoids the cost of validating every claim upfront, relying instead on economically incentivized watchers to detect and dispute invalid assertions. The approach underpins optimistic rollups and many cross-chain bridge designs that prioritize throughput over instant finality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:optimistic-verification", "labels": [ "Optimistic Verification" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "optimization-algorithms", "title": "Optimization Algorithms", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Mathematical procedures for minimising or maximising objective functions, central to training machine learning models. Gradient-based methods (SGD, Adam, RMSprop, AdaGrad) iteratively update model parameters to reduce loss; advanced techniques encompass momentum-based optimisation, adaptive learning rates, second-order methods (L-BFGS, natural gradient), and gradient-free approaches (evolutionary strategies, Bayesian optimisation), addressing non-convexity, saddle points, and high-dimensional parameter-space challenges.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:optimization-algorithms", "labels": [ "Optimization Algorithms", "Optimisation Algorithms" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Convex Optimization", "Hyperparameter Tuning", "Backpropagation", "Gradient Descent" ] }, { "id": "optimization-technique", "title": "Optimization Technique", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A systematic method applied to improve computational performance, memory efficiency, or rendering throughput in metaverse and spatial computing infrastructure. Optimisation techniques span algorithmic improvements (level-of-detail, culling, compression), hardware utilisation strategies (GPU parallelism, cache coherence), and software-level approaches (batching, shader optimisation) that collectively reduce latency and resource consumption.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:optimization-technique", "labels": [ "Optimization Technique" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "oracle-network", "title": "oracle network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An oracle network is a decentralised infrastructure layer composed of independent node operators that collectively fetch, validate, aggregate, and deliver off-chain data \u2014 such as asset prices, weather readings, sports outcomes, or IoT sensor readings \u2014 to smart contracts executing on a blockchain. Because deterministic blockchain ledgers cannot natively make external HTTP requests or access off-chain databases, oracle networks serve as the cryptoeconomically-secured bridge between on-chain logic and real-world state. Node operators are aligned to honest reporting through staking, slashing, and reputation mechanisms, while aggregation techniques such as median computation and time-weighted average pricing reduce susceptibility to individual node manipulation. Oracle networks underpin critical decentralised finance primitives including lending protocols, synthetic assets, prediction markets, insurance products, cross-chain bridges, and real-world asset settlement.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:oracle-network", "labels": [ "Oracle Network", "Decentralised Oracle Network", "Decentralized Oracle Network" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "oracle-service", "title": "Oracle Service", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A service that supplies external data to smart contracts, bridging the gap between on-chain logic and off-chain information such as prices, events or sensor readings. It provides the inputs that contracts cannot read directly from the blockchain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:oracle-service", "labels": [ "Oracle Service" ], "is_subclass_of": [ "Price Oracle" ], "wikilinks": [ "Smart Contract", "DeFi", "Chainlink", "Optimistic Oracle", "Price Oracle", "https://ethereum.org/en/developers/docs/oracles/" ] }, { "id": "oracle", "title": "Oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "In blockchain systems, a mechanism that brings external off-chain information onto the chain so that smart contracts can act on data they cannot observe directly. Oracles form the trust boundary between deterministic on-chain logic and the variable outside world, and may be centralised reporters or decentralised aggregation networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:oracle", "labels": [ "Oracle", "Centralized Oracle", "Decentralized Oracle", "Execution Oracle", "Human Oracle" ], "is_subclass_of": [ "Blockchain Infrastructure", "Network Component (Blockchain)", "Blockchain Domain" ], "wikilinks": [ "Smart Contract", "DeFi", "Price Oracle", "Chainlink", "Blockchain Domain", "https://ethereum.org/en/developers/docs/oracles/" ] }, { "id": "orbit-determination", "title": "Orbit Determination", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbit-determination", "labels": [ "Orbit Determination" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbit-propagation", "title": "Orbit Propagation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbit-propagation", "labels": [ "Orbit Propagation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-conjunction", "title": "Orbital Conjunction", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-conjunction", "labels": [ "Orbital Conjunction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-debris", "title": "Orbital Debris", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "Orbital debris comprises artificial objects in Earth orbit or re-entry that no longer serve a useful function, including defunct spacecraft, abandoned rocket stages, mission-related objects and fragments.[^1] Natural meteoroids are a separate population. Debris ranges from intact tonnes-scale hardware to particles too small to track individually, yet capable of damaging a spacecraft at orbital relative velocity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-debris", "labels": [ "Orbital Debris" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-dynamics", "title": "Orbital Dynamics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-dynamics", "labels": [ "Orbital Dynamics" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-eccentricity", "title": "Orbital Eccentricity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-eccentricity", "labels": [ "Orbital Eccentricity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-element", "title": "Orbital Element", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-element", "labels": [ "Orbital Element" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-inclination", "title": "Orbital Inclination", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-inclination", "labels": [ "Orbital Inclination" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-insertion", "title": "Orbital Insertion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-insertion", "labels": [ "Orbital Insertion" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-manoeuvre", "title": "Orbital Manoeuvre", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-manoeuvre", "labels": [ "Orbital Manoeuvre" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-period", "title": "Orbital Period", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-period", "labels": [ "Orbital Period" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-perturbation", "title": "Orbital Perturbation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-perturbation", "labels": [ "Orbital Perturbation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbital-velocity", "title": "Orbital Velocity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbital-velocity", "labels": [ "Orbital Velocity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orbiter", "title": "Orbiter", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orbiter", "labels": [ "Orbiter" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "orchestration-layer", "title": "Orchestration Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An Orchestration Layer is a software architectural component that coordinates the execution of heterogeneous services, agents, or microservices \u2014 managing task routing, dependency resolution, resource allocation, and fault recovery \u2014 to produce coherent outputs from distributed system components. It acts as the control plane above individual execution units, abstracting their composition into unified workflows.", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:orchestration-layer", "labels": [ "Orchestration Layer", "OrchestrationLayer" ], "is_subclass_of": [ "Orchestration" ], "wikilinks": [] }, { "id": "orchestration", "title": "Orchestration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The automated coordination, scheduling, and lifecycle management of containerised workloads, microservices, and distributed agents across heterogeneous infrastructure. In the context of metaverse and telecollaboration systems, orchestration\u2014exemplified by Kubernetes\u2014ensures that compute-intensive rendering, AI inference, and real-time communication services scale elastically and recover automatically from failures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:orchestration", "labels": [ "Orchestration", "Communication Orchestration", "Data Orchestration" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "order-book-exchange", "title": "Order Book Exchange", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An order book exchange is a trading venue that matches buy and sell orders through a central limit order book, an ordered record of outstanding bids and asks at each price level. A matching engine pairs incoming orders against resting liquidity according to price-time priority, executing trades and updating the book in real time. This model, dominant in traditional equities, futures, and centralised cryptocurrency exchanges, contrasts with automated market makers that price trades algorithmically against pooled liquidity rather than against discrete counter-orders.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:order-book-exchange", "labels": [ "Order Book Exchange" ], "is_subclass_of": [ "Market Microstructure" ], "wikilinks": [] }, { "id": "order-book", "title": "Order Book", "domain": "finance", "domain_name": "Finance", "definition": "An order book is a real-time, continuously updated electronic ledger that aggregates all outstanding limit buy orders (bids) and limit sell orders (asks) for a tradeable asset, organised by price level and, within each level, by time of arrival. A matching engine processes incoming market and limit orders against the resting book using price-time priority, generating trades whenever a bid price meets or exceeds an ask price. The visible depth of resting orders at each price level constitutes the liquidity profile of the market, and the gap between the best bid and best ask defines the bid-ask spread. Order books underpin both centralised exchanges and, increasingly, on-chain decentralised venues constrained by ledger throughput and gas costs.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:order-book", "labels": [ "Order Book", "Central Limit Order Book", "Order Matching" ], "is_subclass_of": [ "Exchange Mechanism" ], "wikilinks": [ "Price Discovery", "Market Making", "Decentralised Exchange", "Exchange Mechanism" ] }, { "id": "ordinals", "title": "Ordinals", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A scheme for numbering individual satoshis by order of issuance and transfer, allowing arbitrary data to be inscribed onto specific satoshis on the Bitcoin blockchain via the witness field of Taproot transactions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:ordinals", "labels": [ "Ordinals", "Ordinals Protocol" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Taproot", "UTXO", "BRC-20", "Bitcoin Script", "Bitcoin" ] }, { "id": "ordinary-differential-equation", "title": "Ordinary Differential Equation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An ordinary differential equation (ODE) is a mathematical equation relating a function of a single independent variable to its derivatives. It describes how a quantity changes continuously with respect to that variable, and its solution is the function or family of functions satisfying the relation, often determined by initial conditions. ODEs are foundational tools for modelling dynamical systems in physics, biology, engineering, and increasingly in machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ordinary-differential-equation", "labels": [ "Ordinary Differential Equation" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [] }, { "id": "organisational-layer", "title": "Organisational Layer", "domain": "governance", "domain_name": "Governance", "definition": "The Organisational Layer is the cross-cutting stratum that represents the internal structure, roles, and processes of an operating body. It sits above the Institutional Layer that grants it standing and supports operational and social activity. It contains organisational units, responsibilities, reporting lines, and internal processes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:organisational-layer", "labels": [ "Organisational Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Institutional Layer", "Operational Layer", "Process Layer", "Organisational Structure", "Division of Labour", "owl:Thing" ] }, { "id": "organisational-learning", "title": "Organisational Learning", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Organisational learning is the process by which a group, team or enterprise creates, retains and transfers knowledge to improve its collective performance over time. It encompasses how experience is encoded into routines, how lessons from success and failure are shared, and how the organisation adapts its mental models and practices in response. As a discipline it bridges knowledge management, continuous improvement and organisational culture.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:organisational-learning", "labels": [ "Organisational Learning" ], "is_subclass_of": [ "Knowledge Management" ], "wikilinks": [] }, { "id": "organisational-resilience", "title": "Organisational Resilience", "domain": "governance", "domain_name": "Governance", "definition": "Organisational resilience is the capacity of an organisation to anticipate, prepare for, respond to and adapt to incremental change and sudden disruptions in order to survive and prosper. It integrates risk management, business continuity, crisis response and adaptive capacity into a coherent governance posture. Unlike narrow continuity planning, it emphasises learning, flexibility and the ability to emerge stronger from adversity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:organisational-resilience", "labels": [ "Organisational Resilience" ], "is_subclass_of": [ "Resilience" ], "wikilinks": [] }, { "id": "organisational-theory", "title": "Organisational Theory", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The study of structural, behavioural, and governance patterns within organisations as they adopt distributed collaboration and immersive technologies. Organisational theory examines how hierarchies, communication channels, decision-making processes, and incentive structures must adapt to support decentralised, asynchronous, and spatially distributed teams operating within metaverse and mixed-reality workspaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:organisational-theory", "labels": [ "Organisational Theory", "OrganisationalTheory" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [ "owl:Thing" ] }, { "id": "organizational-change", "title": "Organizational Change", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Organizational Change is the deliberate process of transforming an organisation's structures, workflows, cultural norms, and operational paradigms. In technology contexts this encompasses adoption of immersive collaboration tools, distributed team models, and AI-augmented workflows, requiring sustained stakeholder engagement and iterative change management strategies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:organizational-change", "labels": [ "Organizational Change" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "origin-server", "title": "Origin Server", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An origin server is the authoritative source of truth for web content, hosting the original, canonical version of resources that a content delivery network and its edge caches replicate and serve to end users. When an edge node lacks a requested resource or its cached copy has expired, it fetches the content from the origin server. By concentrating authoritative content and offloading repeat delivery to caches, the origin-edge model improves performance, scalability, and resilience.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:origin-server", "labels": [ "Origin Server" ], "is_subclass_of": [ "CDN" ], "wikilinks": [] }, { "id": "orthometric-height", "title": "Orthometric Height", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orthometric-height", "labels": [ "Orthometric Height" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "orthorectification", "title": "Orthorectification", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:orthorectification", "labels": [ "Orthorectification" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "osmosis", "title": "Osmosis", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised exchange and automated market maker built in the Cosmos ecosystem that allows cross-chain token swaps and customisable liquidity pools using the Inter-Blockchain Communication protocol.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:osmosis", "labels": [ "Osmosis" ], "is_subclass_of": [ "Decentralized Exchange" ], "wikilinks": [ "Cosmos", "Automated Market Maker", "DeFi", "Liquidity Pool", "Decentralized Exchange" ] }, { "id": "otter-ai", "title": "Otter.ai", "domain": "ai", "domain_name": "Ai", "definition": "An American company that provides automatic speech recognition and transcription software for meetings and conversations. It generates real-time transcripts, summaries, and notes from audio.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:otter-ai", "labels": [ "Otter.ai" ], "is_subclass_of": [ "Speech Recognition" ], "wikilinks": [ "Speech Recognition", "Audio Processing", "Natural Language Processing" ] }, { "id": "ouroboros-consensus", "title": "Ouroboros Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Ouroboros is a provably secure proof-of-stake consensus protocol developed for the Cardano blockchain, the first PoS protocol to be formally verified with cryptographic security proofs equivalent to those of proof-of-work systems. It divides blockchain time into epochs and slots, using a verifiable random function to elect slot leaders from among stake pools proportionally to their delegated stake, enabling secure, energy-efficient block production without trusted setup.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ouroboros-consensus", "labels": [ "Ouroboros Consensus", "Ouroboros Protocol" ], "is_subclass_of": [ "Protocol and Consensus", "Proof of Stake" ], "wikilinks": [ "Blockchain", "Proof of Stake" ] }, { "id": "ouroboros", "title": "Ouroboros", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A family of provably secure proof-of-stake consensus protocols that divide time into epochs and slots and select block producers in proportion to stake.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ouroboros", "labels": [ "Ouroboros" ], "is_subclass_of": [ "Proof of Stake" ], "wikilinks": [ "Proof of Stake", "Consensus Mechanism", "Cardano" ] }, { "id": "outer-space", "title": "Outer Space", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:outer-space", "labels": [ "Outer Space" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "outlier-detection", "title": "Outlier Detection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The identification of observations that deviate so markedly from the rest of a dataset as to arouse suspicion that they were generated by a different mechanism \u2014 via statistical tests, distance and density measures, or learned models; a mature discipline spanning Grubbs' test and the IQR rule through Local Outlier Factor and Isolation Forest, applied both as a data-preprocessing step to protect downstream models and as an analysis goal in itself, since an outlier may be an error to remove or a discovery to investigate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:outlier-detection", "labels": [ "Outlier Detection" ], "is_subclass_of": [ "Anomaly Detection" ], "wikilinks": [ "Anomaly Detection", "Data Preprocessing", "Density Estimation" ] }, { "id": "outpainting", "title": "outpainting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Outpainting is a generative AI technique that extends an existing image beyond its original canvas boundaries by synthesising new, contextually consistent pixel content in the surrounding regions. A masked version of the source image\u2014padded with blank or noise-filled regions\u2014is fed to a conditioned image generation model (typically a latent diffusion model) which fills the extended area while respecting the style, lighting, and semantic content of the original. Outpainting is widely used in content creation pipelines for widening aspect ratios, reconstructing damaged artwork borders, and generating panoramic background extensions for virtual production and 3D scene authoring.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:outpainting", "labels": [ "Outpainting" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "output", "title": "Output", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Output is a component of a blockchain transaction that specifies a recipient address and an amount of value to be transferred, forming the fundamental unit through which cryptocurrency is allocated and ownership is recorded on a distributed ledger. In UTXO-based systems such as Bitcoin, unspent outputs serve as inputs to subsequent transactions, creating a directed graph of value flows secured by digital signatures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:output", "labels": [ "Output" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "over-collateralisation", "title": "Over Collateralisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Over-collateralisation is a risk-management practice in which the value of assets pledged as collateral exceeds the value of the obligation they secure. In decentralised finance it is the standard requirement for permissionless lending and for backing stablecoins, where borrowers must deposit more value than they draw to absorb price volatility. The surplus margin protects lenders and protocols against default by ensuring that collateral can be liquidated for at least the amount owed even when prices move adversely.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:over-collateralisation", "labels": [ "Over Collateralisation", "Over-Collateralisation", "Overcollateralisation", "Overcollateralization" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "over-the-air-update", "title": "Over The Air Update", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An over-the-air (OTA) update is the wireless delivery of new software, firmware, or configuration to a deployed device without physical access. OTA mechanisms package an update, transport it over a network, verify its authenticity and integrity, and apply it safely with rollback protection. They are essential to maintaining, securing, and extending the capabilities of fleets of embedded and connected devices throughout their operational life.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:over-the-air-update", "labels": [ "Over The Air Update", "Over-the-Air Update" ], "is_subclass_of": [ "Infrastructure", "Firmware" ], "wikilinks": [] }, { "id": "overfitting", "title": "Overfitting", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Overfitting is a phenomenon in machine learning where a model learns the training data too precisely\u2014including noise and spurious correlations\u2014resulting in poor generalisation to unseen data. It corresponds to high variance and low bias in the bias-variance tradeoff, and is mitigated through regularisation, dropout, early stopping, and data augmentation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:overfitting", "labels": [ "Overfitting" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "overlay-network", "title": "Overlay Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An overlay network is a logical network built on top of an existing physical or underlay network, where nodes are connected by virtual links that may each traverse many underlying hops. By abstracting away the physical topology, overlays implement custom routing, addressing and services such as peer-to-peer file sharing, content delivery and encrypted tunnelling. They trade some efficiency for flexibility, resilience and the ability to deploy new network behaviour without changing the underlay.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:overlay-network", "labels": [ "Overlay Network" ], "is_subclass_of": [ "Infrastructure", "Networking Infrastructure" ], "wikilinks": [] }, { "id": "ownership-freedom-distributed", "title": "Ownership Freedom distributed", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A socio-technical paradigm establishing user sovereignty through decentralized control mechanisms, cryptographic ownership proofs, and distributed governance models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ownership-freedom-distributed", "labels": [ "Ownership Freedom distributed" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "ownership-record", "title": "Ownership Record", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ownership Record is a persistent entry that attests who holds title to an asset at a given time. On a blockchain it is an immutable, timestamped ledger entry binding an asset identifier to a controlling address, providing tamper-evident provenance and transfer history. Such records underpin tokenized assets like certificates and royalty instruments where verifiable, auditable ownership is essential.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ownership-record", "labels": [ "Ownership Record", "Ownership Records" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "ownership-token", "title": "Ownership Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ownership Token is a blockchain token that represents and confers title to an underlying asset, whether digital or real-world. Typically implemented as a non-fungible or semi-fungible token, holding it in a wallet constitutes provable control of the represented item, and transferring it transfers ownership. Ownership tokens enable trading, collateralization, and programmable rights over digital goods and tokenized property.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ownership-token", "labels": [ "Ownership Token" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "ownership-transfer", "title": "Ownership Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Ownership transfer is the process by which legal or functional control of an asset \u2014 physical, digital, or tokenised \u2014 passes from one party to another, recorded in a registry, ledger, or smart contract. In blockchain and Web3 contexts it typically refers to atomic state transitions that update on-chain ownership records for fungible tokens, NFTs, or tokenised real-world assets, with cryptographic proofs replacing traditional notarial or escrow intermediaries. The mechanism must satisfy atomicity, finality, and auditability requirements.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ownership-transfer", "labels": [ "Ownership Transfer", "Ownership Chain" ], "is_subclass_of": [ "Digital Asset Management" ], "wikilinks": [] }, { "id": "ownership-and-freedom-distributed", "title": "Ownership and Freedom Distributed", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The principle of distributing control and governance of digital assets or data across multiple participants rather than concentrating authority in a single entity. This encompasses decentralised identity, cryptographic ownership proofs, peer-to-peer governance mechanisms, and user-controlled data portability as foundational guarantees for sovereignty in metaverse and Web3 ecosystems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ownership-and-freedom-distributed", "labels": [ "Ownership and Freedom Distributed" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Metaverse 101", "MetaverseDomain" ] }, { "id": "oxford-internet-institute", "title": "Oxford Internet Institute", "domain": "governance", "domain_name": "Governance", "definition": "The Oxford Internet Institute is a department of the University of Oxford that conducts multidisciplinary research on the social and economic effects of the internet and digital technologies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:oxford-internet-institute", "labels": [ "Oxford Internet Institute" ], "is_subclass_of": [ "AI Governance and Ethics", "owl:Thing" ], "wikilinks": [ "AI Governance", "owl:Thing" ] }, { "id": "ozone-layer", "title": "Ozone Layer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ozone-layer", "labels": [ "Ozone Layer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "pbft", "title": "PBFT", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Practical Byzantine Fault Tolerance (PBFT) is a state-machine replication protocol designed by Castro and Liskov (1999) that achieves consensus in asynchronous distributed systems despite up to f arbitrarily faulty (Byzantine) nodes, requiring a total of at least 3f+1 replicas. It proceeds through pre-prepare, prepare, and commit phases to ensure all correct replicas execute the same sequence of operations, providing both safety and liveness under partial synchrony assumptions. PBFT was the first Byzantine fault-tolerant protocol deemed practical for deployed systems, with latency polynomial rather than exponential in the number of nodes. Its communication complexity of O(n\u00b2) limits scalability but makes it highly suitable for small-to-medium permissioned blockchain networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pbft", "labels": [ "PBFT", "PBFT Consensus" ], "is_subclass_of": [ "Byzantine Fault Tolerance" ], "wikilinks": [] }, { "id": "pci-dss", "title": "PCI-DSS", "domain": "finance", "domain_name": "Finance", "definition": "The Payment Card Industry Data Security Standard (PCI-DSS) is a security standard governing the handling of cardholder data by organisations that store, process, or transmit payment card information. It defines a set of requirements covering network security, encryption, access control, monitoring, and vulnerability management to reduce the risk of card data breaches. Compliance is mandated by the major card brands and verified through self-assessment questionnaires or external audits depending on transaction volume.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pci-dss", "labels": [ "PCI-DSS", "PCI DSS" ], "is_subclass_of": [ "Compliance" ], "wikilinks": [] }, { "id": "pddl", "title": "PDDL", "domain": "ai", "domain_name": "Ai", "definition": "The Planning Domain Definition Language (PDDL) is a standardised formal language for encoding automated-planning problems as domains and problem instances. A domain specifies predicates and actions with preconditions and effects, while a problem instance defines objects, an initial state, and a goal. PDDL provides a common input format that lets generic planners compute action sequences, and it underpins much of classical AI task planning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pddl", "labels": [ "PDDL", "PDDL Standard" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "pegasus", "title": "PEGASUS", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pre-training with Extracted Gap-sentences for Abstractive SUmmarization: a pre-training approach specifically designed for abstractive summarisation that masks and predicts entire sentences rather than individual tokens.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:pegasus", "labels": [ "PEGASUS" ], "is_subclass_of": [ "AI Model Architecture", "Natural Language Processing" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "pid-control", "title": "pid control", "domain": "robotics", "domain_name": "Robotics", "definition": "PID (Proportional\u2013Integral\u2013Derivative) control is a closed-loop feedback control algorithm that computes a corrective output by summing three terms derived from the error signal: a proportional term that scales the instantaneous error, an integral term that accumulates past error to eliminate steady-state offset, and a derivative term that reacts to the rate of change of error to anticipate and damp future deviations. The control law u(t) = Kp\u00b7e(t) + Ki\u00b7\u222be(\u03c4)d\u03c4 + Kd\u00b7(de/dt) is parameterised by three tunable gains and applies universally to any plant where a measurable output can be compared against a desired setpoint. Originating in the 1940s through work by Minorsky, Ziegler, and Nichols, PID remains the dominant feedback controller in industrial automation, robotics, aerospace, and embedded systems due to its conceptual simplicity, zero requirement for an explicit plant model, and extensive supporting theory for stability analysis and gain tuning.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:pid-control", "labels": [ "PID Control", "PID Tuning" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "pjm-interconnection", "title": "PJM Interconnection", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A regional transmission organization (RTO) that manages the flow of electric power across a 13-state grid in the Eastern United States.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:pjm-interconnection", "labels": [ "PJM Interconnection" ], "is_subclass_of": [ "Energy Infrastructure" ], "wikilinks": [] }, { "id": "plonk", "title": "PLONK", "domain": "blockchain", "domain_name": "Blockchain", "definition": "PLONK is a zero-knowledge proof system that uses polynomial commitments and a universal trusted setup. It is used to build succinct proofs for verifiable computation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:plonk", "labels": [ "PLONK", "PLONK Proof System" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [ "Cryptography", "Rollup", "Scroll", "Zero-Knowledge Proof", "https://eprint.iacr.org/2019/953", "https://vitalik.eth.limo/general/2019/09/22/plonk.html" ] }, { "id": "posix", "title": "POSIX", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "POSIX (Portable Operating System Interface) is a family of IEEE standards that defines the application programming interface, command-line shells, and utility interfaces for maintaining compatibility between operating systems. By specifying system calls, file-system semantics, process control, and shell behaviour, POSIX allows software written against the standard to be ported across conforming Unix-like systems with minimal modification. It is the canonical contract underpinning portable system software.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:posix", "labels": [ "POSIX" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "psnr-metric", "title": "PSNR Metric", "domain": "ai", "domain_name": "Ai", "definition": "Peak Signal-to-Noise Ratio (PSNR) is an objective image- and video-quality metric expressing the ratio between the maximum possible signal power and the power of distorting noise, computed from the mean squared error between a reference and a degraded signal. Measured in decibels, higher PSNR values indicate closer fidelity to the original. It is widely used to benchmark lossy compression, restoration, and reconstruction algorithms despite its known weak correlation with perceived quality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:psnr-metric", "labels": [ "PSNR Metric" ], "is_subclass_of": [ "Evaluation Metric", "Model Performance" ], "wikilinks": [] }, { "id": "pack-ice", "title": "Pack Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pack-ice", "labels": [ "Pack Ice" ], "is_subclass_of": [ "Drift Ice" ], "wikilinks": [] }, { "id": "package-manager", "title": "Package Manager", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A package manager is a tool that automates the installation, upgrade, configuration and removal of software libraries and their dependencies from curated repositories. It resolves version constraints across a dependency graph and records lockfiles so environments are reproducible. Package managers are foundational infrastructure for modern software development and continuous integration.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:package-manager", "labels": [ "Package Manager" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "packet-compression", "title": "Packet Compression", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The application of lossless or lossy compression algorithms to network packets to reduce transmission bandwidth and latency in real-time data pipelines. In metaverse and spatial computing contexts, packet compression is critical for streaming 3D scene state, avatar motion data, and sensor telemetry with minimal perceptual degradation at constrained network bandwidths.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:packet-compression", "labels": [ "Packet Compression" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "packet-loss-recovery", "title": "Packet Loss Recovery", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Techniques used to recover from or mitigate the loss of data packets during transmission over a network. They include retransmission, forward error correction, and concealment of missing data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:packet-loss-recovery", "labels": [ "Packet Loss Recovery", "Packet Loss Detection" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Forward Error Correction", "Video Streaming", "Network Protocol" ] }, { "id": "packet-switching", "title": "Packet Switching", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Packet switching is a method of data transmission in which messages are divided into discrete packets that are routed independently across a shared network and reassembled at the destination. Each packet carries addressing information allowing intermediate nodes to forward it along varying paths, making efficient use of shared links. It is the foundational technique of the internet, contrasting with circuit switching that dedicates a fixed path for a session.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:packet-switching", "labels": [ "Packet Switching" ], "is_subclass_of": [ "Network Communication" ], "wikilinks": [] }, { "id": "pagerank", "title": "PageRank", "domain": "ai", "domain_name": "Ai", "definition": "PageRank is a link-analysis algorithm that assigns a numerical importance score to each node in a directed graph based on the structure of incoming links, modelling importance as the stationary distribution of a random walk that occasionally teleports to a random node. Originally devised to rank web pages by treating hyperlinks as votes whose weight depends on the ranking of the linking page, it generalises to any graph where influence propagates along edges. The scores are computed iteratively until convergence and are robust to local manipulation because importance flows recursively from important neighbours.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pagerank", "labels": [ "PageRank" ], "is_subclass_of": [ "Network Analysis" ], "wikilinks": [] }, { "id": "paged-attention", "title": "Paged Attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Paged attention is a memory management technique for large language model inference that partitions the key-value cache into fixed-size blocks managed like virtual memory pages. By decoupling logical token positions from physical memory layout it eliminates fragmentation, enables near-zero waste in cache allocation and allows sharing of cached prefixes across requests. Introduced in the vLLM serving system, it substantially increases throughput for high-concurrency inference.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:paged-attention", "labels": [ "Paged Attention", "PagedAttention" ], "is_subclass_of": [ "KV Cache" ], "wikilinks": [] }, { "id": "pairwise-comparison", "title": "Pairwise Comparison", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Pairwise comparison is a method of evaluation in which items are judged two at a time, with each judgement expressing which of the two is preferred or superior on some criterion. Because relative judgements are easier and more reliable for humans than absolute scoring, pairwise comparison is widely used to elicit preferences and to construct rankings from many such local decisions. Statistical models such as the Bradley-Terry model convert collections of pairwise outcomes into latent strength or quality scores. In machine learning it is the dominant feedback format for training reward models and aligning language models with human preferences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pairwise-comparison", "labels": [ "Pairwise Comparison" ], "is_subclass_of": [ "Machine Learning", "Model Performance" ], "wikilinks": [] }, { "id": "pairwise-qf", "title": "Pairwise QF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Pairwise QF (Pairwise-bounded Quadratic Funding) is a variant of quadratic funding that limits the matching subsidy attributable to coordinating groups of contributors by discounting the influence of pairs who repeatedly co-fund. It mitigates collusion and Sybil attacks that plague naive quadratic funding by capping the matching any pair of donors can jointly unlock. The mechanism preserves quadratic funding's democratic weighting of many small contributions while resisting manipulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:pairwise-qf", "labels": [ "Pairwise QF" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "pan-sharpening", "title": "Pan-sharpening", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pan-sharpening", "labels": [ "Pan-sharpening" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "panchromatic-imaging", "title": "Panchromatic Imaging", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:panchromatic-imaging", "labels": [ "Panchromatic Imaging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "pandemic-preparedness-ai", "title": "Pandemic Preparedness AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pandemic Preparedness AI is the class of artificial intelligence systems applied specifically to anticipating, detecting, and mitigating infectious disease outbreaks with pandemic potential. It spans genomic surveillance models that flag novel pathogen variants, epidemiological forecasting that projects transmission dynamics, protein-structure and drug-discovery pipelines that accelerate vaccine and therapeutic design, and early-warning systems that fuse clinical, wastewater, and mobility signals. Distinct from the broader public-health discipline of pandemic preparedness, this class denotes the computational and machine-learning tooling that augments human decision-making, compressing the time between the emergence of a threat and an effective countermeasure.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:pandemic-preparedness-ai", "labels": [ "Pandemic Preparedness AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence", "Public Health", "Biosecurity" ] }, { "id": "pandemic-preparedness", "title": "Pandemic Preparedness", "domain": "governance", "domain_name": "Governance", "definition": "Pandemic preparedness is the set of plans, capabilities, and investments that enable societies to detect, prevent, and respond effectively to large-scale infectious-disease outbreaks. It spans surveillance, diagnostics, vaccine and treatment readiness, healthcare surge capacity, stockpiles, and coordinated governance across borders. As a domain of public-health governance, it aims to reduce the health, social, and economic harm of pandemics through anticipatory action rather than reactive crisis management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pandemic-preparedness", "labels": [ "Pandemic Preparedness" ], "is_subclass_of": [ "Biosecurity" ], "wikilinks": [] }, { "id": "panoptic-segmentation", "title": "Panoptic Segmentation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A computer vision task that unifies semantic segmentation and instance segmentation by assigning every image pixel both a class label and an instance identifier, providing holistic scene parsing that distinguishes countable foreground objects (\"things\") from amorphous background regions (\"stuff\"). Architectures such as Panoptic FPN, Panoptic-DeepLab, and MaskFormer formalise this unified representation.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:panoptic-segmentation", "labels": [ "Panoptic Segmentation", "Panoptic FPN", "Panoptic Quality" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "Instance Segmentation", "MetaverseDomain", "Scene Understanding", "Semantic Segmentation" ] }, { "id": "paolo-tasca", "title": "Paolo Tasca", "domain": "finance", "domain_name": "Finance", "definition": "Paolo Tasca is an economist specialising in digital economics and distributed ledger technology. He founded the Centre for Blockchain Technologies at University College London.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:paolo-tasca", "labels": [ "Paolo Tasca" ], "is_subclass_of": [ "Economics" ], "wikilinks": [ "Distributed Ledger Technology", "Economics" ] }, { "id": "parabolic-flight", "title": "Parabolic Flight", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:parabolic-flight", "labels": [ "Parabolic Flight" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "parallel-computing", "title": "Parallel Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Parallel computing executes many operations simultaneously across multiple processing units to reduce wall-clock time for workloads that can be decomposed into independent or loosely coupled parts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:parallel-computing", "labels": [ "Parallel Computing", "Parallel Computation", "Parallel Programming" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [ "GPU Architecture", "Real-Time Rendering", "Pipeline Parallelism", "CUDA", "GPU Computing", "Distributed Computing" ] }, { "id": "parallel-corpus", "title": "Parallel Corpus", "domain": "ai", "domain_name": "Ai", "definition": "A Parallel Corpus is a collection of texts paired with their translations in one or more other languages, aligned at the sentence or segment level. It provides the supervised training signal for statistical and neural machine-translation systems by exemplifying how meaning maps across languages. The size, quality, and domain coverage of a parallel corpus strongly influence the accuracy of trained translation models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:parallel-corpus", "labels": [ "Parallel Corpus", "Multilingual Corpus", "OPUS Corpus", "Parallel Corpora" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "parallel-processing", "title": "Parallel Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Parallel Processing is a computational paradigm in which multiple calculations or processes are carried out simultaneously by decomposing a problem into sub-tasks that execute concurrently across multiple processor cores, GPUs, or distributed compute nodes. It exploits data parallelism, task parallelism, and pipeline parallelism to reduce wall-clock execution time and increase throughput. Parallel processing is foundational to modern high-performance computing, enabling workloads such as large-scale neural network training, real-time physics simulation, and petabyte-scale data analytics that would be infeasible on sequential hardware.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:parallel-processing", "labels": [ "Parallel Processing" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "parallel-programming-model", "title": "Parallel Programming Model", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A parallel programming model is an abstraction that defines how a program expresses concurrent computation and how that computation maps onto parallel hardware, covering concerns such as task decomposition, data sharing, and synchronisation. Examples include the SIMT model used by GPU architectures, shared-memory threading, and message-passing between distributed processes. The choice of parallel programming model determines how effectively an application can exploit hardware such as GPUs to accelerate throughput.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:parallel-programming-model", "labels": [ "Parallel Programming Model" ], "is_subclass_of": [ "Parallel Computing" ], "wikilinks": [] }, { "id": "parallel-robot", "title": "Parallel Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot whose end-effector is connected to the base by multiple independent, simultaneously actuated kinematic chains (limbs), giving it a closed-loop structure. Parallel robots achieve high rigidity, speed, and accuracy with low moving mass; canonical examples include the Delta robot and Stewart platform used in pick-and-place, machining, and flight simulation applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:parallel-robot", "labels": [ "Parallel Robot" ], "is_subclass_of": [ "Robot Type", "Manipulator Robot" ], "wikilinks": [ "Manipulator Robot", "Robotics" ] }, { "id": "parameter-count", "title": "Parameter Count", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The total number of trainable weights and biases in a neural network, serving as the primary measure of model size and capacity. Parameter count typically ranges from millions to hundreds of billions in modern language models, and governs memory requirements, inference cost, and the upper bound on information that can be encoded; scaling law research relates it to training compute and dataset size.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:parameter-count", "labels": [ "Parameter Count" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "parameter-estimation", "title": "Parameter Estimation", "domain": "robotics", "domain_name": "Robotics", "definition": "Parameter estimation is the process of inferring the unknown parameters of a mathematical or statistical model from observed data. In robotics and control it is central to fitting dynamic models, calibrating sensors, and identifying physical constants from measurements, typically by optimising a likelihood or least-squares criterion. It provides the model coefficients that downstream estimation, control, and prediction algorithms rely upon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:parameter-estimation", "labels": [ "Parameter Estimation" ], "is_subclass_of": [ "System Identification" ], "wikilinks": [] }, { "id": "parameter-governance", "title": "Parameter Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Parameter governance is the on-chain process by which a decentralised protocol's tunable configuration values \u2014 such as fees, collateral ratios, interest-rate curves, and reward emissions \u2014 are proposed, deliberated, voted upon, and enacted without altering the underlying contract code. By exposing safe adjustment levers to token-holder governance, it lets protocols adapt to market conditions while preserving the immutability of core logic. It is a foundational pattern in DeFi and DAO management, distinct from full code upgrades.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:parameter-governance", "labels": [ "Parameter Governance" ], "is_subclass_of": [ "On-chain Governance" ], "wikilinks": [] }, { "id": "parameter-modulation-system", "title": "Parameter Modulation System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A signal processing architecture that uses modulator signals to dynamically control carrier signal parameters like pitch, amplitude, and timbre over time, enabling expressive sound synthesis and complex audio effects through techniques such as FM, AM, and envelope modulation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:parameter-modulation-system", "labels": [ "Parameter Modulation System" ], "is_subclass_of": [ "Content and Assets", "Audio System" ], "wikilinks": [ "Dynamic Sound Design", "Audio System", "metaverse" ] }, { "id": "parameter-server", "title": "Parameter Server", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A distributed systems architecture for large-scale machine learning in which a set of centralised server nodes holds the globally shared model parameters while many worker nodes compute gradients on partitions of the training data, pushing updates to the servers and pulling refreshed parameters back. The design decouples computation from state management, supports synchronous and asynchronous update schemes, and underpinned the first generation of industrial-scale distributed training before decentralised all-reduce approaches became dominant.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:parameter-server", "labels": [ "Parameter Server" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [ "Distributed Computing", "Data Parallelism", "Gradient Aggregation", "Collective Communication" ] }, { "id": "parameter-set", "title": "Parameter Set", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A collection of configurable values that define the behavior, appearance, or operation of a system, algorithm, or model, allowing users to store, recall, and modify settings to achieve different outputs or modes of operation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:parameter-set", "labels": [ "Parameter Set" ], "is_subclass_of": [ "Content and Assets", "Configuration" ], "wikilinks": [ "Reproducibility", "Configuration", "metaverse" ] }, { "id": "parameter-efficient-fine-tuning", "title": "Parameter-Efficient Fine-Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Training techniques that update only a small subset of model parameters during fine-tuning, reducing computational and memory requirements whilst maintaining comparable performance to full fine-tuning. PEFT methods enable adaptation of large models with limited resources by freezing most pre-trained weights and adding or modifying a minimal set of trainable parameters.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:parameter-efficient-fine-tuning", "labels": [ "Parameter-Efficient Fine-Tuning", "Parameter-Efficient Training" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Transfer Learning", "Adapter Modules", "ArtificialIntelligenceDomain", "Full Fine Tuning", "LoRA", "Prompt Tuning" ] }, { "id": "parameter", "title": "Parameter", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Parameter in machine learning and AI is a learnable variable internal to a model whose values are adjusted during training to minimise a loss function, as distinguished from hyperparameters, which are configuration choices set before training begins. In neural networks, parameters encompass weight matrices and bias vectors in each layer; the total parameter count (ranging from thousands in small models to hundreds of billions in large language models) is a primary indicator of model capacity and computational cost. Parameters are initialised randomly or via transfer learning, then updated iteratively through gradient-based optimisation algorithms such as stochastic gradient descent, encoding learned representations of the training distribution.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:parameter", "labels": [ "Parameter", "Learnable Scale Parameter", "Model Parameter" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "parameterize", "title": "Parameterize", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:parameterize", "labels": [ "Parameterize" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "parameterized-data-state", "title": "Parameterized Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:parameterized-data-state", "labels": [ "Parameterized Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "parametric-design-methodology", "title": "Parametric Design Methodology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A design methodology in which objects, environments, or behaviours are specified through adjustable parameters rather than fixed geometry, enabling variant generation and real-time adaptation. In metaverse and spatial computing contexts, parametric approaches underpin voice-and-text-driven CAD primitive creation, procedural content generation, and digital twin model customisation within shared virtual workspaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:parametric-design-methodology", "labels": [ "Parametric Design Methodology", "Parametric", "Parametric Design" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Landvault Create", "NVIDIA Omniverse" ] }, { "id": "parametric-insurance", "title": "Parametric Insurance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Parametric insurance is a form of coverage that pays a predetermined amount when a measurable trigger event crosses a defined threshold, rather than indemnifying assessed losses. Because payouts depend on objective parameters such as rainfall, wind speed, earthquake magnitude, or flight delay, claims can be settled automatically without loss adjustment. On blockchains, parametric insurance is implemented through smart contracts that consume oracle data to verify triggers and disburse funds, enabling fast, transparent, and trust-minimised settlement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:parametric-insurance", "labels": [ "Parametric Insurance" ], "is_subclass_of": [ "Decentralized Finance" ], "wikilinks": [] }, { "id": "parametric-modeling", "title": "Parametric Modeling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A CAD design modology that uses algorithms and adjustable parameters to generate and modify complex 3D geometry, enabling flexible, constraint-driven design where changes to input values automatically propagate throughout the model.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:parametric-design-methodology-modeling", "labels": [ "Parametric Modeling" ], "is_subclass_of": [ "Content and Assets", "3D Modelling" ], "wikilinks": [ "Generative Design", "3D Modeling", "metaverse" ] }, { "id": "parental-controls", "title": "Parental Controls", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software tools and platform features that enable guardians to monitor, filter, and limit children's access to digital content, applications, and online interactions, using AI-powered content analysis and customizable restrictions to promote safe technology use.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:parental-controls", "labels": [ "Parental Controls" ], "is_subclass_of": [ "Governance and Safety", "Digital Safety" ], "wikilinks": [ "Safe Online Environment", "Digital Safety", "metaverse" ] }, { "id": "paris-agreement-article-6", "title": "Paris Agreement Article 6", "domain": "governance", "domain_name": "Governance", "definition": "Paris Agreement Article 6 is the provision of the 2015 Paris Agreement that establishes mechanisms for countries to cooperate in meeting emissions-reduction targets, including international transfer of mitigation outcomes and a centralized crediting market. It defines accounting rules to prevent double-counting when carbon reductions are traded between parties. Article 6 provides the international legal basis for cross-border carbon markets and corresponding adjustments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:paris-agreement-article-6", "labels": [ "Paris Agreement Article 6", "Article 6 Paris Agreement" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "paris-agreement", "title": "Paris Agreement", "domain": "governance", "domain_name": "Governance", "definition": "The Paris Agreement is a legally binding international treaty on climate change, adopted under the United Nations Framework Convention on Climate Change (UNFCCC) at COP21 in December 2015 and entering into force in November 2016. It establishes a long-term global temperature goal, requiring parties to limit the increase in global average temperature to well below 2 degrees Celsius above pre-industrial levels while pursuing efforts to limit warming to 1.5 degrees Celsius. Each signatory submits Nationally Determined Contributions (NDCs) \u2014 self-defined emissions-reduction pledges \u2014 subject to a five-year review and ratchet mechanism designed to progressively strengthen ambition. The treaty also creates a transparency framework for monitoring, reporting, and verification, and provisions for climate finance, technology transfer, and capacity building to support developing nations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:paris-agreement", "labels": [ "Paris Agreement", "Paris Agreement Alignment", "Paris Climate Agreement" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "Sustainability", "Carbon Accounting", "Voluntary Carbon Market", "Carbon Credits" ] }, { "id": "part-of-speech-tagging", "title": "Part-of-Speech Tagging", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Part-of-speech tagging is the natural language processing task of assigning each word in a sentence a grammatical category, such as noun, verb or adjective, based on its definition and surrounding context. It is a foundational preprocessing step for downstream tasks such as parsing and information extraction, and has historically been solved with statistical sequence models, including hidden Markov models, before being subsumed by neural sequence labelling approaches. Accurate tagging disambiguates words whose part of speech depends on context.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:part-of-speech-tagging", "labels": [ "Part-of-Speech Tagging", "Part Of Speech Tagging" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "partial-gravity", "title": "Partial Gravity", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:partial-gravity", "labels": [ "Partial Gravity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "partial-synchrony", "title": "Partial Synchrony", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Partial synchrony is a timing model for distributed systems that sits between the fully synchronous model, where message delays are bounded and known, and the fully asynchronous model, where delays are unbounded. In the partially synchronous model there exists an unknown bound on message delay that eventually holds after some unknown global stabilisation time (GST), or alternatively a known bound that holds only after GST. This model is the theoretical foundation for practical Byzantine fault-tolerant consensus protocols, allowing them to circumvent the FLP impossibility result by guaranteeing safety always and liveness once the network behaves synchronously.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:partial-synchrony", "labels": [ "Partial Synchrony" ], "is_subclass_of": [ "Consensus Protocol" ], "wikilinks": [] }, { "id": "participant-authentication", "title": "Participant Authentication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The process of verifying the identity of users in virtual environments, events, and metaverse platforms through methods including multi-factor authentication, biometrics, blockchain-based identity, and AI-powered verification to prevent impersonation and ensure secure access.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participant-authentication", "labels": [ "Participant Authentication" ], "is_subclass_of": [ "Security and Identity", "Identity Management" ], "wikilinks": [ "Secure Virtual Events", "Identity Management", "metaverse" ] }, { "id": "participant-consent", "title": "Participant Consent", "domain": "governance", "domain_name": "Governance", "definition": "Participant Consent is the explicit, informed agreement obtained from individuals before their voice, image, or data are recorded or processed. In meeting and communication contexts it is both an ethical norm and a legal requirement under many wiretapping and data-protection regimes, often necessitating clear notice and an opportunity to decline. Capturing and recording consent state is a prerequisite for lawful recording and downstream processing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:participant-consent", "labels": [ "Participant Consent" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "participant-coordination", "title": "Participant Coordination", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The synchronization and management of multiple users interacting within shared virtual environments, requiring real-time pose tracking, coordinate system alignment, and collaborative state management to enable seamless multi-user experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participant-coordination", "labels": [ "Participant Coordination" ], "is_subclass_of": [ "Platform and Environment", "Multi-User Systems" ], "wikilinks": [ "Shared Virtual Experiences", "metaverse", "Multi-User Systems" ] }, { "id": "participant-management-system", "title": "Participant Management System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A platform for organizing event attendees through registration, invitation management, check-in tracking, and real-time attendance monitoring, providing tools for ticketing, payment processing, and post-event analytics.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participant-management-system", "labels": [ "Participant Management System" ], "is_subclass_of": [ "Platform and Environment", "Event Management" ], "wikilinks": [ "Event Analytics", "Event Management", "metaverse" ] }, { "id": "participant-protection", "title": "Participant Protection", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Policies, technologies, and practices designed to safeguard users' privacy, data, and safety on digital platforms, encompassing regulatory compliance, privacy-enhancing technologies, and platform safety features that protect against data misuse, harassment, and harmful content.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participant-protection", "labels": [ "Participant Protection" ], "is_subclass_of": [ "Governance and Safety", "Digital Safety" ], "wikilinks": [ "Trust in Digital Platforms", "Digital Safety", "metaverse" ] }, { "id": "participant-recruitment", "title": "Participant Recruitment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Participant recruitment is the process of identifying, screening and enlisting individuals who match a target user profile to take part in usability testing or broader user research activities. It involves defining eligibility criteria, sourcing candidates through panels, screeners or existing user bases, and scheduling sessions while balancing sample diversity against study timeline and budget. The quality of participant recruitment directly affects how representative and generalisable a study's usability findings are.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:participant-recruitment", "labels": [ "Participant Recruitment" ], "is_subclass_of": [ "User Research" ], "wikilinks": [] }, { "id": "participation-framework", "title": "Participation Framework", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured approach to designing stakeholder engagement processes that establish goals, principles, mods, and mechanisms for involving citizens, communities, or users in decision-making, governance, and collaborative activities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participation-framework", "labels": [ "Participation Framework" ], "is_subclass_of": [ "Governance and Safety", "Governance Framework" ], "wikilinks": [ "Inclusive Decision Making", "Governance Framework", "metaverse" ] }, { "id": "participatory-design", "title": "Participatory Design", "domain": "governance", "domain_name": "Governance", "definition": "Participatory Design is a design methodology in which end-users, stakeholders, and affected communities are actively involved as co-designers throughout the design process rather than being passive subjects of research. Originating in Scandinavian workplace democracy movements of the 1970s, it prioritises the experiential knowledge of participants and seeks to produce systems that genuinely serve their needs. In digital technology and AI contexts, it is applied to ensure that systems reflect diverse human values and avoid embedding biases of the designer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:participatory-design", "labels": [ "Participatory Design" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "participatory-policy-making", "title": "Participatory Policy Making", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A governance approach that involves citizens and stakeholders directly in the formulation, design, and eof public policies through mechanisms such as citizen advisory boards, deliberative assemblies, and digital participation platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:participatory-policy-making", "labels": [ "Participatory Policy Making" ], "is_subclass_of": [ "Governance and Safety", "Democratic Governance" ], "wikilinks": [ "Policy Legitimacy", "Democratic Governance", "metaverse" ] }, { "id": "particle-filter", "title": "Particle Filter", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ParticleFilter denotes a class of Sequential Monte Carlo (SMC) algorithms that approximate the posterior probability distribution bel(x_t) = p(x_t | z_{1:t}, u_{1:t}) over the hidden state x_t of a stochastic dynamical system by maintaining a weighted empirical measure {(x_t^(i), w_t^(i))}_{i...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:particle-filter", "labels": [ "Particle Filter" ], "is_subclass_of": [ "AI Technique", "Monte Carlo Methods", "Bayes Filter", "Sequential Monte Carlo", "Monte Carlo Method", "Nonparametric Filter", "Recursive Bayesian Estimator" ], "wikilinks": [ "AI Agent", "AlgorithmLayer", "Auxiliary Variable Method", "Bayesian Inference", "Effective Sample Size", "Extended Kalman Filter", "FastSLAM", "Gaussian Process", "Global Localisation", "Histogram Filter", "ICRA", "IEEE Signal Processing Society", "Importance Sampling", "Importance Weight", "IMU", "IROS", "Kidnapped Robot Recovery", "KLD-Sampling", "Likelihood Function", "Markov Assumption" ] }, { "id": "particle-system", "title": "Particle System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computer graphics technique that simulates fuzzy, chaotic, or fluid phenomena by managing large numbers of small graphical objects with properties like position, velocity, color, and lifetime, controlled by emitters and affected by forces to create effects like fire, smoke, water, and magical e...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:particle-system", "labels": [ "Particle System" ], "is_subclass_of": [ "Display and Rendering", "Visual Effects" ], "wikilinks": [ "Dynamic Visual Effects", "metaverse", "Visual Effects" ] }, { "id": "particle-systems", "title": "Particle Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The collective infrastructure and tooling for creating and managing particle-based visual effects in game engines and VFX software, encompassing emitter configuration, physics integration, and real-time or offline rendering pipelines.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:particle-systems", "labels": [ "Particle Systems" ], "is_subclass_of": [ "Display and Rendering", "VFX Tools" ], "wikilinks": [ "Real Time VFX", "metaverse", "VFX Tools" ] }, { "id": "partition-attack", "title": "Partition Attack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Partition Attack is a network-level attack against a blockchain in which an adversary manipulates routing infrastructure to segment the peer-to-peer network into two or more isolated subgraphs, causing each partition to mine or validate on a separate chain branch. When the partition is healed the shorter branch is discarded, enabling the attacker to waste honest mining power and potentially facilitate double-spend attacks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:partition-attack", "labels": [ "Partition Attack" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Network Component" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "partition-tolerance", "title": "Partition Tolerance", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The property of a distributed system that allows it to continue operating correctly when the network splits into components that cannot communicate with one another, so that messages between nodes are arbitrarily delayed or lost. As the P in the CAP theorem, partition tolerance is effectively mandatory over real networks, forcing designers to choose during a partition between refusing requests to preserve consistency and serving them to preserve availability at the cost of divergent replicas.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:partition-tolerance", "labels": [ "Partition Tolerance" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [ "CAP Theorem", "Fault Tolerance", "Eventual Consistency" ] }, { "id": "pas-2060", "title": "Pas 2060", "domain": "governance", "domain_name": "Governance", "definition": "PAS 2060 is a publicly available specification published by the British Standards Institution that defines the requirements for demonstrating and substantiating carbon neutrality for an organisation, product or activity. It sets out a four-stage process of quantification, reduction, offsetting and documentation, requiring a qualifying explanatory statement and credible offsets for residual emissions. The specification is widely used as a recognised benchmark for verifiable carbon-neutrality claims.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pas-2060", "labels": [ "Pas 2060", "PAS 2060" ], "is_subclass_of": [ "Carbon Neutrality" ], "wikilinks": [] }, { "id": "passive-earth-observation-sensor", "title": "Passive Earth Observation Sensor", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:passive-earth-observation-sensor", "labels": [ "Passive Earth Observation Sensor" ], "is_subclass_of": [ "Earth Observation Sensor" ], "wikilinks": [] }, { "id": "passive-income", "title": "Passive Income", "domain": "finance", "domain_name": "Finance", "definition": "Passive Income is earnings generated from an asset or position with minimal ongoing active effort from the holder, as distinct from wages earned through direct labour. In decentralised finance it typically refers to yield accrued from lending, liquidity provision, or staking, where deployed capital earns returns algorithmically rather than through active trading. The yield rate and risk profile depend on the underlying protocol mechanism generating the return.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:passive-income", "labels": [ "Passive Income" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "passive-remote-sensing", "title": "Passive Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:passive-remote-sensing", "labels": [ "Passive Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "passive-thermal-control", "title": "Passive Thermal Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:passive-thermal-control", "labels": [ "Passive Thermal Control" ], "is_subclass_of": [ "Spacecraft Thermal Control" ], "wikilinks": [] }, { "id": "passkey", "title": "Passkey", "domain": "security", "domain_name": "Security", "definition": "A passkey is a phishing-resistant, passwordless credential built on WebAuthn and FIDO2 public-key cryptography, stored on a device or synced across a user's devices via a platform cloud account. Authentication proceeds by the device signing a server-issued challenge with a private key that never leaves secure hardware, eliminating shared secrets that can be phished, leaked, or reused. Passkeys are backed by biometric or PIN unlock at the device level and are promoted jointly by Apple, Google, and Microsoft as a replacement for passwords. They require a compatible authenticator and relying-party support to function.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:passkey", "labels": [ "Passkey" ], "is_subclass_of": [ "Authentication Mechanism" ], "wikilinks": [] }, { "id": "passthrough-ar", "title": "Passthrough AR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Passthrough AR is a form of augmented reality in which a headset captures the physical environment through external cameras and re-renders that video feed on internal displays, composited with virtual content, rather than allowing direct optical see-through of the real world. It allows headsets originally built for virtual reality to support mixed and augmented reality experiences using the same opaque display hardware, at the cost of added latency and reduced fidelity compared with optical see-through approaches. Passthrough AR is supported by runtime standards such as OpenXR, which expose camera passthrough as a standard feature that applications can request.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:passthrough-ar", "labels": [ "Passthrough AR" ], "is_subclass_of": [ "Augmented Reality" ], "wikilinks": [] }, { "id": "password-authentication", "title": "Password Authentication", "domain": "security", "domain_name": "Security", "definition": "Password authentication is a knowledge-based method of verifying a user's identity by checking a secret string they supply against a stored credential. The stored credential is typically a salted cryptographic hash rather than the plaintext password, so that the secret is never recovered even if the store is compromised. As a single-factor mechanism it is widely deployed but vulnerable to guessing, reuse, phishing, and credential-stuffing attacks, motivating stronger or supplementary methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:password-authentication", "labels": [ "Password Authentication", "Password-Based Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "password-hashing", "title": "Password Hashing", "domain": "security", "domain_name": "Security", "definition": "Password hashing is the practice of transforming user passwords into fixed-length irreversible digests before storage, so that a breach of the credential store does not directly reveal the underlying secrets. Secure schemes deliberately use slow, memory-hard functions combined with a unique per-user salt to defeat precomputation and brute-force attacks. Established algorithms such as Argon2, scrypt, bcrypt and PBKDF2 expose tunable cost parameters that can be raised as hardware improves. Password hashing is a specialised application of key derivation functions oriented towards verifying human-chosen secrets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:password-hashing", "labels": [ "Password Hashing" ], "is_subclass_of": [ "Key Derivation Function" ], "wikilinks": [] }, { "id": "password", "title": "Password", "domain": "security", "domain_name": "Security", "definition": "A memorised secret string presented by a user to prove identity during authentication \u2014 the archetypal 'something you know' factor. Passwords are cheap to deploy but structurally weak: they are guessable, reusable across services, phishable, and exposed en masse by database breaches, which is why modern security practice hashes them with slow salted algorithms, pairs them with additional factors, and increasingly replaces them with cryptographic credentials such as passkeys and hardware security keys.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:password", "labels": [ "Password" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [ "Authentication", "Multi-Factor Authentication", "Cryptographic Key", "Hardware Security Key", "Encryption" ] }, { "id": "passwordless-authentication", "title": "Passwordless Authentication", "domain": "security", "domain_name": "Security", "definition": "Passwordless Authentication encompasses identity verification mechanisms that prove user identity without requiring the user to memorise or enter a shared-secret password. Instead, authentication relies on possession of a hardware token or platform authenticator, biometric characteristics, cryptographic key pairs, or one-time codes delivered through a trusted out-of-band channel. The FIDO2 standard \u2014 comprising the W3C WebAuthn specification and the FIDO Alliance CTAP protocol \u2014 provides the primary open standard for passkey-based passwordless authentication, binding credentials to device hardware and enabling phishing-resistant login flows. By eliminating the shared secret as an authentication factor, passwordless schemes structurally defeat credential-stuffing, password-spray, and phishing attack classes.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:passwordless-authentication", "labels": [ "Passwordless Authentication" ], "is_subclass_of": [ "Authentication Mechanism" ], "wikilinks": [] }, { "id": "patch-embedding", "title": "Patch Embedding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Patch embedding is the mechanism by which a Vision Transformer (ViT) converts a 2D image into a sequence of fixed-size vector representations suitable for processing by a self-attention mechanism. The image is divided into non-overlapping rectangular patches; each patch is flattened and projected to a latent dimension via a learnable linear transformation, yielding a sequence of token embeddings analogous to word embeddings in language models. Positional embeddings are added to encode spatial location, and a classification token is prepended to aggregate global information.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:patch-embedding", "labels": [ "Patch Embedding" ], "is_subclass_of": [ "Machine Learning", "Embedding" ], "wikilinks": [] }, { "id": "patch-management", "title": "Patch Management", "domain": "security", "domain_name": "Security", "definition": "Patch management is the systematic process of acquiring, testing, prioritising, and deploying software updates across an organisation's systems to remediate vulnerabilities and defects. It tracks asset inventory and known vulnerabilities, schedules and stages patches to balance risk against operational disruption, and verifies that fixes are applied. Effective patch management is a core control for reducing the window during which known exploits can be used.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:patch-management", "labels": [ "Patch Management" ], "is_subclass_of": [ "Vulnerability Management" ], "wikilinks": [] }, { "id": "patent", "title": "Patent", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A time-limited exclusive right granted by a sovereign authority to an inventor for a novel, inventive, and industrially applicable invention, in exchange for public disclosure of the invention. In the AI and spatial computing landscape, patents protect novel model architectures, training methods, compression algorithms, and hardware accelerators, with global filings exceeding 3.7 million applications annually as of 2024.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:patent", "labels": [ "Patent", "Patent Law", "Patent Licensing" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "3GPP Standard Development", "5G/6G Patents", "5G Networks", "5G Patent Portfolio", "5G Technology", "5G Technology Patents", "6G Research", "6G Research Patents", "Adaptive Coding Patents", "adaptive communication", "Adaptive Compression", "adaptive image transmission", "Adaptive Systems Patents", "advanced neural techniques", "Adversarial loss", "adversarial optimization", "Adversarial Training Patents", "AI", "AI-Blockchain Integration", "AI encoding" ] }, { "id": "path-planning", "title": "Path Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "The computational process of finding a feasible path for a robot or autonomous agent to move from a start configuration to a goal configuration while avoiding obstacles and satisfying kinematic and dynamic constraints. It determines a sequence of configurations connecting start to goal in the robot's configuration space, and forms the foundational planning layer for navigation, manipulation, and autonomous systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:path-planning", "labels": [ "Path Planning", "Global Path Planning", "PathPlanning", "RB-1016-path-planning", "Shortest Path" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [ "Autonomous Systems", "Autonomous Vehicles", "Environment Model", "Feasibility", "Goal Configuration", "Manipulators", "Mapping", "Mobile Robots", "Obstacles", "Optimality", "Path", "RB-1003-optimal-control", "RB-1007-trajectory-generation", "RB-1017-rrt-algorithm", "RB-1018-dijkstra-algorithm", "RB-1019-obstacle-avoidance", "Start Configuration", "Motion Planning", "Navigation", "RB-1013-localization" ] }, { "id": "path-tracing", "title": "Path Tracing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Path tracing is a physically based rendering algorithm that estimates the rendering equation by stochastically sampling complete light-transport paths from the camera through a scene to light sources. It uses Monte Carlo integration over recursive ray bounces to compute unbiased estimates of global illumination, including soft shadows, indirect lighting, and caustics. Image noise decreases as the square root of the number of samples, making convergence and denoising central practical concerns.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:path-tracing", "labels": [ "Path Tracing" ], "is_subclass_of": [ "Physically Based Rendering" ], "wikilinks": [] }, { "id": "pathfinding-algorithm", "title": "Pathfinding Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A computational algorithm that determines an optimal or near-optimal route between a start and goal node in a graph or spatial environment. Pathfinding algorithms are central to autonomous navigation, game AI, robotics, and metaverse movement systems, with classical examples including Dijkstra's algorithm for shortest paths and A* for heuristic-guided search in spatial graphs.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pathfinding-algorithm", "labels": [ "Pathfinding Algorithm" ], "is_subclass_of": [ "AI Technique", "Algorithm" ], "wikilinks": [ "Algorithm", "Artificial Intelligence" ] }, { "id": "pathfinding", "title": "Pathfinding", "domain": "ai", "domain_name": "Ai", "definition": "Pathfinding is the computation of a route between two points in a graph or space, often the shortest or lowest-cost route. It is widely used in games, robotics, and navigation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:pathfinding", "labels": [ "Pathfinding", "Pathfinding System" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [ "Data Structure", "Navigation", "Pathfinding Algorithm", "Search Algorithm", "https://en.wikipedia.org/wiki/Pathfinding", "https://www.redblobgames.com/pathfinding/a-star/introduction.html" ] }, { "id": "pathology-ai", "title": "Pathology AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial intelligence systems for automated analysis of histopathology slides, cytology specimens, and other pathological images. Pathology AI performs tasks including cancer detection and grading, biomarker quantification, and morphological analysis using whole-slide imaging pipelines and deep learning architectures tailored for gigapixel images at multi-scale resolution.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:pathology-ai", "labels": [ "Pathology AI" ], "is_subclass_of": [ "AI Application", "Medical Imaging AI" ], "wikilinks": [ "Computer Vision", "Medical Imaging AI", "MetaverseDomain", "Radiology AI" ] }, { "id": "patient-risk-stratification", "title": "Patient Risk Stratification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Patient risk stratification is the process of classifying patients into groups by their likelihood of experiencing a particular clinical outcome, such as hospital readmission, disease progression, or adverse drug reaction, using clinical, demographic, and increasingly machine-learned features. It allows clinicians and health systems to prioritise limited resources, such as closer monitoring or preventive intervention, toward the patients most likely to benefit. Modern risk stratification tools combine electronic health record data with predictive models to produce continuously updated risk scores rather than static categorical assessments.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:patient-risk-stratification", "labels": [ "Patient Risk Stratification" ], "is_subclass_of": [ "Healthcare AI" ], "wikilinks": [] }, { "id": "pattern-matching", "title": "Pattern Matching", "domain": "ai", "domain_name": "Ai", "definition": "Pattern matching is the act of checking a sequence or structure against a pattern to detect the presence of constituent components, and, in programming languages, of binding variables to parts of the matched structure. In symbolic AI and computation it underpins rule firing, unification, and term rewriting; in software it provides a declarative way to deconstruct data by shape. It contrasts with statistical pattern recognition by operating on exact, often symbolic, structure rather than learned probabilistic similarity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:pattern-matching", "labels": [ "Pattern Matching" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique" ], "wikilinks": [] }, { "id": "pattern-recognition", "title": "Pattern Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pattern Recognition is the automated identification of regularities, structures, and categories in data using machine learning algorithms, encompassing supervised classification, unsupervised clustering, and feature extraction techniques applied across vision, speech, biometrics, and anomaly detection domains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:pattern-recognition", "labels": [ "Pattern Recognition" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Classification", "Computer Vision", "Deep Learning", "Feature Extraction" ] }, { "id": "paxos", "title": "Paxos", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Paxos is a family of fault-tolerant distributed consensus algorithms first formally described by Leslie Lamport in 1989 and published in 1998, designed to allow a cluster of processes to agree on a single value or sequence of values despite the failure of a minority of participants. The algorithm proceeds through prepare and accept phases orchestrated by a proposer, with acceptors voting to commit proposed values and learners observing the final agreement. Multi-Paxos extends the basic protocol to achieve consensus on a log of commands efficiently, forming the algorithmic foundation of replicated state machines. Paxos and its derivatives, including Raft and HotStuff, underpin virtually every strongly consistent distributed database and coordination service in production use today.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:paxos", "labels": [ "Paxos", "Paxos Consensus", "Paxos Protocol" ], "is_subclass_of": [ "Consensus Algorithm" ], "wikilinks": [] }, { "id": "pay-to-public-key-hash", "title": "Pay To Public Key Hash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Pay-to-Public-Key-Hash (P2PKH) is the classic Bitcoin transaction output script that locks funds to the hash of a recipient's public key rather than to the key itself. Spending requires the spender to supply both the matching public key, which is hashed and compared to the committed hash, and a valid signature over the transaction, providing privacy until spend time and shorter addresses. It is expressed in Bitcoin Script as a fixed opcode sequence and contrasts with pay-to-script-hash and witness output types.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pay-to-public-key-hash", "labels": [ "Pay To Public Key Hash", "Pay-To-Public-Key-Hash", "Pay-to-Public-Key-Hash" ], "is_subclass_of": [ "Bitcoin Script" ], "wikilinks": [] }, { "id": "pay-per-request", "title": "Pay-Per-Request", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Pay-per-request is a pricing and access-control model in which each individual API call or resource request is metered and paid for independently, rather than through a subscription or bulk allocation. It is enabled by protocols such as L402, which gate HTTP responses behind a Lightning Network micropayment, making per-query settlement economically viable for machine-to-machine transactions. It is particularly relevant for AI services that need frictionless, automatable payment for compute-intensive queries.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:pay-per-request", "labels": [ "Pay-Per-Request" ], "is_subclass_of": [ "L402" ], "wikilinks": [] }, { "id": "pay-to-script-hash", "title": "Pay-to-Script-Hash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Pay-to-Script-Hash (P2SH) is a Bitcoin transaction type, introduced in BIP 16, that locks funds to the hash of a redeem script rather than to a public key or full script. The spender must later provide both the original redeem script, whose hash matches the locking output, and the data that satisfies it. P2SH shifts the burden and storage cost of complex spending conditions from the sender to the recipient and underpins multi-signature and other advanced locking schemes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pay-to-script-hash", "labels": [ "Pay-to-Script-Hash", "Pay To Script Hash" ], "is_subclass_of": [ "Bitcoin Script" ], "wikilinks": [] }, { "id": "pay-pal", "title": "PayPal", "domain": "finance", "domain_name": "Finance", "definition": "PayPal is a multinational financial technology company that operates one of the world's largest online payment platforms, enabling consumers and merchants to send, receive, and hold funds across more than 200 markets. Founded in 1998 and spun off from eBay in 2015, it provides digital wallets, payment processing infrastructure, buy-now-pay-later credit products, and cryptocurrency custody services. PayPal's two-sided network connects hundreds of millions of consumer accounts to tens of millions of merchant integrations, making it a foundational layer of global e-commerce payments. In 2023 the company launched PayPal USD (PYUSD), a regulated US dollar stablecoin issued on public blockchains, extending its reach into programmable money and digital asset settlement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:pay-pal", "labels": [ "PayPal" ], "is_subclass_of": [ "Payment Network" ], "wikilinks": [ "Payment Network", "Digital Asset", "Stablecoin" ] }, { "id": "payload-fairing", "title": "Payload Fairing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:payload-fairing", "labels": [ "Payload Fairing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "payment-channel-network", "title": "Payment Channel Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Payment Channel Network is a layer-2 scaling architecture in which participants open bilateral payment channels and route payments across a graph of connected channels without settling each transaction on the underlying blockchain. Funds are locked in multisignature channels, and balances update off-chain through signed state, with only channel opening and closing recorded on-chain. This enables high-throughput, low-fee, near-instant payments while inheriting base-layer security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:payment-channel-network", "labels": [ "Payment Channel Network" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "payment-channel", "title": "Payment Channel", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A payment channel is a cryptographic construct that allows two or more parties to conduct multiple off-chain transactions by exchanging signed commitment messages, with only the channel-opening and channel-closing states recorded on a base-layer blockchain. Channels are secured by multisignature scripts, time-locked contracts, and mutual revocation mechanisms that ensure neither party can unilaterally broadcast a superseded state without penalty. By batching an unbounded number of value transfers into two on-chain transactions, payment channels achieve high throughput and negligible per-payment fees while inheriting the settlement finality and trustlessness of the underlying blockchain. They form the foundational primitive of routed payment networks such as the Lightning Network and of generalised state-channel protocols.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:payment-channel", "labels": [ "Payment Channel", "Payment Channel Protocol", "payment-channel" ], "is_subclass_of": [ "Off-Chain Scaling" ], "wikilinks": [] }, { "id": "payment-channels", "title": "Payment Channels", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Payment channels are off-chain constructs that let two parties exchange many bitcoin transactions while only recording opening and closing balances on the blockchain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:payment-channels", "labels": [ "Payment Channels" ], "is_subclass_of": [ "Payment Channel" ], "wikilinks": [ "Bitcoin", "Smart Contract", "Lightning Network", "Payment Channel" ] }, { "id": "payment-gateway", "title": "Payment Gateway", "domain": "finance", "domain_name": "Finance", "definition": "A Payment Gateway is a technology service that authorises, processes, and routes financial transactions between buyers and merchants, acting as the intermediary between a merchant's point-of-sale or e-commerce system and the acquiring bank or payment network. It encrypts sensitive payment credentials, communicates with card networks and issuing banks to obtain authorisation, and returns the result to the merchant in real time. Modern payment gateways provide APIs, SDKs, fraud detection, currency conversion, and compliance tooling, and increasingly support cryptocurrency payments and programmable settlement rails alongside traditional card and bank transfer methods. Payment gateways are foundational infrastructure for e-commerce and digital commerce ecosystems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:payment-gateway", "labels": [ "Payment Gateway" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [] }, { "id": "payment-infrastructure", "title": "Payment Infrastructure", "domain": "finance", "domain_name": "Finance", "definition": "Payment infrastructure is the aggregate of networks, protocols, clearing houses, settlement systems, and regulatory frameworks that enable the transfer of monetary value between parties. It spans card networks, interbank messaging (SWIFT, ISO 20022), real-time gross settlement (RTGS) systems, and emerging digital-asset rails including stablecoins and central bank digital currencies. The infrastructure defines latency, finality, cost, and access characteristics that shape the global economy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:payment-infrastructure", "labels": [ "Payment Infrastructure", "Global Payment Infrastructure" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [] }, { "id": "payment-network", "title": "Payment Network", "domain": "finance", "domain_name": "Finance", "definition": "A payment network is an infrastructure system of interconnected nodes, protocols, messaging standards, and settlement mechanisms that enables the authenticated transfer of value between participants \u2014 individuals, merchants, financial institutions, or autonomous software agents \u2014 with guarantees on finality, atomicity, and fraud prevention. Payment networks operate at multiple layers: a messaging layer (e.g. SWIFT MT/MX, ISO 20022) carries authenticated payment instructions; a clearing layer nets obligations across participants; and a settlement layer achieves irrevocable finality through central bank reserves or distributed ledger consensus. The architecture spans traditional four-party card networks (Visa, Mastercard), interbank clearing systems (SWIFT, ACH, SEPA, Fedwire), cryptographic peer-to-peer blockchain networks, and second-layer off-chain channel networks such as the Lightning Network, collectively underpinning commerce, financial inclusion, and machine-to-machine value exchange.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:payment-network", "labels": [ "Payment Network", "Bitcoin Payment Network", "Payment Networks" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [] }, { "id": "payment-preimage", "title": "Payment Preimage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A payment preimage is the secret value whose cryptographic hash forms a payment hash used to lock funds in a Hash Time-Locked Contract on the Lightning Network. Revealing the preimage proves that the intended recipient received the payment and simultaneously releases the locked funds along every hop of the payment route. It is the mechanism that lets protocols such as BOLT11 invoices and L402 guarantee atomic, trustless settlement across multi-hop payment channels.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:payment-preimage", "labels": [ "Payment Preimage" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [] }, { "id": "payment-processing", "title": "Payment Processing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Payment processing is the sequence of operations that authorise, capture, clear and settle a monetary transaction between a payer and a payee. It coordinates merchants, payment gateways, networks and settlement institutions to verify funds, manage risk and move value reliably. Modern payment processing spans card rails, digital wallets and blockchain-based settlement, increasingly emphasising speed, fraud detection and interoperability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:payment-processing", "labels": [ "Payment Processing" ], "is_subclass_of": [ "Payment Network" ], "wikilinks": [] }, { "id": "payment-processor", "title": "Payment Processor", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A payment processor is a service that handles the transaction lifecycle between a merchant, the customer's payment instrument, and the settlement networks. It authorises, captures, clears, and settles funds while applying fraud checks, currency conversion, and compliance controls. Processors are foundational components of digital payment infrastructure, abstracting card networks, bank rails, and blockchain settlement behind a single integration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:payment-processor", "labels": [ "Payment Processor" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "payment-protocol", "title": "Payment Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A payment protocol is a defined set of rules and message formats governing how value transfer is initiated, authorised and settled between parties.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:payment-protocol", "labels": [ "Payment Protocol" ], "is_subclass_of": [ "Communication Protocols" ], "wikilinks": [ "Cryptographic Protocols", "Stablecoin", "Cryptocurrency", "Communication Protocols", "https://en.wikipedia.org/wiki/Payment_system", "https://www.bis.org/cpmi/" ] }, { "id": "payment-routing", "title": "Payment Routing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Payment routing is the process of determining a viable path along which value moves from a payer to a payee across one or more intermediaries, payment channels or networks. In layered payment systems such as the Lightning Network it involves finding a sequence of hops with sufficient liquidity and acceptable fees, whereas in conventional rails it selects acquirers, schemes or correspondent banks. Routing decisions balance cost, success probability, latency and privacy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:payment-routing", "labels": [ "Payment Routing" ], "is_subclass_of": [ "Payment Processing" ], "wikilinks": [] }, { "id": "payment-settlement", "title": "Payment Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Payment settlement is the process by which a payment obligation between parties is irrevocably discharged through the transfer of value, completing a transaction so that the recipient gains final, unconditional ownership of funds. Settlement may occur gross or net, in real time or in batches, and across traditional rails or blockchain ledgers, with finality being the property that the transfer can no longer be reversed. In distributed-ledger contexts, settlement is achieved when an on-chain transaction reaches consensus finality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:payment-settlement", "labels": [ "Payment Settlement" ], "is_subclass_of": [ "Settlement" ], "wikilinks": [] }, { "id": "payment-system", "title": "Payment System", "domain": "finance", "domain_name": "Finance", "definition": "A Payment System is an organised set of instruments, procedures, rules, and interbank funds-transfer networks that enables the exchange of monetary value between buyers and sellers, encompassing both traditional fiat rails (card networks, bank transfers, ACH) and digital-native mechanisms such as blockchain token transfers, smart-contract escrow, and central bank digital currencies. Payment systems define how obligations are cleared and settled \u2014 gross or net, in real time or deferred \u2014 while satisfying requirements for security, finality, liquidity efficiency, AML/KYC compliance, and cross-border interoperability. Modern payment infrastructure increasingly operates across layers: a base settlement layer (central bank reserves or a public blockchain), a clearing and messaging layer (SWIFT, ISO 20022, or a decentralised protocol), and an application layer (wallets, point-of-sale terminals, or embedded payment APIs) that exposes value exchange to end users and automated agents.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:payment-system", "labels": [ "Payment System", "National Payment System", "Payment Systems", "Traditional Payment System", "Wholesale Payment System" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [] }, { "id": "payment-token", "title": "Payment Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Payment Token is a cryptocurrency token whose primary purpose is to serve as a medium of exchange and store of value for peer-to-peer and merchant payments, distinguished from utility tokens (access rights) and security tokens (equity/asset claims). Examples include Bitcoin, Litecoin, and CBDC-backed digital currencies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:payment-token", "labels": [ "Payment Token" ], "is_subclass_of": [ "Token and Asset", "Cryptocurrency Token" ], "wikilinks": [ "Blockchain", "Cryptocurrency Token" ] }, { "id": "payments-infrastructure", "title": "Payments Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Payments infrastructure is the aggregate of networks, rails, processors, and settlement systems that move value between parties. It spans card schemes, interbank transfer rails, clearing houses, and blockchain-based settlement layers that enable assets such as stablecoins to circulate. As a backbone system it determines the speed, cost, reach, and finality of monetary transactions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:payments-infrastructure", "labels": [ "Payments Infrastructure" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "pedersen-commitment", "title": "Pedersen Commitment", "domain": "security", "domain_name": "Security", "definition": "A Pedersen commitment is a cryptographic commitment scheme in which a committer binds to a secret value v by computing C = g^v * h^r, where g and h are independent group generators and r is a random blinding factor. The scheme is computationally binding under the discrete logarithm assumption and unconditionally (information-theoretically) hiding, meaning an adversary with unlimited computation cannot determine the committed value from C alone. Crucially, Pedersen commitments are additively homomorphic: the product of two commitments C(v1, r1) * C(v2, r2) equals C(v1+v2, r1+r2), enabling arithmetic on committed values without revealing them. This property makes Pedersen commitments foundational to confidential transactions, range proofs, and zero-knowledge proof systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:pedersen-commitment", "labels": [ "Pedersen Commitment", "Pedersen Commitment Scheme" ], "is_subclass_of": [ "Cryptographic Commitment" ], "wikilinks": [ "Elliptic Curve Cryptography", "Zero-Knowledge Proof", "Cryptographic Commitment" ] }, { "id": "peer-discovery", "title": "Peer Discovery", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocol mechanism by which blockchain nodes locate and connect to other network participants without central coordination, using bootstrap nodes, DNS seeds, gossip-based address propagation, and distributed hash table lookups to establish and maintain a resilient peer-to-peer overlay network.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:peer-discovery", "labels": [ "Peer Discovery", "Decentralised Peer Discovery", "Peer Discovery Protocol" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "peer-learning", "title": "Peer Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Peer learning refers to settings where multiple agents improve by exchanging information or imitating each other rather than relying on a central teacher. In multi-agent systems it covers cooperative protocols for sharing experience.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:peer-learning", "labels": [ "Peer Learning" ], "is_subclass_of": [ "Collaborative Learning" ], "wikilinks": [ "Collaborative Learning", "Multi-Agent Coordination", "Knowledge Sharing", "Multi-Agent Systems", "Reinforcement Learning" ] }, { "id": "peer-review", "title": "Peer Review", "domain": "finance", "domain_name": "Finance", "definition": "Peer review is the evaluation of work, proposals or decisions by qualified individuals of comparable competence, used to assess validity, quality and significance before acceptance or publication. In scholarship it gates journal and conference publication; in finance, governance and standards it underpins independent assessment, audit and consensus-based approval. It is a quality-assurance process intended to surface errors, reduce bias and confer credibility through independent scrutiny.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:peer-review", "labels": [ "Peer Review" ], "is_subclass_of": [ "Quality Assurance" ], "wikilinks": [] }, { "id": "peer-to-peer-payment", "title": "Peer To Peer Payment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Peer-to-peer payment is the direct transfer of monetary value between two parties without a traditional intermediary settling each side of the transaction. In cryptocurrency systems it is realised by signing a transaction that reassigns ownership of digital tokens on a shared ledger, validated by network consensus rather than a bank. The model reduces reliance on centralised clearing and enables programmable, near-instant value exchange.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:peer-to-peer-payment", "labels": [ "Peer To Peer Payment", "Peer-to-Peer Payment" ], "is_subclass_of": [ "Blockchain", "Payment System" ], "wikilinks": [] }, { "id": "peer-to-peer-energy-trading", "title": "Peer-to-Peer Energy Trading", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Peer-to-peer energy trading is a model in which prosumers, households or businesses that generate their own renewable energy, buy and sell surplus electricity directly with one another over a local distribution network, without a centralised utility as intermediary. Transactions are typically coordinated through smart grid infrastructure and settled using blockchain-based platforms that record generation, consumption, and payment automatically. Peer-to-peer energy trading is closely linked to renewable energy certificates, which provide verifiable proof of the source and attributes of the traded energy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:peer-to-peer-energy-trading", "labels": [ "Peer-to-Peer Energy Trading", "PeerToPeerEnergyTrading" ], "is_subclass_of": [ "Smart Grid" ], "wikilinks": [] }, { "id": "peer-to-peer-network", "title": "Peer-to-Peer Network", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A decentralised communication architecture in which participating nodes connect directly to one another without a central coordinator, enabling distributed ledger technology to broadcast transactions, propagate blocks, and maintain a shared state across an open membership set. It underlies the censorship resistance and fault tolerance of blockchain systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:peer-to-peer-network", "labels": [ "Peer-to-Peer Network", "Peer To Peer Network", "Peer to Peer Network", "Peer-to-Peer Computing", "Peer-to-Peer Networking", "Trustless Peer Communication" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "peer-to-peer-trading", "title": "Peer-to-Peer Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Peer-to-peer trading is the direct exchange of assets between two parties without a centralised intermediary holding or matching the order. On blockchains it is typically mediated by smart contracts that escrow assets and enforce settlement atomically, removing custodial risk. It enables markets for tokens, NFTs, and goods to operate trustlessly through code rather than a central exchange.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:peer-to-peer-trading", "labels": [ "Peer-to-Peer Trading", "Peer-to-Peer Lending" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "peerto-peer-protocol", "title": "Peerto Peer Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A peer-to-peer protocol defines the rules by which nodes in a decentralised network discover one another, exchange messages and share data directly, without central intermediaries. In blockchain systems such protocols underpin peer discovery, transaction relay and block propagation, providing the communication layer over which consensus is reached.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:peerto-peer-protocol", "labels": [ "Peerto Peer Protocol", "Peer-to-Peer Protocol" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [ "owl:Thing" ] }, { "id": "peg-mechanism", "title": "Peg Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A peg mechanism is the set of economic and technical arrangements by which a token maintains a stable exchange rate against a reference asset such as a fiat currency or another token. It typically combines collateral backing, minting and redemption rules, and arbitrage incentives so that market deviations from the target price are pushed back towards parity. The design of the peg mechanism determines how robustly a stablecoin or wrapped asset holds its value under stress.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:peg-mechanism", "labels": [ "Peg Mechanism" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [] }, { "id": "peg", "title": "Peg", "domain": "finance", "domain_name": "Finance", "definition": "A peg is a commitment to hold the price of one asset fixed against another \u2014 historically a national currency fixed to gold or to the US dollar, and in decentralised finance a stablecoin held at parity with a fiat currency. Maintaining a peg requires reserves, redemption rights, or algorithmic supply adjustment strong enough to absorb market pressure and sustain arbitrage back to parity; when confidence in these defences fails the peg breaks, as in classic currency crises and stablecoin de-pegging events such as the 2022 TerraUSD collapse.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:peg", "labels": [ "Peg" ], "is_subclass_of": [ "Exchange Rate" ], "wikilinks": [ "Exchange Rate", "Peg Mechanism", "Stablecoin", "Redemption Mechanism", "Monetary Policy" ] }, { "id": "pegasus-spyware", "title": "Pegasus Spyware", "domain": "security", "domain_name": "Security", "definition": "Pegasus is commercial surveillance spyware developed by the NSO Group that covertly compromises smartphones, often via zero-click exploits requiring no user interaction. Once installed it can exfiltrate messages, calls, location, microphone, and camera data, granting near-total device control. It is a prominent case study in state-level digital surveillance and the human-rights risks of the commercial intrusion-software market.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pegasus-spyware", "labels": [ "Pegasus Spyware" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "pendle", "title": "Pendle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Pendle is a decentralised finance protocol that tokenises future yield, splitting yield-bearing assets into separate principal and yield tokens that can be traded.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:pendle", "labels": [ "Pendle", "Pendle Finance" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Smart Contract", "Yield Farming", "Tokenisation", "DeFi" ] }, { "id": "penetration-testing", "title": "Penetration Testing", "domain": "security", "domain_name": "Security", "definition": "Penetration testing is an authorised, structured simulation of real-world adversarial attacks against computing systems, networks, or applications, conducted to identify exploitable vulnerabilities before malicious actors can leverage them. Skilled practitioners \u2014 called penetration testers or ethical hackers \u2014 follow an agreed scope and rules of engagement, systematically applying the same techniques, tools, and thought processes used by genuine threat actors. Findings are documented with severity ratings, proof-of-concept evidence, and remediation guidance, directly informing an organisation's risk management and security improvement programme. The discipline bridges theoretical vulnerability knowledge with practical risk quantification, distinguishing it from automated scanning or passive audit activities.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:penetration-testing", "labels": [ "Penetration Testing", "Penetration Testing Tools" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "Cybersecurity", "Information Security", "Software Testing" ] }, { "id": "perception-layer", "title": "Perception Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Perception Layer is the stratum that converts raw sensor signals into structured observations of the environment. It sits above the Sensor Fusion and Hardware sensing strata and below decision-making layers, providing the interpreted percepts that agents and controllers act on. It contains detection, recognition, segmentation, and state-estimation components.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:perception-layer", "labels": [ "Perception Layer" ], "is_subclass_of": [ "Perception and Sensing", "owl:Thing" ], "wikilinks": [ "Sensor Fusion Layer", "Control Layer", "Agent Layer", "Computer Vision", "State Estimation", "owl:Thing" ] }, { "id": "perception-module", "title": "Perception Module", "domain": "ai", "domain_name": "Ai", "definition": "A perception module is the component of an autonomous or cognitive agent that transforms raw sensory input into structured representations of the environment. It performs sensing, filtering, feature extraction, object detection, and state estimation to produce a world model usable by planning and control. As the agent's interface to reality, its accuracy bounds the quality of all downstream decisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:perception-module", "labels": [ "Perception Module" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "perception-system", "title": "Perception System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Perception System is the sensor processing and environmental understanding component of Autonomous Systems that interprets raw Sensor Data to build a coherent representation of the surrounding environment, including Object Detection, Classification, Tracking, Localization, and Scene Understanding. Perception systems fuse data from multiple Sensor Modalities (Camera, LiDAR, Radar, Ultrasonic Sensors) to create robust environmental models for Autonomous Decision-Making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:perception-system", "labels": [ "Perception System", "Robotic Perception System" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "3D Vision", "4D Perception", "4D Radar", "Adverse Weather", "Agricultural Robots", "AI", "Amazon Robotics", "Apollo Auto", "Attention Mechanisms", "Automotive Edge Computing Consortium", "Autonomous Decision-Making", "Autonomous Mobile Robots", "Autonomous Mobility", "Autonomous Systems", "Autonomous Vehicles", "Autoware", "Bayesian Deep Learning", "Boston Dynamics Spot", "CARLA", "Carmaker" ] }, { "id": "perception", "title": "Perception", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Perception is the computational process by which an intelligent system acquires, processes, and interprets sensory signals \u2014 such as visual, auditory, tactile, or LiDAR data \u2014 to construct an internal, structured representation of the external world. It forms the foundational input stage of the sense\u2013plan\u2013act loop that underpins autonomous agents and robotics, translating high-dimensional raw sensor streams into semantically meaningful features, objects, or scene graphs. Modern AI perception leverages deep neural architectures \u2014 including convolutional networks, vision transformers, and multimodal encoders \u2014 to achieve robust generalisation across varied environments. It is distinct from raw data acquisition (sensing) and from higher-order cognitive reasoning, occupying the middle layer that makes physical-world understanding tractable for downstream decision systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:perception", "labels": [ "Perception", "Depth Perception" ], "is_subclass_of": [ "Cognitive Architecture", "Embodied AI" ], "wikilinks": [ "Sensor", "Robotics", "Computer Vision", "Artificial Intelligence" ] }, { "id": "perceptual-experience", "title": "Perceptual Experience", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The multimodal sensory and cognitive engagement a user undergoes within a telepresence or immersive environment, encompassing visual fidelity, spatial audio, haptic feedback, and sense of presence. Perceptual experience quality determines the degree to which a distributed or virtual interaction is felt as equivalent to physical co-presence, directly impacting collaboration effectiveness and user well-being.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:perceptual-experience", "labels": [ "Perceptual Experience" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "owl:Thing" ] }, { "id": "performance-benchmarks", "title": "Performance Benchmarks", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Performance benchmarks are standardised, reproducible test suites and associated metric sets used to measure, compare, and rank the behavioural characteristics of software systems, hardware platforms, algorithms, or AI models under controlled or representative workload conditions. They quantify dimensions such as latency, throughput, resource utilisation, scalability, accuracy, and energy efficiency, enabling objective evaluation across vendors, versions, and deployment environments. Benchmark suites range from micro-benchmarks targeting isolated components to macro-benchmarks simulating realistic end-to-end workloads. Formal benchmark governance \u2014 through bodies such as SPEC, MLCommons, and TPC \u2014 establishes methodology rules, disclosure requirements, and result auditing to prevent benchmark gaming.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:performance-benchmarks", "labels": [ "Performance Benchmarks", "Performance Benchmarking" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [ "metaverse", "Performance Optimization", "Software Testing" ] }, { "id": "performance-capture", "title": "Performance Capture", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Performance capture records an actor's full body, facial and sometimes finger movement simultaneously so that a single performance drives a digital character's motion and expression.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:performance-capture", "labels": [ "Performance Capture" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Motion Capture", "Avatar", "Skeletal Animation", "Volumetric Video", "Photogrammetry", "Computer Graphics" ] }, { "id": "performance-marketing", "title": "Performance Marketing", "domain": "economics", "domain_name": "Economics", "definition": "Performance marketing is a digital advertising strategy where advertisers pay for specific, measurable actions such as clicks, leads, or completed transactions rather than for impressions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:performance-marketing", "labels": [ "Performance Marketing" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "performance-metrics", "title": "Performance Metrics", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Performance Metrics are quantitative and qualitative measurements used to assess the efficiency, correctness, and resource utilisation of AI models, software systems, distributed platforms, and hardware pipelines. They encompass latency, throughput, accuracy, recall, precision, error rate, memory footprint, and energy consumption, forming the empirical basis for benchmarking, capacity planning, and continuous improvement. In machine learning contexts, performance metrics bridge offline evaluation (held-out test sets) and online monitoring (production dashboards), enabling data-driven decisions about model retraining, architecture changes, and deployment rollback. Standardised metric suites underpin regulatory compliance, SLA enforcement, and comparative research across the field.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:performance-metrics", "labels": [ "Performance Metrics", "Performance Metric" ], "is_subclass_of": [ "Model Evaluation", "AI Technique" ], "wikilinks": [] }, { "id": "performance-monitoring", "title": "Performance Monitoring", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Performance Monitoring is the continuous collection, analysis, and visualisation of metrics describing how a system, application, or model behaves under real workloads. It tracks indicators such as latency, throughput, error rates, resource utilisation, and, for machine-learning systems, predictive quality and drift, surfacing degradation through dashboards and alerts. Performance monitoring is a core observability discipline that enables teams to detect regressions, diagnose bottlenecks, and uphold service-level objectives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:performance-monitoring", "labels": [ "Performance Monitoring" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "performance-optimization", "title": "Performance Optimization", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Performance optimization is the systematic process of enhancing the efficiency, speed, and effectiveness of software systems by tuning code, algorithms, and resource utilization to minimize response time and maximize throughput. It encompasses profiling to identify bottlenecks, algorithmic improvements, caching strategies, parallel processing, and compiler optimizations to meet defined performance targets.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:performance-optimization", "labels": [ "Performance Optimization" ], "is_subclass_of": [ "Optimization Technique" ], "wikilinks": [ "Resource Efficiency", "System Scalability", "User Experience", "ComputationAndIntelligenceDomain", "Technology Domain" ] }, { "id": "performing-arts", "title": "Performing Arts", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Time-based artistic disciplines\u2014including theatre, dance, music, and live performance\u2014that occur in shared space between performers and audience. In spatial computing contexts, performing arts intersect with virtual production, motion capture, and location-based experiences to create hybrid physical-digital performances that extend audience reach and deepen narrative immersion.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:performing-arts", "labels": [ "Performing Arts" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "periapsis", "title": "Periapsis", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:periapsis", "labels": [ "Periapsis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "perigee", "title": "Perigee", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:perigee", "labels": [ "Perigee" ], "is_subclass_of": [ "Periapsis" ], "wikilinks": [] }, { "id": "perimeter-security", "title": "Perimeter Security", "domain": "security", "domain_name": "Security", "definition": "Perimeter security is a network defence model that establishes a fortified boundary between a trusted internal network and untrusted external networks, concentrating controls such as firewalls and gateways at that boundary. It treats the network edge as the principal line of defence, inspecting and filtering traffic crossing into or out of the protected zone. The approach assumes that hosts inside the perimeter are comparatively trustworthy, an assumption that modern architectures increasingly challenge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:perimeter-security", "labels": [ "Perimeter Security" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "permissioned-blockchain", "title": "Permissioned Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A restricted-access distributed ledger controlled by known participants using predefined permissions and pluggable consensus mechanisms, enabling organisations to leverage blockchain technology's immutability, transparency, and smart contract capabilities whilst maintaining regulatory compliance and business confidentiality. Access control is enforced at multiple layers\u2014node permissioning restricting which organisations participate, account permissioning controlling transaction submission, and private channels enabling confidential subsets of participants to conduct transactions invisibly to others.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:permissioned-blockchain", "labels": [ "Permissioned Blockchain", "BC-0029-permissioned-blockchain", "BC-0429-permissioned-blockchain", "Permissioned Blockchain Network", "Permissioned Ledger", "PermissionedBlockchain" ], "is_subclass_of": [ "Distributed Ledger" ], "wikilinks": [ "BC-0001-blockchain", "BC-0120-consensus-mechanism", "BC-0245-proof-of-authority", "BC-0426-hyperledger-fabric", "BC-0427-hyperledger-besu", "BC-0428-enterprise-blockchain-architecture", "BC-0430-private-channels", "BlockchainDomain", "Consensus Mechanism", "Hyperledger Fabric", "HyperledgerFabric", "PrivateChannels", "Quorum Blockchain", "QuorumBlockchain", "R3 Corda", "R3Corda", "SmartContract" ] }, { "id": "permissioned-network", "title": "Permissioned Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Permissioned Network is a distributed ledger architecture in which participation\u2014whether as a validator, transaction submitter, or read-only observer\u2014is restricted to entities that have been explicitly authorised by a governing body or membership protocol. Unlike public blockchains, nodes must satisfy identity verification, legal agreement, or technical credentialing requirements before joining, enabling stronger privacy guarantees, higher throughput, and deterministic finality than open networks while sacrificing censorship resistance. Permissioned networks are the dominant choice for enterprise and consortium deployments such as trade finance, healthcare data exchange, and central-bank digital currency infrastructure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:permissioned-network", "labels": [ "Permissioned Network" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "permissionless-innovation", "title": "Permissionless Innovation", "domain": "governance", "domain_name": "Governance", "definition": "A policy stance holding that new technologies and business models should generally be allowed to develop without prior approval, with regulation applied only to address demonstrated harms. It is often discussed in the context of the internet and digital finance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:permissionless-innovation", "labels": [ "Permissionless Innovation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Decentralization", "Web3", "Governance" ] }, { "id": "permissionless-lending", "title": "Permissionless Lending", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissionless lending is the provision of on-chain credit through open protocols that anyone may access without identity gating or intermediary approval. Lenders deposit assets into smart-contract liquidity pools and borrowers draw against over-collateralised positions, with interest rates set algorithmically by utilisation. It is a foundational DeFi primitive that replaces credit underwriting with collateralisation and code-enforced liquidation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:permissionless-lending", "labels": [ "Permissionless Lending" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "permissionless-network", "title": "Permissionless Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Permissionless Network is a blockchain or distributed ledger system in which any entity may join as a node, submit transactions, and participate in consensus without prior authorisation or identity verification from a central authority. Permissionlessness is a foundational design property of public blockchains such as Bitcoin and Ethereum, enabling censorship resistance and global open access at the cost of requiring Sybil-resistant consensus mechanisms\u2014typically proof-of-work or proof-of-stake\u2014to prevent anonymous actors from subverting the network.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:permissionless-network", "labels": [ "Permissionless Network" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "Network Component" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "permissionless-participation", "title": "Permissionless Participation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissionless participation is the property of a network whereby any actor may join, transact, validate, or contribute without requesting approval from a central authority. It is a defining characteristic of public blockchains, where open access to running nodes and proposing or validating blocks underpins censorship resistance and decentralisation. The property is enforced by economic and cryptographic mechanisms rather than identity vetting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:permissionless-participation", "labels": [ "Permissionless Participation", "Permissionless Compute Contribution", "Permissionless Membership" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "permissionless-trading", "title": "Permissionless Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Permissionless trading refers to the ability to execute financial trades \u2014 buying, selling, or swapping digital assets \u2014 without requiring authorisation from a centralised intermediary, gatekeeper, or identity provider. Enabled by on-chain smart contracts and automated market makers (AMMs), permissionless trading removes the account approval, KYC, and whitelist requirements of traditional financial markets, allowing any wallet address to interact directly with liquidity pools. It is a foundational property of decentralised finance (DeFi) protocols.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:permissionless-trading", "labels": [ "Permissionless Trading" ], "is_subclass_of": [ "Decentralized Finance (DeFi)" ], "wikilinks": [] }, { "id": "perpetual-futures", "title": "Perpetual Futures", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Perpetual futures are derivative contracts that track an underlying asset's price without an expiry or settlement date. A periodic funding-rate payment between long and short holders tethers the contract price to the spot index, replacing the convergence that expiry provides in traditional futures. They are a dominant instrument in crypto derivatives markets, enabling leveraged directional exposure that can be held indefinitely.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:perpetual-futures", "labels": [ "Perpetual Futures" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "perplexity-ai-search-platform", "title": "Perplexity AI Search Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Perplexity AI is a conversational search platform that combines large language models with live web retrieval to produce cited, synthesised answers to natural-language queries. Offered through consumer applications and an API, it applies retrieval-augmented generation to web-scale content, sitting between traditional search engines and generative AI assistants.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:perplexity-ai-search-platform", "labels": [ "Perplexity AI Search Platform", "perplexity" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "persistence-layer", "title": "Persistence Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A persistence layer is the architectural tier responsible for durably storing and retrieving application state across sessions and process restarts. It abstracts the underlying storage technology, databases, object stores, or distributed logs, behind a uniform interface for reading and writing data. In agent and spatial-computing systems it preserves memory, anchors, and context so that state survives beyond a single runtime.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:persistence-layer", "labels": [ "Persistence Layer" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "persistence", "title": "Persistence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The foundational architectural mechanism ensuring data, state, and identity continuity across sessions, platforms, and time within metaverse ecosystems through distributed databases, blockchain ledgers, and file systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:persistence", "labels": [ "Persistence", "Persistence Service", "Polyglot Persistence" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Decentraland", "Microsoft Mesh", "MicrosoftMesh", "The Sandbox", "TheSandbox", "Metaverse", "MetaverseDomain" ] }, { "id": "persistent-ar-placement", "title": "Persistent AR Placement", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The capability to anchor augmented reality content at specific real-world locations that persists across sessions and devices, using cloud-stored spatial anchors and environmental mapping to enable shared, location-based AR experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:persistent-ar-placement", "labels": [ "Persistent AR Placement", "Persistent AR" ], "is_subclass_of": [ "Platform and Environment", "Augmented Reality" ], "wikilinks": [ "Augmented Reality", "metaverse", "Shared AR Experiences" ] }, { "id": "persistent-content-anchoring", "title": "Persistent Content Anchoring", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Persistent content anchoring is the augmented-reality capability of fixing virtual content to a precise real-world location so that it reappears in the same place across sessions, devices, and users. It relies on saving spatial-anchor data, typically derived from visual feature maps, and relocalising against that map on subsequent visits. The technique is essential for shared and durable AR experiences tied to physical space.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:persistent-content-anchoring", "labels": [ "Persistent Content Anchoring" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "persistent-identifier", "title": "Persistent Identifier", "domain": "data", "domain_name": "Data", "definition": "A persistent identifier (PID) is a long-lasting, globally unique reference to a digital object that remains stable even as the object's location or custodian changes. PIDs such as DOIs, Handles, and ARKs are resolved through a managed indirection service that maps the identifier to current metadata and access endpoints. They are foundational to citation, provenance, and interoperability in scholarly and digital-object infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:persistent-identifier", "labels": [ "Persistent Identifier" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "persistent-memory", "title": "Persistent Memory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Persistent Memory (PMEM) is a class of byte-addressable, non-volatile storage technology that occupies the memory bus and exposes its capacity directly to processor load/store instructions, combining the persistence of storage media with latency approaching that of DRAM. It sits in a new tier of the memory hierarchy between volatile DRAM and block-accessed SSDs, enabling applications to retain state across power cycles without the overhead of traditional I/O system calls.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:persistent-memory", "labels": [ "Persistent Memory" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "persistent-research-agent", "title": "Persistent Research Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An autonomous AI agent designed for continuous, long-horizon information gathering and synthesis across multiple sources.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:persistent-research-agent", "labels": [ "Persistent Research Agent" ], "is_subclass_of": [ "Agents" ], "wikilinks": [] }, { "id": "persistent-state", "title": "Persistent State", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The durable retention of application or world state across user sessions, server restarts, and network interruptions, such that virtual environments, user progress, and asset ownership remain consistent over time. In metaverse and multiplayer systems, persistent state is typically achieved through distributed databases, cloud storage synchronisation, or blockchain-based ledgers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:persistent-state", "labels": [ "Persistent State" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "persistent-storage", "title": "Persistent Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Persistent storage refers to any data storage mechanism that retains data independently of the lifecycle of the process or system that created it, surviving power-off events, container restarts, and application failures. It contrasts with ephemeral or in-memory storage whose contents are lost when the host process terminates. Persistent storage encompasses file systems, relational and NoSQL databases, object stores, block volumes, and distributed storage systems, all of which provide durability guarantees through techniques such as write-ahead logging, replication, and erasure coding. It is a foundational concern in cloud-native architectures, stateful microservices, and any system that must maintain reliable long-term data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:persistent-storage", "labels": [ "Persistent Storage", "Durable Storage", "PersistentStorage" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "persona", "title": "Persona", "domain": "ai", "domain_name": "Ai", "definition": "A persona is a defined personality, role, and behavioural profile assigned to a conversational AI to shape its tone, knowledge framing, and interaction style. It encodes traits such as voice, expertise, boundaries, and goals, often via system prompts, so that responses stay consistent and aligned with a product's intent. Personas make chatbots feel coherent and purposeful rather than generic.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:persona", "labels": [ "Persona", "Digital Persona Management" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "personal-agent-runtimes", "title": "Personal Agent Runtimes", "domain": "ai", "domain_name": "Ai", "definition": "Always-on self-hosted agent daemons that operate across chat platforms as persistent personal assistants with event loops, memory, and self-improving capabilities \u2014 includes OpenClaw, Hermes, Khoj, Eliza, Agent Zero, OpenHarness, and AIlice.", "entityType": "Class", "qualityScore": 0.9, "maturity": "emerging", "iri": "urn:ngm:class:personal-agent-runtimes", "labels": [ "Personal Agent Runtimes" ], "is_subclass_of": [ "Agent Harness", "Agent Frameworks", "AI Agent System", "Agent Runtime" ], "wikilinks": [ "Agent Harness", "Agent Memory Layers", "Multi-Agent Orchestration Frameworks", "Internal AI Harness", "External AI Harness", "Agent Development SDKs", "Model Context Protocol", "Tool Use", "Vector Database", "Retrieval-Augmented Generation", "Large Language Model", "Local Language Model", "Agent Event Stream", "Agent Execution Sandboxes", "AI Agent System", "Agentic AI", "Autonomous Operation", "Browser Automation", "Code Execution", "Agent Orchestrator" ] }, { "id": "personal-assistance", "title": "Personal Assistance", "domain": "ai", "domain_name": "Ai", "definition": "Personal assistance is an AI application class in which an agent helps an individual accomplish everyday tasks such as scheduling, information lookup, communication, and device or browser control. It combines natural-language understanding with tool use and memory to act on the user's behalf within their personal context. Modern personal assistants increasingly automate multi-step workflows rather than only answering questions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:personal-assistance", "labels": [ "Personal Assistance", "Personal AI Assistants", "Personal Assistants" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "personal-data-store", "title": "Personal Data Store", "domain": "data", "domain_name": "Data", "definition": "A Personal Data Store (PDS) is a user-controlled software system or vault that aggregates, stores, and selectively discloses an individual's personal data \u2014 including identity attributes, health records, behavioural logs, and transaction histories \u2014 under the direct control of the data subject. It enforces consent-based access through verifiable credentials and cryptographic authorisation mechanisms, enabling fine-grained, revocable sharing with third-party services. PDSs embody the principles of data sovereignty and privacy-by-design, decoupling data custody from service providers and restoring ownership to individuals. They serve as a foundational layer for interoperable digital identity ecosystems, federated personal AI assistants, and regulatory compliance with frameworks such as GDPR and CCPA.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:personal-data-store", "labels": [ "Personal Data Store", "Interoperable Personal Data" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "personal-data", "title": "Personal Data", "domain": "governance", "domain_name": "Governance", "definition": "Personal data is any information relating to an identified or identifiable natural person, known as the data subject. Under data protection regimes such as the GDPR, identifiability can be direct (a name or identification number) or indirect (factors specific to a person's physical, economic, cultural or social identity). Special categories such as health, biometric or political data attract heightened protection. The concept anchors most privacy and data governance obligations, determining when processing rules, consent requirements and individual rights apply.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:personal-data", "labels": [ "Personal Data" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "personal-digital-asset-account", "title": "Personal Digital Asset Account", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Personal accounts are individual user accounts that grant access to digital services, platforms, or financial instruments, uniquely associated with a natural person. In the context of digital assets and cryptocurrency, personal accounts include exchange accounts, self-custody wallets, and HMRC Self Assessment registrations through which an individual reports capital gains from asset disposals. Proper management of personal accounts intersects identity management, authentication, privacy, and tax compliance obligations.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:personal-digital-asset-account", "labels": [ "Personal Digital Asset Account", "personal accounts" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "personal-information-protection-law", "title": "Personal Information Protection Law", "domain": "governance", "domain_name": "Governance", "definition": "The Personal Information Protection Law (PIPL) is China's comprehensive data-protection statute, effective 2021, governing how personal information of individuals in China is collected, processed, and transferred. It establishes consent and necessity requirements, data-subject rights, security obligations, and strict cross-border transfer controls, with extraterritorial reach and significant penalties. PIPL is a cornerstone of Asia-Pacific data-privacy regulation alongside frameworks like the GDPR.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:personal-information-protection-law", "labels": [ "Personal Information Protection Law" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "personalisation", "title": "Personalisation", "domain": "data", "domain_name": "Data", "definition": "Personalisation is the practice of tailoring content, recommendations, services and user experiences to the characteristics, behaviour and preferences of individual users or narrowly defined segments. It draws on collected and inferred data \u2014 profiles, interaction histories and contextual signals \u2014 to deliver differentiated outputs rather than a single uniform experience. Personalisation underpins recommendation systems, targeted marketing and adaptive interfaces, and raises associated privacy, consent and fairness considerations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:personalisation", "labels": [ "Personalisation" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [] }, { "id": "personalised-learning", "title": "Personalised Learning", "domain": "ai", "domain_name": "Ai", "definition": "Personalised learning is an educational approach in which the pace, content, modality, and assessment of instruction are dynamically adapted to the knowledge state, learning style, goals, and preferences of the individual learner. Technology-enabled personalised learning uses data about learner interactions and performance to drive adaptive algorithms that present the most effective next learning experience for each person, contrasting with one-size-fits-all curriculum delivery.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:personalised-learning", "labels": [ "Personalised Learning", "Personalised Education" ], "is_subclass_of": [ "Education Technology" ], "wikilinks": [] }, { "id": "personalized-ai-recommendations", "title": "Personalized AI Recommendations", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A system that leverages user-specific data and behavioral profiles to deliver tailored suggestions for AI tools and workflows.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:personalized-ai-recommendations", "labels": [ "Personalized AI Recommendations" ], "is_subclass_of": [ "Enterprise Ai" ], "wikilinks": [] }, { "id": "personalized-experiences", "title": "Personalized Experiences", "domain": "ai", "domain_name": "Ai", "definition": "Personalized experiences are interactions, content, or services dynamically tailored to an individual's preferences, history, and context. They are produced by models that infer user intent and situation from behavioural signals and contextual data, then adapt recommendations, interfaces, or responses accordingly. Personalization aims to increase relevance and engagement while balancing privacy and the risk of filter bubbles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:personalized-experiences", "labels": [ "Personalized Experiences", "Personalized Experience", "Personalized UX" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "personalized-interaction", "title": "Personalized Interaction", "domain": "ai", "domain_name": "Ai", "definition": "Personalized interaction is a conversational or interface exchange that adapts in real time to the specific user it is engaging, using their context, history, and inferred state. Unlike static personalization of content, it tailors the moment-to-moment dialogue, tone, and follow-ups based on prior turns and behavioural feedback. It is central to adaptive assistants that improve as they learn an individual's patterns.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:personalized-interaction", "labels": [ "Personalized Interaction" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "personalized-virtual-experiences", "title": "Personalized Virtual Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Personalized Virtual Experiences are AI-driven, dynamically adaptive virtual environments, narratives, and interactions that are tailored to the preferences, behaviour history, physiological signals, and declared identity of individual users. They integrate recommendation systems, adaptive content generation, avatar customisation, and contextual sensing to construct unique, user-specific pathways through spatial computing environments such as virtual reality, augmented reality, and metaverse platforms. The personalisation layer continuously models user intent and affect, adjusting scene composition, difficulty, pacing, social matchmaking, and content curation in real time. As a convergence point of machine learning, spatial computing, and human-computer interaction, these systems raise significant concerns around data sovereignty, consent, and algorithmic bias.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:personalized-virtual-experiences", "labels": [ "Personalized Virtual Experiences" ], "is_subclass_of": [ "Virtual Experience" ], "wikilinks": [ "User Engagement", "metaverse", "Virtual Experience" ] }, { "id": "pervasive-computing", "title": "Pervasive Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Pervasive computing \u2014 also termed ubiquitous computing \u2014 is a paradigm in which computational capability is embedded throughout the physical environment and integrated seamlessly into everyday objects and infrastructure, such that computation becomes a background utility rather than a discrete user activity. Coined by Mark Weiser at Xerox PARC in 1991, the vision encompasses context-aware, networked devices that sense, process, and communicate without explicit user interaction. It is the conceptual precursor to the Internet of Things and ambient intelligence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pervasive-computing", "labels": [ "Pervasive Computing", "PervasiveComputing" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [] }, { "id": "peter-todd", "title": "Peter Todd", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Peter Todd is a Bitcoin developer and applied cryptography consultant known for contributions to the Bitcoin protocol and related projects. He has worked on consensus and transaction issues.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:peter-todd", "labels": [ "Peter Todd" ], "is_subclass_of": [ "Bitcoin Core" ], "wikilinks": [ "Bitcoin Network", "Cryptography", "Bitcoin Core", "https://petertodd.org", "https://github.com/petertodd" ] }, { "id": "pharmaceutical-logistics", "title": "Pharmaceutical Logistics", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Pharmaceutical logistics is the specialised management of storage, handling, and transport of medicines and biologics under strict regulatory and environmental controls. It enforces cold-chain integrity, serialisation, and chain-of-custody to preserve product efficacy and prevent counterfeiting or diversion. The discipline combines temperature-controlled distribution with traceability and compliance reporting across the supply chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pharmaceutical-logistics", "labels": [ "Pharmaceutical Logistics" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "pharmaceutical-supply-chain", "title": "Pharmaceutical Supply Chain", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The Pharmaceutical Supply Chain encompasses all entities, processes, and systems involved in the production, storage, distribution, and dispensing of medicinal products, from active pharmaceutical ingredient (API) synthesis through to patient delivery. It is subject to stringent regulatory serialisation and traceability requirements designed to prevent counterfeit medicines from entering the legitimate supply chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:pharmaceutical-supply-chain", "labels": [ "Pharmaceutical Supply Chain" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "pharmaceutical-traceability", "title": "Pharmaceutical Traceability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based pharmaceutical supply chain systems employing unit-level serialisation, immutable audit trails, and automated smart-contract verification to combat counterfeit medicines, enable DSCSA and EU FMD regulatory compliance, and deliver cold-chain monitoring and clinical-trial supply management across consortium networks such as MediLedger.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pharmaceutical-traceability", "labels": [ "Pharmaceutical Traceability", "BC-0442-pharmaceutical-traceability", "Pharmaceutical Verification" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Accenture", "BC-0013-smart-contracts", "BC-0023-zero-knowledge-proofs", "BC-0029-permissioned-blockchain", "BC-0044-supply-chain-management", "BC-0067-hyperledger-fabric", "BC-0432-consortium-blockchain", "BC-0434-blockchain-as-a-service", "BC-0441-provenance-tracking", "BC-0443-food-safety-blockchain", "Chronicled", "Corda", "DHL", "Ethereum Enterprise Alliance", "FarmaTrust", "Internet of Things", "Modum", "Pharmaceutical Commerce Association", "PillPack", "rfXcel" ] }, { "id": "phase-transition", "title": "Phase Transition", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A phase transition is a qualitative, often abrupt, change in the macroscopic behaviour of a system as a control parameter crosses a critical threshold. Originating in statistical physics to describe transformations such as freezing or magnetisation, the concept is now applied to complex and learning systems where a small change in scale, data, or connectivity produces a discontinuous jump in capability. It is closely associated with emergence and critical phenomena.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:phase-transition", "labels": [ "Phase Transition" ], "is_subclass_of": [ "Complex Systems" ], "wikilinks": [] }, { "id": "phase-unwrapping", "title": "Phase Unwrapping", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:phase-unwrapping", "labels": [ "Phase Unwrapping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "philosophy-of-mind", "title": "Philosophy of Mind", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Philosophy of Mind is the branch of philosophy that investigates the nature of mental phenomena \u2014 including consciousness, intentionality, perception, belief, desire, and emotion \u2014 and their relationship to physical processes in the brain and body. It encompasses foundational debates over the mind-body problem, such as Cartesian dualism, physicalism, functionalism, and eliminative materialism, seeking to determine whether mental states are reducible to or identical with neurological states. The field directly informs artificial intelligence and cognitive science by establishing conceptual frameworks for what it means to think, represent, and experience, and by posing challenges such as the hard problem of consciousness and the Chinese Room argument against strong AI. It bridges empirical sciences of the mind with normative and metaphysical questions that formal methods alone cannot resolve.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:philosophy-of-mind", "labels": [ "Philosophy of Mind" ], "is_subclass_of": [ "Cognitive Science", "AI Research Area" ], "wikilinks": [ "Cognitive Science", "Artificial Intelligence", "Reasoning" ] }, { "id": "phishing-resistance", "title": "Phishing Resistance", "domain": "security", "domain_name": "Security", "definition": "Phishing resistance is a property of an authentication mechanism whereby credentials cannot be captured, replayed, or relayed by an attacker impersonating a legitimate service, because the authentication protocol cryptographically binds the exchange to the origin requesting it. Mechanisms such as FIDO2/WebAuthn and hardware security keys achieve this by verifying the requesting origin as part of the cryptographic challenge, so a spoofed site cannot obtain a usable credential. It is increasingly mandated as a baseline requirement for high-assurance authentication standards.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:phishing-resistance", "labels": [ "Phishing Resistance" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "phishing-resistant-authentication", "title": "Phishing Resistant Authentication", "domain": "security", "domain_name": "Security", "definition": "Phishing-resistant authentication is a class of authentication methods designed so that credentials cannot be captured and replayed by an attacker who tricks a user into interacting with a fraudulent site or relay. It achieves this primarily through public-key cryptography combined with origin binding, so that a credential is cryptographically tied to the legitimate service's domain and will not authenticate to an impostor. FIDO2/WebAuthn passkeys and hardware security keys are the canonical implementations, replacing shared secrets such as passwords and one-time codes that remain vulnerable to interception.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:phishing-resistant-authentication", "labels": [ "Phishing Resistant Authentication", "Phishing-Resistant Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "phoenix-wallet", "title": "Phoenix Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Phoenix is a non-custodial Bitcoin Lightning wallet developed by ACINQ, the team behind the Eclair implementation, that automates channel management so that users experience Lightning payments without manually opening or balancing channels. It keeps users in self-custody of their keys while abstracting liquidity provisioning, on-the-fly channel creation, and fee handling, making the Lightning Network accessible to non-technical mobile users.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:phoenix-wallet", "labels": [ "Phoenix Wallet" ], "is_subclass_of": [ "Non-Custodial Wallet" ], "wikilinks": [] }, { "id": "phoenix", "title": "Phoenix", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Phoenix is a self-custodial Bitcoin Lightning wallet developed by ACINQ that manages channels automatically on behalf of the user. It is available as a mobile application.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:phoenix", "labels": [ "Phoenix" ], "is_subclass_of": [ "Wallet" ], "wikilinks": [ "Lightning", "BOLT12", "Lightning Service Provider", "Wallet", "https://phoenix.acinq.co", "https://github.com/ACINQ/phoenix" ] }, { "id": "photogrammetry", "title": "Photogrammetry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A computational technique for reconstructing 3D geometry from overlapping photographic images through mathematical analysis of correspondences, camera poses, and geometric transformations to extract spatial information from 2D image data.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:photogrammetry", "labels": [ "Photogrammetry", "Drone Photogrammetry" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "3D Model Creation", "3D Reconstruction Pipeline", "Asset Digitization", "Camera Calibration", "Camera Models", "Computational Resources", "Computer Vision Algorithms", "Feature Detection", "Image Matching", "Mesh Reconstruction", "Multi-View Geometry", "Multiple Images", "Overlapping Coverage", "Point Cloud Generation", "RealityCaptureDomain", "Reality Capture Workflow", "Siemens + ACM", "Structure from Motion", "Terrain Mapping", "Triangulation" ] }, { "id": "photorealism", "title": "Photorealism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Photorealism is the goal and quality of producing synthetic imagery that is visually indistinguishable from a photograph of a real scene. In computer graphics it is pursued through physically based rendering, accurate light transport such as global illumination and ray tracing, high-fidelity material models, and careful post-processing including tone mapping and colour grading. Achieving photorealism requires faithful simulation of how light interacts with surfaces, volumes, and the camera, and it underpins applications from visual effects and architectural visualisation to immersive virtual environments and synthetic training data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:photorealism", "labels": [ "Photorealism" ], "is_subclass_of": [ "Rendering" ], "wikilinks": [] }, { "id": "photorealistic-rendering", "title": "Photorealistic Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Photorealistic Rendering is the computational process of generating images from 3D scene descriptions that are visually indistinguishable from photographs, achieved through physically accurate simulation of light transport, material properties, camera optics, and atmospheric phenomena. Core algorithms include path tracing and bidirectional path tracing, which apply Monte Carlo integration over the rendering equation to compute global illumination, caustics, subsurface scattering, and volumetric effects. Modern implementations leverage GPU hardware ray-tracing acceleration, physically based rendering (PBR) material models such as the Cook-Torrance BRDF, and AI-driven denoising to make high-fidelity output feasible in interactive and real-time contexts. Photorealistic rendering underpins digital twin visualisation, cinematic virtual production, immersive XR experiences, and AI-generated synthetic training data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:photorealistic-rendering", "labels": [ "Photorealistic Rendering" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "photorealistic-synthesis", "title": "Photorealistic Synthesis", "domain": "ai", "domain_name": "Ai", "definition": "Photorealistic synthesis is the generation of images or scenes that are visually indistinguishable from real photographs. It is achieved with generative models such as diffusion networks and GANs, or with physically based rendering, that reproduce accurate lighting, materials, geometry, and texture detail. It underpins high-fidelity image generation and the creation of realistic 4D and procedural content.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:photorealistic-synthesis", "labels": [ "Photorealistic Synthesis" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "photorealistic-telepresence", "title": "Photorealistic Telepresence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Photorealistic telepresence is remote real-time communication in which participants appear as lifelike, volumetrically accurate representations rather than video tiles or stylised avatars. It reconstructs a person's appearance and motion using techniques such as neural avatars and 3D Gaussian splatting, rendering them convincingly in a shared spatial scene. The goal is a sense of co-presence that closely matches being physically together.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:photorealistic-telepresence", "labels": [ "Photorealistic Telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "photosphere", "title": "Photosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:photosphere", "labels": [ "Photosphere" ], "is_subclass_of": [ "Stellar Atmosphere" ], "wikilinks": [] }, { "id": "physical-hardware", "title": "Physical Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The tangible computing devices and peripherals required for metaverse and XR experiences, including VR headsets, AR glasses, haptic devices, motion controllers, and supporting infrastructure like GPUs and networking equipment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:physical-hardware", "labels": [ "Physical Hardware" ], "is_subclass_of": [ "Content and Assets", "Hardware" ], "wikilinks": [ "Hardware", "Immersive Experiences", "metaverse" ] }, { "id": "physical-layer", "title": "Physical Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Physical Layer (OSI Layer 1) is the foundational stratum of the Open Systems Interconnection reference model defined in ISO/IEC 7498-1 and ITU-T X.200, responsible for raw transmission and reception of unstructured bit streams over a physical communication medium.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:physical-layer", "labels": [ "Physical Layer", "Network Physical Layer", "PhysicalLayer" ], "is_subclass_of": [ "Network and Communication", "Hardware Layer", "OSI Model", "Network Architecture", "Telecommunications", "Signal Processing", "Communications Engineering" ], "wikilinks": [ "3GPP", "3GPP NR", "5G NR", "ASIC Design", "Autonomous Vehicles", "Bluetooth Core Specification", "Bluetooth LE", "Bluetooth SIG", "Cable Plant", "Channel Coding", "Channel Coding Theory", "Channel Equalisation", "Chirp Spread Spectrum", "Clock Recovery", "Co-Packaged Optics", "Communications Engineering", "Connector Standards", "Data Link Layer", "Dispersion Compensation", "DSP" ] }, { "id": "physical-network-hardware", "title": "Physical Network Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The networking infrastructure components that enable connectivity and data transmission for metaverse applications, including routers, switches, access points, edge computing devices, and 5G equipment required for low-latency immersive experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:physical-network-hardware", "labels": [ "Physical Network Hardware" ], "is_subclass_of": [ "Platform and Environment", "Network Infrastructure" ], "wikilinks": [ "Low Latency Connectivity", "metaverse", "Network Infrastructure" ] }, { "id": "physical-presence", "title": "Physical Presence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Physical presence is the perceptual sense of being bodily situated in a remote or virtual environment, including the ability to act on and feel that environment. In telepresence it is produced by combining spatial audio-visual immersion with embodiment cues such as haptic feedback and a controllable remote body or robot. Strong physical presence makes a mediated location feel as though the user is genuinely there.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:physical-presence", "labels": [ "Physical Presence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "physical-security", "title": "Physical Security", "domain": "security", "domain_name": "Security", "definition": "Physical security comprises the measures \u2014 locks, barriers, surveillance, access control and guarding personnel \u2014 that protect facilities, hardware and people from physical threats such as unauthorised entry, theft, tampering and environmental hazards. It is a foundational control layer that complements cybersecurity: a system with strong network defences remains vulnerable if an attacker can gain physical access to servers or storage media. Physical security is a standard requirement for data centres and for safeguarding cryptographic key material in cold storage.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:physical-security", "labels": [ "Physical Security" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "physical-virtual-registration", "title": "Physical Virtual Registration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of aligning and synchronizing virtual content with physical world coordinates using spatial tracking, computer vision, and sensor fusion to ensure accurate overlay of digital objects in augmented reality environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:physical-virtual-registration", "labels": [ "Physical Virtual Registration", "3D Registration", "Physical-Virtual Registration" ], "is_subclass_of": [ "Interaction Technology", "Spatial Computing Paradigm" ], "wikilinks": [ "Accurate AR Overlay", "metaverse", "Spatial Computing" ] }, { "id": "physically-based-rendering", "title": "Physically Based Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physically Based Rendering (PBR) is a rendering methodology that simulates the interaction of light with materials using first-principles optics \u2014 energy conservation, the microfacet BRDF model, Fresnel reflectance equations, and radiometric correctness \u2014 to produce consistent, predictable visual output across arbitrary lighting environments. Surface materials are parameterised through a compact, artist-friendly set of maps (albedo, metalness, roughness, normal, ambient occlusion) that together drive evaluation of the Cook-Torrance or similar BRDF at each pixel. PBR has become the dominant material workflow in real-time engines (Unreal Engine, Unity) and offline path tracers alike, and is codified in the glTF 2.0 metallic-roughness material model, ensuring cross-renderer portability. Its physical correctness makes it the foundation for photorealistic digital twins, metaverse environments, and cinematic VFX pipelines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:physically-based-rendering", "labels": [ "Physically Based Rendering", "Physically Based Rendering Material", "Physically-Based Rendering" ], "is_subclass_of": [ "Rendering Technique" ], "wikilinks": [] }, { "id": "physics-engine", "title": "Physics Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software component that simulates physical interactions, constraints, and dynamics in real-time for virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:physics-engine", "labels": [ "Physics Engine", "PhysicsEngine" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "3D Engine", "Animation System", "Collision Detection System", "Collision Response", "Constraint Solver", "EWG/MSF Taxonomy", "ISO/IEC 23090-3", "Kinematic Animation", "Physical Simulation", "Realistic Interaction", "SIGGRAPH Pipeline WG", "Simulation Environment", "Compute Infrastructure", "ComputeLayer", "Compute Layer", "CreativeMediaDomain", "Game Engine", "Graphics API", "InfrastructureDomain", "Math Library" ] }, { "id": "physics-material", "title": "Physics Material", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A data structure encoding surface interaction properties \u2014 friction coefficients, coefficient of restitution, density, and drag \u2014 that govern object behaviour within a physics simulation engine. Physics materials are distinct from visual rendering materials and are consumed by simulation solvers to produce physically plausible contact responses for metaverse avatars, props, and environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:physics-material", "labels": [ "Physics Material" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "Material System", "Collision Detection", "Metaverse", "Physically-Based Rendering", "Physics Simulation", "Rigid Body Dynamics" ] }, { "id": "physics-simulation-engine", "title": "Physics Simulation Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A physics simulation engine is software that numerically integrates equations of motion over time to compute the movement, collision, and interaction of objects according to physical laws such as Newtonian mechanics, rigid-body dynamics, and contact resolution. It is used in real-time applications (games, XR), offline engineering simulation, and robotics training environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:physics-simulation-engine", "labels": [ "Physics Simulation Engine" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [ "Rigid Body Dynamics", "Simulation Environment", "3D Engine", "Simulation", "https://en.wikipedia.org/wiki/Physics_engine", "https://pybullet.org" ] }, { "id": "physics-simulation", "title": "Physics Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physics Simulation is the computational modelling of physical phenomena \u2014 encompassing rigid-body dynamics, soft-body deformation, fluid behaviour, cloth simulation, collision detection, and constraint solving \u2014 to produce physically plausible behaviour in real-time or offline virtual environments. It applies classical mechanics, continuum mechanics, and numerical integration methods to generate deterministic or probabilistic trajectories of simulated objects and agents. Physics simulation is foundational to interactive 3D applications, robotics training, digital-twin fidelity, and scientific computation, enabling rapid exploration of scenarios that would be costly or dangerous in the physical world. Modern implementations exploit GPU parallelism, hierarchical spatial data structures, and position-based dynamics to meet real-time performance budgets across heterogeneous hardware.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:physics-simulation", "labels": [ "Physics Simulation", "Physics-Based Simulation", "PhysicsModeling" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "physics-based-animation", "title": "Physics-Based Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Animation technique that computes object motion through real-time simulation of physical forces, gravity, collisions, and dynamics to create realistic movement and interactions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:physics-based-animation", "labels": [ "Physics-Based Animation" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "3D Transform System", "Animation Controller", "Cloth Simulation", "Collision Detection System", "Constraint Solver", "Force Integrator", "Numerical Integration", "Physics Simulation Engine", "Ragdoll Physics", "SIGGRAPH Standards", "Soft Body Simulation", "ComputeLayer", "CreativeMediaDomain", "Dynamic Character Animation", "Particle Systems", "Physics Engine", "Real-Time Rendering Pipeline", "Rigid Body Dynamics" ] }, { "id": "physiological-signal-processing", "title": "Physiological Signal Processing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Physiological signal processing is the acquisition, filtering and feature extraction of biological signals, such as heart rate, galvanic skin response, EEG or respiration, to infer a person's internal physiological or affective state. It combines digital signal processing techniques with domain-specific artefact removal, since biosignals are typically low-amplitude and prone to motion and electrical noise. It is a core input stage for affective computing systems that infer emotion or stress from bodily signals.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:physiological-signal-processing", "labels": [ "Physiological Signal Processing" ], "is_subclass_of": [ "Digital Signal Processing" ], "wikilinks": [] }, { "id": "pick-and-place", "title": "Pick and Place", "domain": "robotics", "domain_name": "Robotics", "definition": "Pick and place is a canonical robotic manipulation task in which a manipulator grasps an object at a source location and deposits it at a target location. It chains perception to localise the object, motion planning to reach and grasp it, and controlled placement, often repeated at high speed in structured settings. It is the foundational operation of industrial automation in assembly, packaging, and sorting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pick-and-place", "labels": [ "Pick and Place", "Pick And Place", "Pick and Place Automation", "Pick and Place Robotics" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "pid-controller", "title": "Pid Controller", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A PID Controller (Proportional-Integral-Derivative Controller) is a closed-loop feedback control algorithm that continuously calculates an error value as the difference between a desired setpoint and a measured process variable, then applies corrections based on proportional, integral, and derivative terms. PID controllers are foundational in robotics, autonomous systems, and industrial automation for precisely regulating position, velocity, and force.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:pid-controller", "labels": [ "Pid Controller", "PID Controller" ], "is_subclass_of": [ "Interaction Technology", "rb 0047 feedback control" ], "wikilinks": [ "rb 0047 feedback control" ] }, { "id": "piezometer", "title": "Piezometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:piezometer", "labels": [ "Piezometer" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "pinecone", "title": "Pinecone", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A managed cloud vector database service that stores embeddings and provides scalable approximate nearest neighbour similarity search for machine learning applications.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:pinecone", "labels": [ "Pinecone" ], "is_subclass_of": [ "Vector Database" ], "wikilinks": [ "Embeddings", "Vector Search", "Retrieval-Augmented Generation", "Semantic Search", "Vector Database" ] }, { "id": "pipeline-parallelism", "title": "Pipeline Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Pipeline parallelism splits a computation into ordered stages assigned to different processing units, so that distinct items occupy different stages at once and throughput rises once the pipeline is filled. It is widely used to train large deep learning models across multiple devices by assigning successive layers or micro-batches to separate accelerators.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pipeline-parallelism", "labels": [ "Pipeline Parallelism" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [ "Parallel Computing", "Real-Time Rendering", "GPU Architecture", "Graphics Pipeline", "Distributed Computing" ] }, { "id": "piracy-prevention", "title": "Piracy Prevention", "domain": "security", "domain_name": "Security", "definition": "Piracy prevention is the set of technical and procedural measures that deter and detect unauthorised copying, distribution, and use of digital content. It encompasses encryption, licensing, access control, watermarking, and tamper resistance, typically delivered through digital rights management systems. Its objective is to enforce usage rights while balancing legitimate user access and interoperability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:piracy-prevention", "labels": [ "Piracy Prevention" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "pixel-quality-flag", "title": "Pixel Quality Flag", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pixel-quality-flag", "labels": [ "Pixel Quality Flag" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "pixel-screening", "title": "Pixel Screening", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pixel-screening", "labels": [ "Pixel Screening" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "pixel-shader", "title": "Pixel Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A programmable GPU stage that executes once per rasterised fragment, determining each pixel's final colour and depth by sampling textures, computing lighting models, and applying material properties. Pixel shaders operate in a massively parallel fashion and are the primary site for physically-based rendering calculations in real-time graphics pipelines.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:pixel-shader", "labels": [ "Pixel Shader", "Fragment Shader" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "Normal Mapping", "Compute Shader", "Metaverse", "Physically-Based Rendering", "Texture Mapping", "Vertex Shader" ] }, { "id": "pkce", "title": "Pkce", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "PKCE, Proof Key for Code Exchange, is an extension to the OAuth 2.0 authorisation code flow that protects public clients, such as mobile and single-page applications, against interception of the authorisation code. The client generates a secret code verifier and sends its hashed code challenge when requesting authorisation, then proves possession of the verifier when redeeming the code, so a stolen code cannot be exchanged for tokens. It is now recommended for all OAuth clients, not only public ones.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pkce", "labels": [ "Pkce", "PKCE" ], "is_subclass_of": [ "OAuth 2.0" ], "wikilinks": [] }, { "id": "placing-on-the-market", "title": "Placing on the Market", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The first making available of an AI system or general-purpose AI model on the EU Union market, as defined in EU AI Act Article 3(12). This act triggers the full set of provider obligations under the Act, including risk management, technical documentation, conformity assessment, CE marking, and post-market monitoring, and applies regardless of whether the system is offered for payment or free of charge.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:placing-on-the-market", "labels": [ "Placing on the Market" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "plan-and-execute-pattern", "title": "Plan and Execute Pattern", "domain": "ai", "domain_name": "Ai", "definition": "The plan-and-execute pattern is an agent-architecture design in which a dedicated planner component produces a structured task plan that an executor component then realises step by step using tools. Formalising the strategy as a reusable pattern lets agent frameworks separate high-level reasoning from low-level tool invocation, enabling re-planning and observability. It is widely implemented through function-calling pipelines in LLM agent frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:plan-and-execute-pattern", "labels": [ "Plan and Execute Pattern", "Plan-and-Execute Pattern" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "plan-and-execute", "title": "Plan and Execute", "domain": "ai", "domain_name": "Ai", "definition": "Plan and execute is an agentic reasoning strategy in which an LLM-based agent first generates a multi-step plan for a task and then carries out each step, optionally re-planning when steps fail or new information appears. Separating planning from execution improves coherence on long-horizon tasks compared with purely reactive, single-step prompting. It is a common control pattern for autonomous and computer-use agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:plan-and-execute", "labels": [ "Plan and Execute" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "planet-surface-radiation-environment", "title": "Planet Surface Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planet-surface-radiation-environment", "labels": [ "Planet Surface Radiation Environment" ], "is_subclass_of": [ "Surface Radiation Environment" ], "wikilinks": [] }, { "id": "planet", "title": "Planet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planet", "labels": [ "Planet" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "planetary-atmosphere", "title": "Planetary Atmosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-atmosphere", "labels": [ "Planetary Atmosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-exploration", "title": "Planetary Exploration", "domain": "robotics", "domain_name": "Robotics", "definition": "Planetary exploration is the robotic investigation of planets, moons, and other bodies using rovers, landers, and orbiters operating in remote, unstructured, communication-delayed environments. It demands highly autonomous navigation and perception because round-trip signal delays make teleoperation impractical, requiring robots to plan paths and avoid hazards on their own. It is a demanding proving ground for autonomous ground-robot capability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:planetary-exploration", "labels": [ "Planetary Exploration" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "planetary-formation", "title": "Planetary Formation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-formation", "labels": [ "Planetary Formation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-gearbox", "title": "Planetary Gearbox", "domain": "robotics", "domain_name": "Robotics", "definition": "A planetary gearbox is a gear-reduction mechanism in which several planet gears revolve around a central sun gear inside an outer ring gear. This epicyclic arrangement distributes load across multiple gear meshes, yielding high torque density, coaxial input and output, and compact reduction ratios. It is a core mechanical component in robotic joints and actuators where space and weight are constrained.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:planetary-gearbox", "labels": [ "Planetary Gearbox", "Gearbox" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "planetary-geology", "title": "Planetary Geology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-geology", "labels": [ "Planetary Geology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-habitability", "title": "Planetary Habitability", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-habitability", "labels": [ "Planetary Habitability" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-interior", "title": "Planetary Interior", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-interior", "labels": [ "Planetary Interior" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-magnetosphere", "title": "Planetary Magnetosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-magnetosphere", "labels": [ "Planetary Magnetosphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-nebula", "title": "Planetary Nebula", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-nebula", "labels": [ "Planetary Nebula" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-protection", "title": "Planetary Protection", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-protection", "labels": [ "Planetary Protection" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-ring", "title": "Planetary Ring", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-ring", "labels": [ "Planetary Ring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-rover", "title": "Planetary Rover", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-rover", "labels": [ "Planetary Rover" ], "is_subclass_of": [ "Robot", "Spacecraft" ], "wikilinks": [] }, { "id": "planetary-science", "title": "Planetary Science", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-science", "labels": [ "Planetary Science" ], "is_subclass_of": [ "Space Science" ], "wikilinks": [] }, { "id": "planetary-surface", "title": "Planetary Surface", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-surface", "labels": [ "Planetary Surface" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planetary-volcanism", "title": "Planetary Volcanism", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:planetary-volcanism", "labels": [ "Planetary Volcanism" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "planner-executor-pattern", "title": "Planner-Executor Pattern", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A multi-agent orchestration pattern that separates deciding what to do from doing it: a planner component decomposes a goal into an ordered sequence of concrete steps, and one or more executor components carry those steps out, reporting results back so the plan can proceed or be revised. The split lets each role specialise \u2014 the planner reasons about strategy, dependencies, and ordering over the whole task, while executors focus narrowly on faithfully performing individual steps with the appropriate tools \u2014 and it makes the plan an explicit, inspectable artefact rather than an implicit chain of ad-hoc decisions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:planner-executor-pattern", "labels": [ "Planner-Executor Pattern" ], "is_subclass_of": [ "Multi-Agent Orchestration", "MultiAgentOrchestration" ], "wikilinks": [ "MultiAgentOrchestration", "SupervisorWorkerPattern", "TaskDelegation", "LLMOrchestration" ] }, { "id": "planning-module", "title": "Planning Module", "domain": "ai", "domain_name": "Ai", "definition": "A planning module is a software component within an autonomous system or AI agent responsible for generating sequences of actions to achieve specified goals, given a model of the current world state, available actions, and constraints. It translates high-level objectives into concrete executable task sequences, integrating deliberative reasoning, motion planning, and contingency handling. Planning modules appear in robotic architectures, autonomous vehicles, and large-language-model-based agent systems as a dedicated decision-making subsystem.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:planning-module", "labels": [ "Planning Module" ], "is_subclass_of": [ "AI System Component" ], "wikilinks": [] }, { "id": "planning-and-scheduling", "title": "Planning and Scheduling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Planning and Scheduling is an AI subfield concerned with generating sequences of actions (plans) and allocating resources across time (schedules) to achieve goals while satisfying temporal, resource, and precedence constraints. It encompasses classical, temporal, contingent, and probabilistic planning paradigms, as well as job-shop, project, and vehicle-routing scheduling approaches.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:planning-and-scheduling", "labels": [ "Planning and Scheduling", "Gain Scheduling", "Scheduling", "Scheduling Algorithm" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Constraint Satisfaction", "Optimization", "Artificial Intelligence", "Resource Management", "Robotics", "Search Algorithms", "STRIPS" ] }, { "id": "planning", "title": "Planning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Planning is the cognitive and computational process of deliberating over a set of possible actions and selecting a sequence that transforms an initial state into a desired goal state, given a model of how actions change the world. In artificial intelligence, automated planning formalises this problem using state representations, action schemas with preconditions and effects, and search or optimisation algorithms to synthesise executable plans. Planning encompasses both classical deterministic formulations (e.g. STRIPS, PDDL) and richer variants that handle uncertainty, partial observability, continuous time, resources, and preferences. It underpins autonomous agents, robotics, supply-chain management, and any system that must reason prospectively about future action consequences.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:planning", "labels": [ "Planning", "Autonomous Planning", "Deliberative Planning" ], "is_subclass_of": [ "Automated Planning" ], "wikilinks": [ "Search Algorithm", "Autonomous Agent", "Automated Planning", "Artificial Intelligence" ] }, { "id": "plant-model", "title": "Plant Model", "domain": "robotics", "domain_name": "Robotics", "definition": "A plant model is a mathematical representation of the dynamic system being controlled, mapping control inputs and disturbances to the system's state and outputs over time. Usually expressed as differential equations, transfer functions, or state-space form, it captures how the plant responds so a controller can be designed and tuned against it. Accurate plant models are central to control theory and to model-based control synthesis and simulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:plant-model", "labels": [ "Plant Model" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "plasma", "title": "Plasma", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Plasma is a blockchain scaling framework that builds hierarchical chains of child ledgers anchored to a root chain, processing transactions off the main chain while periodically committing compact state commitments to it. Users retain the ability to exit a child chain back to the root chain by submitting fraud proofs, which preserves the security guarantees of the underlying ledger even if a child chain operator misbehaves. It was an early Layer 2 design that influenced later optimistic and rollup-based scaling approaches.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:plasma", "labels": [ "Plasma" ], "is_subclass_of": [ "Blockchain", "Layer-2 Protocol" ], "wikilinks": [] }, { "id": "plasmapause", "title": "Plasmapause", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:plasmapause", "labels": [ "Plasmapause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "plasmasphere", "title": "Plasmasphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:plasmasphere", "labels": [ "Plasmasphere" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "platform-accountability", "title": "Platform Accountability", "domain": "governance", "domain_name": "Governance", "definition": "Platform accountability is the principle and regulatory expectation that online platforms bear responsibility for the systems, content, and harms arising on their services. It encompasses obligations such as transparency reporting, risk assessment, content-moderation due process, and remedies for illegal or harmful material. Frameworks like the Digital Services Act codify these duties, shifting platforms from passive intermediaries toward governed actors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:platform-accountability", "labels": [ "Platform Accountability" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "platform-as-a-service", "title": "Platform As A Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Platform as a Service (PaaS) is a cloud computing service model that provides a managed execution environment \u2014 including runtime, middleware, databases, and development tools \u2014 over the internet, allowing developers to build, deploy, and scale applications without managing underlying infrastructure. PaaS abstracts operating system and server management from the development team, enabling faster iteration cycles and lower operational overhead. Examples include Heroku, Google App Engine, Microsoft Azure App Service, and AWS Elastic Beanstalk.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:platform-as-a-service", "labels": [ "Platform As A Service", "Platform as a Service", "Platform-as-a-Service" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "platform-economy", "title": "Platform Economy", "domain": "finance", "domain_name": "Finance", "definition": "The Platform Economy describes an economic model in which digital intermediaries \u2014 platforms \u2014 create value by facilitating interactions, transactions, and information exchanges between two or more distinct user groups, typically producers and consumers. Platforms achieve scale through network effects: the value of the platform increases with the number of participants on one or both sides, creating winner-take-most market dynamics. Major platform types include transaction platforms (e-commerce, gig labour), innovation platforms (app stores, developer ecosystems), and content platforms (social media, streaming). Platform operators capture value through transaction fees, data monetisation, advertising, and software licences while externalising costs and risks to platform participants.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:platform-economy", "labels": [ "Platform Economy" ], "is_subclass_of": [ "Digital Platform" ], "wikilinks": [] }, { "id": "platform-engineering", "title": "Platform Engineering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Platform Engineering is the discipline of designing, building, and operating internal developer platforms (IDPs) that abstract infrastructure complexity and provide self-service toolchains, automated CI/CD pipelines, and standardised operational workflows for software engineering teams. It treats the IDP as a product, applying product management techniques to improve developer experience and reduce cognitive load. Platform Engineering sits at the intersection of DevOps, site reliability engineering, and cloud-native infrastructure.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:platform-engineering", "labels": [ "Platform Engineering" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "platform-governance", "title": "Platform Governance", "domain": "governance", "domain_name": "Governance", "definition": "Platform governance encompasses the formal and informal rules, decision-making processes, and control mechanisms that determine how blockchain and decentralised platforms operate, evolve, and distribute power among stakeholders. It includes on-chain mechanisms such as token-weighted voting and smart contract-executed decisions, as well as off-chain processes including community discussions, improvement proposals, and delegated representation structures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:platform-governance", "labels": [ "Platform Governance", "Digital Platform Governance", "Platform Governance Framework" ], "is_subclass_of": [ "Decentralised Governance" ], "wikilinks": [ "Decentralised Decision-Making", "Blockchain" ] }, { "id": "platform-independence", "title": "Platform Independence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Platform independence is the property of software that allows it to run correctly across different operating systems, hardware architectures or devices without modification. It is achieved by isolating platform-specific behaviour behind a hardware abstraction layer, so that application logic addresses a stable interface rather than the underlying system directly. It is a design goal rather than an absolute state, typically achieved to varying degrees depending on how much platform-specific code a system still requires.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:platform-independence", "labels": [ "Platform Independence" ], "is_subclass_of": [ "Portability" ], "wikilinks": [] }, { "id": "platform-layer", "title": "Platform Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Architectural tier providing core platform services including identity, world state management, and asset services upon which metaverse applications are built.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:platform-layer", "labels": [ "Platform Layer", "PlatformLayer" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Application Development", "Asset Portability", "Asset Service", "Authentication System", "Cross-World Interoperability", "EWG/MSF Taxonomy", "Identity Service", "Persistence Service", "Platform Services Layer", "User Identity Management", "World State Service", "API Gateway", "Blockchain", "Data Storage", "Database System", "Infrastructure Architecture", "InfrastructureDomain", "Middleware Layer", "Networking Layer", "Platform Middleware" ] }, { "id": "platform-middleware", "title": "Platform Middleware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A software layer that connects operating systems, applications, and services by providing common capabilities such as API management, message routing, authentication, and data integration to enable seamless communication between diverse systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:platform-middleware", "labels": [ "Platform Middleware" ], "is_subclass_of": [ "Platform and Environment", "Software Infrastructure" ], "wikilinks": [ "metaverse", "Software Infrastructure", "System Interoperability" ] }, { "id": "platform-service", "title": "Platform Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A managed, API-exposed capability unit delivered by a cloud or middleware platform that provides reusable building blocks\u2014such as authentication, storage, messaging, compute, or rendering\u2014to applications built atop that platform. Platform services abstract the operational complexity of underlying infrastructure by encapsulating it behind stable, versioned contracts, enabling developers to compose higher-order application features without provisioning or administering raw resources. They are the foundational unit of Platform-as-a-Service (PaaS) and are central to cloud-native, microservices, and distributed systems architectures. At scale, platform services enforce tenancy boundaries, SLA guarantees, and metered billing, transforming infrastructure capabilities into economically composable software products.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:platform-service", "labels": [ "Platform Service" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "play-to-earn-p2-e", "title": "Play-to-Earn (P2E)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Economic model and gameplay process where users gain real-world value through virtual participation, task completion, and reward distribution mechanisms that convert in-game achievements into tradeable assets.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:play-to-earn-p2-e", "labels": [ "Play-to-Earn (P2E)", "Play To Earn", "Play-to-Earn" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Asset Generation", "Asset Ownership", "Community Growth", "GameFi", "Gameplay Mechanics", "Income Generation", "Metaverse 101", "NFT (Non-Fungible Token)", "Player Engagement", "Player Identity", "Task Completion System", "Value Conversion", "ApplicationLayer", "Blockchain", "Cryptocurrency", "Digital Wallet", "Economic Participation", "Game Engine", "Marketplace", "Reward Distribution" ] }, { "id": "player-agency", "title": "Player Agency", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Player agency is the degree to which a participant's choices meaningfully shape the state, narrative, and outcomes of an interactive virtual experience. High agency means decisions have consequential, non-trivial effects rather than cosmetic ones, giving players authorship over their path through the world. It is a core design value of open-world and non-linear interactive systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:player-agency", "labels": [ "Player Agency" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "player-coach", "title": "Player Coach", "domain": "governance", "domain_name": "Governance", "definition": "A hybrid organizational role that combines direct individual contribution with the coaching and leadership of a team.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:player-coach", "labels": [ "Player Coach" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "player-engagement", "title": "Player Engagement", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Player engagement is the degree to which a game holds a player's attention, motivation, and continued participation over time. It is measured through behavioural signals such as session length, retention, progression, and emotional investment. Game designers and AI systems optimise engagement to sustain player communities and, in token-based games, to drive economic activity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:player-engagement", "labels": [ "Player Engagement" ], "is_subclass_of": [ "Metaverse" ], "wikilinks": [] }, { "id": "player-modelling", "title": "Player Modelling", "domain": "ai", "domain_name": "Ai", "definition": "Player modelling is the computational construction of representations of a player's preferences, skill, behaviour, and emotional state from in-game data. These models enable games to personalise difficulty, content, and recommendations, and to predict future actions. It is a foundational technique for adaptive and AI-driven game systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:player-modelling", "labels": [ "Player Modelling" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "player-tracking", "title": "Player Tracking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Player Tracking is the real-time monitoring of user position, movement, and behavioural analytics within virtual and metaverse environments. It encompasses spatial positioning systems, motion-capture input, gaze and gesture tracking, and analytics pipelines that feed avatar behaviour, personalisation, and platform telemetry.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:player-tracking", "labels": [ "Player Tracking" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "player-two-platform-implementation", "title": "Player Two Platform Implementation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Player Two code is the technical implementation layer of the Player Two platform, comprising a Vue.js front-end, BIP85-derived key management for trustless authentication, a Nostr-relay-based messaging architecture, and middleware whitelist logic. It enables encrypted group collaboration without server-held keys, using derivation path m/44'/1237 for per-user identity.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:player-two-platform-implementation", "labels": [ "Player Two Platform Implementation", "Player Two code" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "player-two", "title": "Player Two", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Player Two is a decentralised AI-character and collaborative storytelling platform combining Nostr-relay messaging, BIP85 key derivation, and generative AI workflows. It enables trustless group interactions with AI-driven waifu characters, token-gated access, and a robot project with on-device edge compute for the ED3N distributed network.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:player-two", "labels": [ "Player Two", "PlayerTwo" ], "is_subclass_of": [ "Software Engineering", "Collaboration Platform" ], "wikilinks": [ "Neocadia", "AI Video", "ComfyUI", "Player Two code" ] }, { "id": "plot", "title": "Plot", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:plot", "labels": [ "Plot" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "plotted-data-state", "title": "Plotted Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:plotted-data-state", "labels": [ "Plotted Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "plutocratic-voting", "title": "Plutocratic Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Plutocratic voting is a governance model in which voting power is proportional to the quantity of tokens, shares, or stake an actor holds, so that wealthier participants exert correspondingly greater influence over collective decisions. It is the default in most token-based on-chain governance systems, where one token equals one vote. While simple and Sybil-resistant by construction, it concentrates control among large holders and can entrench incumbent interests.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:plutocratic-voting", "labels": [ "Plutocratic Voting" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "pneumatic-actuator", "title": "Pneumatic Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "PneumaticActuator is a mechanical device that converts compressed-air energy into controlled mechanical motion \u2014 linear (cylinders, bellows), rotary (vane motors, semi-rotary actuators), or contractile (McKibben muscles, fibre-reinforced elastomers, Festo fluidic muscles) \u2014 and serves as the prim...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:pneumatic-actuator", "labels": [ "Pneumatic Actuator", "Pneumatic Actuation" ], "is_subclass_of": [ "Actuation and Control", "Fluid Power Device", "Actuator", "Robotic Actuator", "Compliant Mechanism", "Soft Robotics" ], "wikilinks": [ "Abaqus FEA", "Actuator", "Agricultural Robotics", "Air Compressor", "Air Supply System", "AMESim", "AMRC Sheffield", "ATEX Directive 2014/34/EU", "Bristol Robotics Laboratory", "BSI BS EN ISO 15552:2021", "Carbon Fibre Reinforcement", "Collaborative Robotics", "Compliant Grasping", "Compliant Mechanism", "Compressed Air", "Compressed Air Supply", "COMSOL Multiphysics", "ControlSystemLayer", "Dielectric Elastomer Actuator", "Directional Control Valve" ] }, { "id": "pneumatic-cylinder", "title": "Pneumatic Cylinder", "domain": "robotics", "domain_name": "Robotics", "definition": "A pneumatic cylinder is a mechanical actuator that converts compressed air pressure into linear mechanical force and motion, producing strokes from a few millimetres to over a metre. Pneumatic cylinders are widely used in industrial robotics, material handling, and assembly systems because of their high force-to-weight ratio, fast response, and inherent compliance compared to hydraulic counterparts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pneumatic-cylinder", "labels": [ "Pneumatic Cylinder" ], "is_subclass_of": [ "Actuation and Control", "Pneumatic Actuator" ], "wikilinks": [ "Pneumatic Actuator", "Robotics" ] }, { "id": "pneumatic-motor", "title": "Pneumatic Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "A Pneumatic Motor is an actuator that converts the energy stored in compressed air into continuous rotational mechanical motion, producing torque and speed proportional to the supply pressure and airflow rate. Pneumatic motors are characterised by high power-to-weight ratios, inherent overload protection through stall behaviour, and suitability for hazardous environments where electrical motors would pose ignition risks. They are widely used in industrial tooling, material handling equipment, and mobile robotics applications requiring lightweight, spark-free drive systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:pneumatic-motor", "labels": [ "Pneumatic Motor" ], "is_subclass_of": [ "Actuation and Control", "Pneumatic Actuator" ], "wikilinks": [ "Pneumatic Actuator", "Robotics" ] }, { "id": "podcast-production", "title": "Podcast Production", "domain": "ai", "domain_name": "Ai", "definition": "Podcast production is the end-to-end process of creating episodic audio programmes, encompassing recording, editing, mixing, mastering, and distribution. AI tools increasingly automate transcription, voice synthesis, noise removal, and chaptering. It is a key application domain for speech and audio machine-learning systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:podcast-production", "labels": [ "Podcast Production" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "point-cloud-generation", "title": "Point Cloud Generation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Point Cloud Generation is the process of producing a set of discrete three-dimensional coordinate samples (points), each representing a position on the surface or within the volume of a physical object or environment, typically augmented with attributes such as colour (RGB), intensity, or surface normals. Generation methods include active sensing (LiDAR, structured light, time-of-flight cameras) and passive photogrammetric reconstruction from overlapping images, producing the fundamental geometric representation used in autonomous navigation, digital twins, and 3D content creation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:point-cloud-generation", "labels": [ "Point Cloud Generation" ], "is_subclass_of": [ "Point Cloud" ], "wikilinks": [] }, { "id": "point-cloud-processing", "title": "Point Cloud Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Point cloud processing is the body of computational algorithms and end-to-end pipelines that transform raw collections of georeferenced 3D coordinate samples \u2014 produced by LiDAR scanners, depth cameras, structured-light systems, or photogrammetric reconstruction \u2014 into structured, semantically meaningful representations suitable for downstream applications such as autonomous navigation, digital twin construction, heritage documentation, and environmental monitoring. Core operations include noise filtering, voxel downsampling, multi-scan registration via Iterative Closest Point (ICP) and feature-based variants, normal estimation, segmentation, surface reconstruction, and compression. These operations are increasingly augmented by deep learning models operating directly on unordered point sets (PointNet family, sparse 3D CNNs), often executed on GPU-accelerated or specialised embedded hardware to meet real-time constraints. The field bridges classical computational geometry with modern neural scene representations such as NeRF and 3D Gaussian Splatting.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:point-cloud-processing", "labels": [ "Point Cloud Processing", "PCL Point Cloud Library", "Point Cloud Processor" ], "is_subclass_of": [ "Point Cloud" ], "wikilinks": [] }, { "id": "point-cloud", "title": "Point Cloud", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A point cloud is a discrete set of data points in three-dimensional coordinate space \u2014 each sample defined by (X, Y, Z) spatial coordinates and optionally augmented with attributes such as colour (RGB), intensity, return number, or surface normal vectors \u2014 acquired through active sensors like LiDAR and time-of-flight cameras or passive photogrammetric reconstruction from overlapping imagery. Point clouds constitute the primary raw geometric representation of physical objects and environments produced by scanning or depth-sensing systems, capturing surface geometry without presupposing mesh topology. They serve as the foundational data structure for downstream spatial-computing workflows including SLAM, 3D reconstruction, digital twin creation, and autonomous navigation, and are processed through operations such as voxelisation, segmentation, surface reconstruction, and registration to produce actionable spatial models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:point-cloud", "labels": [ "Point Cloud", "3D Point Cloud", "Dense Point Cloud", "SfM Point Cloud" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "point-time-locked-contracts", "title": "Point Time-Locked Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Point Time-Locked Contracts (PTLCs) are a Bitcoin Lightning Network payment primitive that use elliptic-curve adaptor signatures and a single secret point, rather than a hash preimage, to enforce conditional, time-bound payment across a route. They replace Hash Time-Locked Contracts with a scheme that reveals no shared hash across hops, improving privacy and eliminating hash-linkage attacks between channels. PTLCs are enabled by Taproot's Schnorr signatures and are proposed as a routing primitive for higher Bitcoin layers such as BTC Layer 3.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:point-time-locked-contracts", "labels": [ "Point Time-Locked Contracts", "Point Time-Locked Contract" ], "is_subclass_of": [ "Hash Time-Locked Contract" ], "wikilinks": [] }, { "id": "pointing-accuracy", "title": "Pointing Accuracy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pointing-accuracy", "labels": [ "Pointing Accuracy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "pointing-stability", "title": "Pointing Stability", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pointing-stability", "labels": [ "Pointing Stability" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "polar-coordinate-system", "title": "Polar Coordinate System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:polar-coordinate-system", "labels": [ "Polar Coordinate System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "polar-orbit", "title": "Polar Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:polar-orbit", "labels": [ "Polar Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "polar-robot", "title": "Polar Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Polar Robot (also called a spherical robot) is an industrial robot configuration with a rotary base joint that provides 360-degree horizontal rotation, an elevated rotary shoulder joint that tilts the arm up and down, and a linear telescoping arm that extends and retracts radially\u2014producing a spherical working envelope defined in spherical coordinates (radius, polar angle, azimuth). This geometry provides a large workspace volume relative to the physical footprint of the arm and was historically common in early industrial automation for tasks such as die casting and forging, though articulated arm designs have largely supplanted it in modern applications due to superior dexterity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:polar-robot", "labels": [ "Polar Robot" ], "is_subclass_of": [ "Robot Type", "Industrial Robot" ], "wikilinks": [ "Industrial Robot", "Robotics" ] }, { "id": "polarimetric-sar", "title": "Polarimetric SAR", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:polarimetric-sar", "labels": [ "Polarimetric SAR" ], "is_subclass_of": [ "Synthetic Aperture Radar" ], "wikilinks": [] }, { "id": "policy-administration-point", "title": "Policy Administration Point", "domain": "security", "domain_name": "Security", "definition": "A Policy Administration Point (PAP) is the component within an attribute-based or policy-based access control architecture responsible for authoring, storing, and distributing access control policies to Policy Decision Points. The PAP provides the administrative interface through which security administrators define rules governing which subjects may access which resources under what conditions. It is distinct from policy enforcement and evaluation components, focusing solely on policy lifecycle management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:policy-administration-point", "labels": [ "Policy Administration Point" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "policy-as-code", "title": "Policy As Code", "domain": "data", "domain_name": "Data", "definition": "Policy as code is the practice of expressing governance, security, and compliance rules in a machine-readable, version-controlled language so that they can be automatically evaluated and enforced. By treating policy as a software artefact, organisations gain testability, auditability, and consistent enforcement across infrastructure, data, and application pipelines. Decisions are computed by policy engines at admission or runtime, replacing manual review with deterministic, repeatable checks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:policy-as-code", "labels": [ "Policy As Code", "Policy as Code", "Policy-as-Code" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "policy-decision-point", "title": "Policy Decision Point", "domain": "security", "domain_name": "Security", "definition": "A Policy Decision Point (PDP) is the logical component in an attribute-based or policy-based access control architecture that evaluates access requests against a set of authorisation policies and returns a permit, deny, or indeterminate decision. The PDP receives a request context\u2014including subject attributes, resource attributes, action, and environment conditions\u2014from a Policy Enforcement Point (PEP), retrieves applicable policies from a Policy Information Point (PIP) or Policy Administration Point (PAP), and applies the XACML combining algorithms or equivalent logic to reach a binding decision. PDPs are the computational core of fine-grained, dynamic authorisation systems used in zero-trust security architectures, identity federation, API gateways, and cloud IAM platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:policy-decision-point", "labels": [ "Policy Decision Point" ], "is_subclass_of": [ "Access Control System" ], "wikilinks": [] }, { "id": "policy-enforcement-point", "title": "Policy Enforcement Point", "domain": "security", "domain_name": "Security", "definition": "A Policy Enforcement Point (PEP) is a runtime architectural component in access control systems that intercepts every request for a protected resource, forwards a structured authorisation query to a Policy Decision Point (PDP), and enforces the returned permit, deny, or obligation decision at the resource boundary. The PEP\u2013PDP separation, first formalised in the OASIS XACML standard, decouples policy logic from enforcement infrastructure and enables centralised policy management across heterogeneous systems. In zero-trust architectures every network transaction passes through a PEP, which may be realised as a reverse proxy, API gateway, service mesh sidecar, network firewall, or identity-aware proxy. The model extends naturally to software-defined perimeters, ABAC deployments, and AI governance enforcement layers where model prompts and outputs must satisfy content policies before reaching end users.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:policy-enforcement-point", "labels": [ "Policy Enforcement Point" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "policy-enforcement", "title": "Policy Enforcement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Mechanisms, architectures, and toolchains that translate declarative security and governance rules into runtime decisions\u2014permitting, denying, mutating, or auditing operations across compute, data, network, and AI systems.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:policy-enforcement", "labels": [ "Policy Enforcement", "PolicyEnforcement" ], "is_subclass_of": [ "Security and Identity", "Access Control", "Compliance Framework", "Governance", "Security Services" ], "wikilinks": [ "Narrative Gold Mine", "SecurityDomain", "Access Control", "Access Control System", "AI Governance", "AI Governance Framework", "AI Risk Register", "AI Trust Risk and Security Management", "AI Trustworthiness", "AML KYC Compliance", "ApplicationLayer", "Audit Trail", "Authentication Service", "Authentication Standards", "Blockchain Compliance", "Blockchain Governance", "Cloud Infrastructure", "Cloud-Native Applications", "Community Governance Model", "Compliance Audit Trail" ] }, { "id": "policy-engine", "title": "Policy Engine", "domain": "governance", "domain_name": "Governance", "definition": "A Policy Engine is an automated software component that evaluates, enforces, and logs governance, access-control, and behavioural rules within a metaverse or spatial computing system. It interprets declarative policy specifications (e.g., XACML, OPA Rego, or bespoke ontology-derived rules) and applies them at runtime to user actions, asset transactions, and inter-platform communications, enabling consistent compliance across distributed, multi-stakeholder virtual environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:policy-engine", "labels": [ "Policy Engine" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "ETSI + OMA3", "MetaverseDomain" ] }, { "id": "policy-framework", "title": "Policy Framework", "domain": "governance", "domain_name": "Governance", "definition": "A policy framework is a structured set of principles, rules, standards, and processes established by a government body, standards organisation, or institution to guide decision-making, regulate behaviour, and ensure accountability within a defined domain. It translates high-level objectives \u2014 such as safety, fairness, or interoperability \u2014 into operational requirements and compliance mechanisms. Policy frameworks provide the normative scaffolding that enables coordinated action across organisations and jurisdictions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:policy-framework", "labels": [ "Policy Framework" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "policy-gradient-methods", "title": "Policy Gradient Methods", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Policy Gradient Methods are a class of reinforcement-learning algorithms that directly optimise a parameterised policy by ascending the gradient of expected cumulative reward. Rather than deriving a policy from a learned value function, they adjust action probabilities to make rewarding behaviour more likely, using estimators such as REINFORCE and actor-critic variants. They naturally handle continuous and stochastic action spaces and underpin modern algorithms like proximal policy optimisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:policy-gradient-methods", "labels": [ "Policy Gradient Methods", "Policy Gradient", "Policy Gradient Method" ], "is_subclass_of": [ "Reinforcement Learning", "AI Technique" ], "wikilinks": [] }, { "id": "policy-information-point", "title": "Policy Information Point", "domain": "security", "domain_name": "Security", "definition": "A Policy Information Point (PIP) is the component in a policy-based access control architecture that retrieves and supplies attribute values needed by a Policy Decision Point to evaluate access requests against policies. The PIP acts as an attribute authority, querying identity stores, databases, and external services to resolve subject attributes, resource properties, and environmental conditions at decision time. It decouples the decision logic from the data sources that inform it.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:policy-information-point", "labels": [ "Policy Information Point" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "policy-layer", "title": "Policy Layer", "domain": "governance", "domain_name": "Governance", "definition": "The Policy Layer is the cross-cutting stratum that encodes machine-readable rules, constraints, and authorisations governing system behaviour. It sits above operational concerns and below the human-facing Governance Layer, translating governance intent into enforceable statements that other layers consult before acting. It contains policy definitions, decision points, and enforcement hooks rather than the data or compute those policies regulate.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:policy-layer", "labels": [ "Policy Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Identity Layer", "Control Layer", "Governance Layer", "Compliance Layer", "Access Control", "Attribute-Based Access Control", "owl:Thing", "NIST (National Institute of Standards and Technology)" ] }, { "id": "policy-optimisation", "title": "Policy Optimisation", "domain": "ai", "domain_name": "Ai", "definition": "Policy optimisation is the process of searching over a parameterised space of decision-making policies to maximise a scalar objective, typically a cumulative reward signal in a reinforcement learning context. It encompasses both gradient-based methods such as proximal policy optimisation and gradient-free approaches such as evolutionary strategies, applied across discrete and continuous action spaces.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:policy-optimisation", "labels": [ "Policy Optimisation" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "policy-simulation", "title": "Policy Simulation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Policy simulation is the use of computational models \u2014 typically agent-based or system-dynamics models \u2014 to project the likely outcomes of a proposed policy before it is enacted, allowing decision-makers to compare interventions under varying assumptions. It represents a population of heterogeneous actors and the rules governing their interactions, then runs the model forward to observe emergent, aggregate effects such as economic output, public health or resource use. Policy simulation is widely used in government, urban planning and public health to reduce the risk of costly real-world policy failures.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:policy-simulation", "labels": [ "Policy Simulation" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "policy-update-magnitude", "title": "Policy Update Magnitude", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Policy Update Magnitude is a measure of how much a reinforcement learning agent's policy changes between successive gradient update steps, typically quantified as the KL divergence between the old and new policy distributions or as the Euclidean norm of the parameter change vector. Controlling this magnitude is essential to training stability: excessively large updates can cause catastrophic performance collapse, whilst excessively small updates slow convergence. Algorithms such as Proximal Policy Optimisation (PPO) and Trust Region Policy Optimisation (TRPO) impose explicit constraints on policy update magnitude to balance exploration, exploitation, and stability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:policy-update-magnitude", "labels": [ "Policy Update Magnitude" ], "is_subclass_of": [ "Reinforcement Learning", "Reinforcement Learning Algorithm" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "policy", "title": "Policy", "domain": "governance", "domain_name": "Governance", "definition": "A policy is an explicit, authoritative statement of intent that guides decisions and prescribes acceptable behaviour or actions within an organisation or system. Policies translate high-level goals into rules, principles and constraints that can be applied consistently and enforced. In computing the term also denotes machine-readable rule sets, such as access-control policies, while in reinforcement learning a policy is a mapping from states to actions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:policy", "labels": [ "Policy" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "political-economy", "title": "Political Economy", "domain": "finance", "domain_name": "Finance", "definition": "The study of how political institutions, the political environment, and economic systems interact and influence one another. It examines how power and policy shape the production and distribution of wealth.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:political-economy", "labels": [ "Political Economy" ], "is_subclass_of": [ "Economics" ], "wikilinks": [ "Economics", "Governance", "Monetary Policy" ] }, { "id": "political-impact-of-ai", "title": "Political Impact of AI", "domain": "governance", "domain_name": "Governance", "definition": "The influence of artificial intelligence on political discourse, policy formation, electoral dynamics, and societal stability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:political-impact-of-ai", "labels": [ "Political Impact of AI" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "polkadot-parachains", "title": "Polkadot Parachains", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Application-specific sovereign blockchains that lease a slot on the Polkadot Relay Chain, inheriting its shared security model and cross-chain messaging infrastructure. Parachains can have custom runtime logic, consensus rules, and token economics whilst relying on the relay chain's validator set for finalisation, enabling heterogeneous specialised chains to interoperate without sacrificing security or decentralisation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:polkadot-parachains", "labels": [ "Polkadot Parachains", "Polkadot Parachain" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Polkadot", "Web3 Foundation", "Blockchain", "BlockchainDomain" ] }, { "id": "polkadot-xcm", "title": "Polkadot XCM", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Polkadot XCM (Cross-Consensus Messaging) is a format and language for communicating intent between consensus systems within and beyond the Polkadot network. Rather than transferring assets directly, XCM expresses instructions that a receiving chain interprets and executes locally, enabling asset transfers, remote calls and governance actions across parachains and the relay chain. It is transport-agnostic, relying on underlying delivery layers such as XCMP and HRMP, and is designed to be extensible and version-negotiated.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:polkadot-xcm", "labels": [ "Polkadot XCM" ], "is_subclass_of": [ "Cross-Chain Communication" ], "wikilinks": [] }, { "id": "polkadot", "title": "Polkadot", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Polkadot is a heterogeneous multi-chain blockchain protocol designed by Gavin Wood and developed by Parity Technologies that enables independent application-specific blockchains \u2014 called parachains \u2014 to connect to a central relay chain, share its pooled security, and exchange messages and assets through the Cross-Consensus Message (XCM) format. Its nominated proof-of-stake consensus mechanism uses DOT token validators to secure the relay chain whilst parachains benefit from shared security without needing to bootstrap their own validator sets. Polkadot is built on the Substrate framework, which allows developers to construct purpose-built WebAssembly-based runtimes supporting forkless on-chain upgrades. The protocol addresses the 'island of value' problem by providing a unified security and interoperability layer for a heterogeneous ecosystem of sovereign blockchains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:polkadot", "labels": [ "Polkadot", "Kagome Polkadot Host", "Polkadot Runtime" ], "is_subclass_of": [ "Blockchain Protocol" ], "wikilinks": [] }, { "id": "polycentric-governance", "title": "Polycentric Governance", "domain": "governance", "domain_name": "Governance", "definition": "Polycentric governance is a theory and practice of institutional arrangement in which authority over a shared domain is distributed across multiple overlapping, semi-autonomous decision-making centres rather than concentrated in a single hierarchy. Developed by Vincent and Elinor Ostrom through empirical study of common-pool resource management, the framework holds that complex social-ecological and sociotechnical systems benefit from diverse, redundant governance layers that can adapt, experiment, and mutually check one another. Each governance unit retains meaningful autonomy while engaging with others through negotiation, conflict-resolution mechanisms, and shared rules. Polycentric governance is increasingly applied to internet regulation, AI governance, and decentralised protocol design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:polycentric-governance", "labels": [ "Polycentric Governance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "polygon-agg-layer", "title": "Polygon AggLayer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Polygon AggLayer (Aggregation Layer) is a cross-chain settlement protocol that unifies liquidity and state across independent Polygon and zk-based chains. It uses zero-knowledge proofs to aggregate proofs from connected chains and enable near-atomic cross-chain transactions while preserving each chain's sovereignty. It aims to make a network of L1s and L2s feel like a single unified chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:polygon-agg-layer", "labels": [ "Polygon AggLayer" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "polygon-id", "title": "Polygon ID", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised identity framework that uses zero-knowledge proofs to let users present verifiable credentials without revealing the underlying data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:polygon-id", "labels": [ "Polygon ID" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [ "Zero-Knowledge Proof", "Identity Management", "Polygon", "Decentralized Identity" ] }, { "id": "polygon-mesh", "title": "Polygon Mesh", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A polygon mesh is a collection of vertices, edges, and faces that defines the shape of a polyhedral object in three-dimensional computer graphics. Faces are usually triangles or quadrilaterals whose connectivity describes a surface, and per-vertex attributes such as normals and texture coordinates support shading and texturing. Polygon meshes are the dominant representation for real-time rendering, modelling, and animation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:polygon-mesh", "labels": [ "Polygon Mesh" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "polygon-zkevm", "title": "Polygon Zkevm", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Polygon zkEVM is an Ethereum Layer 2 scaling solution built as a zero-knowledge rollup that is bytecode-equivalent with the Ethereum Virtual Machine. It batches transactions off-chain and posts validity proofs to Ethereum, inheriting Layer 1 security while reducing gas fees and increasing throughput. Developers can deploy existing EVM smart contracts with minimal changes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:polygon-zkevm", "labels": [ "Polygon Zkevm", "Polygon zkEVM" ], "is_subclass_of": [ "zk-Rollup" ], "wikilinks": [] }, { "id": "polygon", "title": "Polygon", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Polygon is a set of Ethereum scaling protocols and a development platform, originally launched as Matic Network in 2017 and rebranded to Polygon in 2021. It began as a proof-of-stake sidechain that runs in parallel to Ethereum and has expanded into zero-knowledge rollup technology, notably Polygon zkEVM, which executes Ethereum-compatible transactions with validity proofs. The architecture aims to offer lower fees and higher throughput while keeping compatibility with Ethereum tooling. The network's token, originally MATIC and migrating to POL, is used for staking and fees across Polygon chains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:polygon", "labels": [ "Polygon", "Polygon Network", "Polygon PoS" ], "is_subclass_of": [ "Blockchain", "Layer2", "Blockchain Domain" ], "wikilinks": [ "Ethereum", "Proof of Stake", "Decentralised Finance Domain", "Rollup", "zkSync", "Arbitrum", "Blockchain Domain" ] }, { "id": "polynomial-commitment", "title": "Polynomial Commitment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Polynomial Commitment is a cryptographic scheme that lets a prover commit to a polynomial with a short, binding value and later open the commitment at chosen evaluation points without revealing the whole polynomial. Verification of an opening is succinct and the commitment hides the polynomial until opened. Polynomial commitments are a core building block of modern succinct proof systems such as zk-SNARKs and zk-STARKs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:polynomial-commitment", "labels": [ "Polynomial Commitment" ], "is_subclass_of": [ "Cryptographic Commitment" ], "wikilinks": [] }, { "id": "polynomial-interpolation", "title": "Polynomial Interpolation", "domain": "security", "domain_name": "Security", "definition": "Polynomial interpolation is the construction of the unique polynomial of degree at most n-1 that passes through n given data points, most commonly computed via Lagrange or Newton forms. It underlies Shamir secret sharing, where a secret is encoded as the constant term of a random polynomial and reconstructed only when enough evaluated points, or shares, are combined via interpolation. The same technique underlies Reed-Solomon error-correcting codes, which treat message symbols as polynomial coefficients and use redundant evaluated points to recover data despite loss or corruption.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:polynomial-interpolation", "labels": [ "Polynomial Interpolation" ], "is_subclass_of": [ "Interpolation" ], "wikilinks": [] }, { "id": "pool-share", "title": "Pool Share", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A proof-of-work unit submitted by an individual miner to a mining pool server demonstrating that the miner performed a bounded amount of computational work toward finding a valid block hash. Pool shares have a lower difficulty target than the network block target, allowing the pool to credit each contributor proportional work units and distribute block rewards fairly, regardless of which specific miner finds the winning nonce.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:pool-share", "labels": [ "Pool Share" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "pooling-layer", "title": "Pooling Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A parameter-free neural network layer that summarises local regions or whole sets of feature activations into a single value each \u2014 typically the maximum or the mean \u2014 thereby reducing spatial resolution, enlarging receptive fields, and conferring a degree of translation invariance in convolutional networks, or collapsing token sequences into fixed-length vectors in embedding models. Pooling trades fine positional detail for compactness and robustness, and its choice (max, average, global, attention-weighted) materially affects what a representation preserves.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:pooling-layer", "labels": [ "Pooling Layer" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [ "Neural Network", "Convolutional Neural Network", "Embedding Model", "Feature Map" ] }, { "id": "population-health-management", "title": "Population Health Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Population health management is a data-driven approach to improving the health outcomes of a defined group of people by aggregating clinical and social data, stratifying members by risk, and coordinating targeted interventions across the care continuum. It shifts focus from treating individual encounters toward proactively managing the health of whole populations, often under value-based payment models. It relies on electronic health records, analytics and care-coordination workflows to identify needs and direct resources where they yield the greatest benefit.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:population-health-management", "labels": [ "Population Health Management" ], "is_subclass_of": [ "Healthcare Technology" ], "wikilinks": [] }, { "id": "portability", "title": "Portability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The capability for seamless migration of digital assets, identities, and experiences across heterogeneous virtual platforms through standardised formats (glTF, USD), metadata schemas, and blockchain bridges\u2014enabling users to transfer avatars, digital goods, and social connections without vendor lock-in.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:portability", "labels": [ "Portability", "Account Portability", "Cross-World Portability", "Software Portability" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Decentraland", "glTF", "NFT", "Ready Player Me", "ReadyPlayerMe", "USD", "MetaverseDomain" ] }, { "id": "portable-document-format-standard", "title": "Portable Document Format Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Portable Document Format: an ISO-standardised file format (ISO 32000) that encodes documents with fixed layout, fonts, graphics, and metadata in a device-independent, platform-agnostic representation. In AI and knowledge-graph contexts, PDFs serve as primary carriers of academic papers, technical specifications, and legal documents that are ingested via extraction pipelines for training, retrieval-augmented generation, and knowledge-base construction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:portable-document-format-standard", "labels": [ "Portable Document Format Standard", "PDF" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "portable-identity", "title": "Portable Identity", "domain": "security", "domain_name": "Security", "definition": "Portable identity is the property of a digital identity that allows it to be carried by its owner across services, platforms, and trust domains without being locked to any single provider. It relies on user-held credentials and identifiers that any relying party can verify cryptographically. Portability is a core goal of decentralised and self-sovereign identity systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:portable-identity", "labels": [ "Portable Identity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "portal-system", "title": "Portal System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A spatial computing mechanism that enables instantaneous traversal between distinct virtual locations, worlds, or platform environments, rendering a visual aperture through which users can see and enter the destination, thereby facilitating cross-world navigation and interoperability in metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:portal-system", "labels": [ "Portal System", "Intranet Portal" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "portfolio-management", "title": "Portfolio Management", "domain": "finance", "domain_name": "Finance", "definition": "Portfolio Management is the discipline of selecting, weighting and overseeing a collection of financial assets so that their combined risk and return characteristics align with an investor's objectives and constraints. It applies diversification, asset allocation and ongoing rebalancing to balance expected return against tolerable risk over a defined horizon. The practice spans both passive strategies that track benchmarks and active strategies that seek to outperform them.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:portfolio-management", "labels": [ "Portfolio Management" ], "is_subclass_of": [ "Investment Management" ], "wikilinks": [] }, { "id": "portfolio-optimisation", "title": "Portfolio Optimisation", "domain": "finance", "domain_name": "Finance", "definition": "Portfolio optimisation is the quantitative process of selecting asset weights within an investment portfolio to achieve an objective, most commonly maximising expected return for a given level of risk, using tools such as convex optimisation over a covariance matrix of asset returns. Modern portfolio theory, originating with Markowitz, formalised this as a mean-variance optimisation problem. Practical implementations extend the basic formulation with constraints on turnover, sector exposure and transaction costs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:portfolio-optimisation", "labels": [ "Portfolio Optimisation" ], "is_subclass_of": [ "Quantitative Finance" ], "wikilinks": [] }, { "id": "pose-estimation", "title": "Pose Estimation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pose estimation is the computational task of inferring the spatial configuration\u2014position, orientation, and joint angles\u2014of a body or rigid object from image or video data, spanning 2D keypoint localisation on the image plane (pixel-coordinate skeleton graphs), 3D joint position regression in cam...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:pose-estimation", "labels": [ "Pose Estimation", "DensePose", "Object Pose Estimation", "Pose Data", "PoseEstimation", "Robot Pose Estimation" ], "is_subclass_of": [ "AI Technique", "Perception", "Human Computer Interaction", "Computer Vision", "Spatial Computing Paradigm", "Deep Learning" ], "wikilinks": [ "3DPW Benchmark", "4D-Humans", "6DoF Estimation", "Action Recognition", "AIPerceptionDomain", "AlgorithmLayer", "Animation Pipeline", "Autonomous Vehicles", "Avatar Animation", "Bazarevsky et al. 2020 BlazePose MediaPipe", "Benchmark Dataset", "Body Model", "Body Prior", "Body Segmentation", "Body Shape Models", "Bogo et al. 2016 SMPLify Keep It SMPL", "BOP Benchmark", "Camera Calibration", "Cao et al. 2017 OpenPose Part Affinity Fields", "CenterPose" ] }, { "id": "pose-graph", "title": "Pose Graph", "domain": "robotics", "domain_name": "Robotics", "definition": "A pose graph is a graph representation used in robotics where nodes are robot poses (positions and orientations) and edges encode relative spatial constraints derived from odometry or sensor measurements. Pose-graph optimisation finds the configuration of poses that best satisfies all constraints, correcting accumulated drift. It is the backbone of modern graph-based SLAM systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pose-graph", "labels": [ "Pose Graph", "Pose Graph Optimisation" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "pose-tracking", "title": "Pose Tracking", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Pose tracking is the continuous estimation of the position and orientation (6-DoF pose) of a device, user, or body over time. In spatial computing it underpins head, hand, and controller tracking for AR and VR, fusing camera, IMU, and depth data at low latency. Accurate, drift-free pose tracking is essential for stable, comfortable immersive experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:pose-tracking", "labels": [ "Pose Tracking" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "position-dilution-of-precision", "title": "Position Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:position-dilution-of-precision", "labels": [ "Position Dilution of Precision" ], "is_subclass_of": [ "Dilution of Precision" ], "wikilinks": [] }, { "id": "position-sensor", "title": "Position Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A position sensor is a transducer that measures the displacement, angle, or spatial location of an object or joint relative to a reference frame, converting mechanical position into an electrical signal suitable for control system feedback. Position sensors are fundamental to closed-loop robotic and mechatronic systems, enabling precise joint control, end-effector placement, and localisation. They span a wide range of physical measurement principles including optical encoding, magnetic hall-effect sensing, capacitive displacement, and inductive resolver techniques.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:position-sensor", "labels": [ "Position Sensor", "PositionSensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "position-control", "title": "PositionControl", "domain": "robotics", "domain_name": "Robotics", "definition": "A closed-loop feedback control modology that commands a robotic manipulator, actuator, or motion platform to achieve and maintain a desired spatial position or angular orientation by continuously measuring the actual position via encoders, resolvers, or external sensing systems (vision, LIDAR), c...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:position-control", "labels": [ "PositionControl", "Closed-Loop Position Control", "Position Control", "Position Controller", "Position Tracking" ], "is_subclass_of": [ "Actuation and Control", "ControlAlgorithm", "FeedbackControl", "ServoSystem", "MotionControl" ], "wikilinks": [ "ActuatorCommand", "AutomationDomain", "Calibration", "ControlLaw", "ControlTheoryDomain", "ErrorCalculation", "FeedforwardCompensation", "Homing", "IEC 61800 Adjustable Speed Electrical Power Drive Systems", "IEEE Control Systems Society", "ISO 8373 Robotics Vocabulary", "ISO 9283 Manipulating Industrial Robots Performance Criteria", "KinematicModel", "MotorDriver", "PathFollowing", "PickAndPlace", "PositionFeedback", "PositionSensor", "Robot Operating System (ROS) Control Standards", "ServoSystem" ] }, { "id": "positional-audio", "title": "Positional Audio", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Positional Audio is a technique for rendering sound sources at specific locations within a 3D virtual or mixed-reality environment, simulating how sound propagates, attenuates, and is spatially perceived by the listener. It relies on head-related transfer functions (HRTFs) and room acoustics modelling to create realistic auditory presence.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:positional-audio", "labels": [ "Positional Audio" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "positional-encoding", "title": "Positional Encoding", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A technique for injecting information about the relative or absolute position of tokens in a sequence into a neural network, essential for transformer models since self-attention mechanisms are inherently permutation-invariant and lack sequential ordering awareness.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:positional-encoding", "labels": [ "Positional Encoding", "Rotary Positional Embedding", "Rotary Positional Encoding" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "positional-tracking", "title": "Positional Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Positional Tracking is the real-time determination of the position and orientation of a device, body part, or object in three-dimensional space, typically using sensors such as IMUs, cameras, or external emitters. It is a foundational capability for spatial computing systems including virtual reality, augmented reality, and robotics, enabling accurate alignment of digital content with the physical world. Techniques range from inside-out camera-based tracking to outside-in lighthouse systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:positional-tracking", "labels": [ "Positional Tracking" ], "is_subclass_of": [ "Spatial Computing Interaction" ], "wikilinks": [] }, { "id": "positioning-navigation-and-timing", "title": "Positioning Navigation and Timing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:positioning-navigation-and-timing", "labels": [ "Positioning Navigation and Timing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "positive-feedback", "title": "Positive Feedback", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Positive feedback is a control and systems mechanism in which the output of a process amplifies the input that produced it, reinforcing change rather than damping it. While destabilising in control systems, it underlies self-reinforcing dynamics such as network effects, exponential growth, and adoption cascades. Its behaviour contrasts with negative feedback, which seeks equilibrium.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:positive-feedback", "labels": [ "Positive Feedback", "Positive Feedback Mechanisms" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "possibility-unlock", "title": "Possibility Unlock", "domain": "economics", "domain_name": "Economics", "definition": "A market creation mechanism where AI renders new service models operationally viable at scale, enabling the emergence of entirely new markets that were previously impossible due to technical or logistical constraints.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:possibility-unlock", "labels": [ "Possibility Unlock" ], "is_subclass_of": [ "Economic Impact of AI" ], "wikilinks": [] }, { "id": "post-hoc-explanation", "title": "Post Hoc Explanation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Interpretability techniques applied after a machine learning model has been trained, providing explanations for model behaviour and predictions without modifying the model's architecture or requiring retraining.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:post-hoc-explanation", "labels": [ "Post Hoc Explanation", "Post-Hoc Explanation" ], "is_subclass_of": [ "AI Technique", "Explainability" ], "wikilinks": [ "Counterfactual Explanation", "Feature Attribution", "Grad-CAM", "LIME", "Model-Agnostic Explanations", "Saliency Map", "SHAP", "AI Companies", "Explainable AI", "Intrinsic Interpretability", "MetaverseDomain", "Model Interpretability", "Prompt Engineering" ] }, { "id": "post-incident-review", "title": "Post Incident Review", "domain": "governance", "domain_name": "Governance", "definition": "A post-incident review is a structured retrospective conducted after an operational incident to establish what happened, why, and how recurrence can be prevented. Conducted in a blameless manner, it reconstructs the timeline, identifies contributing factors through root-cause analysis, and produces tracked corrective actions and durable lessons. It is a core practice of incident management and site reliability engineering that turns failures into systemic improvement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:post-incident-review", "labels": [ "Post Incident Review", "Post-Incident Review" ], "is_subclass_of": [ "Incident Management" ], "wikilinks": [] }, { "id": "post-processing", "title": "Post Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Post processing is the stage in a real-time or offline rendering pipeline where image-space operations are applied to the fully rasterised or ray-traced framebuffer before it is displayed or composited. Effects\u2014such as bloom, depth of field, motion blur, tone mapping, colour grading, screen-space ambient occlusion (SSAO), temporal anti-aliasing (TAA), and chromatic aberration\u2014are executed as one or more full-screen shader passes that read from and write back to render targets. By operating in screen space rather than on scene geometry, post processing achieves high visual fidelity at comparatively low computational cost, and has become an indispensable component of game engines, virtual production pipelines, and XR head-mounted display rendering stacks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:post-processing", "labels": [ "Post Processing", "Post-Processing", "Post-processing" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "post-training-quantisation", "title": "Post Training Quantisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Post-training quantisation (PTQ) converts a trained full-precision neural network to a lower-precision representation, typically 8-bit integers, without re-running the original training loop. A small calibration dataset is used to estimate the dynamic range of activations so that scale and zero-point parameters can be chosen. PTQ trades a small, usually recoverable, drop in accuracy for substantial reductions in model size and inference cost.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:post-training-quantisation", "labels": [ "Post Training Quantisation", "Post-Training Quantisation" ], "is_subclass_of": [ "Quantisation" ], "wikilinks": [] }, { "id": "post-quantum-cryptography", "title": "Post-Quantum Cryptography", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cryptographic algorithms and protocols designed to be resistant to attacks from both classical and quantum computers, protecting secure communications in the post-quantum era.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:post-quantum-cryptography", "labels": [ "Post-Quantum Cryptography", "Post-Quantum Cryptography Standards", "Post-Quantum Signature", "Post-Quantum-Cryptography", "Quantum-Safe Cryptography" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Algorithm Implementation", "Code-Based Cryptography", "Computational Complexity Theory", "Cryptographic Infrastructure", "Hash-Based Signatures", "Isogeny-Based Cryptography", "Lattice-Based Cryptography", "Long-Term Data Protection", "Multivariate Cryptography", "NIST PQ Standard (2024)", "Quantum-Resistant Encryption", "Secure Key Exchange", "Security Protocol", "Compute Layer", "Cryptographic Key Management", "Data Layer", "Digital Signatures", "Mathematical Hard Problems", "Network Layer", "Physical Layer" ] }, { "id": "posterior-distribution", "title": "Posterior Distribution", "domain": "ai", "domain_name": "Ai", "definition": "A posterior distribution is the probability distribution over unknown quantities after observing data, obtained by combining a prior distribution with the likelihood via Bayes' theorem. It represents updated belief and is the central object of Bayesian inference. In recursive estimators such as Bayes and particle filters, the posterior at each step becomes the basis for the next prediction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:posterior-distribution", "labels": [ "Posterior Distribution" ], "is_subclass_of": [ "Machine Learning Discipline", "Probability Distribution" ], "wikilinks": [] }, { "id": "postgre-sql", "title": "PostgreSQL", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "PostgreSQL is an open-source, ACID-compliant object-relational database management system (ORDBMS) that extends the SQL standard with features such as table inheritance, function overloading, and a rich type system including arrays, JSON, and user-defined types. It supports both relational and document storage paradigms, advanced indexing strategies including GIN, GiST, BRIN, and B-tree, and full-text search, making it a versatile choice for analytical and transactional workloads. Developed from the POSTGRES project at UC Berkeley (1986), it has been maintained as free software since 1996 and is widely regarded as the most standards-compliant open-source relational database.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:postgre-sql", "labels": [ "PostgreSQL" ], "is_subclass_of": [ "Database System" ], "wikilinks": [] }, { "id": "postquantum-cryptography", "title": "Postquantum Cryptography", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Postquantum Cryptography (PQC) is the class of cryptographic algorithms designed to remain secure against adversaries equipped with large-scale quantum computers, replacing public-key schemes such as RSA and elliptic-curve cryptography that are broken by Shor's algorithm. PQC families include lattice-based, hash-based, code-based, and isogeny-based constructions, several of which NIST standardised in 2024 (e.g. CRYSTALS-Kyber, CRYSTALS-Dilithium).", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:postquantum-cryptography", "labels": [ "Postquantum Cryptography" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "potential-field-method", "title": "Potential Field Method", "domain": "robotics", "domain_name": "Robotics", "definition": "The Potential Field Method is a reactive robot navigation technique that models the goal as an attractive potential and obstacles as repulsive potentials, steering the robot along the negative gradient of the summed field. It produces smooth, real-time motion commands directly from sensor readings without explicit global search. Its principal weakness is susceptibility to local minima, where attractive and repulsive forces cancel and the robot stalls short of the goal.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:potential-field-method", "labels": [ "Potential Field Method" ], "is_subclass_of": [ "Path Planning" ], "wikilinks": [] }, { "id": "potentiometric-surface", "title": "Potentiometric Surface", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:potentiometric-surface", "labels": [ "Potentiometric Surface" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "poverty-reduction", "title": "Poverty Reduction", "domain": "finance", "domain_name": "Finance", "definition": "Poverty Reduction encompasses the policies, programmes, and structural changes aimed at decreasing the proportion of individuals living below nationally or internationally defined poverty thresholds, improving access to essential goods, services, and economic opportunities. Approaches span conditional cash transfer schemes, microfinance, labour market reforms, progressive taxation, and targeted public service delivery. It is central to the United Nations Sustainable Development Goals (SDGs), most directly SDG-1 (No Poverty) and SDG-10 (Reduced Inequalities). Successful poverty reduction typically requires complementary progress in health, education, infrastructure, and governance quality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:poverty-reduction", "labels": [ "Poverty Reduction" ], "is_subclass_of": [ "Sustainable Development" ], "wikilinks": [] }, { "id": "power-conditioning-and-distribution", "title": "Power Conditioning and Distribution", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:power-conditioning-and-distribution", "labels": [ "Power Conditioning and Distribution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "power-distribution-unit", "title": "Power Distribution Unit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A power distribution unit (PDU) is a device that delivers and manages electrical power to multiple pieces of equipment within a data centre rack or row. It takes input from an upstream supply such as a UPS or the building electrical feed and distributes regulated, often metered, power across many outlets. Modern intelligent PDUs add remote monitoring, per-outlet switching and environmental sensing to support availability and energy-efficiency goals.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:power-distribution-unit", "labels": [ "Power Distribution Unit" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "power-electronics", "title": "Power Electronics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Power electronics is the engineering discipline concerned with the conversion and control of electrical power using switching semiconductor devices such as MOSFETs, IGBTs and SiC transistors, central to motor drives, power supplies, inverters and energy management systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:power-electronics", "labels": [ "Power Electronics" ], "is_subclass_of": [ "Embedded Systems", "Embedded Systems Domain" ], "wikilinks": [ "Motor Driver", "Electric Motor", "Battery Management System", "Servo Motor", "Embedded Systems Domain" ] }, { "id": "power-grid", "title": "Power Grid", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A power grid is the interconnected network of generation, transmission and distribution infrastructure that delivers electrical power from producers to consumers. It balances supply and demand in real time across high-voltage transmission lines, substations and lower-voltage distribution networks. Modern grids increasingly integrate renewable generation and digital controls to maintain stability and reliability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:power-grid", "labels": [ "Power Grid" ], "is_subclass_of": [ "Power Infrastructure", "Grid Infrastructure" ], "wikilinks": [] }, { "id": "power-infrastructure", "title": "Power Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Power infrastructure is the set of electrical generation, distribution, conditioning and backup systems that deliver reliable energy to computing and physical facilities. In data centres it encompasses utility feeds, power distribution units, uninterruptible supplies and generators that together guarantee continuity under load and fault. It is a foundational dependency for high-availability infrastructure, often co-designed with cooling and capacity planning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:power-infrastructure", "labels": [ "Power Infrastructure" ], "is_subclass_of": [ "Infrastructure", "Data Center" ], "wikilinks": [] }, { "id": "power-management", "title": "power management", "domain": "robotics", "domain_name": "Robotics", "definition": "Power management is the discipline of regulating, distributing, monitoring, and optimising electrical energy flow within computing, robotic, and embedded systems to maximise operational endurance, thermal safety, and efficiency. It spans hardware circuits (DC\u2013DC converters, power distribution units, battery management systems), firmware-level control algorithms (voltage-frequency scaling, clock-gating, duty-cycle regulation), and system-level policies (workload scheduling, energy-aware task allocation) that together balance instantaneous demand against available supply. In robotic and mobile contexts it additionally encompasses energy harvesting, regenerative braking, and state-of-charge estimation to enable untethered autonomous operation. Effective power management is a prerequisite for deploying autonomous systems in resource-constrained field environments and for meeting safety standards governing collaborative human\u2013robot interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:power-management", "labels": [ "Power Management", "Data Centre Power Management", "Power Management Unit" ], "is_subclass_of": [ "Embedded Systems" ], "wikilinks": [] }, { "id": "power-purchase-agreement", "title": "Power Purchase Agreement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Power Purchase Agreement (PPA) is a long-term contract between an electricity generator and a buyer that fixes the price, volume and delivery terms for electrical energy over a defined period. PPAs underpin the financing of generation assets, particularly renewable plants, by guaranteeing a stable revenue stream that de-risks capital investment. They allocate market, volume and curtailment risk between producer and offtaker and may be physical or virtual (financial).", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:power-purchase-agreement", "labels": [ "Power Purchase Agreement" ], "is_subclass_of": [ "Power Systems" ], "wikilinks": [] }, { "id": "power-spectrum", "title": "Power Spectrum", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:power-spectrum", "labels": [ "Power Spectrum" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "power-supply", "title": "Power Supply", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A power supply unit (PSU) is an internal hardware component that converts alternating current (AC) from mains electricity into regulated low-voltage direct current (DC) required by computing equipment's internal components including motherboard, CPU, GPU, and storage devices. In data centre contexts, power supply systems include redundant PSUs, uninterruptible power supplies (UPS), and power distribution units (PDUs) organised in hierarchical redundancy tiers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:power-supply", "labels": [ "Power Supply", "DC Power Supply", "Electrical Power Supply", "Linear Power Supply", "Power Supply Unit", "PowerSupply" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [ "Continuous Operation", "System Reliability", "InfrastructureDomain", "Technology Domain" ] }, { "id": "power-systems", "title": "Power Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Power Systems refers to the interconnected infrastructure of electrical generation, transmission, distribution, and consumption that constitutes a modern electricity network, together with the engineering discipline that analyses, designs, and controls this infrastructure. Key concerns include power flow analysis, voltage stability, frequency regulation, fault analysis, and the integration of variable renewable generation sources into grid operations. Power systems engineering underpins all modern industrial economies and is undergoing significant transformation as decarbonisation mandates, distributed energy resources, and digital control systems reshape traditional architectures. The discipline employs optimisation, control theory, and increasingly machine learning for real-time grid management.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:power-systems", "labels": [ "Power Systems", "Power System" ], "is_subclass_of": [ "Energy and Power" ], "wikilinks": [] }, { "id": "power-usage-effectiveness", "title": "Power Usage Effectiveness", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Power usage effectiveness (PUE) is a metric used to assess the energy efficiency of a data centre, defined as the ratio of total facility energy to the energy delivered to IT equipment. A PUE of 1.0 represents perfect efficiency where all energy reaches computing hardware, while higher values reflect overhead from cooling, power conversion and lighting. It is a widely used benchmark for sustainability and operational efficiency in data centres.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:power-usage-effectiveness", "labels": [ "Power Usage Effectiveness" ], "is_subclass_of": [ "Energy Efficiency" ], "wikilinks": [] }, { "id": "power-efficient-ai", "title": "Power-Efficient AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Power-Efficient AI is a design and optimisation discipline that minimises energy consumption during machine learning inference and training, enabling deployment on battery-powered edge devices, IoT sensors, and mobile platforms with constrained power budgets. It combines hardware-level techniques (dynamic voltage and frequency scaling, power gating, specialised neural processing units) with model-level optimisations (quantisation to INT8/FP16, pruning, knowledge distillation) to achieve high inference throughput per watt. Efficiency is typically measured in TOPS/Watt, and sub-10mW average power envelopes are required for month-to-year deployment lifetimes in wearable and environmental sensing applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:power-efficient-ai", "labels": [ "Power-Efficient AI", "Ultra-Low-Power AI" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "practical-byzantine-fault-tolerance", "title": "Practical Byzantine Fault Tolerance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A consensus algorithm providing high-performance Byzantine state machine replication for asynchronous distributed systems, tolerating up to f faulty nodes in a system of 3f+1 total nodes. PBFT achieves deterministic finality through a three-phase protocol (pre-prepare, prepare, commit) and processes thousands of requests per second with sub-millisecond latency overhead, making it suitable for permissioned blockchain networks requiring strong consistency guarantees.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:practical-byzantine-fault-tolerance", "labels": [ "Practical Byzantine Fault Tolerance", "PracticalByzantineFaultTolerance" ], "is_subclass_of": [ "Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Byzantine Fault Tolerance" ] }, { "id": "practitioner-learning-roadmap-backlog", "title": "Practitioner Learning Roadmap Backlog", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Revision List is a curated, task-tracked syllabus of technologies, frameworks, and concepts that a practitioner aims to learn or revisit, typically structured as a prioritised backlog with completion states. In a data science and AI context, such a list spans programming languages, ML frameworks, cloud platforms, DevOps tooling, and conceptual foundations such as deep learning, NLP, and reinforcement learning. It functions as a personal knowledge-gap audit and learning roadmap.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:practitioner-learning-roadmap-backlog", "labels": [ "Practitioner Learning Roadmap Backlog", "Revision List" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "practitioner-workflow-optimisation-heuristics", "title": "Practitioner Workflow Optimisation Heuristics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Practical heuristics, shortcuts, and workflow optimisations accumulated through practitioner experience with software tools, knowledge-management systems, and AI-assisted development. This category collects actionable guidance that reduces friction in everyday tasks such as Logseq database management, document format conversion, and prompt engineering.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:practitioner-workflow-optimisation-heuristics", "labels": [ "Practitioner Workflow Optimisation Heuristics", "Tips and Tricks" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "github", "LaTeX" ] }, { "id": "pre-production", "title": "Pre Production", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Pre-production is the planning and preparation phase of a creative or software project that occurs before active content creation or development commences. It encompasses conceptual design, narrative structuring, technical specification, asset planning, and prototyping activities that reduce risk and align stakeholder expectations in film, game, and metaverse content pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:pre-production", "labels": [ "Pre Production", "Film Pre-Production", "Pre-Production" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "pre-trained-language-model", "title": "Pre Trained Language Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A pre-trained language model is a neural language model that has first been trained on a large, general corpus using self-supervised objectives such as masked or next-token prediction, then reused as a foundation for many downstream tasks. By learning broadly transferable linguistic and world knowledge during pre-training, it can be adapted with comparatively little task-specific data through fine-tuning, prompting, or instruction tuning. This pre-train-then-adapt paradigm, exemplified by BERT and the GPT family, is the foundation of modern natural language processing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pre-trained-language-model", "labels": [ "Pre Trained Language Model", "Pre-Trained Language Model" ], "is_subclass_of": [ "Language Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "pre-training", "title": "Pre Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The initial training phase where a model learns general representations from large amounts of unlabelled or weakly labelled data before being adapted to specific tasks. Pre-training establishes foundational knowledge that can be transferred across multiple downstream applications, forming the basis for modern large-scale foundation models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:pre-training", "labels": [ "Pre Training", "Language Model Pre-Training", "Large-Scale Pre-Training", "Pre-Training", "Pre-training" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Base models", "Chain of Thought", "Cohere", "documentation", "Ethan Mollick", "few shot", "langchain", "Open Source", "tagging", "Training Modules", "Adoption of Convergent Technologies", "AI Video", "Anthropic Claude", "Apple", "Artificial Intelligence", "Artificial Superintelligence", "ChatGPT", "ComfyUI", "Cortex Agent", "Courses and Training" ] }, { "id": "pre-trained-model", "title": "Pre-Trained Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A pre-trained model is a machine-learning model whose parameters have already been learned on a large, often general-purpose dataset, so that it can be reused as a starting point for downstream tasks. Rather than training from random initialisation, practitioners adapt the pre-trained weights through fine-tuning or use the model directly for inference, transferring learned representations to new problems. Pre-trained models underpin transfer learning and are the practical foundation of modern deep learning across language, vision and multimodal tasks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pre-trained-model", "labels": [ "Pre-Trained Model", "Pre-trained Model" ], "is_subclass_of": [ "Foundation Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "pre-deployment-evaluation", "title": "Pre-deployment Evaluation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pre-deployment evaluation is the structured assessment of an AI model's capabilities, limitations, and risks conducted before it is released or deployed into production use. It typically combines capability benchmarking, red-teaming, and safety testing to detect dangerous capabilities, misuse potential, or unexpected behaviour ahead of exposure to real users. Pre-deployment evaluation is a central mechanism of frontier AI governance frameworks, including commitments made at the Bletchley Declaration.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:pre-deployment-evaluation", "labels": [ "Pre-deployment Evaluation", "Pre-Deployment Evaluation" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "precession", "title": "Precession", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:precession", "labels": [ "Precession" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "precipitation-retrieval", "title": "Precipitation Retrieval", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:precipitation-retrieval", "labels": [ "Precipitation Retrieval" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "precise-point-positioning", "title": "Precise Point Positioning", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:precise-point-positioning", "labels": [ "Precise Point Positioning" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "precision-agriculture", "title": "Precision Agriculture", "domain": "robotics", "domain_name": "Robotics", "definition": "Precision agriculture is a farming-management approach that uses sensing, geolocation, and data analytics to observe and respond to variability within fields at fine spatial resolution. It applies inputs such as water, fertiliser, and pesticide only where and when needed, improving yield and reducing waste. It integrates IoT sensors, satellite and drone imagery, and increasingly autonomous ground robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:precision-agriculture", "labels": [ "Precision Agriculture", "Precision Agriculture System" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "precision-manufacturing", "title": "Precision Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Precision manufacturing is a set of industrial processes and quality systems that produce mechanical components and assemblies within extremely tight dimensional tolerances \u2014 typically sub-micrometre to nanometre scale \u2014 using advanced machining, metrology, and process control techniques. It encompasses CNC machining, ultra-precision grinding, electrical discharge machining (EDM), and micro-fabrication, combined with in-process measurement and statistical process control to ensure conformance to specification. Precision manufacturing is fundamental to the production of optical systems, semiconductor equipment, aerospace components, medical implants, and scientific instruments where dimensional accuracy directly determines system performance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:precision-manufacturing", "labels": [ "Precision Manufacturing", "Precision Engineering" ], "is_subclass_of": [ "Manufacturing Automation" ], "wikilinks": [] }, { "id": "precision-medicine", "title": "Precision Medicine", "domain": "ai", "domain_name": "Ai", "definition": "Precision Medicine utilises artificial intelligence to tailor medical treatment to individual patient characteristics, integrating genomic, proteomic, and clinical data to predict treatment response and stratify patient populations. AI-driven precision medicine enables personalised diagnosis, prognosis, and therapeutic selection based on multi-omic data integration and predictive modelling.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:precision-medicine", "labels": [ "Precision Medicine", "Personalised Medicine", "Personalized Medicine" ], "is_subclass_of": [ "Medical AI" ], "wikilinks": [ "Genomics", "Drug Discovery AI", "Medical AI", "MetaverseDomain" ] }, { "id": "precision-recall-curve", "title": "Precision-Recall Curve", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A diagnostic plot that traces the trade-off between precision and recall across the decision thresholds of a binary classifier, most informative under class imbalance where it focuses performance assessment on the minority positive class.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:precision-recall-curve", "labels": [ "Precision-Recall Curve", "Precision-Recall" ], "is_subclass_of": [ "Confusion Matrix", "Model Performance" ], "wikilinks": [ "Precision", "Recall", "ROC Curve", "F1 Score", "Confusion Matrix" ] }, { "id": "precision", "title": "Precision", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A classification performance metric representing the proportion of positive predictions made by an artificial intelligence model that are actually correct, calculated as the ratio of true positives to all positive predictions (true positives plus false positives), measuring the model's ability to avoid false alarms and providing critical insight into prediction reliability, particularly important in applications where the cost or consequence of false positive errors is significant.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:precision", "labels": [ "Precision" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "False Positive", "Positive Predictive Value", "Precision-Recall Curve", "Specificity", "True Positive", "Accuracy", "Confusion Matrix", "F1 Score", "MetaverseDomain", "Model Performance", "Recall" ] }, { "id": "predicate-logic", "title": "Predicate Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Predicate logic, also called first-order predicate calculus, is a formal system that extends propositional logic with quantifiers and predicates over individual variables, allowing statements about the properties of and relations between objects in a domain of discourse. It introduces the universal and existential quantifiers, enabling expressions such as 'for all x, P(x)' and 'there exists an x such that P(x)'. Predicate logic provides the semantic and syntactic backbone for much of automated reasoning, formal specification, and knowledge representation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:predicate-logic", "labels": [ "Predicate Logic" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "prediction-markets", "title": "Prediction Markets", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Prediction markets are exchanges where participants trade contracts whose payoff depends on the outcome of future events, so the market price reflects an aggregated probability estimate.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:prediction-markets", "labels": [ "Prediction Markets", "Prediction Market" ], "is_subclass_of": [ "Economics", "Economics Domain" ], "wikilinks": [ "Order Book", "DeFi", "Smart Contracts", "Automated Market Maker", "Economics Domain" ] }, { "id": "predictive-analytics", "title": "Predictive Analytics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Predictive Analytics is the discipline of extracting patterns from historical data using statistical algorithms and machine learning models to generate probabilistic forecasts of future events, behaviours, or conditions. It encompasses the full workflow from data ingestion and feature engineering through model selection, training, validation, and operationalised deployment within business intelligence and decision-support systems. The field draws on regression, classification, ensemble methods, and time-series modelling to produce actionable outputs in domains ranging from demand forecasting and risk assessment to anomaly detection and personalised recommendation. It is distinguished from descriptive analytics by its forward-looking orientation and from prescriptive analytics by its focus on prediction rather than optimised action.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:predictive-analytics", "labels": [ "Predictive Analytics" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "predictive-maintenance", "title": "Predictive Maintenance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Predictive Maintenance (PdM) is a condition-based maintenance strategy that uses continuous or periodic monitoring of physical asset health signals\u2014vibration, temperature, acoustic emission, current draw, oil particle count, ultrasound, and corrosion metrics\u2014combined with machine learning inf...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:predictive-maintenance", "labels": [ "Predictive Maintenance" ], "is_subclass_of": [ "AI Application", "Maintenance Strategy", "Condition-Based Maintenance", "Asset Health Management", "Industrial IoT", "Machine Learning Discipline" ], "wikilinks": [ "Aerospace MRO", "AnalyticsLayer", "Anomaly Detection", "Asset Health Management", "Asset Life Extension", "AssetManagementDomain", "C-MAPSS Dataset", "CMMS Integration", "Condition-Based Maintenance", "Convolutional Neural Networks", "EdgeComputingLayer", "EnterpriseIntegrationLayer", "Envelope Analysis", "Fault Classification", "FFT", "Gradient Boosting", "Health Index", "IBM Maximo", "IEC 61499", "IEEE 1451" ] }, { "id": "predictive-modelling", "title": "Predictive Modelling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Predictive modelling is the practice of building statistical or machine-learning models that estimate unknown or future outcomes from observed input features. It spans data preparation, feature engineering, model training and validation, and is evaluated on its generalisation to unseen data. Predictive modelling powers forecasting, classification and risk assessment across applied domains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:predictive-modelling", "labels": [ "Predictive Modelling" ], "is_subclass_of": [ "Machine Learning", "Statistical Modelling" ], "wikilinks": [] }, { "id": "predictive-personalization", "title": "Predictive Personalization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of machine learning models to anticipate individual user preferences and dynamically tailor content, recommendations, interfaces, and experiences before an explicit request is made. Systems combine behavioural analytics, user profiling, and predictive models to deliver contextually relevant personalisation at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:predictive-personalization", "labels": [ "Predictive Personalization" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "predictive-processing", "title": "Predictive Processing", "domain": "ai", "domain_name": "Ai", "definition": "Predictive processing is a theoretical framework in cognitive science and AI in which the brain or agent continuously generates predictions about sensory input and updates internal models to minimise prediction error. Perception, action, and learning are unified as processes of reducing the mismatch between expected and actual signals. It motivates active inference and free-energy approaches to embodied, cognitive systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:predictive-processing", "labels": [ "Predictive Processing", "Predictive Coding" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "preference-aggregation", "title": "Preference Aggregation", "domain": "governance", "domain_name": "Governance", "definition": "The process of combining individual preferences, rankings, or utility functions from multiple stakeholders into a collective or social preference ordering for decision-making purposes. Preference aggregation methods range from simple majority voting to sophisticated social choice mechanisms, each making different assumptions about preference structure and interpersonal comparability. Arrow's impossibility theorem establishes fundamental limits on consistent aggregation. The field is central to voting theory, mechanism design, and collaborative governance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:preference-aggregation", "labels": [ "Preference Aggregation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "preference-data", "title": "Preference Data", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Preference data is a dataset of paired or ranked comparisons in which human annotators indicate which of two or more model outputs they prefer, rather than providing an absolute quality score. It is the primary training signal for reward models used in reinforcement learning from human feedback, since relative judgements are typically easier and more consistent for annotators to produce than calibrated absolute ratings. The quality and diversity of preference data materially shape the behaviour that RLHF instils in a model.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:preference-data", "labels": [ "Preference Data" ], "is_subclass_of": [ "Human Preference" ], "wikilinks": [] }, { "id": "preference-learning", "title": "Preference Learning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A machine learning paradigm that trains models from comparative human judgements (e.g., 'A is better than B') rather than absolute labels or demonstrations, enabling alignment with human values. Preference learning underpins reinforcement learning from human feedback and direct preference optimisation, and is the standard technique for aligning large language models.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:preference-learning", "labels": [ "Preference Learning" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Ethan Mollick", "MetaverseDomain" ] }, { "id": "prefix-tuning", "title": "Prefix Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A parameter-efficient fine-tuning technique that prepends trainable continuous vectors (prefixes) to the key and value matrices at each transformer layer, whilst keeping the pre-trained model parameters frozen. Unlike prompt tuning, which operates only on input embeddings, prefix tuning influences the attention mechanism at every layer, achieving stronger task-specific adaptation with roughly 0.1% of model parameters.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:prefix-tuning", "labels": [ "Prefix Tuning" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "preimage-resistance", "title": "Preimage Resistance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Preimage Resistance is a security property of a cryptographic hash function requiring that, given a hash output h, it is computationally infeasible to find any input m such that H(m) = h. This one-way property is foundational for password hashing, proof-of-work puzzles, and blockchain address derivation, ensuring that knowledge of a public address does not reveal the corresponding private key.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:preimage-resistance", "labels": [ "Preimage Resistance" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "preimage", "title": "Preimage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A preimage is an input value that, when passed through a cryptographic hash function, produces a specified output digest. Preimage resistance, the computational infeasibility of finding such an input given only the digest, is a core security property required of cryptographic hash functions. In protocols such as hash time-locked contracts, revealing the preimage of a previously published hash is used as cryptographic proof that a condition has been satisfied, enabling trustless conditional payments.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:preimage", "labels": [ "Preimage" ], "is_subclass_of": [ "Cryptographic Hash Function" ], "wikilinks": [] }, { "id": "presence-awareness", "title": "Presence Awareness", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Presence awareness is the capability of a collaborative system to track and surface which users are currently active in a shared space or document, along with lightweight state such as their cursor position, selection or focus. It is a foundational building block for real-time collaborative editing, allowing participants to see who else is present and where their attention is directed. Systems typically implement it as a low-latency, frequently updated channel separate from the durable document state itself.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:presence-awareness", "labels": [ "Presence Awareness" ], "is_subclass_of": [ "Awareness" ], "wikilinks": [] }, { "id": "presence-detection", "title": "Presence Detection", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Presence Detection is the real-time determination of whether a user, avatar, or entity is active and spatially located within a digital environment. It combines sensor fusion, computer vision, and network signalling to maintain accurate availability states, enabling responsive social interactions, attention-aware interfaces, and adaptive content delivery in spatial computing systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:presence-detection", "labels": [ "Presence Detection", "PresenceDetection" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "presence-indicator", "title": "Presence Indicator", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A presence indicator is a real-time signal encoding a user's availability, attention, and engagement state in a collaborative system, enabling other participants to determine the optimal channel and urgency of communication before initiating contact.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:presence-indicator", "labels": [ "Presence Indicator" ], "is_subclass_of": [ "Communication Technology", "Ambient Awareness", "Awareness Mechanisms", "Synchronous Collaboration Infrastructure", "Real-Time Communication", "User State Management" ], "wikilinks": [ "Activity Inference", "Activity Monitor", "Activity Monitoring", "AI Prediction", "Always-On Culture", "Ambient Awareness", "Ambient Display", "Authentication", "Availability Scheduler", "Awareness Mechanisms", "Bossware", "Calendar API", "Calendar Data", "Calendar Integration", "Client Activity Tracking", "Communication Orchestration", "Deep Work Protection", "Device Sensors", "DistributedCollaborationDomain", "Distributed State Synchronisation" ] }, { "id": "presence-technology", "title": "Presence Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Presence technology refers to systems and protocols that detect, represent, and communicate the real-time availability and contextual state of users or entities within digital and physical environments. It aggregates signals such as location, device activity, calendar status, and explicit user input to publish a presence indicator consumed by communication, collaboration, and ambient computing applications. In extended reality contexts, presence technology additionally encompasses volumetric capture and avatar fidelity systems that convey embodied social presence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:presence-technology", "labels": [ "Presence Technology" ], "is_subclass_of": [ "Spatial Computing", "Platform and Environment" ], "wikilinks": [] }, { "id": "presence", "title": "Presence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Perceptual state in which a user feels located inside a virtual or mixed environment, experiencing spatial, social, and self presence as a unified phenomenological response to immersive sensory stimulation.", "entityType": "Class", "qualityScore": 0.68, "maturity": "established", "iri": "urn:ngm:class:presence", "labels": [ "Presence", "Conversational Presence", "Microsoft Graph Presence API", "Presence API", "Presence System", "TELE-006-presence", "User Presence" ], "is_subclass_of": [ "Immersive Experience" ], "wikilinks": [ "ACM", "Embodiment", "Engagement", "Field of View", "Frame Rate", "Haptic Device", "Sensory Feedback", "Social Connection", "Visual Display", "ComputeLayer", "Immersive Experience", "InteractionDomain", "Latency", "Self Presence", "Social Presence", "Spatial Presence" ] }, { "id": "presentation-attack-detection", "title": "Presentation Attack Detection", "domain": "security", "domain_name": "Security", "definition": "Presentation attack detection (PAD) is the set of techniques that determine whether a biometric sample is presented by a genuine, live subject or by an artefact intended to spoof the system. Attacks include printed photographs, replayed video, silicone fingerprints, and three-dimensional masks. PAD, also known as liveness or anti-spoofing analysis, is standardised under ISO/IEC 30107 and is essential to the trustworthiness of biometric authentication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:presentation-attack-detection", "labels": [ "Presentation Attack Detection" ], "is_subclass_of": [ "Biometric Authentication" ], "wikilinks": [] }, { "id": "presentation-concluding-synthesis-slide", "title": "Presentation: Conclusion", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The concluding slide deck section of a presentation on generative AI and its societal significance. It synthesises key points about AI's role in society, addresses audience takeaways, and acknowledges contributors, serving as the closing narrative arc for a structured public or academic AI presentation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:presentation-concluding-synthesis-slide", "labels": [ "Presentation: Conclusion" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "presentation-exchange", "title": "Presentation Exchange", "domain": "standards", "domain_name": "Standards", "definition": "A Decentralized Identity Foundation specification defining a data format by which verifiers articulate proof requirements (presentation definitions) and holders describe how submitted credentials satisfy them (presentation submissions). Presentation Exchange is transport-agnostic and credential-format-agnostic, and is the requirement language embedded in protocols such as OpenID for Verifiable Presentations to negotiate selective disclosure of verifiable credentials.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:presentation-exchange", "labels": [ "Presentation Exchange" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard", "Decentralized Identity Foundation", "Verifiable Credentials", "Self Sovereign Identity", "Digital Identity", "Decentralized Identifier" ] }, { "id": "presentation-layer", "title": "Presentation Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Presentation Layer is the topmost stratum of the canonical stack, responsible for rendering application state into a form humans can perceive and act on. It sits directly above the Application Layer and has nothing above it in the stack. It contains user interface components, rendering pipelines, formatting, and interaction handling.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:presentation-layer", "labels": [ "Presentation Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Application Layer", "User Experience Layer", "Human-Computer Interaction", "Information Visualisation", "owl:Thing" ] }, { "id": "pressure-sensor", "title": "Pressure Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "Pressure Sensor is a robotics and autonomous systems concept and a type of robotics.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:pressure-sensor", "labels": [ "Pressure Sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "pretrained-model", "title": "Pretrained Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network or machine learning model that has been trained on a large dataset (often a broad corpus) before being adapted to a specific downstream task through fine-tuning or prompting. Pretrained models encode general representations\u2014linguistic, visual, or multimodal\u2014that can be efficiently transferred, dramatically reducing the data and compute required for specialised applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pretrained-model", "labels": [ "Pretrained Model", "Pretrained Language Model" ], "is_subclass_of": [ "AI Technique", "Model" ], "wikilinks": [ "Artificial Intelligence", "Model" ] }, { "id": "pretrained-weights", "title": "Pretrained Weights", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pretrained Weights are the learned parameter tensors of a neural network obtained by training on a large corpus prior to task-specific adaptation. They encode generalised representations\u2014syntactic, semantic, visual, or domain knowledge\u2014that downstream tasks can exploit through fine-tuning or zero-shot inference, dramatically reducing the data and compute required to achieve strong performance on new problems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:pretrained-weights", "labels": [ "Pretrained Weights" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "pretraining", "title": "Pretraining", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Pretraining is the initial phase of training a model on a large, broad corpus using self-supervised objectives to learn general-purpose representations before any task-specific adaptation. It produces a foundation model whose learned features can be transferred to downstream tasks through fine-tuning or prompting. For language models this typically involves predicting masked or next tokens over vast text collections.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pretraining", "labels": [ "Pretraining" ], "is_subclass_of": [ "Self-Supervised Learning" ], "wikilinks": [] }, { "id": "price-discovery", "title": "Price Discovery", "domain": "finance", "domain_name": "Finance", "definition": "Price discovery is the market mechanism through which asset prices are determined via the continuous interaction of buyers and sellers, incorporating supply/demand dynamics, order flow analysis, bid-ask spread formation, and arbitrage across venues to establish fair market value in real-time.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:price-discovery", "labels": [ "Price Discovery", "Spot Price" ], "is_subclass_of": [ "Market Microstructure" ], "wikilinks": [ "Arbitrage", "Fair Valuation", "FinancialDomain", "Liquidity", "Market Efficiency", "Market Microstructure", "Order Book", "Price Transparency", "MEV" ] }, { "id": "price-oracle", "title": "price oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A price oracle is an on-chain or hybrid data feed that supplies decentralised protocols with reliable, manipulation-resistant market prices for tokens, synthetic assets, and other financial instruments. On-chain price oracles \u2014 such as time-weighted average price (TWAP) feeds derived from automated market maker pool reserves \u2014 are fully decentralised but lag real-time prices. Off-chain oracle networks aggregate prices from multiple centralised and decentralised exchanges before committing them on-chain, offering fresher data at the cost of additional trust assumptions on node operators. Price oracles are foundational to DeFi lending protocols, synthetic asset minting, perpetual futures settlement, and insurance claim adjudication.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:price-oracle", "labels": [ "Price Oracle", "Oracle Price Feeds", "Price Oracle Integration" ], "is_subclass_of": [ "Blockchain Oracle" ], "wikilinks": [] }, { "id": "price-stability", "title": "Price Stability", "domain": "finance", "domain_name": "Finance", "definition": "Price stability is a macroeconomic condition in which the general level of prices for goods and services remains broadly constant over time, with inflation low, stable and predictable. It is the primary mandate of most central banks because it preserves the purchasing power of money, anchors expectations and supports sustainable economic growth. Central banks pursue price stability through monetary policy tools that influence interest rates, money supply and inflation expectations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:price-stability", "labels": [ "Price Stability" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "prime-intellect", "title": "Prime Intellect", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Prime Intellect is an open-source decentralised AI research organisation and GPU compute marketplace founded by Vincent Weisser (CEO, formerly co-founder of VitaDAO and AI lead at Molecule biopharma) and Johannes Hagemann headquartered in San Francisco, operating on 4M total funding across three ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:prime-intellect", "labels": [ "Prime Intellect" ], "is_subclass_of": [ "AI Technique", "AI Research Area", "Distributed Computing", "Federated Learning", "Agent Frameworks", "Decentralised AI", "Open Source AI" ], "wikilinks": [ "Agentic AI", "Akash Network", "AlgorithmLayer", "Apache 2.0 Open Source Licence", "arXiv Preprint Standards", "Asynchronous RL", "Bittensor", "Centralised Training", "Checkpoint Recovery", "Collaborative Model Ownership", "Compute Democratisation", "Compute Exchange", "Decentralised AI", "Decentralised Foundation Model Training", "Decentralised Science", "DiLoCo", "DistributedComputingDomain", "Distributed Training", "ElasticDeviceMesh", "FineWeb-Edu Dataset" ] }, { "id": "prime-meridian", "title": "Prime Meridian", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:prime-meridian", "labels": [ "Prime Meridian" ], "is_subclass_of": [ "Meridian" ], "wikilinks": [] }, { "id": "principal-component-analysis", "title": "Principal Component Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Principal Component Analysis (PCA) is an unsupervised linear technique that transforms correlated variables into a smaller set of uncorrelated components ordered by the variance they capture. The components are the eigenvectors of the data's covariance matrix, and projecting onto the leading components yields a lower-dimensional representation that preserves as much variance as possible. PCA is widely used for dimensionality reduction, noise reduction, visualisation, and feature decorrelation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:principal-component-analysis", "labels": [ "Principal Component Analysis" ], "is_subclass_of": [ "Dimensionality Reduction" ], "wikilinks": [] }, { "id": "principal-agent-problem", "title": "Principal-Agent Problem", "domain": "finance", "domain_name": "Finance", "definition": "The principal-agent problem is the conflict of interest that arises when one party, the agent, is empowered to act on behalf of another, the principal, but the agent's incentives are not perfectly aligned with the principal's interests. It typically emerges under information asymmetry, where the agent has knowledge or discretion the principal cannot fully observe or verify. Corporate governance mechanisms such as monitoring, incentive contracts, and disclosure requirements exist largely to mitigate principal-agent problems.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:principal-agent-problem", "labels": [ "Principal-Agent Problem", "Principal Agent Problem" ], "is_subclass_of": [ "Corporate Governance" ], "wikilinks": [] }, { "id": "prior-distribution", "title": "Prior Distribution", "domain": "ai", "domain_name": "Ai", "definition": "A prior distribution is the probability distribution over unknown quantities that encodes belief before observing data, in Bayesian inference. It is combined with the likelihood via Bayes' theorem to produce the posterior. The choice of prior, ranging from informative to weakly informative or uninformative, encodes assumptions and regularises inference, especially with limited data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:prior-distribution", "labels": [ "Prior Distribution" ], "is_subclass_of": [ "Machine Learning Discipline", "Probability Distribution" ], "wikilinks": [] }, { "id": "priority-ceiling-protocol", "title": "Priority Ceiling Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The priority ceiling protocol is a real-time scheduling and resource-access protocol that prevents unbounded priority inversion and deadlock among tasks sharing mutually exclusive resources. Each resource is assigned a ceiling equal to the highest priority of any task that may lock it, and a task may acquire a resource only if its priority exceeds the ceilings of all currently locked resources. It bounds blocking time, enabling provable schedulability in hard real-time systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:priority-ceiling-protocol", "labels": [ "Priority Ceiling Protocol" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "priority-fee", "title": "Priority Fee", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Priority Fee (also called a miner tip or validator tip) is an optional, user-specified additional payment on top of the base fee in EIP-1559-compatible blockchains, paid directly to the block producer to incentivise preferential inclusion and ordering of a transaction within the next block. By offering a higher tip, users signal urgency and compete for limited block space during periods of network congestion, enabling a market-based transaction prioritisation mechanism. Priority fees are burned alongside the base fee under EIP-1559's fee model only partially \u2014 the base fee is burned while the priority fee flows to the validator, aligning incentives for prompt transaction confirmation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:priority-fee", "labels": [ "Priority Fee" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "priority-queue", "title": "Priority Queue", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A priority queue is an abstract data type in which each element has an associated priority and elements are served in order of priority rather than insertion order. It supports insertion of elements and extraction of the highest- (or lowest-) priority element, and is most commonly implemented with a binary heap. Priority queues underpin many graph and scheduling algorithms where the next item to process is the most urgent one.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:priority-queue", "labels": [ "Priority Queue" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "priority-scheduling", "title": "Priority Scheduling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Priority scheduling is a scheduling discipline in which competing tasks or processes are ordered by an assigned priority value rather than by arrival time, so that higher-priority work pre-empts or is dispatched ahead of lower-priority work. It is typically implemented over a priority queue and is central to real-time computing and robotic control systems, where deadline-critical tasks must run ahead of best-effort work. Variants include static priority, dynamic priority, and rate-monotonic scheduling.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:priority-scheduling", "labels": [ "Priority Scheduling" ], "is_subclass_of": [ "Priority Queue" ], "wikilinks": [] }, { "id": "privacy-architecture", "title": "Privacy Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A systematic design framework that embeds data protection principles into software systems from inception, incorporating privacy-by-design modologies, access controls, anonymization techniques, and compliance mechanisms to safeguard personal information.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:privacy-architecture", "labels": [ "Privacy Architecture" ], "is_subclass_of": [ "Governance and Safety", "System Architecture" ], "wikilinks": [ "Data Privacy", "metaverse", "System Architecture" ] }, { "id": "privacy-budget-management", "title": "Privacy Budget Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Privacy Budget Management is the systematic allocation, tracking, and enforcement of differential privacy parameters (epsilon and delta) across multiple queries or analyses, preventing cumulative privacy loss from exceeding acceptable thresholds over time. It applies composition theorems to bound total privacy expenditure and employs strategies such as fixed allocation, adaptive allocation, and hierarchical budgeting to maximise analytical utility while respecting organisational privacy constraints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:privacy-budget-management", "labels": [ "Privacy Budget Management", "Privacy Budget" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Dwork and Roth (2014)", "Google DP Accounting", "NIST Privacy Framework", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "privacy-by-design", "title": "Privacy By Design", "domain": "governance", "domain_name": "Governance", "definition": "Privacy by Design is a proactive privacy framework and GDPR requirement (Article 25) mandating that data protection be embedded into system architecture, business practices, and technologies from inception rather than bolted on as afterthought, implementing privacy as the default setting. The framework comprises seven foundational principles articulated by Ann Cavoukian \u2014 including proactive prevention, privacy as default, full functionality (positive-sum), end-to-end lifecycle security, visibility and transparency, and respect for user privacy \u2014 requiring engineers and product teams to treat privacy as a core design constraint from the earliest specification stage.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:privacy-by-design", "labels": [ "Privacy By Design", "Data Protection by Design", "Privacy by Design", "Privacy-by-Design" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [ "Cavoukian (2009)", "GDPR Article 25", "ISO 29100", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "privacy-coin", "title": "Privacy Coin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Privacy Coin is a cryptocurrency engineered to conceal the sender, receiver, and amount of transactions, breaking the on-chain linkability that characterises transparent ledgers. Such coins employ techniques like ring signatures, zero-knowledge proofs, confidential transactions, and stealth addresses to deliver fungibility and transaction privacy. Examples in the public domain include Monero and Zcash, which represent contrasting cryptographic approaches to the same confidentiality goal.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:privacy-coin", "labels": [ "Privacy Coin" ], "is_subclass_of": [ "Cryptocurrency" ], "wikilinks": [] }, { "id": "privacy-controls", "title": "Privacy Controls", "domain": "security", "domain_name": "Security", "definition": "Privacy Controls are the technical, organisational, and procedural mechanisms that organisations deploy to protect individuals' personal data from unauthorised access, misuse, or disclosure, and to provide data subjects with meaningful agency over how their information is collected and used. They encompass access restrictions, anonymisation and pseudonymisation techniques, consent management interfaces, data minimisation policies, encryption, and privacy impact assessment processes. Effective privacy controls implement privacy by design principles, embedding data protection into systems from inception rather than applying it retrospectively. Regulatory frameworks such as GDPR mandate specific control categories and assign accountability for their implementation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:privacy-controls", "labels": [ "Privacy Controls" ], "is_subclass_of": [ "Privacy" ], "wikilinks": [] }, { "id": "privacy-disclosure", "title": "Privacy Disclosure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The formal process by which platforms communicate to users what personal data is collected, how it is used, who it is shared with, and what rights users hold over it. In spatial computing and metaverse contexts, privacy disclosure must address novel data types including biometric signals, eye tracking, motion data, and behavioural analytics gathered in immersive environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:privacy-disclosure", "labels": [ "Privacy Disclosure", "Privacy Notice" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "privacy-engineering", "title": "Privacy Engineering", "domain": "security", "domain_name": "Security", "definition": "Privacy Engineering is the systematic application of engineering methods to translate privacy principles and regulatory requirements into concrete technical controls embedded within systems and processes. It operationalises concepts such as data minimisation, purpose limitation, and consent management through design patterns, threat models, and measurable privacy metrics. Techniques include differential privacy for statistical disclosures, homomorphic encryption for computation on sensitive data, and k-anonymity for dataset release. The discipline bridges legal obligations\u2014particularly GDPR and similar frameworks\u2014with software architecture and data pipeline design, treating privacy as a quality attribute alongside performance and security.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:privacy-engineering", "labels": [ "Privacy Engineering" ], "is_subclass_of": [ "Privacy By Design" ], "wikilinks": [] }, { "id": "privacy-enhancing-technologies", "title": "Privacy Enhancing Technologies", "domain": "security", "domain_name": "Security", "definition": "Privacy Enhancing Technologies (PETs) are a family of cryptographic and systems-engineering techniques designed to minimise the collection, use, and disclosure of personal data while still enabling legitimate data processing for analytics, machine learning, and regulatory compliance. The family encompasses Zero-Knowledge Proofs, Differential Privacy, Homomorphic Encryption, Secure Multi-Party Computation, Trusted Execution Environments, anonymisation pipelines, pseudonymisation, and synthetic data generation. PETs implement the data-minimisation and privacy-by-design principles mandated by frameworks such as GDPR and the UK Data Protection Act 2018. They are increasingly deployed by financial institutions, healthcare providers, and government agencies to unlock the utility of sensitive data without exposing individual records, and are recognised by the UK ICO, European Data Protection Board, and US NIST as essential infrastructure for trustworthy data ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:privacy-enhancing-technologies", "labels": [ "Privacy Enhancing Technologies", "Privacy Enhancing Technology", "Privacy-Enhancing Technologies", "Privacy-Enhancing Technology" ], "is_subclass_of": [ "Privacy Preserving Technology" ], "wikilinks": [] }, { "id": "privacy-filter", "title": "Privacy Filter", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A sidecar service (ADR-008) that sanitises agent outputs before emission to Nostr Relay|Nostr relays or Federation Surface|federation surfaces, redacting personally identifiable information (PII), secrets, and regulated data, whilst maintaining data utility for downstream consumers and le...", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:privacy-filter", "labels": [ "Privacy Filter" ], "is_subclass_of": [ "Security and Identity", "Security Layer" ], "wikilinks": [ "ADR-008", "AgenticSystemsDomain", "CCPA Regulation", "Classification Rules", "Compliance with CCPA", "Compliance with GDPR", "Data Anonymisation", "DataGovernanceDomain", "FilteringLayer", "GDPR Regulation", "HIPAA Regulation", "Machine Learning (optional)", "Nostr Relay", "Nostr Relay", "Pattern Matching", "PII Detection", "PII Detection Best Practices", "PrivacyDomain", "Privacy Policy", "Privacy Preservation" ] }, { "id": "privacy-framework", "title": "Privacy Framework", "domain": "governance", "domain_name": "Governance", "definition": "A Privacy Framework is a structured system of policies, technical controls, legal obligations, and operational procedures that collectively govern the lifecycle of personal and behavioural data across collection, storage, processing, sharing, and deletion. Such frameworks operationalise principles including data minimisation, purpose limitation, user consent, accountability, and privacy-by-design, translating regulatory instruments such as GDPR and CCPA into enforceable organisational and engineering practice. In extended-reality, metaverse, and AI-driven environments, privacy frameworks must additionally govern high-sensitivity data streams including biometric signals, spatial telemetry, gaze patterns, and social-graph interactions that have no direct precedent in conventional web privacy regimes. Mature frameworks combine technical mechanisms such as differential privacy, federated learning, zero-knowledge proofs, and homomorphic encryption with governance structures including data-protection impact assessments, privacy officers, and incident-response procedures.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:privacy-framework", "labels": [ "Privacy Framework" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "privacy-impact-assessment-pia", "title": "Privacy Impact Assessment (PIA)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic eprocess that identifies and assesses privacy risks arising from the processing of personal data in metaverse systems, ensuring compliance with data protection regulations and ical standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:privacy-impact-assessment-pia", "labels": [ "Privacy Impact Assessment (PIA)" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Compliance Verification System", "Data Flow Mapping", "Data Protection Management System", "GDPR Compliance", "Impact Analysis Framework", "ISO 29134", "Mitigation Strategy Generator", "Organizational Privacy Policy", "Personal Data Inventory", "Privacy Governance Framework", "Privacy Requirements", "Regulatory Compliance Database", "Risk Assessment Methodology", "Risk Identification Module", "Stakeholder Consultation Engine", "User Trust", "Legal Framework", "MiddlewareLayer", "Privacy By Design", "Risk Mitigation" ] }, { "id": "privacy-impact-assessment", "title": "Privacy Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A structured evaluation process identifying, analysing, and mitigating privacy risks associated with data processing activities, particularly AI systems handling personal information. Mandated by GDPR Article 35 for high-risk processing, it covers necessity assessment, risk identification, severity evaluation, and documentation of technical and organisational controls.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:privacy-impact-assessment", "labels": [ "Privacy Impact Assessment" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [ "GDPR Article 35", "ICO DPIA Code", "ISO 29134", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "privacy-law", "title": "Privacy Law", "domain": "governance", "domain_name": "Governance", "definition": "The body of constitutional provisions, statutes, regulations, and case law that protects individuals' informational and personal privacy \u2014 governing how organisations and states may collect, process, retain, share, and surveil personal data \u2014 anchored internationally in instruments such as Article 8 of the European Convention on Human Rights and operationalised by data protection regimes including the EU and UK GDPR, the California Consumer Privacy Act, and sectoral statutes, which grant data subjects enforceable rights and impose accountability obligations backed by supervisory authorities and substantial penalties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:privacy-law", "labels": [ "Privacy Law" ], "is_subclass_of": [ "Legal Framework" ], "wikilinks": [ "Legal Framework", "GDPR", "Data Protection", "Data Privacy" ] }, { "id": "privacy-mechanism", "title": "Privacy Mechanism", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Privacy Mechanism is a technical method or protocol designed to preserve individuals' data privacy during data collection, processing, or publication. Examples include differential privacy, federated learning, and homomorphic encryption, each providing mathematically grounded guarantees against disclosure of sensitive information.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:privacy-mechanism", "labels": [ "Privacy Mechanism" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "privacy-policy", "title": "privacy policy", "domain": "governance", "domain_name": "Governance", "definition": "A privacy policy is a formal legal disclosure instrument through which an organisation communicates to data subjects the categories of personal data collected, the purposes and legal bases for processing, retention periods, third-party sharing arrangements, and individual rights available under applicable regulations. Such documents are mandated by data protection frameworks including the EU General Data Protection Regulation (GDPR), the UK Data Protection Act 2018, and the California Consumer Privacy Act (CCPA), each requiring clear, accessible language and prominent placement at the point of data collection to constitute valid notice. Privacy policies function both as a transparency mechanism for individuals and as a compliance artefact evidencing an organisation's accountability obligations. As AI systems introduce novel data flows\u2014inference logging, training on user data, automated decision-making\u2014privacy policies must be continuously updated to reflect these processing activities accurately.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:privacy-policy", "labels": [ "Privacy Policy" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "privacy-preservation", "title": "Privacy Preservation", "domain": "security", "domain_name": "Security", "definition": "Privacy preservation is the practice and set of techniques for protecting personal or sensitive information against unauthorised access, inference, or disclosure while still permitting useful processing. It spans cryptographic methods, data minimisation, anonymisation, and policy controls. It is foundational to compliant data systems and to trust in digital services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:privacy-preservation", "labels": [ "Privacy Preservation" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "privacy-preserving-analytics", "title": "Privacy Preserving Analytics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Privacy Preserving Analytics (PPA) is a cross-disciplinary field at the intersection of cryptography, statistics, and data engineering concerned with enabling quantitative analysis of datasets containing sensitive personal or proprietary information without exposing the underlying individual-leve...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:privacy-preserving-analytics", "labels": [ "Privacy Preserving Analytics", "Privacy-Preserving Analytics" ], "is_subclass_of": [ "AI Governance and Ethics", "Data Analysis", "Privacy Enhancing Technologies", "Machine Learning Discipline", "Statistical Computing", "Information Security" ], "wikilinks": [ "Abadi et al 2016 Deep Learning with Differential Privacy", "Abowd 2018 US Census Differential Privacy", "AI Model Training on Sensitive Data", "AlgorithmLayer", "Algorithmic Fairness", "BFV Scheme", "Carlini et al 2021 LLM Training Data Extraction", "CKKS Scheme", "Clinical Trial Data Sharing", "Computational Complexity Theory", "Corrigan-Gibbs & Boneh 2017 PRIO Protocol", "Cross-Institutional Data Collaboration", "Cryptographic Primitives", "CryptographyDomain", "Data Aggregation", "Data Analysis", "Data Governance Framework", "DataScienceDomain", "Dwork & Roth 2014 Algorithmic Foundations of Differential Privacy", "Erlingsson et al 2014 RAPPOR Google Chrome LDP" ] }, { "id": "privacy-preserving-blockchain", "title": "Privacy Preserving Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain architectures and cryptographic protocol suites that conceal transaction metadata \u2014 sender identity, recipient identity, transferred amounts, and smart contract state \u2014 whilst preserving network verifiability, immutability, and auditability for authorised parties through an ensemble of...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:privacy-preserving-blockchain", "labels": [ "Privacy Preserving Blockchain", "BC-0431-privacy-preserving-blockchain", "Blockchain Privacy", "Privacy-Preserving Blockchain", "Privacy-Preserving Ledger" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain", "Blockchain Network", "Cryptographic Protocol", "Privacy Technology", "Distributed Ledger Technology", "Zero-Knowledge System" ], "wikilinks": [ "Action Spending Authority", "Aleo", "Anoma", "Anonymous Transactions", "Anti-Money Laundering", "Ariel Gabizon", "Aztec", "Aztec Protocol", "BAE Systems Applied Intelligence", "Baseline Protocol", "Beam", "Bisq", "BLAKE3", "Blockchain Technology Laboratory", "Brakedown", "Bulletproofs+", "Bulletproofs", "Bulletproofs Range Proof", "Celestia", "Centre for Cryptocurrency Research and Engineering" ] }, { "id": "privacy-preserving-data-mining", "title": "Privacy Preserving Data Mining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Privacy-Preserving Data Mining (PPDM) is a research field and collection of techniques that enable extraction of useful knowledge patterns\u2014including association rules, classifiers, clusters, and outliers\u2014from datasets while preventing disclosure of sensitive individual records. PPDM methods span data perturbation (noise addition, randomisation, synthetic data generation before mining), cryptographic protocols (secure multi-party computation for distributed pattern discovery, homomorphic encryption for encrypted operations), anonymisation transformations (k-anonymity, l-diversity, t-closeness applied before dataset release), and differential-privacy query mechanisms that inject calibrated noise into published pattern outputs. The field navigates an inherent privacy\u2013utility trade-off: stronger privacy guarantees typically reduce pattern accuracy, and compositions of multiple analyses risk cumulative information leakage through inference and re-identification attacks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:privacy-preserving-data-mining", "labels": [ "Privacy Preserving Data Mining" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Agrawal and Srikant (2000)", "GDPR Article 9", "ISO/IEC TR 24027", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "privacy-preserving-data-sharing", "title": "Privacy Preserving Data Sharing", "domain": "data", "domain_name": "Data", "definition": "Privacy Preserving Data Sharing (PPDS) encompasses the set of cryptographic, statistical, and algorithmic techniques that allow multiple parties to exchange, query, or jointly analyse data without disclosing raw sensitive records. Core mechanisms include differential privacy, secure multi-party computation, homomorphic encryption, federated learning, and synthetic data generation, each providing formal or empirical guarantees that individual-level information cannot be inferred. PPDS enables collaborative analytics, AI model training, and regulatory reporting across organisational and jurisdictional boundaries while satisfying privacy regulations such as GDPR and HIPAA. It is a foundational discipline at the intersection of cryptography, distributed systems, and machine learning, increasingly deployed in healthcare, finance, and cross-industry data-sharing consortia.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-data-sharing", "labels": [ "Privacy Preserving Data Sharing", "Privacy-Preserving Data Sharing" ], "is_subclass_of": [ "Data Sharing" ], "wikilinks": [ "Collaborative Analytics", "Data Sharing", "metaverse" ] }, { "id": "privacy-preserving-technology", "title": "Privacy Preserving Technology", "domain": "security", "domain_name": "Security", "definition": "Privacy Preserving Technology (PPT) is the family of cryptographic and computational methods that enable analysis, sharing, and machine learning on sensitive data without exposing raw individual records. Core paradigms include differential privacy (injecting calibrated noise into query outputs), federated learning (training models across distributed silos without centralising data), homomorphic encryption (computing directly on ciphertext), secure multi-party computation (joint computation among mutually distrusting parties), and zero-knowledge proofs (demonstrating a statement's truth without revealing the witness). Together these techniques form the technical foundation for privacy-by-design engineering under regulatory regimes such as GDPR, CCPA, and the EU AI Act, and are increasingly integral to trusted AI pipelines, healthcare analytics, and decentralised identity systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-technology", "labels": [ "Privacy Preserving Technology", "Privacy-Preserving Technology" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "privacy-protection", "title": "Privacy Protection", "domain": "governance", "domain_name": "Governance", "definition": "Privacy Protection encompasses the legal frameworks, technical mechanisms, and organisational practices deployed to safeguard individuals' rights to control the collection, use, storage, and disclosure of their personal data. It spans regulatory instruments such as the GDPR, sector-specific legislation, and constitutional provisions, alongside technical controls including data minimisation, pseudonymisation, encryption, access control, and privacy-enhancing computation. Effective privacy protection requires privacy-by-design principles to be embedded in system architecture from inception rather than retrofitted. It is a foundational requirement for trust in digital services, particularly where sensitive personal, biometric, or health data are processed.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:privacy-protection", "labels": [ "Privacy Protection" ], "is_subclass_of": [ "Data Protection" ], "wikilinks": [] }, { "id": "privacy-regulation", "title": "Privacy Regulation", "domain": "data", "domain_name": "Data", "definition": "Privacy regulation refers to the body of laws and rules that govern how organisations collect, process, store, and share personal data, granting individuals rights over information about them. Such regulation establishes lawful bases for processing, mandates transparency and security obligations, and provides for enforcement and penalties for non-compliance. Prominent examples set global benchmarks for data subject rights, consent, and accountability, shaping how digital products handle personal information across jurisdictions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:privacy-regulation", "labels": [ "Privacy Regulation" ], "is_subclass_of": [ "Data Protection Regulation" ], "wikilinks": [] }, { "id": "privacy-requirements", "title": "Privacy Requirements", "domain": "governance", "domain_name": "Governance", "definition": "Privacy requirements are the documented obligations and constraints a system must satisfy to protect personal data, derived from law, regulation, contracts, and organisational policy. They specify what data may be collected, how it is processed and retained, and the rights afforded to data subjects. They drive system design, data-flow controls, and privacy impact assessments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:privacy-requirements", "labels": [ "Privacy Requirements" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "privacy-utility-tradeoffs", "title": "Privacy Utility Tradeoffs", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Privacy-Utility Tradeoffs represent the fundamental tension in privacy-preserving AI between privacy preservation\u2014protecting sensitive information through techniques such as differential privacy, anonymisation, or encryption\u2014and model utility, which encompasses accuracy and other performance metrics necessary for effective decision-making. This tension is characterised by Pareto frontiers of achievable (privacy, utility) pairs where strengthening privacy typically degrades model performance and vice versa.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:privacy-utility-tradeoffs", "labels": [ "Privacy Utility Tradeoffs" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Dwork and Roth (2014)", "Narayanan and Shmatikov (2008)", "NIST Privacy Framework", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "privacy-and-data-governance", "title": "Privacy and Data Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Privacy and Data Governance is a trustworthiness dimension ensuring AI systems protect personal information, respect data rights, maintain data quality, and implement appropriate access controls throughout data collection, processing, storage, and sharing activities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:privacy-and-data-governance", "labels": [ "Privacy and Data Governance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EDPB Opinion 28/2024", "GDPR", "ISO/IEC 27701", "AIEthicsDomain", "ConceptualLayer", "EU AI Act" ] }, { "id": "privacy-enhancing-computation-pec", "title": "Privacy-Enhancing Computation (PEC)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computational techniques that enable data processing and analysis while preserving privacy through cryptographic mods such as homomorphic encryption, secure multi-party computation, and differential privacy.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:privacy-enhancing-computation-pec", "labels": [ "Privacy-Enhancing Computation (PEC)" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Cryptographic Primitives", "ENISA 2024", "Homomorphic Encryption", "ISO 27559", "Key Management", "NIST PEC Guidelines", "Privacy-Compliant Processing", "Privacy Models", "Secure Computation Protocols", "Zero-Knowledge Proofs", "Confidential Computing", "DataLayer", "Differential Privacy", "MiddlewareLayer", "Privacy Architecture", "Privacy-Preserving Analytics", "Secure Data Sharing", "Secure Multi-Party Computation", "Security Framework", "TrustAndGovernanceDomain" ] }, { "id": "privacy-preserving-ai", "title": "Privacy-Preserving AI", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Privacy-Preserving AI is a family of machine-learning techniques and system architectures that enable models to be trained, validated, and deployed without exposing raw personal or sensitive data to any single party. Core mechanisms include federated learning, differential privacy, homomorphic encryption, and secure multi-party computation, each offering distinct trade-offs between privacy guarantees, computational cost, and model utility. The discipline addresses regulatory requirements (GDPR, HIPAA) as well as ethical imperatives around data minimisation and individual autonomy. By decoupling learning from data centralisation, Privacy-Preserving AI enables collaborative intelligence across organisational and jurisdictional boundaries that would otherwise be closed to data sharing.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:privacy-preserving-ai", "labels": [ "Privacy-Preserving AI" ], "is_subclass_of": [ "Privacy-Preserving" ], "wikilinks": [ "Privacy-Preserving", "Machine Learning", "Federated Learning", "AI Governance" ] }, { "id": "privacy-preserving-authentication", "title": "Privacy-Preserving Authentication", "domain": "security", "domain_name": "Security", "definition": "Privacy-preserving authentication encompasses cryptographic mechanisms that allow a party to prove identity, membership, or credential possession to a verifier without revealing the underlying identity attributes, credential content, or linkage information across sessions. Core techniques include zero-knowledge proofs, anonymous credentials (e.g., U-Prove, BBS+), blind signatures, and selective disclosure, enabling authentication that is both unforgeable and unlinkable. The goal is to satisfy verifier assurance requirements while minimising the personal data exposed in each authentication event.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-authentication", "labels": [ "Privacy-Preserving Authentication", "Privacy Preserving Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "privacy-preserving-compliance", "title": "Privacy-Preserving Compliance", "domain": "governance", "domain_name": "Governance", "definition": "Privacy-preserving compliance is the use of cryptographic and selective-disclosure techniques to satisfy regulatory requirements such as KYC, AML, and audit without exposing the underlying personal data. It lets a party prove a fact (e.g. age, jurisdiction, sanction-list status) to a regulator or counterparty while revealing nothing more. It is central to reconciling decentralised identity and digital-currency systems with financial regulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-compliance", "labels": [ "Privacy-Preserving Compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "privacy-preserving-computation", "title": "Privacy-Preserving Computation", "domain": "security", "domain_name": "Security", "definition": "Privacy-preserving computation is a family of techniques that allow data to be processed or analysed without exposing the underlying plaintext to the computing party. It includes homomorphic encryption, secure multiparty computation, trusted execution environments, federated learning, and differential privacy. These methods enable collaboration and analytics over sensitive data while maintaining confidentiality.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-computation", "labels": [ "Privacy-Preserving Computation", "Privacy Preserving Computation" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "privacy-preserving-dynamic-creative-optimisation", "title": "Privacy-Preserving Dynamic Creative Optimisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Client-side Dynamic Creative Optimisation (DCO) is an approach where personalised advertising content is generated and matched on the user's device rather than on centralised servers. Using locally maintained preference hashes and decentralised distribution protocols (such as Nostr), the system delivers contextually relevant content without exposing personal identity data to advertisers.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-dynamic-creative-optimisation", "labels": [ "Privacy-Preserving Dynamic Creative Optimisation", "Client side DCO" ], "is_subclass_of": [ "Computing and Cloud", "Decentralised Web" ], "wikilinks": [ "Nostr", "Decentralised Web", "Hardware and Edge", "Hyper personalisation", "latent space", "Multimodal", "Nostr protocol", "NVIDIA Omniverse", "Politics, Law, Privacy", "Training and fine tuning" ] }, { "id": "privacy-preserving-identity", "title": "Privacy-Preserving Identity", "domain": "security", "domain_name": "Security", "definition": "Privacy-preserving identity refers to identity management architectures and protocols that enable individuals to prove attributes, credentials, or facts about themselves to verifiers without revealing unnecessary personal information, leveraging cryptographic techniques such as zero-knowledge proofs, selective disclosure, and unlinkable credentials. These systems reconcile strong authentication guarantees with user privacy, countering surveillance by minimising the attack surface of identity data aggregation and preventing cross-context correlation of user activity. They are foundational to self-sovereign identity frameworks, anonymous credential schemes, hardware-backed identity wallets, and privacy-respecting regulatory compliance workflows. The field spans cryptographic research, standardisation bodies (W3C, IETF, ISO), and deployment infrastructure including secure enclaves and decentralised ledgers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-identity", "labels": [ "Privacy-Preserving Identity" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "privacy-preserving-machine-learning", "title": "Privacy-Preserving Machine Learning", "domain": "security", "domain_name": "Security", "definition": "Privacy-preserving machine learning is the set of techniques that train, evaluate, and serve machine-learning models while limiting exposure of sensitive training data and model internals. It combines cryptographic protocols, statistical guarantees, and distributed training architectures to bound what an adversary can learn about individual records. The goal is to retain predictive utility while satisfying confidentiality, regulatory, and trust constraints.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-machine-learning", "labels": [ "Privacy-Preserving Machine Learning" ], "is_subclass_of": [ "Privacy-Enhancing Technologies" ], "wikilinks": [] }, { "id": "privacy-preserving-protocol", "title": "Privacy-Preserving Protocol", "domain": "security", "domain_name": "Security", "definition": "A privacy-preserving protocol is a cryptographic protocol designed to let parties achieve a useful outcome while revealing as little personal or sensitive data as possible. Such protocols use techniques like zero-knowledge proofs, commitments, blind signatures, secure multiparty computation and differential privacy to prove claims, transact or compute jointly without disclosing the underlying inputs. They are central to confidential payments, anonymous credentials, private identity and any system that must reconcile verifiability with data minimisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:privacy-preserving-protocol", "labels": [ "Privacy-Preserving Protocol" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "privacy-preserving", "title": "Privacy-Preserving", "domain": "security", "domain_name": "Security", "definition": "Privacy-preserving describes methods and systems designed to perform useful computation or data analysis while limiting exposure of the underlying personal or sensitive data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:privacy-preserving", "labels": [ "Privacy-Preserving", "Privacy Preserving", "Privacy Preserving Computing", "Privacy Preserving ML", "Privacy-Preserving ML", "Privacy-Preserving Payment", "Privacy-Preserving Payments", "Privacy-Preserving Transactions" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Cryptography", "Privacy-Preserving AI", "Federated Learning" ] }, { "id": "privacy", "title": "Privacy", "domain": "governance", "domain_name": "Governance", "definition": "The protection of personal information and individual autonomy in digital and AI systems, encompassing data minimization, purpose limitation, transparency, consent management, and individual control over how personal data is collected, processed, stored, and shared; grounded in legal frameworks (GDPR, AI Act) and technical mechanisms (encryption, differential privacy, zero-knowledge proofs) across the full data lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:privacy", "labels": [ "Privacy", "Privacy Technique" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [ "back2002hashcash", "bayer2023artificial", "burnham1983rise; @chaum1985security", "callas1998openpgp", "dai1998b", "dwork1992pricing; @jakobsson1999proofs", "Goldenfein Mann 2024", "harari2014sapiens", "lavoie1990prefatory", "Nakamoto2008", "Nostr", "o2021god", "rosenbergmanipulation", "salinCosts; @cypherPunkMailList", "swartz2008guerilla", "szabo1997formalizing", "Cyber Security and Military", "cypherpunk", "Death of the Internet", "Decentralised Web" ] }, { "id": "private-5g-network", "title": "Private 5G Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A private 5G network is a standalone or non-standalone 5G deployment operated within a defined premises, such as a factory, campus or port, using licensed, shared or unlicensed spectrum dedicated to that organisation rather than a public mobile operator. It gives the operator direct control over coverage, latency, security and quality-of-service guarantees, including network slicing tailored to specific industrial applications. It is commonly deployed where deterministic low latency is required, such as in robotics and machine vision.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:private-5g-network", "labels": [ "Private 5G Network" ], "is_subclass_of": [ "5G Network" ], "wikilinks": [] }, { "id": "private-blockchain", "title": "Private Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A private blockchain is a permissioned distributed ledger in which access, participation, and validation rights are restricted to a pre-approved set of known entities, typically operated by a single organisation or a closed consortium. Unlike public blockchains, private blockchains prioritise throughput, deterministic finality, and data confidentiality over open participation, relying on governance frameworks and identity management rather than anonymous proof-of-work mining to secure the network.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:private-blockchain", "labels": [ "Private Blockchain", "Private Chain" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "AML", "AustralianSecuritiesExchange", "CBDC", "Chainlink", "Clique", "Cosmos", "DApp", "DeFi", "Decentralized", "DeutscheBank", "DigitalPound", "ECB", "EnergyTrading", "EVM", "FCA", "FDA", "GasFee", "GDPR", "HashTimeLock", "HIPAA" ] }, { "id": "private-channels", "title": "Private Channels", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Confidential transaction subsets in PermissionedBlockchain|permissioned blockchains where designated participants conduct transactions invisibly to other network members, implementing Encryption|encryption and AccessControl|access control to segregate sensitive business data whilst ma...", "entityType": "Class", "qualityScore": 0.4, "maturity": "draft", "iri": "urn:ngm:class:private-channels", "labels": [ "Private Channels", "BC-0430-private-channels", "PrivateChannels" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BC-0315-zero-knowledge-proof", "BC-0316-secure-multi-party-computation", "BC-0426-hyperledger-fabric", "BC-0427-hyperledger-besu", "BC-0429-permissioned-blockchain", "BC-0431-privacy-preserving-blockchain", "Encryption", "Encryption", "AccessControl", "BlockchainDomain", "ConsensusProtocol", "HyperledgerFabric", "Immutability", "PermissionedBlockchain", "SmartContract" ] }, { "id": "private-inference", "title": "Private Inference", "domain": "security", "domain_name": "Security", "definition": "Private inference is the execution of a machine learning model's forward pass such that neither the input data nor the model's internal state is revealed to the party running the computation, preserving confidentiality of both user data and, where required, model weights. It is commonly implemented using confidential computing hardware such as trusted execution environments, which isolate computation from the host operating system. It enables sensitive applications, such as processing medical or financial data, to use cloud-hosted models without exposing raw inputs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:private-inference", "labels": [ "Private Inference" ], "is_subclass_of": [ "Confidential Computing" ], "wikilinks": [] }, { "id": "private-key", "title": "Private Key", "domain": "security", "domain_name": "Security", "definition": "A Private Key is a secret scalar value generated through cryptographically secure randomness that serves as the sole proof of ownership within asymmetric cryptographic systems. In blockchain contexts it authorises transaction signing via digital signatures, controls access to associated public addresses and funds, and underpins the non-custodial security model where loss or exposure is irreversible.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:private-key", "labels": [ "Private Key", "Spend Key" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "private-set-intersection", "title": "Private Set Intersection", "domain": "security", "domain_name": "Security", "definition": "Private set intersection (PSI) is a cryptographic protocol that allows two or more parties to compute the intersection of their private sets without revealing any element outside that intersection to one another. It is built from techniques such as oblivious transfer, homomorphic encryption, or Bloom filter-based hashing, and is a specialised instance of secure multi-party computation restricted to the set-intersection function. PSI is used in applications such as contact discovery, privacy-preserving advertising measurement, and cross-organisation fraud detection, where parties need to find common records without pooling raw data. Its efficiency has improved substantially with modern oblivious transfer extension protocols, making PSI practical at scale for sets with millions of elements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:private-set-intersection", "labels": [ "Private Set Intersection" ], "is_subclass_of": [ "Multi-Party Computation" ], "wikilinks": [] }, { "id": "private-smart-contract", "title": "Private Smart Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A private smart contract is a smart contract whose inputs, state or logic are concealed from public view while still being verifiably executed and settled on a blockchain. It typically relies on zero-knowledge proofs or confidential computing so that correctness can be checked without revealing the underlying data. Private smart contracts let parties transact programmable agreements with confidentiality comparable to traditional finance while retaining the auditability and finality of public ledgers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:private-smart-contract", "labels": [ "Private Smart Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "privilege-escalation", "title": "Privilege Escalation", "domain": "security", "domain_name": "Security", "definition": "Privilege escalation is the act of gaining rights beyond those originally granted, allowing an attacker or process to perform actions reserved for higher-trust principals. Vertical escalation moves from a lower to a higher privilege level, while horizontal escalation moves laterally to another principal at the same level. It is a pivotal phase in attack chains, typically exploiting misconfiguration, flawed access control or software vulnerabilities.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:privilege-escalation", "labels": [ "Privilege Escalation" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "privileged-access-management", "title": "Privileged Access Management", "domain": "security", "domain_name": "Security", "definition": "Privileged Access Management (PAM) is a cybersecurity discipline and category of technology solutions that controls, monitors, and audits the access rights of users, accounts, and systems with elevated permissions in IT environments. PAM encompasses vaulting of privileged credentials, just-in-time access provisioning, session recording, and anomaly detection for privileged sessions. It addresses the risk that compromised administrator accounts represent the most damaging attack vector in enterprise breaches. PAM solutions enforce the principle of least privilege and provide forensic audit trails required by compliance frameworks such as ISO/IEC 27001 and SOX.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:privileged-access-management", "labels": [ "Privileged Access Management", "Privilege Access Management" ], "is_subclass_of": [ "Identity and Access Management" ], "wikilinks": [] }, { "id": "pro-worker-ai", "title": "Pro-Worker AI", "domain": "economics", "domain_name": "Economics", "definition": "A category of artificial intelligence development and deployment strategies designed to augment human labor and create new tasks rather than primarily displacing existing jobs.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:pro-worker-ai", "labels": [ "Pro-Worker AI" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "probabilistic-finality", "title": "Probabilistic Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Probabilistic finality is a transaction confirmation model characteristic of longest-chain consensus mechanisms where confidence in transaction irreversibility increases exponentially with block depth, approaching but never reaching absolute certainty, with reversal probability decaying as (q/p)^...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:probabilistic-finality", "labels": [ "Probabilistic Finality", "Nakamoto-Style Probabilistic Finality" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain", "Transaction Finality" ], "wikilinks": [ "51% Attack", "Blockchain", "BlockchainDomain", "Consensus Mechanism", "Deterministic Finality", "Longest Chain Rule", "Proof of Work", "Transaction Confirmation", "Transaction Finality" ] }, { "id": "probabilistic-forecasting", "title": "Probabilistic Forecasting", "domain": "ai", "domain_name": "Ai", "definition": "Probabilistic forecasting predicts future quantities as full probability distributions rather than single point estimates, quantifying uncertainty in the prediction. Outputs are typically expressed as predictive intervals, quantiles, or samples. It is essential for risk-aware decision-making in domains such as demand planning, energy, and finance where the cost of error is asymmetric.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:probabilistic-forecasting", "labels": [ "Probabilistic Forecasting", "Probabilistic Forecast" ], "is_subclass_of": [ "Machine Learning Discipline" ], "wikilinks": [] }, { "id": "probabilistic-inference", "title": "Probabilistic Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Probabilistic inference is the computation of the probability of unknown quantities given observed evidence within a probabilistic model. It produces posterior distributions used for prediction and decision making under uncertainty.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:probabilistic-inference", "labels": [ "Probabilistic Inference" ], "is_subclass_of": [ "AI Technique", "Bayesian Inference" ], "wikilinks": [ "Probabilistic Model", "Bayesian Inference", "Uncertainty Quantification", "Variational Inference", "Markov Chain Monte Carlo" ] }, { "id": "probabilistic-model", "title": "Probabilistic Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ProbabilisticModel is an owl:Class within the artificial-intelligence domain representing any computational or mathematical framework that explicitly represents uncertainty over random variables through probability distributions, enabling principled inference, prediction, and decision-making unde...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:probabilistic-model", "labels": [ "Probabilistic Model", "Denoising Diffusion Probabilistic Model", "Probabilistic Models" ], "is_subclass_of": [ "AI Technique", "Neural Network", "Machine Learning Discipline", "Statistical Learning", "Bayesian Inference", "Uncertainty Quantification", "Generative Model" ], "wikilinks": [ "Acquisition Function", "AISTATS", "AlgorithmLayer", "Anomaly Detection", "Automatic Differentiation", "Autonomous Systems", "Bayes Theorem", "Bayesian Analysis Journal", "Bayesian Decision Theory", "Bayesian Deep Learning", "Bayesian Inference", "Bayesian Optimisation", "Belief Propagation", "Causal Inference", "Causal Model", "Climate Modelling", "Computational Graph", "Computational Statistics", "Conformal Prediction", "Deterministic Model" ] }, { "id": "probabilistic-modelling", "title": "Probabilistic Modelling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The practice of representing systems, processes, and observations as probability distributions over possible states rather than as single deterministic values, so that predictions carry calibrated uncertainty. Probabilistic modelling encompasses specifying random variables and their dependencies, choosing priors, performing inference to update beliefs from data, and validating the resulting probabilistic model\u2014underpinning Bayesian statistics, generative machine learning, risk analysis, and scientific simulation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:probabilistic-modelling", "labels": [ "Probabilistic Modelling" ], "is_subclass_of": [ "Computational Modelling" ], "wikilinks": [ "Computational Modelling", "Bayesian Inference", "Uncertainty Quantification" ] }, { "id": "probabilistic-programming", "title": "Probabilistic Programming", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Probabilistic programming is a paradigm in which statistical models are expressed as programs that include random variables and conditioning statements, with inference performed automatically by the language runtime. It lets practitioners specify generative models declaratively while delegating the mechanics of Bayesian inference, such as sampling or variational optimisation, to the system. This separation of model specification from inference enables rapid iteration on complex probabilistic models across science and machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:probabilistic-programming", "labels": [ "Probabilistic Programming" ], "is_subclass_of": [ "Probabilistic Model" ], "wikilinks": [] }, { "id": "probabilistic-reasoning", "title": "Probabilistic Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Probabilistic reasoning is the process of drawing conclusions under uncertainty by representing beliefs as probability distributions and updating them with evidence according to the rules of probability. It uses models such as graphical models and applies inference procedures, often grounded in Bayesian updating, to compute the likelihood of hypotheses given observations. By quantifying uncertainty explicitly, it supports robust decision-making where deterministic logic would be brittle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:probabilistic-reasoning", "labels": [ "Probabilistic Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "probabilistic-risk-assessment", "title": "Probabilistic Risk Assessment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Probabilistic Risk Assessment (PRA) is a systematic methodology for quantifying the likelihood and consequences of adverse events in complex engineered systems. It enumerates accident scenarios, estimates the probability of each contributing failure and combines them to produce numeric risk measures and confidence bounds. PRA underpins safety-critical decision-making in domains such as nuclear, aerospace and critical infrastructure where rare, high-consequence failures must be rigorously characterised.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:probabilistic-risk-assessment", "labels": [ "Probabilistic Risk Assessment" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "probabilistic-roadmap", "title": "Probabilistic Roadmap", "domain": "robotics", "domain_name": "Robotics", "definition": "A probabilistic roadmap (PRM) is a sampling-based motion-planning algorithm that constructs a graph representation of the free configuration space of a robot by randomly sampling collision-free configurations and connecting nearby configurations with local path planners. In a preprocessing phase, many random configurations are sampled and validated against the robot's collision model; valid configurations become nodes and successful local connections become edges. At query time, start and goal configurations are connected to the roadmap and a graph-search algorithm finds a path. PRMs are effective in high-dimensional configuration spaces where deterministic grid-based planners are computationally infeasible.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:probabilistic-roadmap", "labels": [ "Probabilistic Roadmap" ], "is_subclass_of": [ "Robotics", "Navigation and Planning" ], "wikilinks": [] }, { "id": "probabilistic-robotics", "title": "Probabilistic Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Probabilistic robotics is an approach to robot perception and control that represents uncertainty explicitly using probability distributions, and reasons about state and action through Bayesian estimation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:probabilistic-robotics", "labels": [ "Probabilistic Robotics" ], "is_subclass_of": [ "Robotics", "Navigation and Planning", "Robotics Domain" ], "wikilinks": [ "Bayesian Inference", "Probability Theory", "Localization", "SLAM", "Particle Filter", "Robotics Domain" ] }, { "id": "probability-distribution", "title": "Probability Distribution", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A probability distribution is a mathematical function that assigns probabilities to the possible outcomes of a random variable, describing how likely each value or region of values is. Distributions may be discrete (probability mass functions) or continuous (probability density functions), and are characterised by parameters and summary statistics such as mean and variance. They are the central object of probabilistic modelling, underpinning inference, sampling and generative modelling in machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:probability-distribution", "labels": [ "Probability Distribution" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [] }, { "id": "probability-theory", "title": "Probability Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Probability theory is the branch of mathematics that provides a rigorous framework for quantifying and reasoning about uncertainty and random phenomena. Built on measure-theoretic foundations formalised by Andrei Kolmogorov in 1933, it defines probability spaces, random variables, and expectation operators over sigma-algebras to give a unified axiomatic treatment of chance. The theory encompasses both frequentist and Bayesian interpretations, and its core results \u2014 the law of large numbers, the central limit theorem, and the law of total probability \u2014 underpin virtually every quantitative discipline from statistical inference to machine learning and financial mathematics.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:probability-theory", "labels": [ "Probability Theory" ], "is_subclass_of": [ "Measure Theory" ], "wikilinks": [] }, { "id": "procedural-animation", "title": "Procedural Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedural Animation is a computational technique in which character and object motion is synthesised algorithmically at runtime \u2014 through rules, mathematical functions, and physical simulation \u2014 rather than played back from pre-authored keyframe sequences. It encompasses inverse kinematics solvers, physics-based secondary motion, constraint-driven posing, and behaviour-tree-driven locomotion, allowing virtual agents to adapt dynamically to unpredictable environments. Widely used in games, virtual reality, robotics control, and digital twins, procedural animation replaces or augments traditional hand-keyed or motion-captured data with generative motion pipelines. The approach scales cheaply across large numbers of unique characters and environmental configurations that would be prohibitively expensive to author by hand.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:procedural-animation", "labels": [ "Procedural Animation" ], "is_subclass_of": [ "Animation Technique" ], "wikilinks": [] }, { "id": "procedural-audio-generator", "title": "Procedural Audio Generator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "System that produces context-sensitive sound effects algorithmically in real-time, generating audio content through computational rules rather than playing back pre-recorded samples.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:procedural-audio-generator", "labels": [ "Procedural Audio Generator", "Game Audio Procedural Generation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Adaptive Music", "Audio API", "Audio Rendering Pipeline", "Audio Synthesis Engine", "Context Analysis Module", "Dynamic Soundscapes", "Event System", "Interactive Audio", "MPEG-H Audio Standard", "Real-Time Mixer", "Responsive Sound Effects", "Synthesis Algorithms", "Audio Parameters", "ComputeLayer", "Context Awareness System", "CreativeMediaDomain", "Digital Signal Processing", "Parameter Modulation System" ] }, { "id": "procedural-content-generation", "title": "Procedural Content Generation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Algorithmic creation of 3D objects, textures, environments, or complete scenes using computational rules, mathematical functions, or AI models rather than manual authoring.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:procedural-content-generation", "labels": [ "Procedural Content Generation", "Procedural Story Generation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "3D Modeling API", "AI Model Inference Engine", "Asset Synthesizer", "Automated Asset Creation", "Content Creation Pipeline", "Dynamic Environments", "ETSI ARF 010", "Fractal Algorithms", "Infinite World Generation", "L-Systems", "Perlin Noise", "Rule-Based Generator", "Texture Generation System", "Algorithmic Framework", "ComputeLayer", "CreativeMediaDomain", "DataLayer", "Grammar System", "Machine Learning Models", "Noise Function Library" ] }, { "id": "procedural-content", "title": "Procedural Content", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital content\u2014terrain, architecture, textures, narrative events, or game levels\u2014generated algorithmically from rules and random seeds rather than authored by hand. Procedural content enables scalable world-building in metaverse and gaming contexts by producing combinatorially vast, non-repetitive environments at runtime whilst reducing manual asset creation effort.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:procedural-content", "labels": [ "Procedural Content" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "procedural-generation", "title": "Procedural Generation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedural Generation is the algorithmic creation of digital content \u2014 including terrain, geometry, textures, vegetation, buildings, soundscapes, quests, and narrative elements \u2014 using mathematical functions, noise algorithms, grammars, and rule-based systems rather than entirely manual authoring. It enables scalable world-building by deriving arbitrarily large, varied outputs from compact parametric seeds, making it foundational to games, simulation, virtual environments, and generative design workflows. Modern procedural generation incorporates machine-learning-guided heuristics alongside classical stochastic sampling, bridging traditional algorithmic design with contemporary AI-driven content synthesis. The technique trades artist-authoring effort for computational cost while preserving statistical variety and stylistic coherence.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:procedural-generation", "labels": [ "Procedural Generation", "ProceduralGeneration", "Rule-Based Procedural Generation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "procedural-memory", "title": "Procedural Memory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The long-term memory subsystem that stores knowledge of how to perform actions and skills \u2014 as opposed to episodic memory of specific events or semantic memory of facts; in cognitive science it underlies habits and motor skills learnt through repetition, while in cognitive architectures and AI agents it is realised as production rules, learned policies, reusable workflows, or refined system prompts that encode competence rather than content.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:procedural-memory", "labels": [ "Procedural Memory" ], "is_subclass_of": [ "Agent Memory" ], "wikilinks": [ "Agent Memory", "Cognitive Architecture", "Working Memory" ] }, { "id": "procedural-terrain", "title": "Procedural Terrain", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedural Terrain refers to algorithmically generated landscape geometry, typically based on noise functions such as Perlin or simplex noise, fractals, hydraulic erosion simulation, and rule-based placement of features. It enables scalable, varied, and believable virtual worlds without manual authoring of every surface element, supporting open-world game environments and simulation scenarios.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:procedural-terrain", "labels": [ "Procedural Terrain" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "procedural-texture", "title": "Procedural Texture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Algorithmically generated pattern used to simulate surface detail without stored images, computed on-demand using mathematical functions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:procedural-texture", "labels": [ "Procedural Texture", "Procedural Material" ], "is_subclass_of": [ "Display and Rendering", "Render Pipeline" ], "wikilinks": [ "Dynamic Surface Detail", "GPU Shader", "Graphics Processing Unit", "Material System", "Mathematical Functions", "Memory Efficient Texturing", "Noise Function", "Pattern Generator", "Procedural Materials", "Resolution Independent Graphics", "Shader Code", "SIGGRAPH Graphics Glossary", "Texture Coordinates", "ComputeLayer", "CreativeMediaDomain", "Parameter Set", "Rendering Pipeline", "Shader Language" ] }, { "id": "procedural-and-hybrid-4-d", "title": "Procedural and Hybrid 4D", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedural and Hybrid 4D is a compound domain in graphics and creative tooling that unifies two convergent paradigms: (1) procedural generation \u2014 the algorithmic, rule-driven construction of geometry, materials, animation, simulation, and environment data using mathematical operations rather than...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:procedural-and-hybrid-4-d", "labels": [ "Procedural and Hybrid 4D" ], "is_subclass_of": [ "Platform and Environment", "Content and Assets", "Computer Graphics", "3D Modelling", "Visual Effects", "Simulation", "Generative AI" ], "wikilinks": [ "3D Modelling", "4D Content Creation", "4D Gaussian Splatting", "Academy Software Foundation", "AI Animation", "AlgorithmLayer", "Blender Geometry Nodes", "Cascadeur", "Computer Graphics", "ComputerGraphicsDomain", "Computer Vision Training Data", "CUDA", "Deformation Field", "Differentiable Rendering", "Diffusion Model", "Dynamic Scene Reconstruction", "EmberGen", "Film VFX Production", "FLIP Solver", "Game Development" ] }, { "id": "process-automation", "title": "Process Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Process automation is the use of technology to perform repetitive business, operational, or industrial tasks with minimal human intervention, replacing manual steps in workflows with software execution, robotic systems, or rule-based logic engines. It spans a spectrum from simple macro-based automation of individual tasks through robotic process automation (RPA) that mimics user interface interactions, to intelligent automation that integrates machine learning models for decision-making in unstructured data contexts. The goal is to increase throughput, reduce errors, lower labour costs, and free human workers for higher-value activities.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:process-automation", "labels": [ "Process Automation" ], "is_subclass_of": [ "Workflow Automation" ], "wikilinks": [] }, { "id": "process-control", "title": "Process Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Process control is the discipline of regulating continuous industrial processes, such as chemical reactions, temperature, flow, and pressure, to maintain outputs at desired setpoints despite disturbances. It relies on sensors, controllers, and actuators arranged in feedback loops, using techniques from PID control to model predictive control. It is fundamental to manufacturing, energy, and process-industry automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:process-control", "labels": [ "Process Control" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "process-layer", "title": "Process Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Process Layer is the cross-cutting stratum that defines the ordered activities and workflows by which work is carried out. It sits above the Organisational structure that staffs it and supports operational execution. It contains process definitions, workflows, hand-offs, and the rules that sequence activity.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:process-layer", "labels": [ "Process Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Organisational Layer", "Operational Layer", "Coordination Layer", "Business Process Management", "Workflow", "owl:Thing" ] }, { "id": "process-mining", "title": "Process Mining", "domain": "ai", "domain_name": "Ai", "definition": "Process mining is a family of data-driven techniques that reconstruct, analyse, and improve real business processes by extracting knowledge from event logs recorded in enterprise information systems. It encompasses process discovery, which infers a process model from observed event sequences; conformance checking, which compares the discovered behaviour against a reference model; and enhancement, which enriches models with performance and frequency data. By grounding analysis in actual recorded execution rather than idealised documentation, it reveals bottlenecks, deviations, and automation opportunities.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:process-mining", "labels": [ "Process Mining" ], "is_subclass_of": [ "Business Process Management" ], "wikilinks": [] }, { "id": "process-regulation", "title": "Process Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Process regulation is a regulatory approach that specifies how activities must be carried out rather than dictating particular outcomes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:process-regulation", "labels": [ "Process Regulation" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Compliance", "Regulatory Requirements", "Regulation", "https://en.wikipedia.org/wiki/Regulation", "https://www.oecd.org/gov/regulatory-policy/" ] }, { "id": "process-reward-model", "title": "Process Reward Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A process reward model is a reward model that scores the individual intermediate steps of a model's reasoning trajectory rather than only its final answer. By supervising each step of a chain of thought, it provides dense, step-level feedback that guides search and reinforcement learning toward sound reasoning processes. Process reward models contrast with outcome reward models, which assign a single reward based solely on the final result.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:process-reward-model", "labels": [ "Process Reward Model", "Process Reward Models" ], "is_subclass_of": [ "Reward Model", "AI Technique" ], "wikilinks": [] }, { "id": "process", "title": "Process", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A coordinated sequence of activities, state changes, and events that transforms inputs into outputs to achieve a specific goal or outcome.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:process", "labels": [ "Process", "Maintenance Process" ], "is_subclass_of": [ "Computing and Cloud", "Temporal Entity" ], "wikilinks": [ "Automation", "Business Process Management", "Systems Theory", "Temporal Entity", "Orchestration" ] }, { "id": "processed-data-state", "title": "Processed Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:processed-data-state", "labels": [ "Processed Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "processing-hardware", "title": "Processing Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specialized computing components including CPUs, GPUs, and accelerators that execute computational operations for graphics rendering, artificial intelligence workloads, and real-time data processing in immersive digital environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:processing-hardware", "labels": [ "Processing Hardware" ], "is_subclass_of": [ "Display and Rendering", "Physical Hardware" ], "wikilinks": [ "High Performance Computing", "metaverse", "Physical Hardware" ] }, { "id": "processing-lineage", "title": "Processing Lineage", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:processing-lineage", "labels": [ "Processing Lineage" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "procurement", "title": "Procurement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Procurement is the organised process by which organisations acquire goods, services and works from external suppliers, encompassing needs identification, sourcing, tendering, evaluation, contracting and ongoing supplier management. It seeks to obtain the right inputs at the right quality, cost and time while managing risk and ensuring compliance with policy and regulation. Procurement is a key lever for cost control, value creation and resilience across the supply chain.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:procurement", "labels": [ "Procurement" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [] }, { "id": "product-adoption-metrics", "title": "Product Adoption Metrics", "domain": "economics", "domain_name": "Economics", "definition": "Quantitative indicators used to measure the uptake, engagement, and retention of a software product or feature among its target user base.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:product-adoption-metrics", "labels": [ "Product Adoption Metrics" ], "is_subclass_of": [ "AI Adoption" ], "wikilinks": [] }, { "id": "product-design", "title": "Product Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Product Design is the multidisciplinary creative and engineering practice of conceiving, planning, and producing goods, digital interfaces, and services that satisfy user needs, business objectives, and manufacturing or deployment constraints \u2014 integrating human-centred research mods, aesthetic j...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:product-design", "labels": [ "Product Design" ], "is_subclass_of": [ "Content and Assets", "Design Thinking", "Human-Computer Interaction", "Systems Thinking", "Engineering Design", "Creative Practice" ], "wikilinks": [ "Additive Manufacturing", "Adobe Creative Cloud", "Agile Design", "Autodesk Fusion 360", "Brand Guidelines", "Brand Identity", "Brand Management", "BS EN ISO 9241-210", "BSI Design Standards", "Business Strategy", "CAD", "Cognitive Psychology", "Competitive Analysis", "Continuous Discovery", "Creative Practice", "CreativeProcessDomain", "Customer Satisfaction", "Design Operations", "DesignOps", "Design Research" ] }, { "id": "product-liability-directive", "title": "Product Liability Directive", "domain": "governance", "domain_name": "Governance", "definition": "The Product Liability Directive is European Union legislation establishing rules on liability for damage caused by defective products. A revised directive adopted in 2024 extends its scope to software and AI systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:product-liability-directive", "labels": [ "Product Liability Directive" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "AI Safety", "owl:Thing" ] }, { "id": "product-liability", "title": "Product Liability", "domain": "governance", "domain_name": "Governance", "definition": "The area of law concerning the responsibility of producers and sellers for harm caused by defective or unsafe products placed on the market, covering manufacturing defects, design defects, and failure to warn.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:product-liability", "labels": [ "Product Liability", "Vicarious Liability" ], "is_subclass_of": [ "Consumer Protection" ], "wikilinks": [ "Consumer Protection", "Accountability", "Safety", "Risk Management" ] }, { "id": "product-management", "title": "Product Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Product management is the organisational function responsible for guiding a product through its lifecycle by defining vision, strategy, and roadmap based on user needs, business goals, and market conditions. It coordinates design, engineering, and go-to-market work and prioritises features against constraints. AI tools increasingly assist with research synthesis, prioritisation, and roadmap visualisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:product-management", "labels": [ "Product Management" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "product-provenance", "title": "Product Provenance", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Product provenance is the verifiable record of a product's origin and the chain of custody it has passed through from raw material to end consumer. It is established by capturing and linking events, such as manufacture, inspection and transfer, into an auditable trail, increasingly recorded on distributed ledgers to resist tampering. Platforms such as VeChain implement product provenance by anchoring supply-chain traceability events on-chain so that authenticity claims can be independently verified.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:product-provenance", "labels": [ "Product Provenance" ], "is_subclass_of": [ "Supply Chain Traceability" ], "wikilinks": [] }, { "id": "product-recall-management", "title": "Product Recall Management", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-powered systems enabling rapid, precise product recall execution through immutable supply chain traceability, reducing recall scope by 70\u201385% whilst accelerating contamination source identification from days to seconds. Deployed at scale by Walmart (6 days to 2.2 seconds mango tracing), MediLedger, and automotive consortia for food safety, pharmaceuticals, and vehicle component recalls.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:product-recall-management", "labels": [ "Product Recall Management", "Product Recall", "Recall Management" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Automotive Safety", "BC-0188-self-sovereign-identity", "BC-0202-zero-knowledge-proofs", "BC-0319-micropayments", "BC-0440-blockchain-interoperability", "BC-0441-supply-chain-traceability", "BC-0442-certification-and-compliance", "BC-0445-conflict-mineral-tracking", "BC-0446-food-traceability", "BC-0447-pharmaceutical-traceability", "BC-0448-luxury-goods-authentication", "BC-0449-timber-and-forest-products", "BC-0453-ethical-sourcing", "BC-0454-waste-management", "Carrefour", "DSCSA", "FDA Regulations", "Food Safety", "FSMA Section 204", "MediLedger" ] }, { "id": "product-validation", "title": "Product Validation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:product-validation", "labels": [ "Product Validation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "production-design", "title": "Production Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Production Design is the creative and technical discipline responsible for defining the visual aesthetic, spatial environment, and material world of a production \u2014 whether film, game, or virtual world. It spans concept art, environment modelling, asset creation pipelines, and art direction, ensuring visual coherence across all content elements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:production-design", "labels": [ "Production Design" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "production-facility", "title": "Production Facility", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Production Facility is a specialised physical and digital infrastructure for professional metaverse and immersive content creation, encompassing motion-capture stages, volumetric capture rigs, LED volume virtual production sets, GPU render farms, and cloud-based asset management pipelines. It integrates real-time rendering engines with collaborative tools enabling distributed global teams to produce high-fidelity 3D and spatial content.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:production-facility", "labels": [ "Production Facility" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "production-pipeline", "title": "Production Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The end-to-end sequence of tools, processes, and automated stages that transform raw creative or data assets into finished, deployable outputs. In spatial computing contexts a production pipeline typically spans asset creation, format conversion, quality assurance, rendering, and deployment to target platforms (game engines, XR runtimes, or streaming services).", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:production-pipeline", "labels": [ "Production Pipeline", "Creative Production Pipeline" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "production-rules", "title": "Production Rules", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Production rules are condition-action statements of the form 'if antecedent then consequent' that encode procedural and declarative knowledge in symbolic artificial-intelligence systems. A collection of such rules forms a production system whose inference engine repeatedly matches rule conditions against a working memory of facts, selects which eligible rule to apply, and fires it to assert new facts or perform actions. They are the principal knowledge-representation formalism of classical expert systems and rule-based reasoning, valued for transparency and modular editing. Production rules underpin forward- and backward-chaining inference and remain widely used in business rule engines and policy automation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:production-rules", "labels": [ "Production Rules" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "profile-management", "title": "Profile Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Profile Management encompasses the systems and services for creating, storing, updating, and securing user profiles within metaverse and digital platforms. It covers avatar customisation preferences, personal data storage, authentication credentials, and preference settings, enabling persistent personalised experiences across sessions and across interoperable environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:profile-management", "labels": [ "Profile Management" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "profiling", "title": "Profiling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Profiling is the measurement of a running program's resource use -- CPU time, memory allocation, I/O and call frequency -- to locate the specific functions or code paths that dominate cost. Profilers sample or instrument execution to produce call graphs and hotspot reports that guide targeted optimisation rather than guesswork. Profiling data feeds compiler optimisation decisions and forms the evidence base behind performance benchmarks.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:profiling", "labels": [ "Profiling" ], "is_subclass_of": [ "Performance Optimization" ], "wikilinks": [] }, { "id": "profinet", "title": "Profinet", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "PROFINET is an open industrial Ethernet standard for real-time communication between controllers, devices and supervisory systems in factory and process automation. Built on standard Ethernet, it supports cyclic real-time data exchange, isochronous motion control and integration with IT networks, while remaining interoperable through the PROFIBUS and PROFINET International organisation. It is widely used to connect programmable logic controllers to distributed I/O, drives and sensors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:profinet", "labels": [ "Profinet", "PROFINET" ], "is_subclass_of": [ "Industrial Automation" ], "wikilinks": [] }, { "id": "program-synthesis", "title": "Program Synthesis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Program synthesis is the automated generation of source code or executable procedures that satisfy a given specification, expressed as examples, natural language, or formal constraints, without a human writing the implementation by hand.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:program-synthesis", "labels": [ "Program Synthesis" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "programmable-compliance", "title": "Programmable Compliance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Programmable compliance is the encoding of regulatory requirements \u2014 such as transfer restrictions, jurisdiction limits, investor accreditation checks and KYC status \u2014 directly into a token's smart-contract logic so that non-compliant transactions are automatically rejected on-chain. It shifts enforcement from manual, after-the-fact review to automated, pre-transaction validation, and is a defining feature of enterprise token standards used for security tokens. Programmable compliance is particularly relevant to security token offerings, where regulators require ongoing enforcement of transfer eligibility.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:programmable-compliance", "labels": [ "Programmable Compliance" ], "is_subclass_of": [ "Enterprise Token Standards" ], "wikilinks": [] }, { "id": "programmable-finance", "title": "Programmable Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Programmable finance is the encoding of financial logic, agreements, and assets as executable code on blockchains, so that payments, lending, settlement, and compliance execute automatically and verifiably. It is enabled by smart contracts that compose into open, permissionless financial primitives. It underpins decentralised finance and emerging layered scaling on networks such as Bitcoin.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:programmable-finance", "labels": [ "Programmable Finance" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "programmable-logic-controller", "title": "Programmable Logic Controller", "domain": "robotics", "domain_name": "Robotics", "definition": "A programmable logic controller (PLC) is a ruggedised industrial computer designed to control machinery and processes by repeatedly scanning inputs, executing a stored control program, and updating outputs in real time. Built to withstand harsh factory environments, it interfaces with sensors and actuators and is typically programmed in standardised languages such as ladder logic. PLCs are foundational building blocks of industrial automation and supervisory control systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:programmable-logic-controller", "labels": [ "Programmable Logic Controller" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "programmable-money", "title": "programmable money", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Programmable Money is a form of digital currency in which the conditions governing the creation, transfer, and use of monetary units are encoded as executable logic\u2014typically within smart contracts on a public or permissioned blockchain\u2014and enforced autonomously without requiring trusted intermediaries. This enables payment instruments that can embed compliance rules, spending constraints, vesting schedules, or multi-party approval workflows directly into the currency itself, transforming money from a passive bearer instrument into an active, self-enforcing contractual construct. Central bank digital currencies (CBDCs), tokenised commercial bank deposits, DeFi stablecoins, and protocol-native assets are all instances of programmable money at different points on the decentralisation and trust spectrum. The paradigm extends the concept of money from a static store of value and medium of exchange into a substrate for encoding and automating financial relationships.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:programmable-money", "labels": [ "Programmable Money", "Programmable Money Paradigm", "Programmable Payments", "Programmable Revenue" ], "is_subclass_of": [ "Digital Currency" ], "wikilinks": [] }, { "id": "programmable-privacy", "title": "Programmable Privacy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Programmable Privacy is the capability to define custom, application-specific privacy rules for blockchain transactions and smart-contract state using zero-knowledge proofs, allowing developers to selectively disclose or conceal data rather than relying on an all-or-nothing privacy model. It lets a single protocol support both private and public execution paths within the same contract. Protocols such as Aztec expose this as a first-class primitive for confidential smart contracts.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:programmable-privacy", "labels": [ "Programmable Privacy" ], "is_subclass_of": [ "Zero-Knowledge Proof (ZKP)" ], "wikilinks": [] }, { "id": "programmatic-advertising", "title": "Programmatic Advertising", "domain": "data", "domain_name": "Data", "definition": "Programmatic advertising is the automated buying and selling of digital advertising inventory through software platforms and real-time auctions, replacing manual insertion orders with algorithmic matching of advertiser bids to publisher ad slots. It encompasses demand-side platforms, supply-side platforms, ad exchanges, and data management systems that collectively orchestrate the delivery of targeted advertisements at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:programmatic-advertising", "labels": [ "Programmatic Advertising" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "programming-language", "title": "Programming Language", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A formal notation system used to specify algorithms, data structures, and computational models for AI and software applications, characterised by defined syntax and semantics; prominent examples include Python for machine-learning ecosystems, Julia for high-performance numerical computing, and domain-specific languages embedded in ML frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:programming-language", "labels": [ "Programming Language", "Cairo Programming Language", "Forth Programming Language", "Functional Programming", "General-Purpose Programming Language", "Programming Languages", "Scala Programming Language" ], "is_subclass_of": [ "Software Engineering", "AI Technique" ], "wikilinks": [ "Compiler Optimization", "Domain-Specific Languages", "Julia Language", "Python Ecosystem" ] }, { "id": "programming-paradigm", "title": "Programming Paradigm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Programming Paradigms in AI represent fundamental styles and approaches to structuring code for artificial intelligence systems. Key paradigms include imperative (procedural, object-oriented), declarative (functional, logic-based), and differentiable programming, each shaping how models are specified, composed, and optimised across symbolic and neural approaches.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:programming-paradigm", "labels": [ "Programming Paradigm" ], "is_subclass_of": [ "AI Technique", "owl:Thing" ], "wikilinks": [ "Differentiable Programming", "Functional Programming", "owl:Thing", "Software Engineering", "Symbolic AI" ] }, { "id": "progressive-disclosure-harnesses", "title": "Progressive Disclosure Harnesses", "domain": "ai", "domain_name": "Ai", "definition": "Formats, runtimes, and patterns that reveal context, tools, or instructions to agents in layers \u2014 index first, details on demand \u2014 to control token usage and improve agent focus \u2014 includes MCP-Zero, ToolGen, ToolRAG, langgraph-bigtool, agents.md, and awesome-cursorrules.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:progressive-disclosure-harnesses", "labels": [ "Progressive Disclosure Harnesses" ], "is_subclass_of": [ "Agent Harness", "AI Infrastructure", "Cognitive Architecture" ], "wikilinks": [ "Agent Harness", "Agent Memory Layers", "Harness Configuration Packs", "Internal AI Harness", "External AI Harness", "Context Window", "Model Context Protocol", "Tool Use", "Function Calling", "Large Language Models", "Retrieval-Augmented Generation", "Embeddings", "Vector Database", "Agent Loop", "Multi-Agent System", "Agentic Workflow", "Prompt Engineering", "Information Retrieval", "Semantic Search", "Agentic RAG" ] }, { "id": "progressive-download", "title": "Progressive Download", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Progressive Download is a content delivery technique in which assets\u20143D models, textures, audio, or video\u2014are streamed and partially consumed before the full file is retrieved. It enables early playback or rendering from incomplete data, reduces perceived loading latency in metaverse environments, and underpins level-of-detail streaming and lazy asset loading strategies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:progressive-download", "labels": [ "Progressive Download" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "progressive-web-app", "title": "Progressive Web App", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Progressive Web App (PWA) is a web application built with standard web technologies that delivers an app-like experience by being installable, working offline, and supporting capabilities such as background synchronisation and push notifications. PWAs use service workers and a web app manifest to bridge the gap between web pages and native mobile applications, allowing a single codebase to run across platforms. They are designed to be reliable on poor networks, fast to load, and engaging enough to live on a device home screen.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:progressive-web-app", "labels": [ "Progressive Web App" ], "is_subclass_of": [ "Web Technology" ], "wikilinks": [] }, { "id": "prohibited-ai-practice", "title": "Prohibited AI Practice", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Prohibited AI Practices are categories of AI deployment banned under EU AI Act Article 5 (effective 2 February 2025), covering subliminal manipulation, exploitation of vulnerability groups, social scoring by public or private bodies, and real-time remote biometric identification in publicly accessible spaces. Violations carry fines up to \u20ac35 million or 7% of global annual turnover.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:prohibited-ai-practice", "labels": [ "Prohibited AI Practice" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "project-agora", "title": "Project Agora", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Project Agora is a cross-border wholesale payments initiative led by the Bank for International Settlements with several central banks and private financial institutions. It explores integrating tokenised commercial-bank deposits with tokenised central-bank money on a unified programmable ledger to improve cross-border settlement. It is a flagship experiment in the tokenisation of the monetary and financial system.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:project-agora", "labels": [ "Project Agora", "Agora Forum", "Project Agor\u00e1" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "project-dunbar", "title": "Project Dunbar", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Project Dunbar was a multi-central-bank experiment exploring a shared platform for multiple central bank digital currencies to settle cross-border transactions directly.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:project-dunbar", "labels": [ "Project Dunbar" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [ "Central Bank Digital Currency", "Cross-Border Payments", "Payment Network", "Stablecoin" ] }, { "id": "project-management", "title": "Project Management", "domain": "governance", "domain_name": "Governance", "definition": "Project Management is the disciplined application of knowledge, skills, tools, and techniques to project activities in order to meet defined requirements within constraints of scope, time, cost, and quality. It encompasses initiation, planning, execution, monitoring and control, and closure phases, employing methodologies ranging from waterfall and PRINCE2 to agile frameworks such as Scrum and Kanban, and supporting the delivery of technology, construction, research, and organisational change initiatives.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:project-management", "labels": [ "Project Management", "Project Management System" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "project-nexus", "title": "Project Nexus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Project Nexus is a Bank for International Settlements initiative to interlink domestic instant-payment systems across countries through a standardised hub, enabling fast, low-cost cross-border retail payments. Rather than building a single global system, it provides a common protocol that connects existing national fast-payment rails. It is relevant to discussions of cross-border CBDC interoperability and payment-system modernisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:project-nexus", "labels": [ "Project Nexus" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "project-rosalind", "title": "Project Rosalind", "domain": "finance", "domain_name": "Finance", "definition": "A joint experiment by the Bank of England and the Bank for International Settlements Innovation Hub that explored how an application programming interface layer could support retail central bank digital currency payments. It tested programmability and interoperability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:project-rosalind", "labels": [ "Project Rosalind" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [ "Central Bank Digital Currency", "Bank of England" ] }, { "id": "project-m-bridge", "title": "Project mBridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A multi-central-bank initiative testing a shared distributed ledger platform for cross-border payments settled in central bank digital currencies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:project-m-bridge", "labels": [ "Project mBridge" ], "is_subclass_of": [ "Cross-Border Settlement" ], "wikilinks": [ "Wholesale CBDC", "Atomic Settlement", "BIS", "Cross-Border Payments", "Cross-Border Settlement" ] }, { "id": "projected-coordinate-reference-system", "title": "Projected Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:projected-coordinate-reference-system", "labels": [ "Projected Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "projection-mapping", "title": "Projection Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Projection Mapping is a spatial AR display technique that aligns projected imagery precisely onto three-dimensional physical surfaces, transforming arbitrary geometry into dynamic visual canvases. It requires real-time rendering of geometry-corrected frames, spatial calibration, and surface modelling, enabling immersive installations, virtual production environments, and location-based XR experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:projection-mapping", "labels": [ "Projection Mapping" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "projective-geometry", "title": "Projective Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Projective geometry is the branch of geometry concerned with properties of figures that are invariant under projective transformations, where points at infinity are treated on equal footing with ordinary points. Using homogeneous coordinates, it provides the mathematical foundation for modelling how three-dimensional scenes project onto image planes. It underpins camera models, multi-view geometry, and the reconstruction of structure from images in computer vision and spatial computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:projective-geometry", "labels": [ "Projective Geometry" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "prometheus", "title": "Prometheus", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Prometheus is an open-source monitoring and alerting system that collects time-series metrics by periodically scraping HTTP endpoints exposed by instrumented targets. It stores samples in a local time-series database, queries them with the PromQL language, and evaluates alerting rules whose firing alerts are dispatched through a separate Alertmanager. A graduated Cloud Native Computing Foundation project, it is a de-facto standard for monitoring containerised and cloud-native systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:prometheus", "labels": [ "Prometheus" ], "is_subclass_of": [ "Monitoring" ], "wikilinks": [] }, { "id": "prompt-caching", "title": "Prompt Caching", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Prompt caching is an inference optimisation in which the key-value attention state for a shared, unchanging prefix of a prompt -- such as a system prompt or long context document -- is computed once and reused across subsequent requests, avoiding redundant computation. It reduces latency and cost for workloads that repeatedly send the same long context with only a short suffix varying between calls. It builds directly on KV cache mechanisms and is a common lever in context engineering for large language model applications.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:prompt-caching", "labels": [ "Prompt Caching" ], "is_subclass_of": [ "Caching" ], "wikilinks": [] }, { "id": "prompt-engineering-reference-library", "title": "Prompt Engineering Reference Library", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A curated collection of high-quality prompts, system instructions, and prompt engineering patterns for directing large language model behaviour across tasks including image generation, code synthesis, transcript processing, and writing style control. Serves as a practitioner reference library for reproducible AI-assisted workflows.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:prompt-engineering-reference-library", "labels": [ "Prompt Engineering Reference Library", "bestprompts" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "prompt-engineering", "title": "Prompt Engineering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Prompt Engineering is the systematic discipline of designing, structuring, and optimising natural language or structured inputs (prompts) to elicit desired behaviours, outputs, and reasoning traces from large language models (LLMs) and other generative AI systems, spanning a hierarchy of techniqu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:prompt-engineering", "labels": [ "Prompt Engineering" ], "is_subclass_of": [ "AI Technique", "In-Context Learning", "AI Application", "Natural Language Processing", "Human-Computer Interaction", "Machine Learning Discipline", "AI Alignment", "Instruction Following", "Applied Artificial Intelligence" ], "wikilinks": [ "AlgorithmLayer", "Anthropic Prompt Engineering Documentation 2024", "APILayer", "Applied Artificial Intelligence", "Automatic Prompt Optimisation", "Bai et al. 2022 Constitutional AI", "Brown et al. 2020 GPT-3 Few-Shot Learning", "Chain of Thought", "DSPy", "Fernando et al. 2023 PromptBreeder", "Few-Shot Prompting", "Greshake et al. 2023 Prompt Injection", "HumanComputerInteractionDomain", "In-Context Learning", "Instruction Following", "Khattab et al. 2024 DSPy", "Kojima et al. 2022 Zero-Shot Reasoners", "Lu et al. 2022 Fantastically Ordered Prompts", "Min et al. 2022 ICL Role of Demonstrations", "NCSC UK AI Security Guidelines 2024" ] }, { "id": "prompt-injection", "title": "prompt injection", "domain": "security", "domain_name": "Security", "definition": "Prompt injection is a class of adversarial attacks against large language model (LLM) systems in which attacker-controlled text embedded within the model's input context overrides or subverts the developer-specified system prompt, causing the model to follow attacker instructions instead of its intended operating constraints. Direct prompt injection occurs when a user submits malicious instructions directly; indirect prompt injection occurs when the model retrieves or processes external content such as web pages, documents, or tool outputs that contain embedded adversarial instructions. As LLMs are increasingly deployed in agentic pipelines with tool-calling and autonomous action capabilities, prompt injection constitutes a critical security boundary violation that can lead to data exfiltration, privilege escalation, and unauthorised actions on external systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:prompt-injection", "labels": [ "Prompt Injection" ], "is_subclass_of": [ "Adversarial Attack" ], "wikilinks": [] }, { "id": "prompt-template", "title": "Prompt Template", "domain": "ai", "domain_name": "Ai", "definition": "A Prompt Template is a reusable, parameterised text structure that combines static instructional scaffolding with dynamic variable slots to produce consistent, well-formed inputs for large language models. Templates encode best-practice prompt engineering patterns \u2014 role declarations, task specifications, output format constraints, and chain-of-thought scaffolding \u2014 in a form that can be instantiated with runtime values such as user queries, document snippets, or API responses. They are the primary unit of composition in LLM orchestration frameworks such as LangChain, LlamaIndex, and DSPy, and they serve as version-controlled artefacts that enable systematic A/B testing of prompt variations. Effective prompt templates balance specificity (providing enough context to constrain model behaviour) with generality (accommodating the full range of valid inputs).", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:prompt-template", "labels": [ "Prompt Template" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "prompt-tuning", "title": "Prompt Tuning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Prompt Tuning is a parameter-efficient fine-tuning method that learns continuous soft prompt embeddings prepended to the input sequence, whilst keeping all pre-trained model weights frozen. It optimises task-specific prompts in the embedding space using gradient descent, requiring as little as 0.01% of model parameters and enabling efficient multi-task deployment from a single frozen backbone.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:prompt-tuning", "labels": [ "Prompt Tuning" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "Dec 19th, 2023", "DSPy", "Ethan Mollick", "optimization", "organisation", "productivity", "project management", "ArtificialIntelligenceDomain", "cloud computing", "ComfyUI", "knowledge management", "Melvin Carvalho", "Stable Diffusion" ] }, { "id": "proof-of-personhood", "title": "Proof Of Personhood", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of personhood is a mechanism for establishing that a participant in a digital system is a unique, real human being without necessarily revealing their full identity. It addresses Sybil resistance by ensuring one person cannot cheaply masquerade as many, which is critical for fair voting, airdrops and universal-basic-income style distributions. Approaches range from biometric uniqueness checks to social-graph and in-person verification protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:proof-of-personhood", "labels": [ "Proof Of Personhood", "Proof of Personhood" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "proof-of-reserve", "title": "Proof Of Reserve", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of reserve is a verification practice in which a custodian or exchange demonstrates that it holds assets sufficient to cover its customer liabilities. It typically combines an on-chain attestation of owned assets with a Merkle-tree commitment to the liability set, allowing users to verify inclusion of their balance without revealing others. Robust schemes also prove liabilities to establish solvency, often with auditor or zero-knowledge support.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:proof-of-reserve", "labels": [ "Proof Of Reserve", "Proof of Reserve" ], "is_subclass_of": [ "Blockchain", "Governance and Regulation" ], "wikilinks": [] }, { "id": "proof-of-spacetime", "title": "Proof Of Spacetime", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of Spacetime (PoSt) is a cryptographic consensus mechanism in which a participant repeatedly proves that it has continuously stored a specific set of data over a period of time, rather than expending computation as in proof of work. Used by decentralised storage networks such as Filecoin, it combines storage proofs with time-based challenges so that providers cannot reclaim space without losing the ability to answer. This ties block production and rewards to verifiable, useful storage capacity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:proof-of-spacetime", "labels": [ "Proof Of Spacetime", "Proof of Spacetime" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "proof-of-stake-sustainability", "title": "Proof Of Stake Sustainability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of Stake Sustainability refers to the environmental, economic, and governance characteristics of Proof of Stake consensus mechanisms that make them a viable long-term alternative to energy-intensive Proof of Work systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "draft", "iri": "urn:ngm:class:proof-of-stake-sustainability", "labels": [ "Proof Of Stake Sustainability" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Sustainability", "Consensus Mechanism", "Environmental Technology" ], "wikilinks": [ "AI Energy Optimisation", "ASIC", "Avalanche", "Cambridge Centre for Alternative Finance", "Cosmos", "Crypto Climate Accord", "Delegation Protocol", "Distributed Ledger Technology", "Energy Consumption", "Environmental Technology", "Ethereum Foundation", "IEEE P2418.5", "Institutional Adoption", "Layer 1", "Liquid Staking", "Network Security", "Polkadot", "Slashing Conditions", "Staking Mechanism", "Sustainable Virtual Infrastructure" ] }, { "id": "proof-of-work", "title": "Proof Of Work", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A consensus mechanism requiring network participants to expend significant computational effort solving a cryptographic puzzle before appending a new block to the blockchain. The difficulty of the puzzle self-adjusts to maintain a target block interval, making chain rewriting computationally prohibitive and providing Sybil resistance through physical resource expenditure.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:proof-of-work", "labels": [ "Proof Of Work", "Cryptographic Proof of Work", "Proof of Work", "Proof-of-Work", "ProofOfWork", "proof-of-work" ], "is_subclass_of": [ "Consensus Protocol", "Blockchain Entity", "ConsensusProtocol" ], "wikilinks": [ "51% attacks", "AntPool", "ASIC", "Avalanche", "Back, A. (2002)", "Bitcoin Cash", "Bitmain", "block hash", "Block Reward (Mining Incentive)", "Block Reward (Mining Incentive)", "Block Reward (Mining Incentive)", "Block size limits", "Blockchain Technology Laboratory", "Bonneau, J., Miller, A., Clark, J., Narayanan, A., Kroll, J. A., & Felten, E. W. (2015)", "Cambridge Centre for Alternative Finance (2019-2025)", "Canaan", "carbon credits", "chain split", "Chia", "CleanSpark" ] }, { "id": "proof-system", "title": "Proof System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Proof System is a formal cryptographic or logical framework that enables a prover to convince a verifier that a statement is true, often without revealing the underlying witness or private data. Modern cryptographic proof systems include zero-knowledge proofs, interactive proof systems, and SNARKs, which are foundational to privacy-preserving authentication and blockchain scalability. They provide mathematical guarantees of soundness and completeness under well-defined computational assumptions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:proof-system", "labels": [ "Proof System" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "proof-theory", "title": "Proof Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Proof theory is the branch of mathematical logic that studies formal proofs as syntactic objects \u2014 their structure, rules of inference, and the transformations between them \u2014 independent of the semantic models that make statements true. It underpins automated theorem proving, type theory, and formal verification, providing the rules by which a machine can check or search for a valid derivation of a conclusion from axioms. Results such as cut-elimination and normalisation are central to its use in computer science.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:proof-theory", "labels": [ "Proof Theory" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "proof-of-authority", "title": "Proof of Authority", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A permissioned blockchain consensus mechanism in which a fixed set of pre-approved validators with verified real-world identities are authorised to produce and validate blocks. PoA sacrifices decentralisation for high throughput and fast finality, making it suitable for enterprise and consortium blockchains where participant trust can be established off-chain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:proof-of-authority", "labels": [ "Proof of Authority", "Proof-of-Authority", "ProofOfAuthority" ], "is_subclass_of": [ "Protocol and Consensus", "Proof-Based Consensus" ], "wikilinks": [ "Blockchain", "Digital Twin", "Proof-Based Consensus" ] }, { "id": "proof-of-history", "title": "Proof of History", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of History is a cryptographic clock mechanism that timestamps transactions using a sequential, verifiable delay function (SHA-256 hash chain), establishing a tamper-evident historical record. This allows Solana validators to process transactions in parallel without requiring round-trip consensus on ordering, dramatically increasing throughput.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:proof-of-history", "labels": [ "Proof of History", "Proof-of-History" ], "is_subclass_of": [ "Protocol and Consensus", "Proof-Based Consensus" ], "wikilinks": [ "Blockchain", "Proof-Based Consensus" ] }, { "id": "proof-of-publication", "title": "Proof of Publication", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of publication is a cryptographic guarantee that a given message has been published to, and ordered within, an append-only medium visible to all relevant parties, such that the publisher cannot later equivocate or hide it. Together with single-use seals it forms one of the two foundations of client-side validation: the publication medium (most commonly the Bitcoin blockchain, but also a Nostr relay set or other consensus system) provides ordering and non-equivocation, while seals bind specific messages to spendable objects. It is the property that lets off-chain smart-contract systems trust that a seal closure was witnessed once and only once.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:proof-of-publication", "labels": [ "Proof of Publication", "Proof Of Publication", "Proof-of-Publication" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "proof-of-reserves", "title": "Proof of Reserves", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof of reserves is a cryptographic auditing technique by which a custodial entity, such as a cryptocurrency exchange or stablecoin issuer, demonstrates that it holds sufficient assets to cover its customer liabilities. The asset side is typically attested by publishing on-chain wallet ownership, while the liability side is committed using a Merkle tree so that individual customers can verify their balance is included without exposing the full ledger. More advanced schemes combine these with zero-knowledge proofs to prove solvency while preserving the confidentiality of total liabilities and individual balances.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:proof-of-reserves", "labels": [ "Proof of Reserves", "Proof Of Reserves" ], "is_subclass_of": [ "Blockchain", "Cryptographic Primitive (Blockchain)" ], "wikilinks": [] }, { "id": "proof-of-stake", "title": "Proof of Stake", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A consensus mechanism allowing blockchains to validate transactions and create new blocks based on the number of tokens held or staked by network participants. Validators are selected proportionally to their stake, replacing the energy-intensive mining of Proof of Work with a capital-cost security model that achieves deterministic or probabilistic finality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:proof-of-stake", "labels": [ "Proof of Stake", "Proof Of Stake", "Proof-of-Stake", "Proof-of-Stake Consensus", "ProofOfStake", "proof-of-stake" ], "is_subclass_of": [ "Proof-Based Consensus" ], "wikilinks": [ "Blockchain", "Proof-Based Consensus" ] }, { "id": "proof-based-consensus", "title": "Proof-Based Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof-Based Consensus is the family of blockchain consensus mechanisms in which participants must provide a verifiable cryptographic or computational proof to earn the right to propose or validate blocks. This family includes proof-of-work (computational puzzle), proof-of-stake (economic stake), delegated proof-of-stake, proof-of-authority, proof-of-history, and related variants.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:proof-based-consensus", "labels": [ "Proof-Based Consensus" ], "is_subclass_of": [ "Protocol and Consensus", "Consensus Mechanism" ], "wikilinks": [ "Blockchain", "Consensus Mechanism" ] }, { "id": "proof-of-work-energy-consumption", "title": "Proof-of-Work Energy Consumption", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proof-of-work energy consumption is the electricity used by miners performing the computational hashing that secures proof-of-work blockchains such as Bitcoin. Because security scales with aggregate hashpower, energy use rises with price and competition, making it a central environmental and policy concern. It is tracked by indices that estimate network-wide power draw and carbon footprint.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:proof-of-work-energy-consumption", "labels": [ "Proof-of-Work Energy Consumption" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "propagation-delay", "title": "Propagation Delay", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Propagation delay is the time a signal takes to travel from sender to receiver across a transmission medium, determined by the physical distance divided by the signal's propagation speed. Bounded by the speed of light and reduced in copper or fibre by the medium's refractive properties, it sets a hard floor on network latency that no amount of bandwidth can remove. Propagation delay is one of several additive components of end-to-end latency, alongside transmission, queuing, and processing delays.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:propagation-delay", "labels": [ "Propagation Delay" ], "is_subclass_of": [ "Network Latency" ], "wikilinks": [] }, { "id": "propellant-feed-system", "title": "Propellant Feed System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:propellant-feed-system", "labels": [ "Propellant Feed System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "propellant-tank", "title": "Propellant Tank", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:propellant-tank", "labels": [ "Propellant Tank" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "propellant", "title": "Propellant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:propellant", "labels": [ "Propellant" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "property-graph", "title": "Property Graph", "domain": "data", "domain_name": "Data", "definition": "A property graph is a graph data structure in which both nodes and edges carry labels and an arbitrary set of key-value properties. Edges are directed and uniquely identifiable, allowing multiple parallel relationships of different types between the same pair of nodes. It is the data model behind many native graph databases and is queried with traversal languages such as Cypher and Gremlin.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:property-graph", "labels": [ "Property Graph" ], "is_subclass_of": [ "Graph Data Model" ], "wikilinks": [] }, { "id": "property-registry", "title": "Property Registry", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based land title recording systems employing immutable distributed ledgers, cryptographic signatures, and timestamp verification to create tamper-proof property ownership records, reduce fraud, and accelerate transaction processing from 30\u201390 days to 72 hours. Dubai processed 188,000+ transactions worth AED 625 billion in 2024; Georgia was the first country to adopt blockchain land administration at national scale.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:property-registry", "labels": [ "Property Registry" ], "is_subclass_of": [ "Token and Asset", "Real Estate Tokenization" ], "wikilinks": [ "BC-0142-smart-contract", "BC-0432-consortium-blockchain", "BC-0456-self-sovereign-identity", "BC-0457-decentralized-identifiers", "BC-0458-verifiable-credentials", "BC-0493-real-estate-tokenization", "BlockchainDomain" ] }, { "id": "property-rights", "title": "Property Rights", "domain": "governance", "domain_name": "Governance", "definition": "Property rights are the legally and institutionally recognised entitlements that determine who may use, transfer, exclude others from, or derive value from a resource \u2014 whether physical, digital, or intellectual. They form the foundational incentive structure of market economies, specifying the bundle of rights (usus, fructus, abusus) attached to an asset. In digital and blockchain contexts, property rights extend to on-chain ownership records, smart-contract-enforced exclusivity, NFT-based title deeds, and virtual land parcels, requiring mechanisms for interoperability, provenance tracking, and dispute resolution that operate across jurisdictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:property-rights", "labels": [ "Property Rights" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "owl:Thing" ] }, { "id": "property-schema", "title": "Property Schema", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A formal specification of the object properties, data properties, and annotation properties that relate classes within an ontology, including domain and range constraints, cardinality axioms, and logical characteristics such as transitivity and symmetry. Property schemas are the mechanism by which OWL2 ontologies encode typed, machine-readable relations between entities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:property-schema", "labels": [ "Property Schema" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "property-system", "title": "Property System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Property System is a software architecture pattern for attaching typed key-value attributes to entities within a metaverse or game engine context, enabling data binding, runtime configuration, and dynamic object behaviour without hard-coded class inheritance. It underpins avatar customisation, world-object state management, and reactive UI data flows in spatial computing platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:property-system", "labels": [ "Property System" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "property", "title": "Property", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In the context of AI systems and ontological engineering, a Property is a named attribute or characteristic that describes a measurable or observable quality of an entity, concept, or system component. Properties may be data properties (mapping an individual to a literal value such as a number or string) or object properties (relating an individual to another individual within the knowledge graph). In AI safety and governance contexts, properties such as fairness, robustness, and interpretability are the formal targets of evaluation and certification requirements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:property", "labels": [ "Property", "System Property" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "proportional-control", "title": "Proportional Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Proportional Control is a feedback control strategy in which the corrective output applied to an actuator is directly proportional to the current error\u2014the difference between the desired setpoint and the measured process variable. It is the foundational component of PID controllers, providing immediate, scaled response to deviations but typically leaving a steady-state offset that requires integral or derivative terms to eliminate.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:proportional-control", "labels": [ "Proportional Control", "Proportional Flow Control", "Proportional Pressure Control" ], "is_subclass_of": [ "Actuation and Control", "Feedback Control" ], "wikilinks": [ "Feedback Control", "Robotics" ] }, { "id": "proportional-valve", "title": "Proportional Valve", "domain": "robotics", "domain_name": "Robotics", "definition": "A proportional valve is a fluid-power control element that modulates flow rate, pressure, or direction continuously and proportionally in response to an electrical command signal, in contrast to on/off solenoid valves that switch between two discrete states. The valve's spool or poppet position is controlled by a proportional solenoid or voice-coil actuator whose force output is linearly related to the applied current, enabling smooth, variable control of hydraulic or pneumatic systems.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:proportional-valve", "labels": [ "Proportional Valve" ], "is_subclass_of": [ "Fluid Power Device" ], "wikilinks": [] }, { "id": "proposal-distribution", "title": "Proposal Distribution", "domain": "ai", "domain_name": "Ai", "definition": "A proposal distribution is an auxiliary distribution used in Monte Carlo methods to generate candidate samples when sampling directly from a target distribution is infeasible. In importance sampling, Metropolis-Hastings, and particle filters it determines where samples are drawn, and its closeness to the target governs efficiency and variance. A poorly chosen proposal causes sample degeneracy or slow mixing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:proposal-distribution", "labels": [ "Proposal Distribution" ], "is_subclass_of": [ "Machine Learning Discipline", "Monte Carlo Methods" ], "wikilinks": [] }, { "id": "proposal-system", "title": "Proposal System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Multi-stage governance workflow mechanism enabling DAO community members to initiate, discuss, deliberate, vote upon, and execute protocol changes through structured processes that combine off-chain deliberation (forums, governance calls), cryptographic signalling via off-chain snapshot votes, and binding on-chain execution through smart contracts, with threshold requirements (token holdings, quorum) balancing permissionless participation against governance efficiency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:proposal-system", "labels": [ "Proposal System", "ProposalSystem" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "Aave", "Aave Improvement Proposals", "BC-0142-smart-contract", "BC-0142-smart-contract", "BC-0461-decentralized-autonomous-organization", "BC-0461-decentralized-autonomous-organization", "BC-0462-on-chain-voting", "BC-0462-on-chain-voting", "BC-0463-governance-token", "BC-0463-governance-token", "BC-0464-treasury-management", "Compound", "Compound Governor Bravo", "MakerDAO", "MakerDAO Executive Votes", "Snapshot", "Snapshot Off-Chain Voting", "Tally Governance", "Uniswap", "Uniswap Governance Process" ] }, { "id": "propositional-logic", "title": "Propositional Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Propositional logic, also known as sentential logic or zeroth-order logic, is a formal system in which the basic units are whole propositions that are either true or false, combined using logical connectives such as conjunction, disjunction, negation, implication, and biconditional. It studies the truth-functional behaviour of compound statements and provides the foundation on which richer logics, including predicate logic, are built. Its decidability and clear semantics make it central to circuit design, satisfiability solving, and automated reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:propositional-logic", "labels": [ "Propositional Logic" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "proprietary-ai-video", "title": "Proprietary AI Video", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Proprietary AI Video denotes the class of closed-source, commercially operated Generative AI systems that synthesise temporally-coherent moving imagery conditioned on text prompts, reference images, existing video clips, audio signals, or structured control inputs (depth maps, pose skeletons,...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:proprietary-ai-video", "labels": [ "Proprietary AI Video" ], "is_subclass_of": [ "AI Application", "Model Training", "AI Video", "Generative AI", "Large-Scale Pretrained Foundation Model", "Diffusion Models", "Computer Vision" ], "wikilinks": [ "Advertising and Marketing", "APIServiceLayer", "Classifier Free Guidance", "ComputerVisionDomain", "CreativeAIDomain", "Diffusion Transformer", "FoundationModelLayer", "GenerativeAIDomain", "Open Source AI", "Runway", "Text-to-Video Generation", "Agentic Internet", "AI Adoption", "AI companions", "AI-GroundedDomain", "AI Liability", "AI Risks", "AI Safety", "AI Video", "AnimateDiff" ] }, { "id": "proprietary-evals", "title": "Proprietary Evals", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Internal, non-public evaluation benchmarks and datasets used by AI developers to measure and optimize model performance on specific, often domain-specific, tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:proprietary-evals", "labels": [ "Proprietary Evals" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [] }, { "id": "proprietary-format", "title": "Proprietary Format", "domain": "standards", "domain_name": "Standards", "definition": "A proprietary format is a data or file format whose specification is controlled by a single vendor and is not openly published or freely implementable. Access to and interpretation of the data typically depend on the vendor's software, which can constrain interoperability, archival longevity, and user control. Proprietary formats stand in contrast to open standards and are a common source of vendor lock-in.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:proprietary-format", "labels": [ "Proprietary Format" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "proprietary-image-generation", "title": "Proprietary Image Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Proprietary Image Generation refers to the class of closed-source, commercially operated text-to-image and multimodal-to-image generative AI systems whose model weights, training data, and inference pipelines are withheld from public release, typically delivered as API endpoints or subscription i...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:proprietary-image-generation", "labels": [ "Proprietary Image Generation" ], "is_subclass_of": [ "AI Application", "Generative AI", "Text-to-Image Generation", "Diffusion Model", "Multimodal AI", "Commercial AI Systems" ], "wikilinks": [ "Adobe Content Authenticity Initiative", "Adobe Creative Cloud", "Adobe Firefly", "Advertising Production", "Autoregressive Image Decoding", "Brand Asset Generation", "C2PA Content Provenance Standards", "ChatGPT Integration", "Classifier-Free Guidance", "CLIP Text Conditioning", "Commercial AI Systems", "Concept Art", "ControlNet", "Creative AI", "Creative Content Generation", "CreativeTechnologyDomain", "DALL-E 3", "Diffusion Model", "Digital Art", "Discord API" ] }, { "id": "proprietary-large-language-models", "title": "Proprietary Large Language Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Proprietary Large Language Models (PLLMs) are closed-weight, commercially deployed Foundation Models trained on multi--token corpora by well-resourced AI laboratories, where model weights, training data composition, and architectural specifics are withheld from the public under commercial lic...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:proprietary-large-language-models", "labels": [ "Proprietary Large Language Models" ], "is_subclass_of": [ "AI Model Architecture", "Transformer Architecture", "Cognitive AI", "Large-Scale Pretrained Foundation Model", "Large Language Models", "Natural Language Processing", "Generative AI" ], "wikilinks": [ "Chain of Thought", "CommercialAIDomain", "Edge AI", "EnterpriseIntegrationLayer", "FoundationModelDomain", "Human Feedback Data", "InferenceAPILayer", "Inference Infrastructure", "ISO IEC 42001", "Large Scale Compute", "NaturalLanguageProcessingDomain", "NIST AI RMF", "Open Source AI", "ProductLayer", "Retrieval Augmented Generation", "RLHF", "Safety Evaluation", "Safety Testing", "Self Hosted AI", "System Prompt" ] }, { "id": "proprietary-model", "title": "Proprietary Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A machine learning model whose weights, architecture, and training data are owned by a specific entity and distributed under restrictive licensing terms, as opposed to open-source or open-weight models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:proprietary-model", "labels": [ "Proprietary Model" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "proprietary-protocol", "title": "Proprietary Protocol", "domain": "standards", "domain_name": "Standards", "definition": "A proprietary protocol is a communication or data-exchange protocol owned and controlled by a single organisation, whose specification is closed, licensed, or undocumented rather than openly published. Such protocols can enable tight vendor integration and rapid iteration but tend to inhibit interoperability and create dependence on the controlling vendor. They contrast with open standards, which are publicly specified and freely implementable to promote cross-vendor compatibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:proprietary-protocol", "labels": [ "Proprietary Protocol" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "proprietary-software", "title": "Proprietary Software", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Proprietary software is software whose source code and usage rights are owned and controlled by a vendor, distributed under restrictive licences that limit copying, modification and redistribution. Users typically receive only compiled binaries and a licence granting specified use, while the owner retains the underlying intellectual property. It contrasts with open source, where source code is freely available under permissive or copyleft terms, and is often associated with commercial licensing and vendor control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:proprietary-software", "labels": [ "Proprietary Software" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "proprietary-specification", "title": "Proprietary Specification", "domain": "standards", "domain_name": "Standards", "definition": "A proprietary specification is a technical specification controlled by a single vendor or consortium that retains exclusive rights over its definition, evolution and use. Access may require licensing, non-disclosure or membership, and the owner can change it unilaterally. It contrasts with an open standard governed by a neutral body under royalty-free or fair terms, and it tends to increase the risk of vendor lock-in for adopters.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:proprietary-specification", "labels": [ "Proprietary Specification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "proprietary-video", "title": "Proprietary Video", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Proprietary Video denotes the class of video coding formats, digital rights management (DRM) systems, adaptive bitrate streaming protocols, and broadcast transmission standards whose specification, licensing, or enforcement mechanism is controlled by a corporation, patent pool, or standards body ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:proprietary-video", "labels": [ "Proprietary Video" ], "is_subclass_of": [ "Network and Communication", "Legal and Regulatory", "Content Delivery Network", "Video Codec Standards", "Digital Rights Management", "Streaming Protocols", "Broadcast Standards", "Intellectual Property", "Patent Licensing" ], "wikilinks": [ "4K UHD Delivery", "Addressable Advertising", "Advertising and Marketing", "Alliance for Open Media", "Amazon Prime Video", "Apple TV Plus", "ARM TrustZone", "ATSC", "ATSC 3.0", "ATSC Standards", "AV1", "AVS3", "BBC iPlayer", "Blu-ray Disc Association", "Blu-ray Disc Standards", "BroadcastDomain", "Broadcast Standards", "Broadcast Television", "BT Sport", "CMAF" ] }, { "id": "proprioception", "title": "Proprioception", "domain": "robotics", "domain_name": "Robotics", "definition": "Proprioception is a robot's internal sense of its own configuration and motion, derived from sensors that report joint angles, velocities, motor torques, body orientation and contact forces. Distinct from exteroceptive sensing of the external world, proprioceptive feedback lets a system estimate its pose and dynamics without external references, which is essential for stable balance, compliant interaction and dead-reckoning. It provides the fast, low-latency state signal that closed-loop motor controllers depend upon.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:proprioception", "labels": [ "Proprioception" ], "is_subclass_of": [ "Robot Perception", "Perception and Sensing" ], "wikilinks": [] }, { "id": "proprioceptive-sensing", "title": "Proprioceptive Sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Proprioceptive sensing is a robot's perception of its own internal state, such as joint angles, motor torques, link velocities, and body orientation, as distinct from exteroceptive sensing of the external environment. It relies on encoders, force/torque sensors, and inertial measurement units. It is essential for balance, whole-body control, and safe interaction, especially in legged and humanoid robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:proprioceptive-sensing", "labels": [ "Proprioceptive Sensing" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "proprioceptive-sensor", "title": "Proprioceptive Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "ProprioceptiveSensor is a transducer or sensing system that measures a robot's internal physical state \u2014 encompassing joint angle, angular velocity, linear and angular acceleration, motor torque, drive current, strain, and contact force \u2014 without reference to external landmarks or environmental f...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:proprioceptive-sensor", "labels": [ "Proprioceptive Sensor" ], "is_subclass_of": [ "Perception and Sensing", "Actuation and Control", "Sensor", "Robotic Subsystem", "Control Feedback Element", "State Estimation Component", "Mechatronic Component" ], "wikilinks": [ "Actuator", "BiSS-C Protocol", "CAN Bus", "Control Feedback Element", "ControlLayer", "ControlSystemsDomain", "Domain Randomisation", "EtherCAT", "Exoskeleton Control", "Hall Effect Sensing", "I2C", "IEC 61800 Adjustable Speed Drives", "IEEE 1451 Sensor Standard", "Inertial Measurement Unit", "ISO 9283 Robot Performance Standard", "Joint Encoder", "Legged Locomotion", "LiDAR", "Mechatronic Component", "MechatronicsDomain" ] }, { "id": "propulsion-efficiency", "title": "Propulsion Efficiency", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:propulsion-efficiency", "labels": [ "Propulsion Efficiency" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "propulsion-power-processing-unit", "title": "Propulsion Power Processing Unit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:propulsion-power-processing-unit", "labels": [ "Propulsion Power Processing Unit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "prosthetics", "title": "Prosthetics", "domain": "robotics", "domain_name": "Robotics", "definition": "Prosthetics is the field concerned with the design, fitting, and control of artificial devices that replace missing or impaired body parts to restore form and function. Modern prosthetic limbs increasingly incorporate actuators, sensors, and control systems, with advanced devices using myoelectric or neural signals to drive movement intuitively. The discipline draws on biomechanics, human-robot interaction, and assistive technology to match device behaviour to human intent and gait. It contrasts with exoskeletons and orthotics, which augment or support existing limbs rather than replacing them.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:prosthetics", "labels": [ "Prosthetics" ], "is_subclass_of": [ "Assistive Technology" ], "wikilinks": [] }, { "id": "protected-attributes", "title": "Protected Attributes", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Protected attributes are personal characteristics, such as race, gender, age, disability, or religion, that are legally or ethically safeguarded against discriminatory treatment in automated decision-making systems. Fairness metrics and bias mitigation techniques use protected attributes as the basis for measuring whether a model's predictions differ systematically across groups defined by those characteristics. Handling protected attributes correctly, including deciding whether to use them directly or as held-out variables for auditing, is central to building fair and legally compliant machine learning systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:protected-attributes", "labels": [ "Protected Attributes", "Protected Attribute" ], "is_subclass_of": [ "Fairness Metrics" ], "wikilinks": [] }, { "id": "protein-structure-prediction", "title": "Protein Structure Prediction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Protein structure prediction is the computational determination of a protein's three-dimensional folded structure from its amino-acid sequence. Deep-learning systems such as AlphaFold, built on attention-based architectures, achieved near-experimental accuracy and transformed structural biology. It is a landmark AI application with deep impact on drug discovery, molecular biology, and rational enzyme design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:protein-structure-prediction", "labels": [ "Protein Structure Prediction" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "proteus-effect", "title": "Proteus Effect", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Proteus Effect is the phenomenon whereby the visual characteristics of a user's digital avatar unconsciously shape that user's own behaviour, attitudes, and self-perception within virtual environments. Named after the shape-shifting Greek deity, it was empirically demonstrated by Yee and Bailenson (2007): users assigned taller or more attractive avatars exhibited corresponding shifts in confidence and social openness, while users embodying outgroup avatars showed reduced implicit bias. The effect arises through the intersection of self-perception theory, stereotype activation, and embodied cognition, and has direct implications for therapeutic applications, metaverse platform design, and ethical avatar-assignment practices.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:proteus-effect", "labels": [ "Proteus Effect" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Avatar Identity", "Digital Embodiment", "Virtual Psychology", "Metaverse", "Social Presence", "Virtual Group Dynamics" ] }, { "id": "proto-danksharding", "title": "Proto-Danksharding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Proto-danksharding is an Ethereum scaling upgrade, specified in EIP-4844, that introduces blob-carrying transactions to provide cheap, temporary data availability for layer-2 rollups. It adds a new transaction type carrying large binary data blobs that are not accessible to the Ethereum Virtual Machine and are pruned after a short retention period, with KZG commitments proving their contents. It is an incremental step towards the full danksharding design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:proto-danksharding", "labels": [ "Proto-Danksharding" ], "is_subclass_of": [ "Data Availability" ], "wikilinks": [] }, { "id": "protobuf", "title": "Protobuf", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Protocol Buffers (Protobuf) is a language-neutral, platform-neutral mechanism for serialising structured data using a compact binary wire format defined by an interface description in a .proto schema. A code generator produces typed accessors in many languages from that schema, enabling efficient, forwards- and backwards-compatible message exchange. It is the default payload format for gRPC and is widely used for high-throughput inter-service communication and storage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:protobuf", "labels": [ "Protobuf", "Protocol Buffers" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "protocol-buffer", "title": "Protocol Buffer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Protocol Buffer is a technology infrastructure concept and a type of Data Format.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:protocol-buffer", "labels": [ "Protocol Buffer", "Protocol Buffers" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "protocol-compatibility", "title": "Protocol Compatibility", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Protocol compatibility is the property that two or more implementations of a communication protocol can correctly exchange messages and interpret them consistently, whether they are identical versions or different versions designed to interwork. It requires agreement on message framing, semantics, and versioning strategy, often achieved through published technical standards. Protocol compatibility is a precondition for interoperability between independently developed systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:protocol-compatibility", "labels": [ "Protocol Compatibility" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "protocol-design", "title": "Protocol Design", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Protocol design is the systems-engineering discipline of specifying the message formats, state transitions, and interaction rules that govern communication between independent components. It balances correctness, efficiency, extensibility, and security concerns before a formal specification is written. Protocol design decisions - such as framing, versioning, and error-handling strategy - directly shape the reliability and interoperability of the resulting system.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:protocol-design", "labels": [ "Protocol Design" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [] }, { "id": "protocol-development", "title": "Protocol Development", "domain": "standards", "domain_name": "Standards", "definition": "Protocol development is the process of designing, specifying, implementing and iterating a technical communication or consensus protocol, from initial specification through reference implementation to formal standardisation. It typically proceeds through draft proposals, interoperability testing and governance review before a protocol is adopted as a compatibility standard. In blockchain ecosystems, protocol development is frequently funded through public-goods funding mechanisms that reward contributors to shared infrastructure.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:protocol-development", "labels": [ "Protocol Development" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "protocol-governance", "title": "Protocol Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocol Governance encompasses the processes, structures, and decision-making mechanisms by which changes to communication or consensus protocols are proposed, evaluated, ratified, and implemented across a decentralised participant base. Unlike traditional software governance, protocol governance must achieve coordination among parties with heterogeneous interests and no central authority, often relying on off-chain social processes (improvement proposals, mailing lists, developer calls) and on-chain voting mechanisms (token-weighted ballots, validator multisig). The choice of governance model has direct consequences for protocol security, decentralisation, upgrade velocity, and stakeholder alignment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:protocol-governance", "labels": [ "Protocol Governance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "protocol-layer", "title": "Protocol Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The ProtocolLayer represents the abstraction level of protocol specifications, implementations, communication standards, distributed algorithms, and coordination mechanisms that define how system components interact in blockchain and distributed systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:protocol-layer", "labels": [ "Protocol Layer", "ProtocolLayer" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "protocol-specification", "title": "Protocol Specification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A protocol specification is a precise, unambiguous document that defines how independent systems must exchange messages to interoperate, covering message formats, encodings, sequencing, state machines, error handling and timing. It serves as the authoritative contract that implementers follow so that conforming systems built by different parties can communicate reliably. Well-formed specifications separate the wire format and behaviour from any particular implementation, and they are typically published and maintained by standards bodies to ensure long-term interoperability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:protocol-specification", "labels": [ "Protocol Specification" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "protocol-stack", "title": "Protocol Stack", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A protocol stack is the layered set of network protocols that cooperate to provide communication services, where each layer offers services to the layer above and uses services of the layer below. Canonical examples are the OSI seven-layer model and the TCP/IP suite. Layering isolates concerns such as physical transmission, routing, transport reliability, and application semantics, enabling modular interoperable networking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:protocol-stack", "labels": [ "Protocol Stack", "IoT Protocol Stack" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "protocol-sustainability", "title": "Protocol Sustainability", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocol sustainability is the capacity of a blockchain or DeFi protocol to remain economically viable, secure, and well-governed over the long term without relying on unsustainable token emissions or external subsidies. It depends on durable revenue, balanced tokenomics, treasury resilience, and active governance that can fund maintenance and development. It contrasts with growth models that depend on continuous inflationary incentives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:protocol-sustainability", "labels": [ "Protocol Sustainability" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "protocol-upgrade", "title": "Protocol Upgrade", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A protocol upgrade is a coordinated change to a blockchain's consensus rules, transaction format, or feature set, deployed through mechanisms such as hard forks, soft forks, or on-chain governance votes. It requires network participants, including validators, miners, and node operators, to adopt new client software to remain in consensus with the upgraded chain. Protocol upgrades are typically preceded by proposal review, testnet deployment, and community signalling through processes such as snapshot governance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:protocol-upgrade", "labels": [ "Protocol Upgrade" ], "is_subclass_of": [ "Protocol Governance" ], "wikilinks": [] }, { "id": "protocol-upgrades", "title": "Protocol Upgrades", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocol upgrades are coordinated modifications to the consensus rules, transaction format, or feature set of a blockchain network, deployed through soft forks, hard forks, or activation mechanisms. They allow a decentralised system to evolve while preserving network continuity and, where possible, backward compatibility. Because no central authority can impose changes, upgrades typically require broad stakeholder coordination and on-chain or social signalling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:protocol-upgrades", "labels": [ "Protocol Upgrades", "Forkless Upgrades", "Protocol Upgrade" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "protocol-owned-liquidity", "title": "Protocol-Owned Liquidity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Protocol-owned liquidity (POL) is a DeFi treasury strategy in which a protocol acquires and holds its own trading-pair liquidity rather than renting it from third-party liquidity providers through mercenary incentives. By owning the liquidity, the protocol earns trading fees, gains durable market depth, and reduces dependence on emissions that can trigger capital flight. It was popularised by bonding mechanisms in OlympusDAO-style designs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:protocol-owned-liquidity", "labels": [ "Protocol-Owned Liquidity", "Protocol-Owned Liquidity Bonding" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "protocol", "title": "Protocol", "domain": "standards", "domain_name": "Standards", "definition": "A protocol is an agreed set of rules and message formats that govern how two or more parties exchange information or coordinate behaviour. It defines the syntax, semantics, timing and error handling needed for systems to interoperate reliably without prior bespoke arrangement. Protocols underpin networking, distributed systems and many forms of machine-to-machine and human-to-machine interaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:protocol", "labels": [ "Protocol", "Open Standards" ], "is_subclass_of": [ "Open Standards" ], "wikilinks": [] }, { "id": "protostar", "title": "Protostar", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:protostar", "labels": [ "Protostar" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "prototyping", "title": "Prototyping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Prototyping is the practice of building early, often incomplete representations of a product, system, or experience to explore ideas, test assumptions, and gather feedback before committing to full development. Prototypes range from low-fidelity sketches and wireframes to high-fidelity interactive or physical models, and they shorten the cycle between concept and validated learning. In spatial computing and product design it is central to iterative, user-centred development.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:prototyping", "labels": [ "Prototyping" ], "is_subclass_of": [ "Design Thinking" ], "wikilinks": [] }, { "id": "provenance-ontology-prov-o", "title": "Provenance Ontology (PROV-O)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "W3C standard ontology for representing and interchanging provenance information, capturing the origin, attribution, derivation, and lifecycle of digital entities through formal Entity-Activity-Agent relationships.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:provenance-ontology-prov-o", "labels": [ "Provenance Ontology (PROV-O)", "PROV-O", "PROV-O Ontology" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Activity Model", "Agent Model", "Attribution Model", "Audit Trails", "Blockchain Provenance", "Data Lineage Tracking", "Derivation Chains", "Entity Model", "Generation Events", "Influence Patterns", "Linked Data Platform", "Ontology Reasoner", "Qualified Relations", "RDF Store", "Scientific Reproducibility", "Semantic Reasoning Engine", "Trust Verification", "Usage Events", "W3C PROV Data Model", "W3C PROV-O Recommendation" ] }, { "id": "provenance-standard", "title": "Provenance Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical specifications and protocols for recording immutable ownership history and authenticity verification of digital assets through blockchain-based token IDs, contract addresses, and metadata that establish chain of custody from creation to present ownership.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:provenance-standard", "labels": [ "Provenance Standard", "W3C PROV" ], "is_subclass_of": [ "Standards and Interoperability", "Digital Asset Standards" ], "wikilinks": [ "Asset Authenticity", "Digital Asset Standards", "metaverse" ] }, { "id": "provenance-tracking", "title": "Provenance Tracking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Systems employing immutable distributed ledgers, IoT sensors, and smart contracts to create comprehensive audit trails tracking entities (products, data, AI model outputs) from origin to consumer or end-use, enabling rapid traceability, counterfeit prevention, regulatory compliance, and ethical sourcing verification across food safety, pharmaceuticals, luxury goods, and AI governance contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:provenance-tracking", "labels": [ "Provenance Tracking", "BC-0441-provenance-tracking", "Energy Provenance Tracking", "Provenance Tracker" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [ "Arianee", "Artory", "AWS Managed Blockchain", "Azure Blockchain Service", "BC-0013-smart-contracts", "BC-0029-permissioned-blockchain", "BC-0067-hyperledger-fabric", "BC-0432-consortium-blockchain", "BC-0434-blockchain-as-a-service", "BC-0442-pharmaceutical-traceability", "BC-0443-food-safety-blockchain", "BC-0444-luxury-goods-authentication", "BC-0445-conflict-mineral-tracking", "Blockchain in Transport Alliance", "Corda", "Cosmos", "Distributed Ledger Technology", "Everledger", "GS1", "GS1 Blockchain Working Group" ] }, { "id": "provenance-verification", "title": "Provenance Verification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Computational process for validating the origin, authenticity, and chain of custody of digital assets through metadata analysis and distributed ledger records.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:provenance-verification", "labels": [ "Provenance Verification" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Asset Authentication", "Authenticity Checking", "Chain of Custody Tracking", "Compliance Auditing", "ETSI ARF 010", "ISO 19115", "Ledger Record Verification", "Metadata Validation", "Ownership Validation", "Timestamp Authority", "Trust Establishment", "W3C PROV-O", "Asset Management", "Blockchain", "Cryptographic Verification", "DataLayer", "Digital Signatures", "Identity Management", "InfrastructureDomain", "Metadata Standards" ] }, { "id": "provenance", "title": "Provenance", "domain": "data", "domain_name": "Data", "definition": "The documented record of the origin, history and chain of custody of data or assets, used to establish authenticity, trust and accountability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:provenance", "labels": [ "Provenance", "AI Model Provenance", "Digital Provenance", "Provenance Attestation", "Provenance Chain", "Provenance Record", "Provenance Recorder", "Verifiable Provenance" ], "is_subclass_of": [ "Data Provenance" ], "wikilinks": [ "Audit Trail", "Trust", "Provenance Tracking", "Supply Chain", "Data Provenance" ] }, { "id": "provider-agnostic-dev-environment-tool", "title": "Provider-Agnostic Dev Environment Tool", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "DevPod is an open-source, provider-agnostic tool for creating reproducible cloud development environments defined by devcontainer specifications. It abstracts over compute providers \u2014 including local Docker, Kubernetes clusters, cloud VMs, and managed container services \u2014 allowing developers to spin up identical, pre-configured workspaces on any infrastructure without vendor lock-in. DevPod acts as a portable alternative to managed cloud IDE products such as GitHub Codespaces or GitPod, enabling teams to self-host development environments on their own Kubernetes or cloud infrastructure.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:provider-agnostic-dev-environment-tool", "labels": [ "Provider-Agnostic Dev Environment Tool", "devpod" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "provider", "title": "Provider", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A natural or legal person, public authority, agency or other body that develops an AI system or a general-purpose AI model, or that has an AI system or a general-purpose AI model developed, and places it on the market or puts the AI system into service under its own name or trademark, whether for payment or free of charge.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:provider", "labels": [ "Provider", "API Provider", "Service Provider" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "proxemics", "title": "Proxemics", "domain": "robotics", "domain_name": "Robotics", "definition": "The study and application of interpersonal and human-robot spatial relationships \u2014 including intimate, personal, social, and public distance zones \u2014 to design robotic systems that behave in spatially appropriate ways. Proxemics-aware robots adjust their approach trajectories, velocities, and stopping distances in response to detected human zones, improving perceived safety, comfort, and social acceptability in shared environments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:proxemics", "labels": [ "Proxemics" ], "is_subclass_of": [ "Human-Robot Interaction", "Robotics" ], "wikilinks": [ "HRI", "Robotics", "RoboticsDomain", "Social Robotics" ] }, { "id": "proximal-policy-optimisation", "title": "Proximal Policy Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A reinforcement learning algorithm that updates policies through incremental steps whilst constraining how much the policy can change via a clipped surrogate objective, preventing destabilising updates. PPO is the dominant RL algorithm used in reinforcement learning from human feedback (RLHF) for fine-tuning language models to align with human preferences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:proximal-policy-optimisation", "labels": [ "Proximal Policy Optimisation", "Proximal Policy Optimization" ], "is_subclass_of": [ "AI Technique", "Reinforcement Learning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "proximity-detection", "title": "Proximity Detection", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Proximity Detection is the computational process of determining the spatial distance and relative position between entities\u2014users, objects, or agents\u2014within a virtual or physical environment in real time. In metaverse and robotics contexts it underpins social-distance awareness, collision avoidance, interaction triggering, and context-sensitive content delivery. Implementations range from bounding-volume overlap tests and signed-distance fields in 3D engines to sensor-fusion pipelines combining lidar, ultrasound, and camera inputs in robotic systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:proximity-detection", "labels": [ "Proximity Detection" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "proximity-search", "title": "Proximity Search", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Geospatial query algorithms that locate nearby points of interest within a specified radius using coordinate-based indexing mods such as geohashing and quadtree structures for efficient location-based service applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:proximity-search", "labels": [ "Proximity Search" ], "is_subclass_of": [ "Content and Assets", "Geospatial Technology" ], "wikilinks": [ "Nearby Discovery", "Geospatial Technology", "metaverse" ] }, { "id": "proximity-sensor", "title": "Proximity Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An electronic device that detects the presence or distance of nearby objects without physical contact, using capacitive, inductive, ultrasonic, optical, or magnetic principles. Proximity sensors are foundational to robotic collision avoidance, autonomous navigation, and spatial mapping in both physical and hybrid physical-digital environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:proximity-sensor", "labels": [ "Proximity Sensor", "ProximitySensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "proximity", "title": "Proximity", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:proximity", "labels": [ "Proximity" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "proxy-pattern", "title": "Proxy Pattern", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Proxy Pattern is a smart-contract design technique that separates a contract's persistent storage and address from its executable logic, allowing the logic to be upgraded without migrating state or changing the public address. A lightweight proxy contract holds the state and forwards calls via delegatecall to a swappable implementation contract. The pattern enables upgradeable contracts but introduces storage-layout, initialisation and access-control hazards that must be managed carefully.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:proxy-pattern", "labels": [ "Proxy Pattern" ], "is_subclass_of": [ "Design Pattern" ], "wikilinks": [] }, { "id": "prudential-regulation-authority", "title": "Prudential Regulation Authority", "domain": "finance", "domain_name": "Finance", "definition": "The Prudential Regulation Authority (PRA) is the United Kingdom regulator responsible for the prudential supervision of banks, building societies, credit unions, insurers and major investment firms. Operating as part of the Bank of England, it sets capital, liquidity and governance standards to promote the safety and soundness of individual firms and to protect insurance policyholders. The PRA shares the UK's twin-peaks regulatory model with the conduct-focused Financial Conduct Authority.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:prudential-regulation-authority", "labels": [ "Prudential Regulation Authority" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "prudential-regulation", "title": "Prudential Regulation", "domain": "finance", "domain_name": "Finance", "definition": "Prudential regulation is the body of supervisory rules and oversight aimed at ensuring the safety and soundness of individual financial institutions and the stability of the wider financial system. It sets requirements for capital adequacy, liquidity, governance and risk management, and grants supervisors powers to monitor, intervene and resolve failing firms. Microprudential measures protect depositors and counterparties of single firms, while macroprudential measures address risks to the system as a whole.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:prudential-regulation", "labels": [ "Prudential Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "pruned-node", "title": "Pruned Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Pruned Node is a blockchain full node that validates the complete chain history during initial block download but subsequently discards spent transaction outputs and old block data beyond a configurable retention window, reducing on-disk storage requirements by orders of magnitude while retaining full validation capability for new blocks. Pruning enables operators with limited storage (e.g. 5\u201310 GB rather than 500+ GB for Bitcoin's full history) to participate in consensus verification without trusting third parties, unlike SPV light nodes that skip validation entirely. The pruned node can no longer serve historical block data to peers, constraining its contribution to network archival.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:pruned-node", "labels": [ "Pruned Node" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "Network Component" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "pruning", "title": "Pruning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pruning is a model compression technique that removes redundant or low-importance parameters from a neural network to reduce its size and computational cost while preserving accuracy. It ranges from unstructured pruning of individual weights to structured pruning of whole neurons, channels or attention heads. Pruned models are typically fine-tuned to recover any lost accuracy and can be deployed with lower memory and latency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pruning", "labels": [ "Pruning" ], "is_subclass_of": [ "Model Compression", "AI Technique" ], "wikilinks": [] }, { "id": "pseudonymisation", "title": "Pseudonymisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Pseudonymisation is a data de-identification technique, defined in GDPR Article 4(5), that processes personal data such that it can no longer be attributed to a specific data subject without additional information held separately under technical and organisational safeguards. It replaces direct identifiers\u2014names, national identification numbers, email addresses\u2014with pseudonyms such as tokens, encrypted identifiers, or keyed hashes, preserving data utility for analytics and machine learning while reducing but not eliminating re-identification risk. Unlike full anonymisation, pseudonymisation is reversible by an authorised party holding the supplementary mapping or key material.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:pseudonymisation", "labels": [ "Pseudonymisation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "GDPR Article 25", "GDPR Article 32", "GDPR Article 4(5)", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "pseudonymity", "title": "Pseudonymity", "domain": "security", "domain_name": "Security", "definition": "Pseudonymity is the property of operating under a persistent, stable identifier that is not directly linked to a participant's real-world legal identity, enabling consistent reputation and accountability without exposing personal information. Unlike full anonymity \u2014 where individual actions cannot be attributed to any consistent actor \u2014 pseudonymity preserves linkability across interactions under the chosen identifier while severing the mapping to biological or legal selfhood. The pseudonymous identifier may be a username, cryptographic public key, or blockchain address; its strength depends on the isolation of contextual signals that could enable de-anonymisation. Pseudonymity is foundational to privacy-preserving communication, decentralised identity systems, and selective-disclosure credential architectures.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:pseudonymity", "labels": [ "Pseudonymity" ], "is_subclass_of": [ "Privacy" ], "wikilinks": [ "Ring Signature", "Decentralized Identifier", "Privacy" ] }, { "id": "pseudorandom-function", "title": "Pseudorandom Function", "domain": "security", "domain_name": "Security", "definition": "A pseudorandom function (PRF) is a keyed family of functions whose outputs are computationally indistinguishable from those of a truly random function to any adversary lacking the key. PRFs are a foundational primitive in modern cryptography, providing the security guarantee that underlies message authentication codes, key derivation, and many symmetric protocols. Given the same key and input a PRF is deterministic, yet without the key its outputs reveal no exploitable structure. Closely related is the pseudorandom permutation, which adds invertibility and models block ciphers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:pseudorandom-function", "labels": [ "Pseudorandom Function", "Pseudo-Random Function" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "psychoacoustics", "title": "Psychoacoustics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Psychoacoustics is the scientific study of how humans perceive sound, relating physical properties of acoustic signals such as frequency, intensity and timing to subjective sensations of pitch, loudness, timbre and spatial location. It characterises perceptual phenomena including auditory masking, critical bands and localisation cues, and explains the limits and biases of the human auditory system. Its findings underpin perceptual audio coding, spatial audio rendering and hearing-aid design.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:psychoacoustics", "labels": [ "Psychoacoustics" ], "is_subclass_of": [ "Perception", "Audio Spatialization" ], "wikilinks": [] }, { "id": "psychological-phenomenon", "title": "Psychological Phenomenon", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Psychological Phenomenon, in the context of spatial computing, denotes any observable, reproducible effect on cognition, emotion, or behaviour that arises from interaction with immersive virtual environments. Established examples include the Proteus Effect (avatar-driven behavioural conformity), simulator sickness (sensorimotor conflict), flow states, social presence, and embodiment illusions. Understanding these phenomena is foundational to ethical metaverse design, therapeutic XR applications, and the study of human-computer interaction at scale.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:psychological-phenomenon", "labels": [ "Psychological Phenomenon" ], "is_subclass_of": [ "Software Engineering", "Digital Human Avatar Representation" ], "wikilinks": [] }, { "id": "psychological-research", "title": "Psychological Research", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scientific investigation of mental health applications within virtual reality and metaverse environments, examining therapeutic benefits for conditions including PTSD, anxiety, and depression, while assessing potential risks such as addiction and attention impairment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:psychological-research", "labels": [ "Psychological Research" ], "is_subclass_of": [ "Content and Assets", "Research Methods" ], "wikilinks": [ "Evidence Based VR Treatment", "metaverse", "Research Methods" ] }, { "id": "psychology", "title": "Psychology", "domain": "ai", "domain_name": "Ai", "definition": "Psychology is the scientific study of mind and behaviour, encompassing cognitive, developmental, social, and clinical subfields that examine how individuals perceive, learn, reason, and act.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:psychology", "labels": [ "Psychology" ], "is_subclass_of": [ "Artificial Intelligence", "AI Research Area" ], "wikilinks": [] }, { "id": "public-access", "title": "Public Access", "domain": "data", "domain_name": "Data", "definition": "Public access is a data-governance and infrastructure model in which digital resources, collections, or services are made openly available to the general public without restrictive authentication or paywalls. In cultural-heritage and digital-curation contexts it underpins open repositories, online catalogues, and reading-room digitisation programmes. It balances openness against rights management, preservation constraints, and equitable availability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:public-access", "labels": [ "Public Access" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "public-blockchain", "title": "Public Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A permissionless, decentralised distributed ledger network in which any party may participate, validate transactions, and inspect the chain state without requiring authorisation. Public blockchains such as Bitcoin and Ethereum achieve censorship resistance and immutability through global state replication and open consensus mechanisms, at the cost of reduced transaction throughput and higher energy consumption relative to permissioned alternatives.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:public-blockchain", "labels": [ "Public Blockchain", "PublicBlockchain" ], "is_subclass_of": [ "Distributed Ledger Technology" ], "wikilinks": [ "User Sovereignty", "UserSovereignty", "Bitcoin", "Cardano", "Censorship Resistance", "CensorshipResistance", "Ethereum", "Immutability", "MetaverseDomain", "PermissionedBlockchain" ] }, { "id": "public-consultation", "title": "Public Consultation", "domain": "governance", "domain_name": "Governance", "definition": "Public consultation is a formal process by which government bodies or organisations solicit input from affected individuals and groups before finalising a policy, regulation, or major decision, typically through published proposals, comment periods, and hearings. It is a specific, structured mechanism of stakeholder participation intended to widen legitimacy and surface concerns that internal deliberation might miss. Public consultation outcomes are commonly published alongside a summary of how the feedback shaped the final decision.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:public-consultation", "labels": [ "Public Consultation" ], "is_subclass_of": [ "Stakeholder Participation" ], "wikilinks": [] }, { "id": "public-goods-funding", "title": "public goods funding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Public goods funding encompasses the mechanisms, institutions, and coordination protocols that allocate capital toward resources which are non-excludable and non-rival \u2014 goods whose consumption by one party does not diminish availability to others and from which no party can be practically excluded. In open digital ecosystems, this addresses a structural market failure: rational actors under-invest in shared infrastructure because they can free-ride on others' contributions. Blockchain-native primitives such as quadratic funding, retroactive grants, and DAO-managed treasuries have revived and extended classical public economics, enabling transparent, on-chain allocation of capital toward open-source software, protocol research, and collective knowledge commons. The field sits at the intersection of mechanism design, cryptoeconomics, and welfare economics, driving continuous innovation in sybil resistance, matching fund architecture, and governance legitimacy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:public-goods-funding", "labels": [ "Public Goods Funding", "Continuous Public Goods Funding" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "public-goods", "title": "Public Goods", "domain": "governance", "domain_name": "Governance", "definition": "Public goods are goods or services that are non-excludable and non-rival in consumption, meaning that no one can be effectively prevented from using them and one person's use does not diminish availability to others. Because individuals cannot be charged directly, public goods are prone to under-provision through the free-rider problem and typically require collective or government funding. Examples include clean air, national defence, public knowledge, and open-source software.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:public-goods", "labels": [ "Public Goods" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "public-health", "title": "Public Health", "domain": "governance", "domain_name": "Governance", "definition": "The organised science and practice of protecting and improving the health of populations rather than individuals, encompassing disease surveillance, epidemiology, vaccination programmes, sanitation, food and water safety, health promotion, and emergency preparedness; delivered through governmental and international institutions, it works predominantly through prevention and policy, and increasingly depends on data infrastructure, computational modelling, and cross-sector coordination to detect and respond to threats at population scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:public-health", "labels": [ "Public Health" ], "is_subclass_of": [ "Healthcare" ], "wikilinks": [ "Healthcare", "Epidemiological Modelling", "Food Safety" ] }, { "id": "public-key-infrastructure", "title": "Public Key Infrastructure", "domain": "security", "domain_name": "Security", "definition": "Public Key Infrastructure (PKI) is the integrated set of roles, policies, hardware, software, and procedures used to create, manage, distribute, use, store, and revoke digital certificates and manage public-key encryption. PKI binds public keys to verified entity identities through a hierarchical trust chain anchored by Certificate Authorities (CAs), enabling secure authentication, data integrity, and confidential communication across distributed and internet-scale systems. It forms the foundational security layer for TLS/HTTPS, code signing, S/MIME email encryption, VPN access, and emerging decentralised identity frameworks. PKI standards are governed principally by IETF RFCs (X.509, PKIX), NIST guidelines, CA/Browser Forum Baseline Requirements, and ISO/IEC 27001 family controls.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:public-key-infrastructure", "labels": [ "Public Key Infrastructure", "Public-Key Infrastructure", "PublicKeyInfrastructure" ], "is_subclass_of": [ "Cryptographic Infrastructure" ], "wikilinks": [] }, { "id": "public-key", "title": "Public Key", "domain": "security", "domain_name": "Security", "definition": "The publicly shareable component of an asymmetric key pair, derived from the private key via elliptic curve or RSA mathematics, used to derive wallet addresses, verify digital signatures, and enable encrypted communication without transmitting secret material.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:public-key", "labels": [ "Public Key" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "public-opinion-on-ai", "title": "Public Opinion on AI", "domain": "governance", "domain_name": "Governance", "definition": "The aggregate attitudes, trust levels, and sentiment of the general population toward artificial intelligence technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:public-opinion-on-ai", "labels": [ "Public Opinion on AI" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [] }, { "id": "public-perception-of-ai", "title": "Public Perception of AI", "domain": "governance", "domain_name": "Governance", "definition": "The aggregate societal attitudes, beliefs, and sentiments regarding the impact, risks, and benefits of artificial intelligence as measured by surveys and demographic studies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:public-perception-of-ai", "labels": [ "Public Perception of AI" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "public-policy", "title": "Public Policy", "domain": "governance", "domain_name": "Governance", "definition": "Public policy is the set of decisions, principles, and courses of action that governments and public institutions adopt to address societal problems and pursue collective goals. It encompasses the formulation, adoption, implementation, and evaluation of measures spanning law, regulation, spending, and service delivery, informed by evidence and shaped by competing interests and values. In digital contexts it determines how states govern technology, data, infrastructure, and rights.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:public-policy", "labels": [ "Public Policy" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "public-trust-in-ai", "title": "Public Trust In Ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Public trust in AI is the degree of confidence that individuals and society place in artificial-intelligence systems to behave reliably, fairly, and in their interests. It is shaped by transparency, accountability, demonstrated safety, and the alignment of systems with human values. Sustaining public trust is widely seen as a precondition for broad and beneficial adoption of AI.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:public-trust-in-ai", "labels": [ "Public Trust In Ai", "Public Trust in AI" ], "is_subclass_of": [ "Trustworthy AI" ], "wikilinks": [] }, { "id": "public-wealth-fund", "title": "Public Wealth Fund", "domain": "economics", "domain_name": "Economics", "definition": "A sovereign or national investment vehicle that pools capital to invest in diversified assets, distributing returns directly to citizens to share in economic growth.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:public-wealth-fund", "labels": [ "Public Wealth Fund" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "public-key-cryptography", "title": "Public-Key Cryptography", "domain": "security", "domain_name": "Security", "definition": "An asymmetric cryptographic system using mathematically related key pairs \u2014 a public key for encryption or signature verification and a private key for decryption or signing \u2014 enabling secure communication, digital signatures, and authentication without requiring shared secrets. In blockchain systems it underpins wallet addresses, transaction signing, and identity verification.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:public-key-cryptography", "labels": [ "Public-Key Cryptography", "Public Key Cryptography" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "publish-subscribe-pattern", "title": "Publish-Subscribe Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A messaging design pattern in which message producers (publishers) emit typed events to named topics or channels without direct knowledge of consumers, and message consumers (subscribers) declare interest in specific topics to receive matching events asynchronously. A broker or event bus mediates delivery, fully decoupling senders from receivers in space, time, and control flow. This architectural separation enables independent scaling, fault isolation, and runtime addition or removal of participants without coordination. Pub/Sub underpins event-driven architectures, stream processing systems, and reactive microservices.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:publish-subscribe-pattern", "labels": [ "Publish-Subscribe Pattern", "Data-Centric Publish-Subscribe", "Publish Subscribe", "Publish-Subscribe Messaging" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Message Queue", "Distributed Systems", "Software Architecture" ] }, { "id": "pull-request", "title": "Pull Request", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A pull request is a proposal to merge a set of changes from one branch into another, packaged so that collaborators can review, discuss, and verify the work before it is integrated. It bundles a diff, a description, and a thread of review comments, and typically triggers automated checks that must pass before merge. Pull requests are the central unit of collaboration in distributed version control workflows, making change proposals visible, reviewable, and auditable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:pull-request", "labels": [ "Pull Request" ], "is_subclass_of": [ "Code Review" ], "wikilinks": [] }, { "id": "pulsar", "title": "Pulsar", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pulsar", "labels": [ "Pulsar" ], "is_subclass_of": [ "Neutron Star" ], "wikilinks": [] }, { "id": "pulse-width-modulation", "title": "Pulse Width Modulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Pulse-width modulation (PWM) is a technique that controls the average power delivered to a load by rapidly switching a signal between on and off and varying the proportion of on-time, the duty cycle. Because the switching frequency is high relative to the load's response, the load reacts to the average value, yielding efficient, near-lossless control of voltage, current, or position. It is ubiquitous in motor drives, power conversion, and digital generation of analogue-like outputs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:pulse-width-modulation", "labels": [ "Pulse Width Modulation", "Pulse-Width Modulation" ], "is_subclass_of": [ "Robotics", "Actuation and Control" ], "wikilinks": [] }, { "id": "pulse-code-modulation", "title": "Pulse-Code Modulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Pulse-code modulation (PCM) is a method for digitally representing sampled analogue signals by periodically measuring amplitude and quantising each sample to a fixed number of bits. It is the baseline uncompressed digital audio format underlying most audio signal processing pipelines and audio systems, from which compressed codecs are derived. Its two key parameters, sample rate and bit depth, jointly determine the fidelity and file size of the represented signal.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:pulse-code-modulation", "labels": [ "Pulse-Code Modulation" ], "is_subclass_of": [ "Audio Signal Processing" ], "wikilinks": [] }, { "id": "pulsed-plasma-thruster", "title": "Pulsed Plasma Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pulsed-plasma-thruster", "labels": [ "Pulsed Plasma Thruster" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "purchasing-power", "title": "Purchasing Power", "domain": "finance", "domain_name": "Finance", "definition": "Purchasing power is the quantity of goods and services that a unit of currency can buy at a given point in time. It declines as prices rise, so it is inversely related to inflation and is commonly tracked through price indices such as the consumer price index. Real purchasing power, adjusted for price changes, is central to comparing incomes, wages and asset values across time and between economies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:purchasing-power", "labels": [ "Purchasing Power" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "pure-proof-of-stake", "title": "Pure Proof of Stake", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A proof-of-stake consensus variant, exemplified by Algorand, that randomly selects block proposers and committee members from all token holders proportional to their stake weight using cryptographic sortition, without delegation mechanisms, enabling high decentralisation and Byzantine fault tolerance with immediate finality.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:pure-proof-of-stake", "labels": [ "Pure Proof of Stake", "Algorand Pure Proof of Stake" ], "is_subclass_of": [ "Protocol and Consensus", "Proof of Stake" ], "wikilinks": [ "Blockchain", "Proof of Stake" ] }, { "id": "push-notification", "title": "Push Notification", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A push notification is a server-initiated message delivered to a client device or application without an explicit client request, enabling real-time alerts about events, state changes, or incoming communications. It relies on persistent connections or platform push services that maintain a delivery channel even when the application is backgrounded. It is a foundational primitive for presence, messaging, and event-driven user engagement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:push-notification", "labels": [ "Push Notification", "Notification Message" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "pushbroom-scanner", "title": "Pushbroom Scanner", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:pushbroom-scanner", "labels": [ "Pushbroom Scanner" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "py-torch", "title": "PyTorch", "domain": "ai", "domain_name": "Ai", "definition": "PyTorch is an open-source deep learning framework developed by Meta AI Research that provides a dynamic computation graph, automatic differentiation via autograd, and tight integration with Python for flexible model development and research. It has become the dominant framework in academic machine learning research and is widely used in production via TorchServe and TorchScript. PyTorch's tensor operations are GPU-accelerated through CUDA, and its ecosystem encompasses libraries such as TorchVision, TorchAudio, and PyTorch Lightning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:py-torch", "labels": [ "PyTorch", "TorchScript" ], "is_subclass_of": [ "Machine Learning Framework" ], "wikilinks": [] }, { "id": "pyodide-knowledge-graph-node-enumerator", "title": "Pyodide Knowledge Graph Node Enumerator", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Python to list nodes is a Logseq Pyodide script that iterates over all pages in the knowledge graph, filters those with the `public:: true` metadata property, and returns a comma-separated list of public page names. It demonstrates direct integration of Python scripting with the Logseq Plugin API for programmatic knowledge-graph introspection and inventory generation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:pyodide-knowledge-graph-node-enumerator", "labels": [ "Pyodide Knowledge Graph Node Enumerator", "Python to list nodes" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "pyodide-rag-corpus-builder-script", "title": "Pyodide RAG Corpus Builder Script", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Python to build RAG data page is a Logseq Pyodide script that enumerates all pages with the `public:: true` property, loads their block content, applies text-cleaning routines (URL removal, bracket stripping, special-character normalisation), and appends the cleaned content to a dedicated FULLRAG page. This page then serves as the consolidated corpus for local Retrieval-Augmented Generation pipelines, enabling semantic search over the entire public knowledge graph without external data transfer.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:pyodide-rag-corpus-builder-script", "labels": [ "Pyodide RAG Corpus Builder Script", "Python to build RAG data page" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "..." ] }, { "id": "python-3", "title": "Python 3", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Python 3 is the current major series of the Python programming language, a high-level, dynamically typed, interpreted language emphasising readability and a large standard library. It is the dominant language for machine learning, data science, automation, and robotics scripting, supported by an extensive ecosystem of scientific and AI packages. Python 3 introduced Unicode-by-default strings and other breaking changes relative to the discontinued Python 2 series.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:python-3", "labels": [ "Python 3", "Python 3.6+" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "python-programming-language", "title": "Python Programming Language", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Python is a high-level, interpreted, general-purpose programming language emphasising code readability and a clean syntax. It has become the dominant language for machine learning and data science due to its extensive ecosystem of numerical and scientific libraries. Python supports multiple programming paradigms including procedural, object-oriented, and functional styles, and its dynamic typing and interactive REPL accelerate experimentation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:python-programming-language", "labels": [ "Python Programming Language" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "python-pytorch-deep-learning-stack", "title": "Python PyTorch Deep Learning Stack", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Python and PyTorch together form the dominant open-source stack for deep learning research and production. Python provides the high-level scripting environment and ecosystem (NumPy, Hugging Face, spaCy), whilst PyTorch supplies dynamic computational graph execution, autograd differentiation, and GPU acceleration via CUDA, enabling rapid prototyping and deployment of neural network models.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:python-pytorch-deep-learning-stack", "labels": [ "Python PyTorch Deep Learning Stack", "PyTorch", "Python and PyTorch" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "ComfyUI", "Large Language Models", "Machine Learning", "Nostr protocol", "Python and PyTorch" ] }, { "id": "python-runtime", "title": "python runtime", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Python runtime is the interpreter process and associated standard library environment responsible for compiling Python source to bytecode, executing that bytecode in a virtual machine, managing heap memory through reference-counting combined with a cyclic garbage collector, and resolving module imports via the package search path. CPython\u2014the canonical C-language reference implementation\u2014uses a Global Interpreter Lock (GIL) that serialises bytecode execution across threads, making multi-processing the preferred concurrency model for CPU-bound workloads. The Python runtime underpins the dominant AI/ML toolchain including PyTorch, TensorFlow, and the Hugging Face ecosystem, making its performance characteristics, extension mechanisms, and packaging conventions central concerns for AI infrastructure engineering.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:python-runtime", "labels": [ "Python Runtime" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "python-sample1", "title": "Python Sample1", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Python Sample1 is a Logseq Python code block that demonstrates the Logseq Plugin API via Pyodide: it defines helper utilities for inspecting JavaScript objects, appending timestamped entries to a named Logseq page, and searching all pages for a given term using the Logseq search API. The sample illustrates the pattern of bridging Python scripting with the Logseq knowledge-graph API for automated page discovery and log management.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:python-sample1", "labels": [ "Python Sample1" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "python-sample2", "title": "Python Sample2", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Python Sample2 is a minimal Logseq JavaScript/Pyodide evaluation block used as a stub execution test. It demonstrates the evalblock mechanism that allows inline code execution within the Logseq knowledge graph, serving as a baseline template for more complex Python integrations that interact with the Logseq Plugin API.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:python-sample2", "labels": [ "Python Sample2" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "python", "title": "python", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Python is a high-level, dynamically typed, interpreted programming language designed by Guido van Rossum with an explicit emphasis on code readability, expressive syntax, and developer productivity. Its reference implementation, CPython, executes code via a bytecode interpreter protected by the Global Interpreter Lock (GIL), whilst third-party runtimes such as PyPy provide JIT compilation for CPU-bound workloads. Python's extensive standard library and the PyPI ecosystem\u2014encompassing NumPy, Pandas, PyTorch, TensorFlow, Scikit-learn, and the Hugging Face stack\u2014have established it as the dominant language for machine learning research, data science, scientific computing, and AI infrastructure automation. Its clean syntax, interactive tooling via Jupyter notebooks, and first-class cloud-platform support have made it the lingua franca of artificial intelligence development globally.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:python", "labels": [ "Python" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "q-learning", "title": "Q Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Q-learning is a model-free, off-policy reinforcement learning algorithm that learns the value of taking a given action in a given state by iteratively updating an action-value (Q) function towards a bootstrapped Bellman target. Because it learns the optimal action-value function regardless of the policy being followed, it converges to optimal behaviour without requiring a model of the environment's dynamics. It is a foundational algorithm extended by deep Q-networks for high-dimensional problems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:q-learning", "labels": [ "Q Learning", "Q-Learning" ], "is_subclass_of": [ "Reinforcement Learning", "Reinforcement Learning Algorithm" ], "wikilinks": [] }, { "id": "qbft", "title": "QBFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "QBFT is a Byzantine fault tolerant consensus algorithm used in enterprise Ethereum clients to provide immediate finality among a known set of validators. It tolerates a minority of faulty or malicious validators.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:qbft", "labels": [ "QBFT", "QBFT Consensus", "QBFT Protocol" ], "is_subclass_of": [ "Byzantine Fault Tolerance" ], "wikilinks": [ "Consensus Algorithm", "Byzantine Fault Tolerance", "Distributed Ledger Technology", "Ethereum" ] }, { "id": "qlo-ra", "title": "QLoRA", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An extension of LoRA (Low-Rank Adaptation) that combines 4-bit NormalFloat quantisation of frozen base model weights with full-precision trainable low-rank adapter matrices. QLoRA additionally employs double quantisation and paged optimisers to achieve extreme memory efficiency, enabling fine-tuning of 65B-parameter models on a single 48GB GPU without significant performance degradation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:qlo-ra", "labels": [ "QLoRA" ], "is_subclass_of": [ "AI Technique", "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "qr-code", "title": "QR Code", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A QR code (Quick Response code) is a two-dimensional matrix barcode that encodes data in a grid of black and white modules readable by a camera from any orientation. It carries substantially more information than a one-dimensional barcode and includes Reed-Solomon error correction that allows reliable decoding despite damage or partial occlusion. QR codes are standardised under ISO/IEC 18004 and are widely used to bridge physical objects to digital resources, including payment requests, product identifiers and verifiable credentials.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:qr-code", "labels": [ "QR Code" ], "is_subclass_of": [ "Traceability" ], "wikilinks": [] }, { "id": "quic", "title": "QUIC", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "QUIC is a general-purpose transport layer protocol standardised as RFC 9000 by the IETF, running atop UDP rather than TCP in order to eliminate head-of-line blocking inherent in byte-stream semantics. It integrates TLS 1.3 cryptographic handshaking into the connection establishment phase, reducing round-trip latency to as little as zero additional RTT for resumed sessions, and multiplexes independent byte-streams so that packet loss on one stream does not stall others. Developed initially at Google and later standardised as the foundation for HTTP/3 (RFC 9114), QUIC also supports seamless connection migration across changing IP addresses, making it particularly suited to mobile and lossy network environments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:quic", "labels": [ "QUIC", "HTTP/3 QUIC", "IETF QUIC", "IETF RFC 9000 QUIC", "QUIC Protocol", "QUIC Transport" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "Communication Protocol", "Scalability", "Network Architecture" ] }, { "id": "qt-framework", "title": "Qt Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Qt is a cross-platform C++ application and UI framework providing widgets, a declarative QML language, signals-and-slots event handling, and abstractions for graphics, networking, and threading. It enables a single codebase to target desktop, mobile, and embedded platforms with native look and feel. It is widely used to build performant desktop clients and 3D-capable interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:qt-framework", "labels": [ "Qt Framework" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "quadcopter", "title": "Quadcopter", "domain": "robotics", "domain_name": "Robotics", "definition": "A vertical take-off and landing (VTOL) aerial robot with four fixed-pitch rotors arranged symmetrically around a central frame, achieving attitude control and thrust modulation by differentially varying rotor speeds. Quadcopters are mechanically simple (no swashplate), highly manoeuvrable, and statically stable in hover, making them the dominant platform for consumer UAVs, autonomous aerial robotics research, and inspection applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:quadcopter", "labels": [ "Quadcopter" ], "is_subclass_of": [ "Robot Type", "Multirotor UAV" ], "wikilinks": [ "Multirotor UAV", "Robotics" ] }, { "id": "quadratic-funding", "title": "quadratic funding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Quadratic Funding (QF) is a mechanism-design method for subsidising public goods in which a matching pool is distributed proportionally to the square of the sum of square roots of individual contributions to each project, ensuring that projects with many small donors receive proportionally greater matching than projects with few large donors. Formalised by Vitalik Buterin, Zoe Hitzig, and E. Glen Weyl in the 2018 paper 'Liberal Radicalism', QF is derived from welfare-economics theory as an approximation to the socially optimal provision of public goods under privately observed preferences. It has been deployed on-chain via Gitcoin Grants, where smart contracts automate matching calculations against a community-funded subsidy pool, though it requires robust Sybil resistance mechanisms such as decentralised identity verification to prevent gaming.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:quadratic-funding", "labels": [ "Quadratic Funding" ], "is_subclass_of": [ "Mechanism Design" ], "wikilinks": [] }, { "id": "quadratic-programming", "title": "Quadratic Programming", "domain": "robotics", "domain_name": "Robotics", "definition": "Quadratic programming (QP) is a class of mathematical optimisation problems in which a quadratic objective function is minimised subject to linear equality and inequality constraints. It is a convex optimisation problem when the quadratic term is positive semidefinite, admitting efficient and globally optimal solvers. QP is foundational to model-based control, trajectory optimisation, and constrained robotic motion generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:quadratic-programming", "labels": [ "Quadratic Programming", "Nonlinear Programming" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "quadratic-voting", "title": "Quadratic Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Collective decision-making mechanism where the cost of casting n votes on any single proposal equals n\u00b2 voice credits, formalising preference-intensity revelation whilst resisting plutocratic capture \u2014 developed by Steven Lalley and E.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:quadratic-voting", "labels": [ "Quadratic Voting", "Quadratic Voting Analytics" ], "is_subclass_of": [ "Governance and Regulation", "Public Goods Funding", "Mechanism Design", "Voting Theory", "Governance Protocol", "Decentralised Autonomous Organisation", "Welfare Economics" ], "wikilinks": [ "Allo Protocol", "Allo Protocol v2 GitHub", "Anti-Plutocratic Governance", "BlockScience QF Risk Parameters 2022", "BrightID", "Buterin 2019 On Collusion", "Buterin 2019 Pairwise QF Ethresear", "Buterin-Hitzig-Weyl 2019 Liberal Radicalism Management Science", "Circom", "Civic Technology", "clr.fund Protocol", "clr.fund Protocol v1.1", "Commit-Reveal Scheme", "Common Pool Resources", "Commons Stack Mechanism Comparison 2022", "Community-Led Allocation", "DeFi Governance", "Decentralised Autonomous Organisation", "Digital Democracy", "Edinburgh BLT EPSRC Year 1 Report 2024" ] }, { "id": "quadruped-robot", "title": "Quadruped Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotic platform that locomotes on four legs using gaits inspired by mammals such as dogs and cats. Quadruped robots employ reinforcement learning, model-predictive control, and whole-body dynamics to achieve robust locomotion over rough terrain, enabling deployment in inspection, search-and-rescue, and logistics applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:quadruped-robot", "labels": [ "Quadruped Robot" ], "is_subclass_of": [ "Robot Type", "Legged Robot" ], "wikilinks": [ "Legged Robot", "Robotics" ] }, { "id": "qualified-custodian", "title": "Qualified Custodian", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A qualified custodian is a regulated financial institution authorised to hold client assets, including digital assets, under fiduciary and supervisory standards set by securities regulators. It provides segregated accounts, independent audit and statutory protections that distinguish it from informal custody arrangements. For institutional investors, using a qualified custodian is often a regulatory precondition for holding crypto assets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:qualified-custodian", "labels": [ "Qualified Custodian" ], "is_subclass_of": [ "Digital Asset Custody" ], "wikilinks": [] }, { "id": "qualified-electronic-signature", "title": "Qualified Electronic Signature", "domain": "security", "domain_name": "Security", "definition": "A qualified electronic signature (QES) is the highest-assurance class of electronic signature defined by the European Union's eIDAS regulation, created with a qualified signature-creation device and backed by a qualified certificate issued by an accredited trust service provider. It is legally equivalent to a handwritten signature across the EU and provides strong guarantees of signer identity and document integrity. QES underpins cross-border legal acts requiring the highest evidentiary weight.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:qualified-electronic-signature", "labels": [ "Qualified Electronic Signature" ], "is_subclass_of": [ "Electronic Signature" ], "wikilinks": [] }, { "id": "quality-assurance", "title": "Quality Assurance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Quality Assurance (QA) is the systematic discipline of establishing and maintaining defined standards of correctness, reliability, safety, and fitness-for-purpose across the full lifecycle of software systems, AI models, and digital infrastructure. It encompasses planned and systematic activities \u2014 including requirements analysis, process audits, test design, validation, verification, and continuous monitoring \u2014 that prevent defects from reaching production rather than merely detecting them after the fact. In AI contexts, QA extends beyond functional correctness to encompass model fairness, adversarial robustness, data quality, and distributional-shift monitoring, increasingly mandated by governance frameworks such as the EU AI Act and ISO/IEC 42001. Effective QA integrates with DevOps and MLOps pipelines through automated gates that enforce quality thresholds before any artefact is promoted to the next deployment stage.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:quality-assurance", "labels": [ "Quality Assurance", "Quality Assurance System", "Quality Assurance Testing", "Quality Assurance Toolchain", "Quality Assurance Workflows", "QualityAssurance" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "quality-control", "title": "Quality Control", "domain": "ai", "domain_name": "Ai", "definition": "Quality control is the operational process of inspecting, testing, and verifying products or outputs against defined specifications to detect and remove defects before delivery. In manufacturing it spans dimensional inspection, surface-defect detection, and statistical process control, increasingly automated through machine vision. It reduces scrap, ensures conformance, and provides feedback for process improvement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:quality-control", "labels": [ "Quality Control", "Industrial Quality Control" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "quality-estimation", "title": "Quality Estimation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The task of predicting the quality of machine translation output without access to reference translations, at sentence, word, or document level, enabling production systems to route low-confidence translations to human post-editing, filter parallel corpora, and gate automated publication \u2014 in contrast to reference-based metrics that require gold translations for scoring.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:quality-estimation", "labels": [ "Quality Estimation" ], "is_subclass_of": [ "Machine Translation" ], "wikilinks": [ "Machine Translation", "COMET Metric", "Back-Translation", "BLEU Score" ] }, { "id": "quality-mask", "title": "Quality Mask", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:quality-mask", "labels": [ "Quality Mask" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "quality-of-service", "title": "Quality Of Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Quality of Service (QoS) is a set of network management techniques, policies, and protocols that prioritise, shape, and guarantee specified performance characteristics\u2014including latency, jitter, throughput, and packet loss\u2014for distinct traffic classes traversing shared network infrastructure. QoS mechanisms operate at multiple OSI layers, using traffic classification, queuing disciplines, scheduling algorithms, and admission control to meet differentiated service-level objectives. In real-time systems such as XR, telepresence, and cloud-native distributed applications, QoS is essential for maintaining acceptable user experience under variable load. Standardised frameworks including IntServ, DiffServ, and IEEE 802.1p provide interoperable models for end-to-end QoS across heterogeneous networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:quality-of-service", "labels": [ "Quality Of Service", "Quality of Service", "Quality of Service Manager", "Quality of Service Policy", "QualityOfService" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "quality-screening", "title": "Quality Screening", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:quality-screening", "labels": [ "Quality Screening" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "quality-standard", "title": "Quality Standard", "domain": "robotics", "domain_name": "Robotics", "definition": "A Quality Standard is a formally published specification that defines measurable requirements for quality management, process control, testing methodology, and conformance verification in robot manufacturing and deployment contexts. Quality standards provide the normative benchmarks against which robotic systems are assessed for reliability, safety, and fitness for purpose across their operational lifecycle.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:quality-standard", "labels": [ "Quality Standard" ], "is_subclass_of": [ "Safety and Standards", "Robot Standard" ], "wikilinks": [ "Robot Standard", "Robotics" ] }, { "id": "quantexa", "title": "Quantexa", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantexa is a company that provides data analytics software for connecting and analysing data to support decision intelligence, including fraud and financial crime detection.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:quantexa", "labels": [ "Quantexa" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [ "Data Analytics", "Fraud Detection", "Anti-Money Laundering" ] }, { "id": "quantisation", "title": "quantisation", "domain": "ai", "domain_name": "Ai", "definition": "Quantisation is a model compression technique that reduces the numerical precision of a neural network's weights and activations from high-precision floating-point formats used during training (typically FP32 or BF16) to lower bit-widths such as INT8, INT4, FP8, or binary representations. This precision reduction decreases memory footprint and accelerates matrix operations on hardware with native low-precision arithmetic units, enabling deployment of large models on memory-constrained or power-limited devices with minimal accuracy degradation. The two primary paradigms are post-training quantisation (PTQ), which applies precision reduction after training using a calibration dataset, and quantisation-aware training (QAT), which simulates low-precision arithmetic during the forward pass so the model adapts its weight distributions to minimise quantisation error. Advanced algorithms such as GPTQ, AWQ, and SmoothQuant extend these approaches to very large language models and diffusion models.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:quantisation", "labels": [ "Quantisation", "INT8 Quantisation", "Quantization" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "quantitative-easing", "title": "Quantitative Easing", "domain": "finance", "domain_name": "Finance", "definition": "Quantitative easing (QE) is an unconventional monetary policy in which a central bank purchases large quantities of longer-dated financial assets, typically government bonds, to inject liquidity into the financial system and lower long-term interest rates. It is deployed when short-term policy rates are already near zero and conventional rate cuts are exhausted. By expanding the central bank's balance sheet and the monetary base, QE aims to stimulate lending, asset prices and aggregate demand.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:quantitative-easing", "labels": [ "Quantitative Easing" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "quantitative-finance", "title": "Quantitative Finance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Quantitative finance is the application of mathematical models, statistical methods, and computational techniques to financial markets for pricing, risk management, and trading. It draws heavily on stochastic calculus, probability theory, and optimisation to value derivatives, model asset dynamics, and construct portfolios. It is the analytical backbone of derivatives desks, algorithmic trading, and risk analytics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:quantitative-finance", "labels": [ "Quantitative Finance" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "quantum-ai", "title": "Quantum AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantum AI is a research area at the intersection of quantum computing and artificial intelligence, applying quantum mechanical phenomena \u2014 superposition, entanglement, and interference \u2014 to accelerate or qualitatively improve machine learning algorithms, combinatorial optimisation, and probabilistic inference. Quantum circuits can in principle represent exponentially large state spaces with polynomial resources, offering potential speed-ups for specific learning and search tasks beyond classical limits. The field encompasses variational quantum algorithms, quantum neural networks, quantum-enhanced sampling, and the use of quantum hardware as accelerators within classical AI pipelines.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:quantum-ai", "labels": [ "Quantum AI" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "quantum-computation-paradigm", "title": "Quantum Computation Paradigm", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A paradigm of computation that exploits quantum-mechanical phenomena\u2014superposition, entanglement, and interference\u2014to perform certain calculations exponentially faster than classical computers. Quantum computing holds particular relevance for cryptography, optimisation, simulation of molecular systems, and potentially accelerating machine learning workloads, though near-term devices remain limited by qubit decoherence and error rates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:quantum-computation-paradigm", "labels": [ "Quantum Computation Paradigm", "Quantum Computing", "Quantum-Computing", "QuantumComputing" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "quantum-computing", "title": "Quantum Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The engineering discipline and technology stack that performs computation using quantum-mechanical phenomena \u2014 superposition, entanglement and interference \u2014 over registers of qubits, spanning hardware platforms (superconducting circuits, trapped ions, photonics, neutral atoms), error correction, and algorithms such as Shor's factoring and Grover's search that offer provable or conjectured speed-ups over classical computation for specific problem classes including cryptanalysis, simulation of quantum systems and combinatorial optimisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:quantum-computing", "labels": [ "Quantum Computing" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [ "Quantum Computation Paradigm", "Qubit", "Computational Complexity Theory", "Combinatorial Optimisation" ] }, { "id": "quantum-cryptography", "title": "Quantum Cryptography", "domain": "security", "domain_name": "Security", "definition": "Quantum cryptography uses principles of quantum mechanics, such as the no-cloning theorem and measurement disturbance, to perform cryptographic tasks whose security rests on physical law rather than computational hardness. Its best-known application, quantum key distribution, lets two parties establish a shared secret with detectable eavesdropping. It is distinct from post-quantum cryptography, which is classical algorithms resistant to quantum attack.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:quantum-cryptography", "labels": [ "Quantum Cryptography" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "quantum-error-correction", "title": "Quantum Error Correction", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Quantum error correction is the set of techniques that protect quantum information against decoherence and operational noise by encoding a logical qubit redundantly across many physical qubits. Stabiliser measurements detect errors without collapsing the encoded state, allowing the system to diagnose and reverse bit-flip and phase-flip faults. It is the central prerequisite for fault-tolerant quantum computation, where logical error rates can be driven arbitrarily low provided physical error rates fall below a threshold.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:quantum-error-correction", "labels": [ "Quantum Error Correction" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [] }, { "id": "quantum-gate", "title": "Quantum Gate", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A reversible, unitary operation applied to a small number of qubits that transforms their joint quantum state, serving as the fundamental building block of quantum circuits in the gate-based model of quantum computation. Quantum gates are the analogue of classical logic gates but, being unitary matrices acting on complex state vectors, they can create superposition and entanglement, and any quantum algorithm can be decomposed into sequences drawn from a small universal gate set.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:quantum-gate", "labels": [ "Quantum Gate" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [ "Quantum Computation Paradigm", "Quantum Mechanics", "Quantum Error Correction", "Linear Algebra" ] }, { "id": "quantum-key-distribution", "title": "Quantum Key Distribution", "domain": "security", "domain_name": "Security", "definition": "A method of securely sharing cryptographic keys using the principles of quantum mechanics, such that any eavesdropping disturbs the transmission and can be detected. It provides key exchange whose security rests on physics rather than computational hardness.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:quantum-key-distribution", "labels": [ "Quantum Key Distribution" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [ "Quantum Computing", "Post-Quantum Cryptography", "Cryptography" ] }, { "id": "quantum-mechanics", "title": "Quantum Mechanics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Quantum mechanics is the physical theory describing matter and energy at atomic and subatomic scales, where observable quantities are discrete and systems are represented by state vectors in a complex Hilbert space evolving under unitary dynamics. Core principles such as superposition, entanglement, measurement collapse and the uncertainty relation depart sharply from classical intuition. It provides the foundational substrate for quantum computing, quantum cryptography and the engineering of qubit-based information systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:quantum-mechanics", "labels": [ "Quantum Mechanics" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [] }, { "id": "quantum-network-node", "title": "Quantum Network Node", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical device utilizing quantum mechanics principles to enable quantum key distribution (QKD) or entanglement transmission for ultra-secure communication channels.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:quantum-network-node", "labels": [ "Quantum Network Node" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Classical Communication Interface", "Cryogenic Cooling", "Entanglement Distribution", "Environmental Isolation Chamber", "ITU-T QKD Series", "Optical Fiber Connection", "Optical Switch", "Post-Quantum Security", "Quantum Communication Network", "Quantum Cryptography", "Quantum Key Distribution", "Quantum Light Source", "Quantum Memory Unit", "Single Photon Detector", "Ultra-Secure Communication", "Vibration Isolation", "InfrastructureDomain", "Network Infrastructure", "Physical Layer", "Power Supply" ] }, { "id": "quantum-threat-to-cryptography", "title": "quantum threat to cryptography", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Existential risk to public-key cryptographic systems (ECDSA, RSA) from sufficiently powerful quantum computers capable of running Shor's algorithm. Q-Day\u2014the point at which quantum advantage breaks deployed cryptography\u2014threatens the security foundations of blockchain networks, TLS, and digital identity systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:quantum-threat-to-cryptography", "labels": [ "Quantum Threat to Cryptography" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "quasar", "title": "Quasar", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:quasar", "labels": [ "Quasar" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "quasi-monte-carlo", "title": "Quasi Monte Carlo", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quasi-Monte Carlo (QMC) is a family of numerical integration and sampling methods that replace the pseudo-random points of classical Monte Carlo with deterministic low-discrepancy sequences. By spreading sample points more evenly across the integration domain, QMC achieves faster asymptotic convergence than standard Monte Carlo for many smooth, moderate-dimensional integrals. It is widely used in computational finance, computer graphics, and uncertainty quantification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:quasi-monte-carlo", "labels": [ "Quasi Monte Carlo", "Quasi-Monte Carlo" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [] }, { "id": "quasi-direct-drive", "title": "Quasi-Direct Drive", "domain": "robotics", "domain_name": "Robotics", "definition": "Quasi-direct drive (QDD) is an actuation architecture that pairs a high-torque electric motor with a low-ratio gearbox to achieve high force fidelity, backdrivability, and impact tolerance while retaining usable torque density. By minimising gear reduction it preserves transparency between the motor and the load, enabling accurate force sensing and control without dedicated torque sensors. It is a dominant approach in legged and dynamic robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:quasi-direct-drive", "labels": [ "Quasi-Direct Drive", "Quasi Direct Drive Actuator" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "quaternion-math", "title": "Quaternion Math", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Quaternion Math is the application of Hamilton's four-dimensional number system (q = w + xi + yj + zk) to represent and interpolate 3D rotations and orientations in spatial computing. Quaternions avoid gimbal lock inherent to Euler angles, enable smooth spherical linear interpolation (SLERP), and are computationally efficient for composing rotations in real-time rendering engines, robotics kinematics, and XR head-tracking pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:quaternion-math", "labels": [ "Quaternion Math", "Quaternion", "Quaternion Mathematics", "Quaternion Representation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "qubit", "title": "Qubit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The fundamental unit of quantum information: a two-level quantum system whose state is a complex linear superposition \u03b1|0\u27e9 + \u03b2|1\u27e9 of computational basis states, capable of entanglement with other qubits and collapsing probabilistically on measurement; physically realised as superconducting transmon circuits, trapped-ion energy levels, photon polarisation, neutral-atom states or electron spins, and the resource whose coherence time, gate fidelity and count determine what a quantum computer can achieve.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:qubit", "labels": [ "Qubit" ], "is_subclass_of": [ "Quantum Computation Paradigm" ], "wikilinks": [ "Quantum Computation Paradigm", "Quantum Computing", "Information Theory" ] }, { "id": "query-encoder", "title": "Query Encoder", "domain": "ai", "domain_name": "Ai", "definition": "A query encoder is a neural model component that maps a search query into a dense vector embedding within a shared semantic space, enabling similarity comparison against encoded documents. In dual-encoder retrieval architectures it is paired with a document encoder, allowing fast approximate-nearest-neighbour search over precomputed passage embeddings. It is a core building block of dense retrieval and retrieval-augmented generation pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:query-encoder", "labels": [ "Query Encoder" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "query-engine", "title": "Query Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A query engine is the software component that parses, plans, optimises, and executes declarative queries against one or more data stores, returning result sets to callers. It transforms a high-level query into an efficient physical execution plan using statistics, indexes, and cost models, and may operate over a single database, a data warehouse, or federated sources, forming the computational core of database and analytics systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:query-engine", "labels": [ "Query Engine" ], "is_subclass_of": [ "Database System" ], "wikilinks": [] }, { "id": "query-interface", "title": "Query Interface", "domain": "ai", "domain_name": "Ai", "definition": "A query interface is a defined boundary \u2014 programmatic, visual, or linguistic \u2014 through which users or systems express requests for data retrieval, graph traversal, or knowledge discovery against a data store or knowledge base, receiving structured results in return. Query interfaces range from formal declarative languages such as SQL and SPARQL to REST endpoints, vector similarity APIs, and natural-language question-answering systems built on large language models. They abstract the internal organisation of a data source behind a stable contract, enabling diverse clients to access data without knowledge of storage implementation details. Well-designed query interfaces balance expressiveness, performance, and ease of use.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:query-interface", "labels": [ "Query Interface" ], "is_subclass_of": [ "Data Access Interface" ], "wikilinks": [] }, { "id": "query-key-value", "title": "Query Key Value", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The three fundamental components of the attention mechanism introduced by Vaswani et al. (2017): a Query vector representing the current information need, Key vectors representing available information descriptors, and Value vectors containing the content to retrieve. Attention weights are computed via scaled dot-product similarity between queries and keys, then applied to values to produce context-aware output representations.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:query-key-value", "labels": [ "Query Key Value", "Query Key Value Projection", "Query-Key-Value Projection" ], "is_subclass_of": [ "AI Model Architecture", "Attention Mechanism" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "query-optimiser", "title": "Query Optimiser", "domain": "data", "domain_name": "Data", "definition": "A query optimiser is the component of a database management system that transforms a declarative query into an efficient physical execution plan. It enumerates candidate plans, estimates their cost using statistics about data distribution and access paths, and selects the plan expected to minimise resource usage. Cost-based optimisers rely on cardinality estimation and index awareness, while rule-based optimisers apply heuristic transformations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:query-optimiser", "labels": [ "Query Optimiser" ], "is_subclass_of": [ "Query Engine" ], "wikilinks": [] }, { "id": "query-parser", "title": "Query Parser", "domain": "data", "domain_name": "Data", "definition": "A query parser is the component of a database or graph engine that converts a query written in a query language into a structured, executable representation, typically an abstract syntax tree or query plan. It performs lexical analysis and syntactic validation before handing the parsed structure to an optimiser or execution engine, applying the same parsing theory used in general-purpose compilers. Query parsers are required by relational databases and graph databases alike to translate declarative query text into an internal form the engine can execute.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:query-parser", "labels": [ "Query Parser" ], "is_subclass_of": [ "Compiler" ], "wikilinks": [] }, { "id": "query-processing", "title": "Query Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Query processing is the set of activities a database system performs to translate a declarative query into an efficient execution that returns the requested data. It spans parsing and semantic analysis, logical and physical query optimisation, plan selection, and execution against stored data and indexes. Effective query processing is central to database performance, determining how quickly results are produced and how system resources are used.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:query-processing", "labels": [ "Query Processing" ], "is_subclass_of": [ "Database System" ], "wikilinks": [] }, { "id": "query-processor", "title": "Query Processor", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A system component responsible for parsing, optimising, and executing queries against a data store or knowledge base, including spatial queries in 3D environments. Query processors translate declarative query expressions into efficient execution plans, leveraging indexing structures such as octrees, k-d trees, R-trees, and scene graph hierarchies to minimise retrieval latency.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:query-processor", "labels": [ "Query Processor", "Query Execution Engine" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "query-strategy", "title": "Query Strategy", "domain": "ai", "domain_name": "Ai", "definition": "A query strategy is the policy by which an active-learning system selects which unlabelled instances to request labels for, aiming to maximise model improvement per labelling cost. Common strategies include uncertainty sampling, query-by-committee, and expected model change. It is the decision-making core that makes active learning more label-efficient than passive supervised learning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:query-strategy", "labels": [ "Query Strategy" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "query-vector", "title": "Query Vector", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Query Vector is a dense numerical representation of a search query produced by an embedding model, enabling similarity-based retrieval in a high-dimensional vector space. It is matched against stored document or passage embeddings using distance metrics such as cosine similarity or inner product, forming the core retrieval mechanism in semantic search and retrieval-augmented generation systems. Query vectors encode the semantic intent of a query independent of exact keyword overlap, allowing conceptually related results to be surfaced even when surface-level vocabulary differs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:query-vector", "labels": [ "Query Vector" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "question-answering", "title": "Question Answering", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Question Answering (QA) is the NLP task of automatically generating accurate answers to natural language questions posed by users, either by extracting answers from text passages (extractive QA) or generating free-form responses (generative QA). QA systems employ reading comprehension models, retrieval-augmented generation, and knowledge reasoning to power applications including search engines, virtual assistants, and customer support.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:question-answering", "labels": [ "Question Answering", "Complex Question Answering", "Document Question Answering" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "scalability", "\ud83e\udd16", "Dialogue System", "Information Retrieval", "machine learning", "MetaverseDomain", "Natural Language Processing", "Research Tools" ] }, { "id": "quorum-blockchain", "title": "Quorum Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An enterprise-grade permissioned ereum fork, initiated by JPMorgan Chase's Blockchain Centre of Excellence and open-sourced under Apache 2.0 that extends the ereum protocol with private transaction envelopes managed by the Tessera privacy manager (successor to Constellation, implemented v...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:quorum-blockchain", "labels": [ "Quorum Blockchain", "QuorumBlockchain" ], "is_subclass_of": [ "Network Component", "Consortium Blockchain", "Ethereum Smart Contract Platform", "Permissioned Blockchain", "Enterprise Blockchain", "Distributed Ledger" ], "wikilinks": [ "Alastria Network", "Apache License 2.0", "B\u00fcnz et al 2018 Bulletproofs", "Bank of England 2024 RTGS Renewal Synchronisation Specification", "BIS 2023 Project Mariana Cross-CBDC FX", "BIS 2024 Project Agor\u00e1 Unified Ledger", "Cambridge CCAF 2024 Global Cryptoasset Benchmarking Study", "Castro Liskov 1999 Practical BFT", "CBDC Infrastructure", "Clique Consensus", "ConsenSys", "ConsenSys 2025 Linea ZK Private EVM Research", "ConsenSys GoQuorum Documentation", "ConsensusLayer", "Cross-Border Payments", "Daml Ledger", "DigitalAssetDomain", "Digital Securities Settlement", "Dinh et al 2017 BLOCKBENCH", "DistributedLedgerDomain" ] }, { "id": "quorum-mechanism", "title": "Quorum Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A quorum mechanism is a governance rule that requires a minimum threshold of participation or stake to be reached before a vote or decision is considered valid and binding. In on-chain governance it guards against capture by small, motivated minorities and ensures decisions reflect sufficient stakeholder engagement. Quorum thresholds are often expressed as a fraction of total voting power or eligible participants.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:quorum-mechanism", "labels": [ "Quorum Mechanism" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "quorum-system", "title": "Quorum System", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A quorum system is a collection of subsets of a distributed set of replicas, where any two subsets are guaranteed to intersect, used to coordinate read and write operations so that consistency is preserved despite failures. By requiring operations to gather acknowledgements from a quorum rather than every replica, the system tolerates a bounded number of faulty or unreachable nodes while still ensuring that conflicting operations observe one another. Quorum systems are foundational to distributed consensus, replicated databases, and Byzantine fault-tolerant protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:quorum-system", "labels": [ "Quorum System" ], "is_subclass_of": [ "Distributed Consensus" ], "wikilinks": [] }, { "id": "quorum-threshold", "title": "Quorum Threshold", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A quorum threshold is the minimum number or fraction of participants (nodes, validators, signatories, or voters) that must agree or respond for a distributed system or governance process to reach a valid decision. It is a foundational parameter in consensus protocols, distributed databases, and blockchain governance, balancing liveness (the system can make progress) against safety (decisions reflect genuine majority will).", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:quorum-threshold", "labels": [ "Quorum Threshold" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "quorum", "title": "Quorum", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A quorum is the minimum number or proportion of participants\u2014nodes, validators, voters, or signers\u2014that must concur or be present for a decision, transaction, or consensus round to be considered valid in a distributed system or governance process. In distributed computing, quorum systems are collections of node subsets with the intersection property: any two quorums share at least one member, preventing contradictory decisions across network partitions. In blockchain and DAO governance, quorum thresholds set the participation floor required before a vote or proposal carries binding weight, balancing decision liveness against resistance to minority capture. Threshold signature schemes and multi-signature wallets operationalise quorum as an m-of-n approval requirement that eliminates single points of failure in key custody.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:quorum", "labels": [ "Quorum", "Distributed Computing", "Majority Quorum", "Quorum Fraction", "Signer Quorum", "Signing Quorum" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "r3-corda", "title": "R3 Corda", "domain": "blockchain", "domain_name": "Blockchain", "definition": "R3 Corda is a purpose-built permissioned distributed ledger platform designed exclusively for regulated financial services, implementing a \"need-to-know\" peer-to-peer architecture in which transaction data is shared solely amongst parties with legitimate business interest, eliminating global stat...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:r3-corda", "labels": [ "R3 Corda", "R3Corda" ], "is_subclass_of": [ "Network Component", "Blockchain", "Distributed Ledger", "Permissioned Blockchain", "Enterprise Blockchain", "Financial Services Technology", "Smart Contract Platform" ], "wikilinks": [ "AMQP", "Apache Kafka", "Asset Tokenisation", "Atomic Settlement", "Bank of England", "BIS CPMI", "BIS Innovation Hub", "Certificate Authority", "ConsensusLayer", "Contract Code", "CorDapp", "Corda Node", "Corda Vault", "Cross-Border Payment", "Cryptographic Hash Function", "DAML Ledger", "Database Technology", "Delivery Versus Payment", "Deterministic JVM", "Digital Securities" ] }, { "id": "raft-consensus", "title": "raft consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Raft is a distributed consensus algorithm designed as a more understandable alternative to Paxos, decomposing consensus into three relatively independent sub-problems: leader election, log replication, and safety. A Raft cluster elects a single leader by majority vote during which followers grant a term-limited mandate; the leader receives all client writes, appends them to its log, and replicates them to followers, committing entries once a quorum acknowledges receipt. Raft guarantees that committed entries are never lost as long as a majority of nodes remain connected, providing crash fault tolerance but not Byzantine fault tolerance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:raft-consensus", "labels": [ "RAFT Consensus", "Raft Consensus" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "rag-pipeline", "title": "RAG Pipeline", "domain": "ai", "domain_name": "Ai", "definition": "A RAG pipeline is the end-to-end software architecture that implements retrieval-augmented generation, encompassing document ingestion and chunking, embedding generation, vector store indexing, query-time retrieval, context assembly, and language model generation to produce grounded, verifiable responses from external knowledge sources. It operationalises the RAG paradigm as a deployable, maintainable system with distinct stages, each subject to independent optimisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rag-pipeline", "labels": [ "RAG Pipeline" ], "is_subclass_of": [ "Retrieval-Augmented Generation" ], "wikilinks": [] }, { "id": "ransac", "title": "RANSAC", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "RANSAC (Random Sample Consensus) is an iterative, robust estimation algorithm that fits a model to data containing a large fraction of outliers. It repeatedly draws a minimal random sample, fits a candidate model, and counts the inliers that agree within a tolerance, retaining the model with the largest consensus set. RANSAC is foundational in computer vision for estimating geometric relationships such as homographies, fundamental matrices, and camera poses from noisy feature correspondences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ransac", "labels": [ "RANSAC" ], "is_subclass_of": [ "Feature Matching" ], "wikilinks": [] }, { "id": "rb-1002-closedloopcontrol", "title": "RB 1002 closedloopcontrol", "domain": "robotics", "domain_name": "Robotics", "definition": "RB 1002 Closed-Loop Control is an ontology term in the NarrativeGoldmine robotics hierarchy denoting the general class of feedback-based control architectures in which sensor measurements of a system's output are continuously fed back to the controller to correct deviations from a desired set-point. As a superclass it subsumes specific strategies including PID control, model-predictive control, adaptive control, and state-estimation-driven controllers used in robotic actuation and manipulation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-1002-closedloopcontrol", "labels": [ "RB 1002 closedloopcontrol" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "rb-1013-localization", "title": "RB 1013 localization", "domain": "robotics", "domain_name": "Robotics", "definition": "RB 1013 localization is a robotics navigation concept addressing the problem of a mobile robot determining its pose within a known or partially known environment. It encompasses probabilistic state estimation techniques such as particle filters, extended Kalman filters, and scan-matching algorithms applied to sensor data from LiDAR, wheel odometry, and IMUs to produce continuous pose estimates suitable for autonomous navigation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:rb-1013-localization", "labels": [ "RB 1013 localization", "Localization", "RB-1013-localization" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [ "owl:Thing" ] }, { "id": "rb-1016-pathplanning", "title": "RB 1016 pathplanning", "domain": "robotics", "domain_name": "Robotics", "definition": "RB 1016 Path Planning is an ontology term in the NarrativeGoldmine robotics hierarchy representing the superclass of algorithms that compute collision-free trajectories from a start configuration to a goal configuration in a robot's configuration space. It encompasses graph-search methods (A*, Dijkstra), sampling-based planners (RRT, PRM), and reactive local-planning strategies (DWA, potential fields), all of which underpin autonomous navigation in structured and unstructured environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-1016-pathplanning", "labels": [ "RB 1016 pathplanning" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [ "owl:Thing" ] }, { "id": "rb-1007-trajectory-generation", "title": "RB-1007-trajectory-generation", "domain": "robotics", "domain_name": "Robotics", "definition": "RB-1007 Trajectory Generation is a robotics knowledge-base entry cataloguing the algorithms, mathematical representations, and computational methods used to synthesise smooth, dynamically feasible motion trajectories for robotic systems, specifying how a robot must move through configuration space or task space from an initial to a goal state while satisfying kinematic, dynamic, and environmental constraints. It covers polynomial spline interpolation, time-optimal trajectory planning, jerk-limited profiles, and learned trajectory generation approaches, forming the bridge between high-level task planning and low-level joint control execution.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rb-1007-trajectory-generation", "labels": [ "RB-1007-trajectory-generation", "Trajectory Generation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "rb-1008-odometry", "title": "RB-1008-odometry", "domain": "robotics", "domain_name": "Robotics", "definition": "RB-1008-odometry refers to the odometric positioning subsystem of the RB-1008 wheeled mobile robot platform, encompassing the encoder-based wheel odometry algorithms, sensor fusion pipelines, and ROS navigation stack integration used to estimate the robot's pose (position and orientation) relative to a starting frame. The system computes incremental displacement from differential or omnidirectional wheel encoders, corrects for wheel slip and encoder quantisation errors, and fuses encoder data with inertial measurement unit (IMU) readings to produce a continuous odometric estimate used as the prior for simultaneous localisation and mapping (SLAM) and autonomous navigation. Accurate odometry is essential for maintaining coherent map frames and for the navigation stack's local cost-map updates between LIDAR scan matches.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rb-1008-odometry", "labels": [ "RB-1008-odometry" ], "is_subclass_of": [ "Odometry" ], "wikilinks": [] }, { "id": "rdf-data-model", "title": "RDF Data Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The RDF data model represents information as triples of subject, predicate and object, forming a directed labelled graph. It is the foundation of the Resource Description Framework used to publish structured data on the web.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rdf-data-model", "labels": [ "RDF Data Model" ], "is_subclass_of": [ "Semantic Web Linked Data Standard" ], "wikilinks": [ "Semantic Web", "Knowledge Graph", "Linked Data", "SPARQL", "Ontology" ] }, { "id": "rdf-schema", "title": "RDF Schema", "domain": "standards", "domain_name": "Standards", "definition": "A World Wide Web Consortium vocabulary that extends the Resource Description Framework with the means to describe classes, properties and their hierarchies. It provides a lightweight modelling layer for typing resources and defining property domains and ranges.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rdf-schema", "labels": [ "RDF Schema" ], "is_subclass_of": [ "RDF" ], "wikilinks": [ "RDF", "Reasoning", "OWL 2 Web Ontology Language", "Vocabulary", "Knowledge Representation", "https://www.w3.org/TR/rdf-schema/" ] }, { "id": "rdf-store", "title": "RDF Store", "domain": "data", "domain_name": "Data", "definition": "A database designed to store and query data expressed as Resource Description Framework triples, typically supporting the SPARQL query language. Also called a triplestore, it manages subject-predicate-object statements rather than rows and tables.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rdf-store", "labels": [ "RDF Store" ], "is_subclass_of": [ "RDF" ], "wikilinks": [ "RDF", "SPARQL", "Knowledge Graph", "Data Integration", "Linked Data", "Semantic Interoperability", "https://www.w3.org/TR/sparql11-query/" ] }, { "id": "rdf-triple-store", "title": "RDF Triple Store", "domain": "data", "domain_name": "Data", "definition": "An RDF triple store is a purpose-built database for storing and querying data as subject-predicate-object triples following the Resource Description Framework model. It supports the SPARQL query language and often provides reasoning over ontologies to infer implicit facts. Triple stores are the backbone of semantic-web, knowledge-graph, and linked-data applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rdf-triple-store", "labels": [ "RDF Triple Store" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "rdf", "title": "rdf", "domain": "data", "domain_name": "Data", "definition": "The Resource Description Framework (RDF) is a W3C-standardised graph data model in which all statements are expressed as subject\u2013predicate\u2013object triples, where each component is identified by an IRI (Internationalised Resource Identifier) or, for literal values, a typed or language-tagged string. RDF triples compose into directed, labelled graphs that can be serialised in multiple syntaxes including Turtle, N-Triples, JSON-LD, N-Quads, and RDF/XML, and queried via the SPARQL protocol and query language. As the foundational layer of the Semantic Web and Linked Data ecosystems, RDF enables machine-readable knowledge representation that supports federated querying, ontological reasoning with OWL2, and cross-source interoperability through shared vocabulary IRIs. The open-world assumption distinguishes RDF from closed-world relational models, allowing independent datasets to be merged without schema negotiation.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:rdf", "labels": [ "RDF", "RDF Resource", "RDF/RDFS", "W3C RDF Standard" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "research-summary", "title": "RESEARCH_SUMMARY", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive synthesis document aggregating research findings, literature reviews, and academic insights across Metaverse, Blockchain, Artificial Intelligence, and Robotics domains. It distils empirical evidence, theoretical frameworks, mathematical foundations, algorithm comparisons, implementation guidance, and future research directions for practitioners and stakeholders working across these converging technological fields.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:research-summary", "labels": [ "RESEARCH_SUMMARY" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "ArtificialIntelligence", "Blockchain", "Metaverse", "MetaverseDomain", "Robotics" ] }, { "id": "rest-api", "title": "rest api", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A REST API (Representational State Transfer Application Programming Interface) is a web service interface conforming to Roy Fielding's six architectural constraints \u2014 client\u2013server separation, statelessness, cacheability, uniform interface, layered system, and optional code-on-demand \u2014 originally formalised in his 2000 doctoral dissertation. Resources are uniquely addressed by URIs, manipulated through standard HTTP verbs (GET, POST, PUT, PATCH, DELETE), and represented in negotiated formats such as JSON or XML. The uniform interface constraint \u2014 encompassing resource identification, manipulation through representations, self-descriptive messages, and hypermedia as the engine of application state (HATEOAS) \u2014 is the defining characteristic that separates REST from earlier RPC and SOAP-based architectures. REST APIs have become the dominant integration surface across cloud platforms, AI model serving, microservices ecosystems, and public developer APIs precisely because they align with existing HTTP infrastructure including caches, proxies, load balancers, and API gateways.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:rest-api", "labels": [ "REST API", "HTTP REST API", "REST API Integration", "REST API Pattern" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "rest", "title": "REST", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "REST (Representational State Transfer) is an architectural style for distributed hypermedia systems, defined by Roy Fielding, that uses stateless client-server communication over HTTP with a uniform resource-based interface, cacheable responses, and layered system constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rest", "labels": [ "REST", "HTTP REST", "REST Architectural Style" ], "is_subclass_of": [ "Web Standards" ], "wikilinks": [ "HTTP", "Payment Protocol", "Microservices", "Web Standards", "https://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm", "https://developer.mozilla.org/en-US/docs/Glossary/REST" ] }, { "id": "restful-api", "title": "RESTful API", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A RESTful API is a web application programming interface that follows the Representational State Transfer architectural style, exposing resources identified by URIs and manipulated through a uniform set of HTTP methods. It is stateless, treats responses as representations of resource state, and uses standard status codes and media types so clients and servers can evolve independently. RESTful design favours predictable, cacheable, hypermedia-driven interactions over the web.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:restful-api", "labels": [ "RESTful API" ], "is_subclass_of": [ "Web API" ], "wikilinks": [] }, { "id": "rfc-2119-should-normative-keyword", "title": "RFC 2119 SHOULD Normative Keyword", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SHOULD is an RFC 2119 normative keyword indicating that a particular behaviour or implementation choice is strongly recommended but not absolutely required. In specification and standards documents, SHOULD implies that valid reasons may exist in particular circumstances to deviate from the stated guidance, but the implementer must understand the implications and weigh the trade-offs carefully before choosing a different course. It contrasts with MUST (mandatory) and MAY (optional).", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:rfc-2119-should-normative-keyword", "labels": [ "RFC 2119 SHOULD Normative Keyword", "SHOULD" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "rfc-5280", "title": "RFC 5280", "domain": "security", "domain_name": "Security", "definition": "RFC 5280 is the IETF standard that defines the Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) profile. It specifies the structure, encoding, and processing rules for X.509 v3 certificates and v2 CRLs, including the required and optional extensions and the certification path validation algorithm. The document is the normative reference that governs how PKI implementations issue, encode, and validate certificates on the public Internet.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rfc-5280", "labels": [ "RFC 5280" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "rfc-8785-canonical-json", "title": "RFC 8785 Canonical JSON", "domain": "standards", "domain_name": "Standards", "definition": "RFC 8785 specifies the JSON Canonicalization Scheme, a method for producing a deterministic serialisation of JSON data to support cryptographic operations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rfc-8785-canonical-json", "labels": [ "RFC 8785 Canonical JSON" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IETF", "Technical Standard" ] }, { "id": "rfc-8785", "title": "RFC 8785", "domain": "standards", "domain_name": "Standards", "definition": "An IETF Request for Comments defining the JSON Canonicalization Scheme (JCS), a method for producing a deterministic serialisation of JSON data. It supports consistent hashing and signing of JSON.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rfc-8785", "labels": [ "RFC 8785" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "IETF", "Technical Standard" ] }, { "id": "rfid", "title": "RFID", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Radio-Frequency Identification (RFID) is a wireless technology that uses electromagnetic fields to automatically identify and track tags attached to objects. Tags contain electronically stored information and can be read without requiring line of sight, enabling automated data capture across distances from a few centimetres to several metres.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:rfid", "labels": [ "RFID" ], "is_subclass_of": [ "Communication Network" ], "wikilinks": [] }, { "id": "rgb-protocol", "title": "RGB Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "RGB Protocol is a client-side validation smart-contract system built on Bitcoin and the Lightning Network that enables issuance, transfer, and programmable logic over tokenised assets (fungible and non-fungible) without placing contract state on the public blockchain. Contract state is stored in off-chain client-held directed acyclic graphs (DAGs), with only cryptographic commitments anchored to Bitcoin UTXOs via single-use seals, inheriting Bitcoin's security and censorship resistance while providing scalability, privacy, and programmability unavailable to purely on-chain designs. The AluVM virtual machine executes contract validation logic deterministically off-chain, and Lightning Network channels enable instant, low-fee RGB asset transfers. Developed by the LNP/BP Association from 2018 onward, RGB reached production stability in 2024 with standardised schemas RGB20 (fungible tokens) and RGB21 (non-fungible tokens).", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:rgb-protocol", "labels": [ "RGB Protocol", "RGB Specification", "RGB-Protocol" ], "is_subclass_of": [ "Blockchain Protocol" ], "wikilinks": [] }, { "id": "rgb-and-client-side-validation", "title": "RGB and Client Side Validation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "RGB and Client-Side Validation (CSV) is a Bitcoin-native Layer 3 smart contract system and cryptographic architecture invented by Dr. Peter Todd (foundational client-side validation theory, 2016\u20132019) and brought to production implementation by Dr.", "entityType": "Class", "qualityScore": 0.54, "maturity": "established", "iri": "urn:ngm:class:rgb-and-client-side-validation", "labels": [ "RGB and Client Side Validation", "RGB and Client-Side Validation" ], "is_subclass_of": [ "Protocol and Consensus", "Client-Side Validation Theory", "Network Component", "Bitcoin Proof-of-Work Protocol", "Smart Contracts", "Layer 3", "Off-Chain Protocol", "UTXO Model" ], "wikilinks": [ "AI agent", "AI Agent Asset Ownership", "AI Agent Coordination", "algorithmic trading", "AluVM", "AluVM Specification", "AluVM Validation", "API", "asset", "Asset Interoperability", "Atomic Swaps", "BiTMASK", "Bifrost Protocol", "BitVM", "BitcoinDomain", "Bitcoin Domain", "Bitcoin Layer 3 Ecosystem", "Bitcoin-Native DeFi", "Bitcoin Network", "Bitcoin Protocol" ] }, { "id": "rgb-d-camera", "title": "RGB-D Camera", "domain": "robotics", "domain_name": "Robotics", "definition": "An RGB-D camera is a sensor that captures both colour (RGB) imagery and per-pixel depth (D) information in a single aligned frame. Depth is typically derived from structured light, time-of-flight, or stereo techniques, yielding a dense 3D representation of the scene. RGB-D cameras are widely used in robotics for perception, mapping, and manipulation because they combine appearance and geometry cheaply.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rgb-d-camera", "labels": [ "RGB-D Camera", "RGB-D Cameras" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "rgb", "title": "RGB", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A protocol for issuing and transferring assets and running smart contracts on Bitcoin using client-side validation and single-use seals anchored to the chain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:rgb", "labels": [ "RGB", "RGB Blackpaper" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Client-Side Validation", "UTXO", "Lightning Network", "Smart Contract", "Bitcoin" ] }, { "id": "rlhf", "title": "rlhf", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Reinforcement Learning from Human Feedback (RLHF) is a training methodology that aligns large language models and other AI systems with human preferences by first collecting human comparison judgements between model outputs, training a reward model on those judgements, and then optimising the language model policy against the reward model using reinforcement learning \u2014 typically Proximal Policy Optimisation with a KL-divergence penalty to prevent reward hacking. RLHF enables models to be steered towards outputs that human annotators prefer for helpfulness, harmlessness, and honesty, going beyond what is achievable with supervised fine-tuning on static demonstration data alone. The technique was popularised by OpenAI's InstructGPT work and underlies the alignment pipeline of models such as ChatGPT, Claude, and Gemini. Variants including Direct Preference Optimisation, Constitutional AI, and RLAIF (Reinforcement Learning from AI Feedback) extend or simplify the original three-stage pipeline.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:rlhf", "labels": [ "RLHF", "RLHF Alignment", "RLHF Pipeline" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [] }, { "id": "rmsnorm", "title": "rmsnorm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Root Mean Square Layer Normalisation (RMSNorm) is a neural network normalisation technique that re-scales each activation vector by the inverse of the root mean square of its elements, eliminating the mean-subtraction and bias terms present in standard Layer Normalisation. By removing the re-centring computation, RMSNorm reduces operational cost while achieving comparable training stability and generalisation performance to Layer Normalisation in transformer architectures. RMSNorm has been adopted as the default normalisation layer in several state-of-the-art large language models including LLaMA and Mistral, reflecting both its empirical effectiveness and its computational efficiency on GPU hardware.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rmsnorm", "labels": [ "RMSNorm" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "rmsprop", "title": "RMSProp", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A stochastic gradient descent variant, proposed by Geoffrey Hinton in his 2012 Coursera lectures, that divides each parameter's update by the root of an exponentially weighted moving average of its squared gradients, giving every parameter its own adaptive step size; it cured AdaGrad's vanishing learning rate, made recurrent networks practical to train, and supplied the second-moment machinery later absorbed into Adam.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:rmsprop", "labels": [ "RMSProp" ], "is_subclass_of": [ "Optimisation Algorithm" ], "wikilinks": [ "Optimisation Algorithm", "Adaptive Learning Rate", "Adam Optimiser", "Stochastic Gradient Descent" ] }, { "id": "roc-curve", "title": "ROC Curve", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A graphical performance evaluation tool for binary classification models that plots the True Positive Rate (Recall/Sensitivity) against the False Positive Rate across all possible classification thresholds, visualising the trade-off between correctly identifying positive instances and incorrectly classifying negative instances as positive, enabling threshold selection, model comparison, and assessment of a classifier's discriminative ability independent of class distribution or threshold choice.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:roc-curve", "labels": [ "ROC Curve" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [ "diffie1976new", "False Positive Rate", "gayoso2018secure", "harris2020flood", "model comparison", "Model Comparison", "Precision-Recall Curve", "schnorr1989efficient", "Sensitivity", "Specificity", "Threshold selection", "Threshold Selection", "True Positive Rate", "AUC", "Confusion Matrix", "MetaverseDomain", "Model Performance", "Retrieval Augmented Generation - RAG" ] }, { "id": "ros-2", "title": "ros 2", "domain": "robotics", "domain_name": "Robotics", "definition": "ROS 2 (Robot Operating System 2) is an open-source robotics middleware framework developed by Open Robotics that provides a standardised publish-subscribe communication layer built on the DDS (Data Distribution Service) standard, along with a comprehensive ecosystem of drivers, libraries, and tools for sensor integration, motion planning, simulation, and hardware abstraction. ROS 2 supersedes ROS 1 with support for real-time execution, multi-robot systems, production-grade security via DDS-Security, and native Windows and macOS compatibility. It has become the de facto standard software framework for research and increasingly for commercial robotic platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ros-2", "labels": [ "ROS 2", "ROS 2 Humble", "ROS2" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "ros-navigation-stack", "title": "ROS Navigation Stack", "domain": "robotics", "domain_name": "Robotics", "definition": "The ROS Navigation Stack is a collection of software packages within the Robot Operating System (ROS) framework that provides a robot with the capability to move autonomously through an environment by combining map building, localisation, global path planning, local obstacle avoidance, and motor control into an integrated pipeline. It abstracts sensor inputs (laser scan, odometry, IMU), maintains a 2D occupancy grid map and cost layers, runs AMCL (Adaptive Monte Carlo Localisation) for pose estimation, and uses A* or Dijkstra for global plans combined with DWA (Dynamic Window Approach) or TEB (Timed Elastic Band) for local reactive navigation. The Navigation Stack is the de facto standard for mobile ground-robot navigation in research and production robotics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ros-navigation-stack", "labels": [ "ROS Navigation Stack", "ROS 2 Navigation" ], "is_subclass_of": [ "Robot Operating System" ], "wikilinks": [] }, { "id": "ros-rep", "title": "ROS REP", "domain": "robotics", "domain_name": "Robotics", "definition": "A ROS Enhancement Proposal (REP) is a design document that defines conventions and standards for the Robot Operating System ecosystem, analogous to internet RFCs or Python PEPs. REPs codify agreements such as coordinate-frame conventions, units, naming, and message interfaces so that independently developed robotics components interoperate. They provide the normative reference that aligns platforms and libraries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ros-rep", "labels": [ "ROS REP", "REP 103", "REP Standards", "ROS 2 REPs" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "ros-industrial", "title": "ROS-Industrial", "domain": "robotics", "domain_name": "Robotics", "definition": "ROS-Industrial is an open-source project that extends the Robot Operating System with libraries, drivers, and interfaces for industrial manufacturing automation. It provides standardised drivers for industrial robot arms, motion-planning integration, and tools to bring advanced ROS capabilities into factory settings. It bridges research-grade robotics software with production-grade industrial hardware.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ros-industrial", "labels": [ "ROS-Industrial", "ROS Industrial Consortium" ], "is_subclass_of": [ "Robotics", "Robot Operating System" ], "wikilinks": [] }, { "id": "ros", "title": "ROS", "domain": "robotics", "domain_name": "Robotics", "definition": "ROS (Robot Operating System) is an open-source middleware framework providing publish-subscribe communication, build tooling and reusable libraries for constructing robot software from distributed, composable components.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ros", "labels": [ "ROS", "ROS 1", "ROS Node", "ROS TF Framework", "ROS sensor_msgs", "Robot Operating System" ], "is_subclass_of": [], "wikilinks": [ "DDS Middleware", "Middleware", "Autonomous Navigation", "Sensor Fusion", "micro-ROS", "Robot Operating System" ] }, { "id": "rrt-algorithm", "title": "RRT Algorithm", "domain": "robotics", "domain_name": "Robotics", "definition": "A sampling-based path planning algorithm that incrementally builds a tree of collision-free configurations by randomly sampling the configuration space and connecting samples to the nearest existing tree node. It efficiently explores high-dimensional spaces and is probabilistically complete.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:rrt-algorithm", "labels": [ "RRT Algorithm", "RB-1017-rrt-algorithm", "RRT" ], "is_subclass_of": [ "Navigation and Planning", "Path Planning", "RB-1016-path-planning", "Sampling-Based Method", "RRT-Star", "RRT-Connect", "Informed RRT-Star" ], "wikilinks": [ "Autonomous Vehicles", "Collision Checker", "Configuration Space", "Informed RRT-Star", "Manipulators", "Mobile Robots", "Probabilistic Completeness", "Randomization", "RB-1016-path-planning", "RB-1018-dijkstra-algorithm", "RRT-Connect", "RRT-Star", "Sampling-Based Method", "Steven LaValle 1998", "Tree Structure", "A-Star Algorithm", "Collision Detection", "Motion Planning", "Nearest Neighbor Search", "Path Planning" ] }, { "id": "rtab-map", "title": "RTAB-Map", "domain": "robotics", "domain_name": "Robotics", "definition": "RTAB-Map (Real-Time Appearance-Based Mapping) is an open-source library for RGB-D, stereo, and lidar graph-based SLAM with a memory-management scheme that bounds computation for large-scale, long-term operation. Its core is an appearance-based loop-closure detector that recognises previously visited places to correct accumulated drift. It is widely used on ground robots to build consistent metric and topological maps.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rtab-map", "labels": [ "RTAB-Map" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "rtcp-feedback", "title": "RTCP Feedback", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "RTCP feedback is the control-channel mechanism of the RTP Control Protocol by which receivers report reception quality, such as packet loss, jitter, and round-trip time, back to senders during real-time media streaming. Senders use these reports, along with extensions like NACK, PLI, and REMB, to adapt encoding bitrate and recover from loss. It is the feedback loop that enables congestion-aware, resilient audiovisual transport.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rtcp-feedback", "labels": [ "RTCP Feedback" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "rtmpose", "title": "RTMPose", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of real-time multi-person pose estimation models from OpenMMLab's MMPose project (Jiang et al., 2023) that pairs a lightweight CSPNeXt convolutional backbone with a SimCC coordinate-classification head, treating keypoint localisation as classification over discretised horizontal and vertical bins rather than heatmap regression. RTMPose achieves strong COCO accuracy at real-time speeds on CPU, GPU, and mobile targets, and serves as the teacher and architectural basis for distilled variants such as DWPose.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:rtmpose", "labels": [ "RTMPose" ], "is_subclass_of": [ "Pose Estimation" ], "wikilinks": [ "Pose Estimation", "DWPose", "Convolutional Neural Network", "Computer Vision" ] }, { "id": "rabbitmq", "title": "Rabbitmq", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "RabbitMQ is an open-source message broker that implements the Advanced Message Queuing Protocol (AMQP) and related messaging standards, routing messages between producers and consumers through exchanges, bindings, and queues. It supports flexible routing topologies, message acknowledgement, durability, and clustering for high availability. RabbitMQ is widely used to decouple services and enable reliable asynchronous communication in distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:rabbitmq", "labels": [ "Rabbitmq", "RabbitMQ" ], "is_subclass_of": [ "Message Broker" ], "wikilinks": [] }, { "id": "rack-and-pinion-actuator", "title": "Rack and Pinion Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A rack and pinion actuator is a mechanical transmission mechanism that converts rotary motion from a pinion gear into linear motion along a toothed rack, commonly used in robotics and industrial automation for precise linear positioning. The pinion rotates against the rack to produce controlled translational displacement, with speed and force determined by gear ratio and motor torque. It offers high stiffness, repeatability, and scalability for long-stroke linear axes in robotic manipulators and CNC systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:rack-and-pinion-actuator", "labels": [ "Rack and Pinion Actuator" ], "is_subclass_of": [ "Actuation and Control", "Electric Linear Actuator" ], "wikilinks": [ "Electric Linear Actuator", "Robotics" ] }, { "id": "radar-altimetry", "title": "Radar Altimetry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radar-altimetry", "labels": [ "Radar Altimetry" ], "is_subclass_of": [ "Active Remote Sensing" ], "wikilinks": [] }, { "id": "radar-foreshortening", "title": "Radar Foreshortening", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radar-foreshortening", "labels": [ "Radar Foreshortening" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radar-instrument", "title": "Radar Instrument", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radar-instrument", "labels": [ "Radar Instrument" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radar-layover", "title": "Radar Layover", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radar-layover", "labels": [ "Radar Layover" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radar-shadow", "title": "Radar Shadow", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radar-shadow", "labels": [ "Radar Shadow" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radar", "title": "Radar", "domain": "robotics", "domain_name": "Robotics", "definition": "Radar is a sensing technology that transmits radio waves and measures their reflections to determine the range, velocity, and angle of objects; it is used in robotics, autonomous vehicles, aviation, and defence for reliable perception under all weather and lighting conditions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:radar", "labels": [ "Radar", "Radar Sensing", "Synthetic Aperture Radar" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [ "Signal Processing", "Perception System", "Sensor Fusion", "Lidar", "Sensors", "Sensor" ] }, { "id": "radiation-belt", "title": "Radiation Belt", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiation-belt", "labels": [ "Radiation Belt" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiation-dosimetry", "title": "Radiation Dosimetry", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiation-dosimetry", "labels": [ "Radiation Dosimetry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiation-hardness-assurance", "title": "Radiation Hardness Assurance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiation-hardness-assurance", "labels": [ "Radiation Hardness Assurance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiation-shielding", "title": "Radiation Shielding", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiation-shielding", "labels": [ "Radiation Shielding" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiative-transfer-model", "title": "Radiative Transfer Model", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiative-transfer-model", "labels": [ "Radiative Transfer Model" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiative-zone", "title": "Radiative Zone", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiative-zone", "labels": [ "Radiative Zone" ], "is_subclass_of": [ "Stellar Interior" ], "wikilinks": [] }, { "id": "radicle-decentralised-code-forge", "title": "Radicle Decentralised Code Forge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Radicle is a decentralised, peer-to-peer code collaboration platform built on Git that eliminates dependence on centralised forges such as GitHub or GitLab. It uses a content-addressed, cryptographically signed data model where repositories are identified by public keys rather than server-hosted URLs, enabling sovereign code hosting, offline collaboration, and censorship-resistant software development. Radicle integrates with Ethereum-based smart contracts for governance and funding, positioning it at the intersection of blockchain infrastructure and open-source software tooling.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:radicle-decentralised-code-forge", "labels": [ "Radicle Decentralised Code Forge", "Radicle" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "\ud83e\udd16" ] }, { "id": "radio-access-network", "title": "Radio Access Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A radio access network (RAN) is the part of a mobile telecommunications system that connects user devices to the core network over the air interface via base stations and antennas. It manages radio resource allocation, modulation, and handover across cells. Modern architectures such as Open RAN disaggregate hardware and software to enable multi-vendor, virtualised deployments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:radio-access-network", "labels": [ "Radio Access Network" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "radio-astronomy", "title": "Radio Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radio-astronomy", "labels": [ "Radio Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radio-frequency-spectrum", "title": "Radio Frequency Spectrum", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The radio frequency spectrum is the range of electromagnetic frequencies used to transmit information wirelessly, conventionally spanning roughly three kilohertz to three hundred gigahertz. It is a finite, shared natural resource divided into bands that are allocated and licensed by regulators to services such as broadcasting, mobile telephony, and wireless data, with propagation characteristics and capacity varying systematically with frequency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:radio-frequency-spectrum", "labels": [ "Radio Frequency Spectrum" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "radio-frequency", "title": "Radio Frequency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Radio Frequency (RF) refers to the portion of the electromagnetic spectrum, roughly 3 kHz to 300 GHz, used to transmit information wirelessly by modulating alternating current onto an antenna. It is the physical-layer medium underlying cellular networks, Wi-Fi, Bluetooth, and low-power wide-area networks such as LoRa. RF characteristics, including frequency band, transmit power, and propagation behaviour, determine a wireless technology's range, bandwidth, and power consumption trade-offs.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:radio-frequency", "labels": [ "Radio Frequency" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "radio-transceiver", "title": "Radio Transceiver", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A radio transceiver is a hardware device that combines a transmitter and receiver in a single unit to send and receive radio-frequency signals over a shared channel. It performs modulation, demodulation, amplification, and frequency conversion, forming the physical link layer of wireless devices. Transceivers underpin IoT nodes, mobile devices, and short-range wireless links.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:radio-transceiver", "labels": [ "Radio Transceiver" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "radio-frequency-communication", "title": "Radio-frequency Communication", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radio-frequency-communication", "labels": [ "Radio-frequency Communication" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radioisotope-power-system", "title": "Radioisotope Power System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radioisotope-power-system", "labels": [ "Radioisotope Power System" ], "is_subclass_of": [ "System" ], "wikilinks": [] }, { "id": "radiology-ai", "title": "Radiology AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Radiology AI comprises artificial intelligence systems designed for automated interpretation and quantitative analysis of radiological imaging modalities \u2014 X-ray, CT, MRI, and ultrasound. These systems perform lesion detection, organ segmentation, and structured reporting at radiologist-level accuracy, integrating with PACS workflows and validated through prospective clinical trials.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:radiology-ai", "labels": [ "Radiology AI" ], "is_subclass_of": [ "AI Application", "Medical Imaging AI" ], "wikilinks": [ "DICOM", "Medical Imaging AI", "MetaverseDomain", "Pathology AI" ] }, { "id": "radiometer", "title": "Radiometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiometer", "labels": [ "Radiometer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiometric-accuracy", "title": "Radiometric Accuracy", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiometric-accuracy", "labels": [ "Radiometric Accuracy" ], "is_subclass_of": [ "Measurement Accuracy" ], "wikilinks": [] }, { "id": "radiometric-calibration", "title": "Radiometric Calibration", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiometric-calibration", "labels": [ "Radiometric Calibration" ], "is_subclass_of": [ "Calibration" ], "wikilinks": [] }, { "id": "radiometric-resolution", "title": "Radiometric Resolution", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiometric-resolution", "labels": [ "Radiometric Resolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "radiometric-spacecraft-navigation", "title": "Radiometric Spacecraft Navigation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:radiometric-spacecraft-navigation", "labels": [ "Radiometric Spacecraft Navigation" ], "is_subclass_of": [ "Spacecraft Navigation" ], "wikilinks": [] }, { "id": "raft", "title": "Raft", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Raft is a distributed consensus algorithm designed explicitly for understandability, introduced by Diego Ongaro and John Ousterhout at USENIX ATC 2014 as a more comprehensible alternative to the Paxos family of protocols. Raft decomposes the consensus problem into three relatively independent sub-problems: leader election, log replication, and safety. A Raft cluster maintains a replicated log of commands through a strong leader that serialises all writes; followers replicate the leader's log entries and redirect client requests. Leader election uses randomised timeouts to avoid split votes. Raft has become the dominant consensus algorithm in modern distributed systems infrastructure, underpinning etcd, CockroachDB, TiKV, Consul, and many other widely deployed systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:raft", "labels": [ "Raft", "Raft Algorithm", "Raft Protocol" ], "is_subclass_of": [ "Consensus Algorithm" ], "wikilinks": [] }, { "id": "ragdoll-physics", "title": "Ragdoll Physics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Ragdoll physics is a procedural animation technique that simulates the limp, physically reactive motion of an articulated character body using a system of rigid bodies connected by constrained joints. Instead of playing pre-authored animation, the character's limbs respond dynamically to gravity, collisions and impulses via a physics engine. It is widely used in games and interactive media to produce believable falls, impacts and death animations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ragdoll-physics", "labels": [ "Ragdoll Physics" ], "is_subclass_of": [ "Physics-Based Animation" ], "wikilinks": [] }, { "id": "raid", "title": "Raid", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "RAID (Redundant Array of Independent Disks) is a data storage virtualisation technology that combines multiple physical drives into one or more logical units to improve performance, capacity, or fault tolerance. Different RAID levels trade off redundancy, write performance and usable capacity using techniques such as striping, mirroring and parity. It protects against individual drive failure but is complementary to, not a substitute for, backups.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:raid", "labels": [ "Raid", "RAID" ], "is_subclass_of": [ "Data Storage" ], "wikilinks": [] }, { "id": "rainbow-wallet", "title": "Rainbow Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A self-custodial Ethereum wallet application focused on a consumer-friendly interface for holding tokens and interacting with decentralised applications. The user controls the private keys directly rather than entrusting them to a custodian.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rainbow-wallet", "labels": [ "Rainbow Wallet" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [ "Private Key", "Self-Custody", "Non-Fungible Token", "Wallet", "Ethereum", "Digital Wallet", "https://rainbow.me/" ] }, { "id": "raised-bank", "title": "Raised Bank", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:raised-bank", "labels": [ "Raised Bank" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "random-forest", "title": "Random Forest", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A random forest is an ensemble learning method that constructs many decision trees and aggregates their predictions, typically by majority vote for classification or averaging for regression. Each tree is trained on a bootstrap sample of the data and considers a random subset of features at each split, which decorrelates the trees and reduces variance. The resulting model is robust, resistant to overfitting, and provides built-in estimates of feature importance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:random-forest", "labels": [ "Random Forest" ], "is_subclass_of": [ "Decision Tree", "AI Technique" ], "wikilinks": [] }, { "id": "random-number-generation", "title": "Random Number Generation", "domain": "security", "domain_name": "Security", "definition": "The process of generating unpredictable and statistically random values for cryptographic operations, serving as a critical security primitive for key generation, nonces, and protocol initialization; distinguishes between true randomness from physical entropy sources (TRNG) and pseudo-randomness from deterministic algorithms seeded with entropy (CSPRNG).", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:random-number-generation", "labels": [ "Random Number Generation", "Secure Random Number Generation" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain" ], "wikilinks": [ "Commitment Scheme", "Verifiable Random Function", "Asymmetric Encryption", "Blockchain", "Cryptographic Protocol", "Cryptography", "Hash Function", "Key Derivation Function", "Zero-Knowledge Proof" ] }, { "id": "random-number-generator", "title": "Random Number Generator", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A random number generator (RNG) produces sequences of numbers that lack predictable pattern, either through deterministic algorithms seeded from an initial state (pseudo-random) or from physical entropy sources (true random). RNGs are essential to stochastic sampling, Monte Carlo methods, simulation, and cryptographic key generation. Quality is judged by statistical uniformity, period length, and, for security uses, unpredictability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:random-number-generator", "labels": [ "Random Number Generator", "Cryptographically Secure Pseudorandom Number Generator", "Cryptographically Secure Random Number Generator", "Pseudorandom Number Generator", "Secure Random Number Generator", "True Random Number Generator" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "random-oracle-model", "title": "Random Oracle Model", "domain": "security", "domain_name": "Security", "definition": "The Random Oracle Model (ROM) is a theoretical framework in cryptography that idealises a cryptographic hash function as a publicly accessible, truly random function: for every distinct input the oracle returns an independent, uniformly random output, while repeated queries return the same value. Introduced formally by Bellare and Rogaway in 1993, the ROM enables security proofs for practical schemes such as RSA-OAEP, ECDSA, and Schnorr signatures that cannot currently be proved secure under standard computational assumptions alone. Because no concrete hash function perfectly instantiates a random oracle, ROM proofs are treated as strong heuristic evidence rather than unconditional guarantees, and pathological counterexamples exist that are ROM-secure but concretely insecure under any hash instantiation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:random-oracle-model", "labels": [ "Random Oracle Model", "Random Oracle" ], "is_subclass_of": [ "Security and Identity", "Cryptographic Proof Model" ], "wikilinks": [ "Hash Function", "Cryptographic Hash Function", "Cryptography" ] }, { "id": "random-sampling", "title": "Random Sampling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Random sampling is a method of selecting a subset of items from a population such that every element has a known, non-zero probability of being chosen, with selections governed by chance rather than judgement. It is the foundation of statistical inference, allowing properties of a population to be estimated from a representative sample while quantifying uncertainty. In machine learning it underpins data partitioning, stochastic optimisation and Monte Carlo estimation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:random-sampling", "labels": [ "Random Sampling" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "random-search", "title": "Random Search", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Random search is a hyperparameter-optimisation method that samples hyperparameter configurations at random from specified distributions over the search space, rather than evaluating a fixed grid. For a given evaluation budget it often outperforms grid search because it explores more distinct values of the most influential hyperparameters. It is simple, parallelisable, and a strong baseline for automated tuning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:random-search", "labels": [ "Random Search" ], "is_subclass_of": [ "Hyperparameter Optimisation", "Hyperparameter Tuning" ], "wikilinks": [] }, { "id": "random-variable", "title": "Random Variable", "domain": "data", "domain_name": "Data", "definition": "A measurable function that assigns a numerical value to each outcome in the sample space of a random experiment, providing the formal bridge between abstract probability spaces and quantitative analysis. Random variables may be discrete or continuous, are fully characterised by their probability distributions, and underpin expectation, variance, covariance, and the entropy measures central to information theory and statistical machine learning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:random-variable", "labels": [ "Random Variable" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [ "Probability Theory", "Probability Distribution", "Covariance Matrix", "Entropy", "Statistics", "Stochastic Process" ] }, { "id": "random-walk", "title": "Random Walk", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A random walk is a stochastic process describing a path formed by a sequence of random steps, each step's direction and size drawn from a probability distribution independent of the path's prior history in the simplest (Markovian) case. It underlies graph embedding techniques such as node2vec and DeepWalk, which sample walks over a graph to learn vector representations of vertices, and is foundational to the theory of Markov chains. Its long-run behaviour \u2014 recurrence, transience, and diffusion rate \u2014 depends on the dimensionality and structure of the underlying space.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:random-walk", "labels": [ "Random Walk", "Random Walks" ], "is_subclass_of": [ "Stochastic Process" ], "wikilinks": [] }, { "id": "randomised-controlled-trial", "title": "Randomised Controlled Trial", "domain": "governance", "domain_name": "Governance", "definition": "A randomised controlled trial (RCT) is an experimental study design in which participants are randomly assigned to an intervention or a control group to estimate the causal effect of the intervention while minimising bias and confounding. Randomisation balances known and unknown covariates across groups, and the controlled comparison isolates the treatment effect from background trends. RCTs are widely regarded as the methodological gold standard for causal inference in medicine and policy evaluation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:randomised-controlled-trial", "labels": [ "Randomised Controlled Trial" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "range-proof", "title": "Range Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A range proof is a cryptographic zero-knowledge protocol that allows a prover to convince a verifier that a committed secret value lies within a specified numeric interval, without revealing the value itself. It is essential to confidential transaction systems, where amounts are hidden inside commitments yet must be proven non-negative to prevent inflation via negative-value forgery. Modern constructions such as Bulletproofs achieve compact, logarithmic-sized proofs without a trusted setup, making range proofs practical for on-chain privacy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:range-proof", "labels": [ "Range Proof" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [] }, { "id": "ranked-choice-voting", "title": "Ranked-Choice Voting", "domain": "governance", "domain_name": "Governance", "definition": "Ranked-choice voting is a voting method in which voters rank candidates in order of preference, and votes are redistributed in rounds by eliminating the lowest-ranked candidate until one candidate holds a majority. It is studied within social choice theory as an alternative to plurality voting intended to reduce spoiler effects and better reflect aggregate voter preference. It is used in some collective decision-making systems, including governance mechanisms for decentralised organisations.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:ranked-choice-voting", "labels": [ "Ranked-Choice Voting" ], "is_subclass_of": [ "Voting System" ], "wikilinks": [] }, { "id": "ransomware", "title": "Ransomware", "domain": "security", "domain_name": "Security", "definition": "A class of malware that denies victims access to their data or systems \u2014 typically by encrypting files with attacker-held keys, and increasingly by exfiltrating data for extortion \u2014 and demands payment, usually in cryptocurrency, for restoration or non-disclosure. Operated today as a service economy with affiliates, initial-access brokers, and leak sites, ransomware is among the most financially damaging cyber threats and the principal stress test of an organisation's backup, incident response, and disaster recovery posture.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:ransomware", "labels": [ "Ransomware" ], "is_subclass_of": [ "Malware" ], "wikilinks": [ "Malware", "Encryption", "Data Breach", "Disaster Recovery", "Incident Response", "Bitcoin" ] }, { "id": "rapid-ai-deployment", "title": "Rapid AI Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Rapid AI Deployment refers to accelerated processes for taking AI models from development into production, prioritising speed-to-market through streamlined MLOps pipelines, automated testing, and pre-built infrastructure. It encompasses practices such as continuous delivery of model updates, containerised inference serving, and automated monitoring to minimise the time between model training and live operation while managing associated risks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:rapid-ai-deployment", "labels": [ "Rapid AI Deployment" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "rapid-prototyping", "title": "Rapid Prototyping", "domain": "ai", "domain_name": "Ai", "definition": "Rapid Prototyping is an iterative product development approach that emphasises the construction of quick, low-fidelity or medium-fidelity artefacts\u2014physical mock-ups, interactive wireframes, or functional code spikes\u2014to test hypotheses with users and stakeholders before committing to full implementation. The methodology aims to compress the feedback loop between ideation and validated learning, reducing the risk of building features that do not meet user needs. It draws on lean startup principles, agile sprint structures, and design thinking frameworks, integrating user testing results directly into successive prototype iterations. Modern generative AI tools further accelerate rapid prototyping by auto-generating code scaffolds, UI components, and content variants.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rapid-prototyping", "labels": [ "Rapid Prototyping", "Iterative Prototyping", "Rapid Model Development", "Rapid Visual Prototyping" ], "is_subclass_of": [ "Design Thinking" ], "wikilinks": [] }, { "id": "rapidly-exploring-random-tree", "title": "Rapidly Exploring Random Tree", "domain": "robotics", "domain_name": "Robotics", "definition": "A Rapidly Exploring Random Tree (RRT) is a sampling-based motion-planning algorithm that incrementally builds a space-filling tree by drawing random samples from the configuration space and extending the tree toward each sample. It efficiently explores high-dimensional spaces while respecting kinematic and obstacle constraints, and its variants such as RRT* add asymptotic optimality. RRT is widely used for robot path planning where the free space is too large or complex for grid-based search.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rapidly-exploring-random-tree", "labels": [ "Rapidly Exploring Random Tree", "Rapidly-Exploring Random Tree" ], "is_subclass_of": [ "Sampling Based Planning" ], "wikilinks": [] }, { "id": "raspberry-pi", "title": "Raspberry Pi", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Raspberry Pi is a family of low-cost, credit-card-sized single-board computers built around ARM systems-on-chip, designed originally for education and now ubiquitous in hobbyist, prototyping, and edge-computing deployments. It exposes GPIO pins for hardware interfacing and runs full Linux distributions, making it a versatile platform for home automation, robotics controllers, and small servers. Its affordability and large community make it a default choice for edge experimentation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:raspberry-pi", "labels": [ "Raspberry Pi" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "rasterization", "title": "Rasterization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The dominant real-time rendering technique that converts 3D geometric primitives \u2014 predominantly triangles \u2014 into a 2D pixel representation by determining per-pixel coverage, depth, and colour through a GPU-accelerated pipeline of vertex processing, primitive assembly, scan conversion, fragment shading, and output merging. Rasterization trades photorealistic accuracy for deterministic, high-throughput performance suitable for interactive applications.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:rasterization", "labels": [ "Rasterization", "Rasterisation" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "3D Rendering", "Level of Detail", "Metaverse", "Pixel Shader", "Ray Tracing", "Vertex Shader" ] }, { "id": "rate-limiting", "title": "Rate Limiting", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Rate limiting is a traffic-management technique that controls the number of requests a client may make to a service within a defined time window, rejecting or queuing excess requests. It protects backend resources from overload, enforces fair usage and quota policies, and mitigates abuse such as brute-force and denial-of-service attacks. Commonly implemented at API gateways using token-bucket or sliding-window algorithms, it is a core mechanism for the resilience and stability of networked systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rate-limiting", "labels": [ "Rate Limiting" ], "is_subclass_of": [ "Infrastructure", "Network Infrastructure" ], "wikilinks": [] }, { "id": "rational-agent", "title": "Rational Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A rational agent is an agent that selects actions expected to maximise its performance measure given its knowledge, percept history and available actions, a foundational concept in decision theory and classical AI. Rationality is typically formalised through an explicit utility function over outcomes, and mechanism design and multi-agent systems often assume agents behave rationally when reasoning about incentives. The concept contrasts with agents that act on fixed rules or bounded heuristics rather than expected-utility optimisation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:rational-agent", "labels": [ "Rational Agent" ], "is_subclass_of": [ "Agent" ], "wikilinks": [] }, { "id": "raw-data", "title": "Raw Data", "domain": "data", "domain_name": "Data", "definition": "Raw data is unprocessed information as originally collected from a source, before cleaning, transformation, aggregation or annotation. It may be noisy, inconsistent, redundant or incomplete, and typically lacks the structure and quality guarantees of processed data. Raw data forms the input to data pipelines, where it is ingested, validated and transformed into usable, analysis-ready forms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:raw-data", "labels": [ "Raw Data" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "ray-marching", "title": "Ray Marching", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Ray Marching is a rendering technique in which a ray is incrementally stepped through a scene, evaluating a signed distance field (SDF) at each step to determine proximity to geometry. The step size adapts to the SDF value (sphere tracing), enabling efficient rendering of implicit surfaces, volumetric effects, soft shadows, and ambient occlusion that are impractical with triangle-based rasterisation. It is widely implemented in GPU shader programs and is foundational to procedural 3D scene generation in real-time and offline contexts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ray-marching", "labels": [ "Ray Marching", "Volumetric Ray Marching" ], "is_subclass_of": [ "Display and Rendering", "Rendering Technique" ], "wikilinks": [] }, { "id": "ray-tracing", "title": "Ray Tracing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Ray Tracing is a physically-based light-transport simulation technique in computer graphics that generates images by casting rays from a virtual camera through each image-plane pixel into a three-dimensional scene, recursively computing colour values by evaluating ray-geometry intersections, surf...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:ray-tracing", "labels": [ "Ray Tracing", "Ray Tracing Core", "Real-Time Ray Tracing" ], "is_subclass_of": [ "Display and Rendering", "Computer Graphics", "Light Transport", "Physically-Based Rendering", "Global Illumination", "Rendering", "Monte Carlo Methods" ], "wikilinks": [ "Acceleration Structure", "AI Upscaling", "AlgorithmLayer", "Ambient Occlusion", "Ambient Occlusion Maps", "Architectural Visualisation", "Baked Lightmaps", "Bidirectional Path Tracing", "Bitterli et al. 2020 ReSTIR SIGGRAPH TOG", "BRDF", "BVH Acceleration Structure", "BVH Construction", "Camera Model", "Caustics", "CD Projekt Red NVIDIA Cyberpunk 2077 Overdrive 2023", "Computer Graphics", "ComputerGraphicsDomain", "Denoiser", "Denoising", "DirectX" ] }, { "id": "raycast-system", "title": "Raycast System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Raycast System is the interaction subsystem in a spatial computing or XR engine that casts geometric rays from a controller, gaze direction, or cursor into the 3D scene to detect intersections with scene objects, enabling selection, pointing, UI interaction, and collision queries. It forms the primary input-resolution layer in VR/AR user interfaces, translating physical or tracked user intent into object picks within the scene graph.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:raycast-system", "labels": [ "Raycast System" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "rdf-xml", "title": "Rdf Xml", "domain": "data", "domain_name": "Data", "definition": "RDF/XML is the original World Wide Web Consortium serialisation format for the Resource Description Framework, expressing RDF triples as XML documents. It encodes subjects, predicates, and objects using XML elements and attributes, making RDF data parseable by generic XML tooling. Although verbose and historically awkward to read, it remains a standardised interchange format alongside more concise alternatives such as Turtle and JSON-LD.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rdf-xml", "labels": [ "Rdf Xml", "RDF/XML" ], "is_subclass_of": [ "RDF" ], "wikilinks": [] }, { "id": "rdma", "title": "Rdma", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Remote Direct Memory Access (RDMA) is a networking capability that lets one computer read from or write to the memory of another without involving either machine's operating system or CPU on the data path. By bypassing kernel buffering and copying, RDMA delivers very low latency and high throughput, which is essential for the collective communication patterns of large-scale distributed training. It is exposed through fabrics such as InfiniBand and RoCE and underpins high-performance computing clusters.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rdma", "labels": [ "Rdma", "RDMA" ], "is_subclass_of": [ "Networking", "Network and Communication" ], "wikilinks": [] }, { "id": "re-entry", "title": "Re-entry", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:re-entry", "labels": [ "Re-entry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "re-act-pattern", "title": "ReAct Pattern", "domain": "ai", "domain_name": "Ai", "definition": "ReAct (Reasoning and Acting) is a prompting and agent-control pattern in which a language model interleaves explicit reasoning traces with action steps such as tool calls, observing the results before reasoning again. This loop lets the model decompose tasks, gather information, and self-correct rather than answering in a single pass. It is a foundational design for tool-using LLM agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:re-act-pattern", "labels": [ "ReAct Pattern", "ReAct Agent Pattern" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "re-act", "title": "react", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ReAct (Reasoning and Acting) is a prompting paradigm for large language models that interleaves free-form reasoning traces (Thought steps) with structured external action calls in a single output sequence, enabling models to dynamically plan multi-step tasks, observe the results of tool invocations, and revise their reasoning accordingly. The framework produces an alternating Thought\u2013Action\u2013Observation loop that grounds model inference in real-world feedback, contrasting with pure chain-of-thought prompting (reasoning without action) and pure action-only agents (action without transparent reasoning). Originally demonstrated by Yao et al. (2022) on knowledge-intensive QA and interactive decision benchmarks, ReAct has become the foundational interaction primitive underpinning modern agentic AI systems, tool-augmented LLMs, and multi-agent orchestration frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:re-act", "labels": [ "ReAct", "ReAct Framework", "React" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "re-lu-activation", "title": "ReLU Activation", "domain": "ai", "domain_name": "Ai", "definition": "The Rectified Linear Unit (ReLU) is a nonlinear activation function defined as f(x) = max(0, x), outputting the input directly when positive and zero otherwise. Its simplicity, sparse activation, and non-saturating gradient for positive inputs make it the default activation in most deep neural networks, mitigating the vanishing-gradient problem. Variants such as Leaky ReLU and GELU address its dead-neuron limitation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:re-lu-activation", "labels": [ "ReLU Activation", "ReLU" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "react-prompting", "title": "React Prompting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "ReAct prompting is a prompting strategy that interleaves verbal reasoning traces with discrete actions, enabling a language model to think step by step while interacting with external tools or environments. The model alternates between generating a thought, taking an action such as a search or API call, and observing the result, then folding that observation back into subsequent reasoning. This synergy of reasoning and acting reduces hallucination and grounds the model's conclusions in retrieved evidence.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:react-prompting", "labels": [ "React Prompting", "ReAct Prompting" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "reaction-wheel", "title": "Reaction Wheel", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reaction-wheel", "labels": [ "Reaction Wheel" ], "is_subclass_of": [ "Actuator" ], "wikilinks": [] }, { "id": "reactive-control", "title": "Reactive Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Reactive Control is a paradigm of robot control architecture in which sensor inputs are mapped directly to actuator outputs through fast, pre-compiled stimulus-response rules, without constructing an explicit world model. Associated with Brooks's subsumption architecture and behaviour-based robotics, it enables low-latency responses to environmental perturbations and is typically combined with deliberative or hybrid planners in practical autonomous systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:reactive-control", "labels": [ "Reactive Control" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "reactive-planning", "title": "Reactive Planning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An approach to agent control that selects actions directly from the current perceived situation rather than constructing and executing a complete plan in advance, trading long-horizon optimality for immediate responsiveness so that agents can act robustly in dynamic, uncertain, or partially observable environments where deliberative plans would be invalidated before they finish executing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:reactive-planning", "labels": [ "Reactive Planning" ], "is_subclass_of": [ "Automated Planning" ], "wikilinks": [ "Automated Planning", "Classical Planning", "Behaviour Tree", "Subsumption Architecture" ] }, { "id": "reactive-systems", "title": "Reactive Systems", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Reactive systems are software systems designed around asynchronous message passing and event propagation, responding to events as they occur rather than following a fixed, sequential control flow. They are commonly built using an event-driven architecture, in which discrete events drive computation, and the publish-subscribe pattern, which decouples event producers from consumers. This design supports responsiveness, resilience and elasticity under variable load.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:reactive-systems", "labels": [ "Reactive Systems", "Reactive System" ], "is_subclass_of": [ "Event Driven Architecture" ], "wikilinks": [] }, { "id": "read-receipts", "title": "Read Receipts", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Read receipts are delivery-confirmation signals that indicate whether a message has been seen by its recipient, typically displayed as a checkmark, avatar, or timestamp. They provide senders with visibility into message consumption, reducing uncertainty in asynchronous communication. Their use in distributed collaboration can improve follow-up timing but must be balanced against recipient privacy expectations.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:read-receipts", "labels": [ "Read Receipts" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "ready-player-me", "title": "Ready Player Me", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Ready Player Me (RPM) is a cross-application avatar platform that enables users to create a single personalised 3D humanoid avatar and deploy it across hundreds of games, social VR environments, and virtual worlds via an open SDK and REST API. Founded in 2021 by Wolf3D, it provides a browser-based avatar creator, glTF-compliant avatar meshes, morph-target facial animation, and an OAuth-based identity layer so that the same avatar persona persists across compatible applications. By decoupling avatar creation from any single platform, RPM acts as a shared identity infrastructure for the emerging spatial internet, addressing avatar fragmentation across [[Metaverse]] ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ready-player-me", "labels": [ "Ready Player Me" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Avatar", "Metaverse", "Digital Identity" ] }, { "id": "real-analysis", "title": "Real Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Real Analysis is the branch of mathematics that rigorously studies the real numbers, sequences, series, limits, continuity, differentiation and integration. It provides the formal foundations underlying calculus, establishing theorems on convergence and the structure of the real line. Real analysis underpins measure theory, functional analysis and the probabilistic and optimisation theory used in machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:real-analysis", "labels": [ "Real Analysis" ], "is_subclass_of": [ "Calculus", "AI Research Area" ], "wikilinks": [] }, { "id": "real-estate-tokenization", "title": "Real Estate Tokenization", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based systems employing security tokens (ERC-3643, ERC-1400, ERC-20 standards) to represent fractional ownership interests in real estate properties, enabling automated dividend distribution through smart contracts, reducing investment minimums from tens of thousands to \u00a350, and creati...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:real-estate-tokenization", "labels": [ "Real Estate Tokenization", "Tokenised Real Estate" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "BC-0432-consortium-blockchain", "BC-0456-self-sovereign-identity", "BC-0458-verifiable-credentials", "BC-0463-governance-token", "BC-0478-securities-regulation", "BC-0488-licensing-requirements", "BC-0494-property-registry", "BlockchainDomain" ] }, { "id": "real-time-character-animation", "title": "Real Time Character Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of generating and rendering character movements instantaneously during gameplay or interactive experiences, utilizing rigging systems, motion capture data, and procedural animation to create lifelike digital characters that respond dynamically to user input and environmental conditions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:real-time-character-animation", "labels": [ "Real Time Character Animation", "Real-Time Character Animation" ], "is_subclass_of": [ "Content and Assets", "3D Animation" ], "wikilinks": [ "3D Animation", "Interactive Experiences", "metaverse" ] }, { "id": "real-time-collaboration", "title": "Real Time Collaboration", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Real-time collaboration is the practice and technology of multiple users working concurrently on shared content with changes propagated and merged with minimal latency. It relies on synchronisation algorithms that resolve concurrent edits while preserving each participant's intent. It underpins collaborative editors, shared design tools, and multi-user virtual environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-collaboration", "labels": [ "Real Time Collaboration", "Real-Time Collaboration", "Real-time Collaboration" ], "is_subclass_of": [ "Distributed Collaboration", "Synchronous Collaboration" ], "wikilinks": [] }, { "id": "real-time-digital-twin-synchronization", "title": "Real Time Digital Twin Synchronization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The continuous bidirectional data exchange between physical industrial assets and their virtual replicas in the industrial metaverse, enabling real-time monitoring, predictive maintenance, and scenario testing through IoT sensor integration and cloud-based rendering.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:real-time-digital-twin-synchronization", "labels": [ "Real Time Digital Twin Synchronization", "Real-Time Digital Twin Synchronization" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin Technology" ], "wikilinks": [ "Virtual Real Interaction", "Digital Twin Technology", "metaverse" ] }, { "id": "real-time-graphics", "title": "Real Time Graphics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time graphics is the field of computer graphics concerned with generating and displaying images fast enough to produce interactive, continuously updating visuals, typically at frame rates of sixty frames per second or higher. It prioritises low and predictable latency over photorealistic fidelity, relying on hardware acceleration and approximate rendering techniques. It is the foundation for games, simulations and immersive spatial computing experiences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-graphics", "labels": [ "Real Time Graphics", "Real-Time Graphics" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "real-time-messaging", "title": "Real Time Messaging", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Real-time messaging is the exchange of messages between participants with sub-second latency, so that communication feels instantaneous and conversational. It relies on persistent, bidirectional transport such as WebSockets and on patterns like publish-subscribe to fan out events to many recipients. Real-time messaging underpins chat, presence, notifications and live collaboration across distributed systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-messaging", "labels": [ "Real Time Messaging", "Real-Time Messaging" ], "is_subclass_of": [ "Real-Time Communication" ], "wikilinks": [] }, { "id": "real-time-systems", "title": "Real Time Systems", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time systems are computing systems whose correctness depends not only on logical results but also on the time at which those results are produced. They must respond to events within defined timing constraints, classified as hard, firm or soft depending on the consequences of a missed deadline. Such systems are central to control, robotics, simulation and immersive applications where late results are useless or dangerous.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-systems", "labels": [ "Real Time Systems", "Real-Time Systems" ], "is_subclass_of": [ "Embedded System", "Platform and Environment" ], "wikilinks": [] }, { "id": "real-time", "title": "Real Time", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-Time Computing is the discipline of designing computing systems, operating environments, and algorithmic frameworks in which program correctness depends not only on the logical result of computation but also on the time at which those results are produced, enforcing temporal constraints \u2014 de...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:real-time", "labels": [ "Real Time", "Real-Time", "Real-Time System" ], "is_subclass_of": [ "Computing and Cloud", "Operating System", "Safety-Critical Systems", "Embedded Systems", "Operating Systems", "Control Systems", "Cyber Physical Systems" ], "wikilinks": [ "Abeni and Buttazzo 1998 Constant Bandwidth Server", "Audsley et al 1993 Static Priority Scheduling", "Automotive Control", "Autonomous Systems", "Autonomous Vehicles", "Avionics", "Baker 1991 Stack Resource Policy", "Baruah et al 2012 Multiprocessor Schedulability", "Best Effort Networking", "Burns and Davis 2022 Mixed Criticality Survey ACM", "Burns and Wellings 2022 Real-Time Systems 5th ed", "Buttazzo 2024 Hard Real-Time Computing Systems 4th ed", "ComputerScienceDomain", "Constant Bandwidth Server", "Control Systems", "Corbet et al 2024 PREEMPT_RT Linux 6.12", "Cyber Physical Systems", "Davis and Burns 2011 Multiprocessor Real-Time Survey ACM", "Davis and Cucu-Grosjean 2019 Probabilistic Timing Analysis", "DDS" ] }, { "id": "real-world-asset", "title": "Real World Asset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A real-world asset (RWA) is a tangible or off-chain financial asset, such as real estate, commodities, invoices, bonds or equities, whose ownership or economic rights are represented on a blockchain as a token. Tokenisation links the on-chain representation to legal claims and custody arrangements in the physical or traditional financial world. RWAs bridge decentralised finance with regulated, conventional asset markets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:real-world-asset", "labels": [ "Real World Asset", "Real-World Asset" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "real-esrgan", "title": "Real-ESRGAN", "domain": "ai", "domain_name": "Ai", "definition": "Real-ESRGAN is an image super-resolution model that upscales and restores low-quality images, with particular attention to real-world degradations. It extends the ESRGAN architecture using a generative adversarial network trained on synthetic degradation data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:real-esrgan", "labels": [ "Real-ESRGAN", "ESRGAN Upscaling" ], "is_subclass_of": [ "Generative Adversarial Network" ], "wikilinks": [ "Generative Adversarial Network", "Convolutional Neural Network", "Image Generation", "Computer Vision", "Deep Learning" ] }, { "id": "real-time-3d-graphics", "title": "Real-Time 3D Graphics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-Time 3D Graphics is the rendering of three-dimensional scenes at interactive frame rates, typically 30 to 120 frames per second, so that the displayed image responds immediately to changes in viewpoint or scene state. It relies on rasterization pipelines and GPU hardware to meet strict per-frame time budgets, trading off some visual fidelity against the offline techniques used in cinematic rendering. It underlies video games, XR applications, and real-time visualisation tools.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:real-time-3d-graphics", "labels": [ "Real-Time 3D Graphics" ], "is_subclass_of": [ "3D Rendering" ], "wikilinks": [] }, { "id": "real-time-ai-inference", "title": "Real-Time AI Inference", "domain": "ai", "domain_name": "Ai", "definition": "Real-time AI inference is the execution of trained machine learning model forward passes within latency bounds tight enough to support interactive or time-critical applications, typically measured in milliseconds to tens of milliseconds. It requires co-optimisation of model architecture, runtime software, and hardware accelerators \u2014 including GPUs, NPUs, and dedicated AI ASICs \u2014 to meet throughput and latency targets whilst maintaining acceptable accuracy. Applications include autonomous vehicles, real-time video analysis, voice assistants, and spatial computing overlays.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:real-time-ai-inference", "labels": [ "Real-Time AI Inference" ], "is_subclass_of": [ "AI Inference", "AI Infrastructure" ], "wikilinks": [] }, { "id": "real-time-ai", "title": "Real-Time AI", "domain": "ai", "domain_name": "Ai", "definition": "Real-Time AI refers to the application of artificial intelligence inference within strict latency bounds \u2014 typically sub-second to sub-millisecond \u2014 required for time-sensitive tasks such as autonomous driving, robotics control, live speech processing, financial trading, and augmented reality. It integrates specialised hardware (NPUs, GPUs, FPGAs), optimised model representations (quantisation, pruning, TensorRT/ONNX), and edge deployment architectures to ensure deterministic response times. The distinction from batch AI lies in the hard or soft real-time constraints that govern system correctness.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:real-time-ai", "labels": [ "Real-Time AI" ], "is_subclass_of": [ "Real-Time Computation" ], "wikilinks": [] }, { "id": "real-time-analytics", "title": "Real-Time Analytics", "domain": "data", "domain_name": "Data", "definition": "Real-time analytics is the processing and analysis of data immediately as it is generated, producing insights and triggering actions with sub-second to low-second latency. It relies on stream-processing engines, in-memory computation, and windowed aggregation rather than batch ETL. It enables responsive dashboards, anomaly detection, and adaptive systems that act on fresh data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-analytics", "labels": [ "Real-Time Analytics", "Real Time Analytics" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "real-time-animation", "title": "Real-Time Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time animation is the generation or playback of character and object motion at interactive frame rates, computed live rather than pre-rendered, so that the animation can respond immediately to input or simulation state. It relies on techniques such as animation retargeting to map captured or authored motion onto different skeletons on the fly, and on motion capture as a common source of the underlying motion data. It is a core requirement for games, virtual avatars and interactive XR experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:real-time-animation", "labels": [ "Real-Time Animation" ], "is_subclass_of": [ "Animation" ], "wikilinks": [] }, { "id": "real-time-captioning", "title": "Real-Time Captioning", "domain": "ai", "domain_name": "Ai", "definition": "Real-time captioning is the continuous generation of synchronised text captions from live audio using streaming automatic speech recognition, with latency low enough that captions appear as speech occurs. It is used for live broadcasts, video conferencing, and accessibility services for deaf and hard-of-hearing users. It requires low-latency acoustic and language models tuned to trade a small amount of accuracy for responsiveness.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:real-time-captioning", "labels": [ "Real-Time Captioning" ], "is_subclass_of": [ "Automatic Speech Recognition" ], "wikilinks": [] }, { "id": "real-time-collaborative-editing", "title": "Real-Time Collaborative Editing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time collaborative editing is the capability for multiple users to concurrently modify a shared document or data structure and see each other's changes near-instantaneously with automatic conflict resolution. It depends on synchronisation algorithms such as operational transformation or conflict-free replicated data types to maintain a consistent merged state across replicas. It underpins shared documents, whiteboards, and collaborative virtual workspaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-collaborative-editing", "labels": [ "Real-Time Collaborative Editing", "Real-time Collaborative Editor" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "real-time-communication", "title": "Real-Time Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time communication (RTC) is the exchange of audio, video, and arbitrary data between two or more endpoints with latency low enough to support interactive, synchronous engagement \u2014 typically under 150 ms end-to-end for voice and video. It encompasses the protocols, codecs, signalling mechanisms, and network transport layers that jointly minimise delay, jitter, and packet loss while adapting to dynamic network conditions. Modern RTC systems span peer-to-peer browser sessions via WebRTC, carrier-grade VoIP infrastructure built on SIP and RTP, and real-time data channels used in collaborative applications, gaming, and distributed control systems. Quality of experience is governed by congestion control algorithms, forward error correction, jitter buffering, and adaptive bitrate strategies.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:real-time-communication", "labels": [ "Real-Time Communication", "Real-Time Communication Protocols" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "WebRTC", "Video Compression", "Wireless Connectivity", "Communication Protocol" ] }, { "id": "real-time-computation", "title": "Real-Time Computation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time computation refers to computational processes that must produce correct outputs within specified, externally imposed time constraints\u2014where the correctness of a result depends not only on its logical accuracy but on its delivery before a deadline. Hard real-time systems guarantee deadline satisfaction under all conditions, soft real-time systems tolerate occasional deadline misses, and firm real-time systems discard late results as worthless. Applications span industrial control, autonomous vehicles, robotics, financial trading, and interactive media. Real-time computation requires deterministic execution paths, bounded memory allocation, and often specialised hardware or operating system schedulers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:real-time-computation", "labels": [ "Real-Time Computation" ], "is_subclass_of": [ "Real Time" ], "wikilinks": [] }, { "id": "real-time-computing", "title": "Real-Time Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time computing is a computational paradigm in which the correctness of a system depends not only on producing logically correct results but also on producing them within specified timing constraints. Systems are classified as hard real-time (where a missed deadline constitutes a system failure, e.g. aircraft fly-by-wire), firm real-time (where late results are useless but non-catastrophic), or soft real-time (where occasional deadline misses degrade quality rather than cause failure, e.g. multimedia streaming). Achieving real-time guarantees requires deterministic scheduling, bounded interrupt latency, and careful resource management throughout the entire software stack.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:real-time-computing", "labels": [ "Real-Time Computing" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Real-Time Operating System", "Embedded Systems", "Latency", "owl:Thing" ] }, { "id": "real-time-control-loop", "title": "Real-Time Control Loop", "domain": "robotics", "domain_name": "Robotics", "definition": "A real-time control loop is a cyclic sense-compute-actuate process that runs at a fixed, deterministic frequency to regulate a physical or cyber-physical system within strict timing deadlines. Missing a deadline can cause instability or unsafe behaviour, so loops require real-time scheduling and bounded computation. They are the operational core of robotics, motion control, and industrial automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-control-loop", "labels": [ "Real-Time Control Loop" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "real-time-control", "title": "Real-Time Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Real-Time Control is the design and implementation of control systems that must respond to sensor inputs and actuator commands within bounded, deterministic time deadlines, where missing a deadline constitutes system failure. Hard real-time systems\u2014common in aircraft fly-by-wire, automotive ABS, and surgical robotics\u2014require guaranteed worst-case execution times measured in microseconds to milliseconds. Soft real-time systems tolerate occasional deadline misses with degraded performance rather than catastrophic failure. Real-time control relies on real-time operating systems (RTOS), dedicated hardware co-processors, and carefully bounded software to ensure predictable timing under all operating conditions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:real-time-control", "labels": [ "Real-Time Control", "Real Time Control", "Real-Time Control System", "Real-Time Control Systems" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "real-time-data-access", "title": "Real-Time Data Access", "domain": "data", "domain_name": "Data", "definition": "Real-time data access is the capability to query and retrieve current, up-to-date data from underlying systems with minimal delay between data generation and availability to consumers. It is typically achieved through data virtualization layers, streaming pipelines, or optimised database query paths that avoid batch-processing latency. Real-time data access underpins applications such as live dashboards, operational monitoring, and time-sensitive decision-making that cannot tolerate stale data.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:real-time-data-access", "labels": [ "Real-Time Data Access" ], "is_subclass_of": [ "Data Virtualization" ], "wikilinks": [] }, { "id": "real-time-data-processing", "title": "Real-Time Data Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time data processing is the continuous ingestion, transformation, and analysis of data immediately as it is produced, delivering results within strict latency bounds rather than in scheduled batches. It underpins applications that must react to events as they happen, such as fraud detection, monitoring, and live analytics, and depends on low-latency pipelines, stream-processing engines, and event-driven architectures. Its defining constraint is bounded end-to-end latency between data arrival and actionable output.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-data-processing", "labels": [ "Real-Time Data Processing" ], "is_subclass_of": [ "Stream Processing" ], "wikilinks": [] }, { "id": "real-time-gross-settlement", "title": "Real-Time Gross Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Real-time gross settlement (RTGS) is an interbank payment mechanism in which funds transfers between institutions are settled individually and irrevocably, transaction by transaction, in central-bank money as each instruction is processed. Because settlement is gross rather than netted and occurs continuously throughout the day, RTGS eliminates settlement risk between counterparties at the moment of transfer. National central banks typically operate RTGS systems for large-value, time-critical payments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:real-time-gross-settlement", "labels": [ "Real-Time Gross Settlement", "Real Time Gross Settlement" ], "is_subclass_of": [ "Payment System" ], "wikilinks": [] }, { "id": "real-time-inference-engine", "title": "Real-Time Inference Engine", "domain": "ai", "domain_name": "Ai", "definition": "A real-time inference engine is an optimised runtime that executes trained machine-learning models with low, predictable latency to serve predictions within interactive or streaming time budgets. It applies techniques such as operator fusion, quantisation, batching, and hardware acceleration to meet throughput and latency targets. It is the serving layer that turns offline-trained models into responsive online services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-inference-engine", "labels": [ "Real-Time Inference Engine" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "real-time-inference-at-edge", "title": "Real-Time Inference at Edge", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The execution of machine learning inference on edge devices under deterministic latency constraints, typically P99 latency below 10\u2013100 ms, to support safety-critical and time-sensitive applications. Achieves real-time performance through hardware accelerators (NPUs, FPGAs, ASICs), model compression, and priority scheduling without reliance on cloud round-trips.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:real-time-inference-at-edge", "labels": [ "Real-Time Inference at Edge" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "real-time-inference", "title": "Real-Time Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Real-time inference is the execution of a trained machine-learning model to produce predictions within strict, low-latency time bounds suitable for interactive or streaming applications. It demands optimised serving infrastructure, efficient model formats, and often hardware acceleration to meet sub-second or millisecond response targets. Real-time inference enables responsive AI features such as recommendations, fraud scoring, and perception in autonomous systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:real-time-inference", "labels": [ "Real-Time Inference", "Real Time Inference" ], "is_subclass_of": [ "Model Serving", "Model Deployment" ], "wikilinks": [] }, { "id": "real-time-interpretation", "title": "Real-Time Interpretation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Real-time interpretation is the simultaneous conversion of spoken language from a source language into a target language with minimal delay, combining speech recognition, machine translation and speech synthesis in a continuous pipeline. Unlike batch machine translation, it must produce output incrementally as the source speech arrives, trading some accuracy for low latency. It underpins live captioning, simultaneous conference interpretation and voice-based cross-lingual communication tools.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:real-time-interpretation", "labels": [ "Real-Time Interpretation" ], "is_subclass_of": [ "Speech Recognition" ], "wikilinks": [] }, { "id": "real-time-language-translation", "title": "Real-Time Language Translation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The use of artificial intelligence-powered natural language processing to automatically translate spoken or written communication between languages during live telepresence interactions with sub-second latency, enabling cross-lingual collaboration without human interpreters through neural machin...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:real-time-language-translation", "labels": [ "Real-Time Language Translation", "TELE-105-real-time-language-translation" ], "is_subclass_of": [ "Communication Technology", "ArtificialIntelligence" ], "wikilinks": [ "CrossLingualCollaboration", "NeuralMachineTranslation", "TELE-002-telecollaboration", "TELE-020-virtual-reality-telepresence", "TELE-106-speech-to-speech-translation", "TELE-107-ai-meeting-assistants", "ArtificialIntelligence", "NaturalLanguageProcessing" ] }, { "id": "real-time-monitoring", "title": "Real-Time Monitoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time monitoring is the continuous collection, processing, and presentation of operational telemetry \u2014 encompassing metrics, logs, and traces \u2014 with sufficiently low latency that resulting insights can drive immediate human decisions or automated responses without batch delay. It integrates instrumentation agents, stream-processing pipelines, and visualisation dashboards to maintain a live, actionable picture of system health across software, hardware, network, and physical domains. The discipline spans IT operations, industrial control, MLOps observability, and smart infrastructure, where the defining criterion is actionability: data must arrive within a window that permits meaningful intervention before a fault, breach, or degradation cascades. Real-time monitoring is architecturally distinct from batch analytics in its emphasis on bounded latency, stateful windowing, and continuous alerting over persistent connections.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-monitoring", "labels": [ "Real-Time Monitoring" ], "is_subclass_of": [ "Real-time Processing" ], "wikilinks": [] }, { "id": "real-time-networking", "title": "Real-Time Networking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time networking is network infrastructure and protocol design oriented around minimising latency and jitter so that data reaches recipients within tight time budgets, as required by interactive applications such as social VR and telecollaboration. It typically combines low-latency transport protocols, regional server placement and prioritised message delivery to keep round-trip times within the tens of milliseconds needed for a sense of co-presence. Metaverse platforms depend on real-time networking to synchronise avatar movement, voice and shared object state across participants.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:real-time-networking", "labels": [ "Real-Time Networking" ], "is_subclass_of": [ "Real-Time Communication" ], "wikilinks": [] }, { "id": "real-time-operating-system", "title": "Real-Time Operating System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Real-Time Operating System (RTOS) is an operating system designed to service computational tasks within guaranteed, bounded time constraints, providing temporal determinism that makes it the foundational software substrate for safety-critical and time-sensitive embedded applications. RTOSes implement priority-based preemptive scheduling, bounded interrupt latency, and synchronisation primitives (semaphores, mutexes, message queues) that ensure predictable worst-case task-switching behaviour irrespective of system load. They are classified as hard real-time \u2014 where any deadline miss constitutes a system failure \u2014 or soft real-time, where occasional misses cause graceful degradation rather than catastrophic failure. Deployed in medical devices, industrial controllers, automotive electronics, avionics, and robotics, RTOSes underpin every domain in which a missed computational deadline can cause physical harm, financial loss, or mission failure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:real-time-operating-system", "labels": [ "Real-Time Operating System", "Real Time Operating System" ], "is_subclass_of": [ "Operating System" ], "wikilinks": [] }, { "id": "real-time-prediction", "title": "Real-Time Prediction", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Real-Time Prediction is the generation of a model's output within a latency budget tight enough to inform an immediate decision, typically single-digit to low double-digit milliseconds, as opposed to batch inference computed ahead of need. It requires a serving infrastructure optimised for low-latency, high-throughput requests rather than raw computational efficiency alone. Applications include fraud detection, recommendation, and real-time bidding, where the value of a prediction decays rapidly with delay.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:real-time-prediction", "labels": [ "Real-Time Prediction" ], "is_subclass_of": [ "Model Serving" ], "wikilinks": [] }, { "id": "real-time-rendering-engine", "title": "Real-Time Rendering Engine", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A real-time rendering engine generates images from 3D scene descriptions fast enough to sustain interactive frame rates, typically 30 to 120+ frames per second. It coordinates GPU rasterisation or ray tracing, shading, lighting, and post-processing within a fixed per-frame time budget. It is the visual core of game engines, virtual production, and immersive metaverse experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-rendering-engine", "labels": [ "Real-Time Rendering Engine" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "real-time-rendering-pipeline", "title": "Real-Time Rendering Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Sequence of GPU processes converting 3D scene data into visual frames at interactive rates (typically 30-120+ FPS).", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:real-time-rendering-pipeline", "labels": [ "Real-Time Rendering Pipeline" ], "is_subclass_of": [ "Display and Rendering", "Render Pipeline" ], "wikilinks": [ "3D Models", "Dynamic Lighting", "Fragment Shading", "Frame Buffer Operations", "Geometry Processing", "GPU Driver", "Graphics Processing Unit", "Graphics Rendering System", "Interactive 3D Graphics", "ISO/IEC 23090-3 (MPEG-I)", "Memory Management", "Real-Time Visualization", "Shaders", "Textures", "Vertex Processing", "ComputeLayer", "CreativeMediaDomain", "Game Engine", "Graphics API", "Immersive Experiences" ] }, { "id": "real-time-rendering", "title": "Real-Time Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time rendering is the sub-field of computer graphics focused on producing and analyzing images at interactive frame rates, typically using a graphics processing unit (GPU) to transform 3D scene data into 2D display output within milliseconds. The graphics pipeline processes geometry, applies textures and lighting, and rasterizes the final image fast enough to support interactive applications such as video games, simulations, and virtual reality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:real-time-rendering", "labels": [ "Real-Time Rendering", "Real-Time 3D Rendering", "Real-time Background Rendering", "Real-time Rendering" ], "is_subclass_of": [ "Three Dimensional Graphics" ], "wikilinks": [ "Interactive Visualization", "Video Games", "Augmented Reality", "CreativeMediaDomain", "ETSI_Domain_Immersive_Experiences", "Technology Domain", "Virtual Reality" ] }, { "id": "real-time-settlement", "title": "Real-Time Settlement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time settlement is the immediate, final transfer of value between parties at the moment of transaction, eliminating the delay and counterparty risk of deferred net settlement cycles. In payments and securities it collapses the gap between trade and final exchange of funds. Distributed ledgers and central bank digital currencies enable atomic, near-instant settlement around the clock.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:real-time-settlement", "labels": [ "Real-Time Settlement", "24/7 Settlement", "Fast Settlement" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "real-time-signal-processing", "title": "Real-Time Signal Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time signal processing is the analysis and transformation of a continuous or sampled signal within strict timing constraints, so that output is produced before the deadline required by the consuming system. It is implemented on dedicated hardware such as FPGAs and DSPs, which offer deterministic, low-latency execution unattainable on general-purpose operating systems. It is a prerequisite for haptic feedback and other interactive systems where perceptible delay between input and response degrades usability.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:real-time-signal-processing", "labels": [ "Real-Time Signal Processing" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "real-time-streaming", "title": "Real-Time Streaming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time streaming is the continuous delivery of data, audio, or video as an ongoing flow consumed incrementally with minimal latency rather than downloaded in full beforehand. It uses low-latency transport protocols and buffering strategies to balance smoothness against delay. It underpins live media, event streams, and interactive low-latency experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-streaming", "labels": [ "Real-Time Streaming", "Live Data Streaming", "Real-Time Data Streaming", "Real-Time Media Streaming" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "real-time-synchronisation", "title": "Real-Time Synchronisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Real-time synchronisation is the continuous propagation of state changes across distributed participants so that all observers converge on a consistent, up-to-date view with minimal delay. It combines low-latency transport, conflict resolution, and clock or causal ordering to keep replicas aligned. It is essential for collaborative tools, multiplayer environments, and digital-twin mirroring.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-synchronisation", "labels": [ "Real-Time Synchronisation", "Real-time Synchronization" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "real-time-vfx", "title": "Real-Time VFX", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Real-time VFX is the generation of visual effects such as particles, fluids, fire, and dynamic simulations rendered live at interactive frame rates rather than pre-baked offline. It leverages GPU compute, particle systems, and procedural techniques to produce reactive effects within a game or virtual-production frame budget. It enables interactive spectacle in games, live events, and metaverse experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:real-time-vfx", "labels": [ "Real-Time VFX" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "real-world-asset-tokenisation", "title": "real-world asset tokenisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Real-world asset tokenisation (RWA) is the process of issuing blockchain-based digital tokens that represent legally enforceable ownership or economic claims over physical or financial assets \u2014 including real estate, government bonds, private credit, commodities, and infrastructure. Ownership rights, transfer restrictions, and distribution logic are encoded in smart contracts, enabling fractional ownership and 24/7 secondary-market liquidity. The process requires a legal wrapper such as a special purpose vehicle or trust, oracle infrastructure to price off-chain assets on-chain, and compliance with applicable securities regulation in each issuing jurisdiction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:real-world-asset-tokenisation", "labels": [ "Real-World Asset Tokenisation", "RWA Tokenisation", "Real World Asset Tokenisation", "Real World Asset Tokenization", "Real-World Asset Tokenization" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "real-time-kinematic-positioning", "title": "Real-time Kinematic Positioning", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:real-time-kinematic-positioning", "labels": [ "Real-time Kinematic Positioning" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "real-time-processing", "title": "Real-time Processing", "domain": "data", "domain_name": "Data", "definition": "Real-time Processing is a computational model where data is processed immediately upon arrival or generation, with minimal latency between input and output. Systems respond to events within strict time constraints (typically microseconds to seconds), enabling immediate decision-making through continuous stream-oriented processing rather than deferred batch operations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:real-time-processing", "labels": [ "Real-time Processing", "Real-Time Processing" ], "is_subclass_of": [ "Data Processing", "Artificial Intelligence" ], "wikilinks": [ "Anomaly Detection", "Autonomous Vehicles", "Stream Processing", "Artificial Intelligence", "Batch Processing", "Data Processing", "Digital Twin", "Edge Computing", "Event-Driven Architecture", "Predictive Maintenance" ] }, { "id": "real-time-transcription", "title": "real-time transcription", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Real-time transcription is the automated conversion of streaming audio to text with latency sufficiently low (typically below 500 ms word latency) to support synchronous human use cases such as live captioning, voice-controlled interfaces, meeting assistance, and broadcast subtitling. It requires streaming automatic speech recognition (ASR) architectures that produce partial and final hypotheses incrementally as audio frames arrive, rather than processing complete utterances offline. Modern systems combine acoustic models (typically based on conformer or whisper-encoder architectures), language models for hypothesis rescoring, and punctuation/formatting post-processors to produce readable output in real time.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:real-time-transcription", "labels": [ "Real-time Transcription", "Automatic Transcription" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "real-time-translation", "title": "Real-time Translation", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Real-time translation is the automatic conversion of spoken or written language content into one or more target languages with latency low enough to sustain live human communication, defined operationally as end-to-end processing delay below milliseconds for speech-to-speech pipelines and below 5...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:real-time-translation", "labels": [ "Real-time Translation" ], "is_subclass_of": [ "Communication Technology", "Machine Translation", "Natural Language Processing", "Distributed Collaboration Enablers", "Accessibility Tools", "Speech Technology", "Communication Infrastructure" ], "wikilinks": [ "Accent Adaptation", "AccessibilityDomain", "Accessibility Tools", "Accessible Communication", "Alan Turing Institute", "Audio Buffering", "Audio Capture", "Automatic Speech Recognition", "Average Lagging Metric", "Batch Machine Translation", "Beam Search Decoding", "BLEU Score", "BLEURT", "CCAligned Dataset", "chrF", "Code-switching", "COMET", "COMET Metric", "Communication Infrastructure", "Conference Interpreting" ] }, { "id": "reality-capture-system", "title": "Reality Capture System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Physical hardware system comprising 3D scanners, LIDAR sensors, photogrammetry cameras, and associated equipment for acquiring spatial and visual data from real-world environments to create digital representations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:reality-capture-system", "labels": [ "Reality Capture System" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "3D Model Generation", "3D Scanner", "Depth Sensor", "Environment Reconstruction", "ETSI GR ARF 010", "ISO/IEC 17820", "LIDAR Sensor", "Photogrammetry Camera", "Point Cloud Processing", "Point Cloud Processor", "Tracking System", "Visual Representation", "CreativeMediaDomain", "Data Processing Hardware", "Digital Twin Creation", "Digital Twin Creation Pipeline", "Motion Capture Rig", "PhysicalLayer", "Spatial Calibration", "Storage Infrastructure" ] }, { "id": "reality-capture-workflow", "title": "Reality Capture Workflow", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A reality capture workflow is the end-to-end pipeline that converts real-world objects, people, or environments into digital 3D assets through sensing, alignment, reconstruction, and optimisation stages. It typically chains scanning techniques such as photogrammetry, LiDAR, or volumetric capture with mesh generation, texturing, and retopology steps. The workflow matters because asset fidelity and downstream usability depend on disciplined sequencing of capture, processing, and cleanup.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reality-capture-workflow", "labels": [ "Reality Capture Workflow" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "reality-capture", "title": "Reality Capture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Reality capture is the process of digitising the physical world into accurate, measurable 3D representations using sensors such as LiDAR scanners, photogrammetric camera arrays, and depth cameras, producing point clouds, textured meshes, or digital twin models. It encompasses the full workflow from on-site data acquisition to processing, georeferencing, and delivery of survey-grade or visualisation-grade 3D assets. Reality capture is foundational to construction, heritage preservation, urban digital twins, and spatial computing content pipelines.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:reality-capture", "labels": [ "Reality Capture", "RealityCapture" ], "is_subclass_of": [ "ETSI_Domain_Reality_Capture" ], "wikilinks": [] }, { "id": "reality-eth-oracle", "title": "Reality ETH Oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Reality.eth is an Ethereum-based crowdsourced oracle that resolves arbitrary questions by escalating bonds, allowing anyone to post or challenge answers until an economically final result is reached. It feeds resolved outcomes into smart contracts and is widely used to settle governance proposals and prediction-market events. The escalation-game design lets it answer subjective or off-chain questions that price-feed oracles cannot.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reality-eth-oracle", "labels": [ "Reality ETH Oracle", "Reality.eth Oracle Integration" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "reality-virtuality-continuum", "title": "Reality Virtuality Continuum", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A theoretical framework introduced by Milgram and Kishino describing the continuous spectrum between purely physical reality and fully virtual environments, with augmented reality and augmented virtuality as intermediate mixed reality states along this scale.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:reality-virtuality-continuum", "labels": [ "Reality Virtuality Continuum", "Milgram-Kishino Continuum", "Reality-Virtuality Continuum" ], "is_subclass_of": [ "Platform and Environment", "XR Framework" ], "wikilinks": [ "Cross Reality Experiences", "metaverse", "XR Framework" ] }, { "id": "reality-eth", "title": "Reality.eth", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Reality.eth is an Ethereum-based oracle that answers questions by crowdsourcing responses with an escalating bond mechanism, allowing economic dispute resolution. It is used to bring real-world facts on chain.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:reality-eth", "labels": [ "Reality.eth" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Smart Contract", "Ethereum", "SafeSnap", "DAO Governance" ] }, { "id": "realtime-collaboration", "title": "Realtime Collaboration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Realtime Collaboration is the capability for multiple geographically distributed users to co-create, co-edit, and synchronise shared digital artefacts\u2014documents, 3D scenes, code, or virtual environments\u2014with sub-second latency, such that all participants observe consistent state simultaneously. Technically, it requires low-latency networking, conflict-resolution mechanisms such as CRDTs or operational transforms, and state-synchronisation protocols to reconcile concurrent edits without data loss.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:realtime-collaboration", "labels": [ "Realtime Collaboration" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "realtime-communication", "title": "Realtime Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Realtime communication denotes the class of systems, protocols, and architectures that deliver synchronous, low-latency exchange of audio, video, text, and arbitrary data between two or more participants such that end-to-end delay remains below perceptual thresholds \u2014 typically under 150 ms one-way for voice and under 50 ms for interactive haptics or gaming. It encompasses technologies such as WebRTC, VoIP, SIP, XMPP, and WebSocket-based signalling, unified under the common requirement that transport, codec, and signalling planes cooperate to minimise jitter and packet loss. Realtime communication differs from asynchronous messaging in that session state must be continuously negotiated and media flows must be sustained for the interaction to remain coherent. It constitutes foundational infrastructure for distributed collaboration, virtual presence, telemedicine, and immersive social environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:realtime-communication", "labels": [ "Realtime Communication" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "reanalysis", "title": "Reanalysis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reanalysis", "labels": [ "Reanalysis", "Reanalyze" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reanalyzed-data-state", "title": "Reanalyzed Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reanalyzed-data-state", "labels": [ "Reanalyzed Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reasoning-engine", "title": "Reasoning Engine", "domain": "ai", "domain_name": "Ai", "definition": "A reasoning engine is a software component that derives new conclusions from a body of knowledge by applying logical inference rules, probabilistic methods, or learned heuristics. Classical reasoning engines operate over symbolic knowledge bases using forward or backward chaining, description-logic subsumption, or constraint solving, while modern neuro-symbolic and LLM-based engines combine learned language representations with structured tool use and search. Reasoning engines power expert systems, semantic-web query answering, automated planning, and multi-step problem solving in agentic AI systems. In the DreamLab mesh the reasoning engine is Whelk, an OWL 2 EL classifier that derives entailments and rejects contradictions before they enter the shared knowledge graph.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:reasoning-engine", "labels": [ "Reasoning Engine" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "reasoning-models", "title": "Reasoning Models", "domain": "ai", "domain_name": "Ai", "definition": "Reasoning models are large language models trained or post-trained to generate extended intermediate chains of thought before producing a final answer, trading inference-time compute for higher accuracy on complex tasks. They are typically optimised with reinforcement learning on verifiable problems in mathematics, coding, and logic. By spending more tokens deliberating, they outperform standard models on multi-step problems where single-pass generation fails.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reasoning-models", "labels": [ "Reasoning Models" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "reasoning-trace", "title": "Reasoning Trace", "domain": "ai", "domain_name": "Ai", "definition": "A reasoning trace is the recorded sequence of intermediate thoughts, tool calls, and decisions a model produces while solving a task, distinct from its final output. It serves as both a working scratchpad that improves accuracy and an audit artefact for inspecting how an agent reached a conclusion. Traces underpin debugging, evaluation, and self-correction in agentic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reasoning-trace", "labels": [ "Reasoning Trace" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "reasoning", "title": "Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Reasoning in artificial intelligence encompasses the computational processes by which systems derive conclusions, formulate plans, solve problems, and generate explanations from knowledge, data, and prior context. On this graph, unqualified reasoning names the symbolic form: a reasoner deriving entailments that necessarily follow from the ontology's axioms. LLM chain-of-thought is always qualified as such.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:reasoning", "labels": [ "Reasoning", "AI Reasoning", "Arithmetic Reasoning", "Deliberate Reasoning" ], "is_subclass_of": [ "AI Technique", "Artificial General Intelligence", "Cognitive AI", "Machine Learning Discipline", "Knowledge Representation", "Natural Language Processing", "Formal Verification" ], "wikilinks": [ "AIME Benchmark", "AlgorithmLayer", "ARC-AGI", "Beam Search", "Benchmarks", "Chain-of-Thought Prompting", "CognitiveScienceDomain", "Extended Thinking", "Formal Logic", "FormalMethodsDomain", "Formal Verification", "GPQA Benchmark", "Hallucination", "Information Theory", "Knowledge Graphs", "Lean", "Mathematical Reasoning", "ModelLayer", "Neuro-Symbolic AI", "Outcome Reward Model" ] }, { "id": "recall", "title": "Recall", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A classification performance metric representing the proportion of actual positive instances that an artificial intelligence model correctly identifies, calculated as the ratio of true positives to all actual positives (true positives plus false negatives), measuring the model's completeness in detecting positive cases, particularly critical in applications where missing positive instances (false negatives) carries significant cost or consequences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:recall", "labels": [ "Recall" ], "is_subclass_of": [ "Model Performance" ], "wikilinks": [ "False Negative", "Precision-Recall Curve", "Sensitivity", "Specificity", "True Positive Rate", "Accuracy", "Confusion Matrix", "F1 Score", "MetaverseDomain", "Model Performance", "Precision", "ROC Curve" ] }, { "id": "receiver-autonomous-integrity-monitoring", "title": "Receiver Autonomous Integrity Monitoring", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:receiver-autonomous-integrity-monitoring", "labels": [ "Receiver Autonomous Integrity Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "receiver-clock-bias", "title": "Receiver Clock Bias", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:receiver-clock-bias", "labels": [ "Receiver Clock Bias" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reciprocal-rank-fusion", "title": "Reciprocal Rank Fusion", "domain": "ai", "domain_name": "Ai", "definition": "Reciprocal Rank Fusion (RRF) is a rank-aggregation method that combines multiple ranked result lists by summing the reciprocal of each document's rank across lists, weighted by a small smoothing constant. It requires no score calibration between systems, making it ideal for merging lexical and vector retrieval results in hybrid search. Its robustness and parameter simplicity have made it a default fusion technique in modern retrieval pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reciprocal-rank-fusion", "labels": [ "Reciprocal Rank Fusion" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "recommendation-engine", "title": "Recommendation Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A recommendation engine is a system that filters, ranks, and surfaces content or items most relevant to an individual user by analysing preferences, behaviour history, and contextual signals. It employs collaborative filtering, content-based filtering, or hybrid deep-learning approaches to personalise discovery at scale across e-commerce, media, and metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:recommendation-engine", "labels": [ "Recommendation Engine" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "recommendation-system", "title": "Recommendation System", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A recommendation system is an information filtering infrastructure that predicts and surfaces items, content, or actions likely to be of interest to a specific user, based on behavioural history, explicit preferences, item features, or combinations thereof. It encompasses collaborative filtering approaches that exploit user-item interaction patterns, content-based methods that match item attributes to user profiles, and hybrid models that combine multiple signals. Modern recommendation systems employ deep learning, graph neural networks, and large language models to capture complex preference patterns at scale. They are commercially critical infrastructure in e-commerce, streaming media, social networks, and digital advertising.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:recommendation-system", "labels": [ "Recommendation System", "AI Recommendation System", "Recommendation Algorithms", "RecommendationSystem" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "recommendation-systems", "title": "Recommendation Systems", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Recommendation Systems are information filtering systems that predict a user's preference for items\u2014products, content, services\u2014and surface the most relevant items from a large catalogue. They are categorised into collaborative filtering (leveraging the behaviour of similar users), content-based filtering (matching item attributes to user profiles), and hybrid approaches that combine both. Modern large-scale recommendation systems employ deep learning architectures, embedding models, and two-tower neural networks trained on implicit feedback signals such as clicks, watch-time, and purchases. Recommendation systems are among the highest-impact machine learning applications in commercial technology, driving substantial fractions of revenue at platforms such as Netflix, Amazon, YouTube, and Spotify, while also raising significant concerns about filter bubbles, engagement maximisation harms, and algorithmic amplification of misinformation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:recommendation-systems", "labels": [ "Recommendation Systems" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Application" ], "wikilinks": [] }, { "id": "reconfigurable-computing", "title": "Reconfigurable Computing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Reconfigurable computing is a computing paradigm in which hardware logic itself is reconfigured after fabrication to match the structure of a specific workload, combining the flexibility of software with performance closer to fixed-function hardware. Field-programmable gate arrays are the principal enabling technology, allowing logic blocks and interconnects to be reprogrammed to implement custom pipelines, accelerators or entire processor designs. It sits between general-purpose processors and application-specific integrated circuits, trading some raw throughput for the ability to adapt hardware behaviour after deployment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:reconfigurable-computing", "labels": [ "Reconfigurable Computing" ], "is_subclass_of": [ "Hardware Acceleration" ], "wikilinks": [] }, { "id": "reconstructed-data-state", "title": "Reconstructed Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reconstructed-data-state", "labels": [ "Reconstructed Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reconstruction", "title": "Reconstruction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reconstruction", "labels": [ "Reconstruction", "Reconstruct" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "record-keeping-system", "title": "Record Keeping System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Record Keeping System, in the blockchain domain, is an immutable, append-only ledger in which transactions or state changes are cryptographically hashed, linked via Merkle trees, and validated by a consensus mechanism to provide a tamper-evident audit trail. It replaces or augments traditional centralised databases in applications requiring transparent provenance, non-repudiation, and multi-party data integrity\u2014including asset registries, supply-chain tracking, and compliance logging.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:record-keeping-system", "labels": [ "Record Keeping System", "Record-Keeping System" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "owl:Thing" ] }, { "id": "record-keeping", "title": "Record Keeping", "domain": "governance", "domain_name": "Governance", "definition": "Record keeping is the systematic retention of transaction, customer, and decision records so that activities can be reconstructed and verified by auditors or regulators. In financial compliance it mandates preserving identity documentation, transaction details, and risk assessments for prescribed periods. Robust record keeping is a precondition for meeting anti-money-laundering and travel-rule obligations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:record-keeping", "labels": [ "Record Keeping", "Informal Record Keeping" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "records-management", "title": "Records Management", "domain": "standards", "domain_name": "Standards", "definition": "Records management is the systematic control of an organisation's records throughout their lifecycle \u2014 from creation and capture through classification, storage and use, to retention, archiving and eventual disposal. It applies policies and standards, notably ISO 15489, to ensure that records remain authentic, reliable and accessible for as long as they are required for business, legal and regulatory purposes. Effective records management supports accountability, compliance and efficient information governance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:records-management", "labels": [ "Records Management" ], "is_subclass_of": [ "Information Governance" ], "wikilinks": [] }, { "id": "rectified-flow", "title": "rectified flow", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Rectified Flow is a generative modelling framework that trains a velocity field to transport samples along straight-line trajectories in the ODE sense between a source noise distribution and a target data distribution. By reflow iterations\u2014repeatedly pairing coupled samples and re-training\u2014the learned trajectories become increasingly linear, minimising the number of function evaluations (NFEs) required at inference time. The approach unifies flow matching and score-based diffusion models, offering competitive image and video generation quality with substantially faster sampling compared to standard diffusion schedules.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rectified-flow", "labels": [ "Rectified Flow", "Rectified Flow Training" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "recurrent-neural-network", "title": "Recurrent Neural Network", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural network architecture that processes sequential data by maintaining a hidden state across time steps, allowing information from earlier inputs to influence later outputs. Key variants include Long Short-Term Memory networks and Gated Recurrent Units, which address vanishing gradient problems and underpin sequence modelling tasks such as speech recognition, language modelling, and time-series forecasting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:recurrent-neural-network", "labels": [ "Recurrent Neural Network", "Recurrent Neural Networks" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "activation functions", "Adam optimizer", "AI Knowledge Graph", "AI/ML Knowledge Graph", "AI Summit Manchester", "Anomaly detection", "Anomaly Detection", "Approximation Theory", "ARM Cortex-M", "artificial neural networks", "AstraZeneca", "Attention-augmented RNNs", "attention mechanisms", "Automatic Speech Recognition", "Backpropagation Through Time", "biases", "Bidirectional RNN", "Biomedical Applications", "But what is a neural network? (3Blue1Brown)", "Causal RNN Analysis" ] }, { "id": "recursive-self-improvement", "title": "recursive self-improvement", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Recursive self-improvement (RSI) is a theoretical AI capability in which a system autonomously identifies and implements modifications to its own architecture, training procedure, or objective function that result in increased performance or capability, such that each improvement cycle enables further improvements in a self-amplifying feedback loop. This process is conjectured to be a pathway to artificial superintelligence if left unconstrained, as successive capability doublings could occur faster than human oversight can track. RSI is a central concern in AI safety research because systems capable of recursive self-improvement may rapidly violate designers' assumptions about capability bounds and alignment properties.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:recursive-self-improvement", "labels": [ "Recursive Self-Improvement" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "red-dwarf", "title": "Red Dwarf", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:red-dwarf", "labels": [ "Red Dwarf" ], "is_subclass_of": [ "Main-sequence Star" ], "wikilinks": [] }, { "id": "red-giant", "title": "Red Giant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:red-giant", "labels": [ "Red Giant" ], "is_subclass_of": [ "Star" ], "wikilinks": [] }, { "id": "red-teaming", "title": "Red Teaming", "domain": "security", "domain_name": "Security", "definition": "The practice of deliberately attempting to elicit harmful, biased, or undesired outputs from AI systems to identify vulnerabilities and weaknesses. Red teaming involves structured adversarial testing where human evaluators or automated systems probe for failure modes across security, ethics, and alignment dimensions, informing safety improvements and deployment decisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:red-teaming", "labels": [ "Red Teaming", "AI Red Teaming", "Adversarial Red Teaming", "Red Team Testing", "Red-Teaming" ], "is_subclass_of": [ "Adversarial Testing" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "redemption-mechanism", "title": "Redemption Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A redemption mechanism is the process by which a token holder can exchange a stablecoin or wrapped asset back into the underlying reserve asset it represents, such as fiat currency or collateral. It is the structural counterpart of minting and is central to maintaining a stablecoin's peg: credible redeemability allows arbitrageurs to profit from deviations between market price and par value, pulling the price back to the peg. Redemption mechanisms vary from direct par redemption with the issuer to on-chain burn-and-release contracts.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:redemption-mechanism", "labels": [ "Redemption Mechanism" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [] }, { "id": "redress-mechanism", "title": "Redress Mechanism", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A redress mechanism is a structured process that lets individuals affected by an automated or algorithmic decision contest it, seek an explanation, and obtain correction or remedy when harm or error has occurred. It operationalises accountability in AI governance by combining channels for complaint, human review of contested outcomes, and remediation such as reversal, compensation, or model adjustment. Effective redress requires transparency about how decisions are made, due process for the affected person, and feedback loops that surface systemic problems for harm mitigation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:redress-mechanism", "labels": [ "Redress Mechanism" ], "is_subclass_of": [ "Accountability" ], "wikilinks": [] }, { "id": "redress-procedure", "title": "Redress Procedure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Redress Procedure is a formalised mechanism enabling individuals or groups adversely affected by AI system decisions to challenge those decisions, seek explanations, request human review, and obtain remedies including correction, compensation, or policy changes. Redress procedures embody accountability by ensuring that consequential algorithmic decisions\u2014in employment, credit, housing, criminal justice, and public services\u2014remain contestable rather than treated as unappealable automated verdicts. Effective procedures must be accessible (understandable without legal expertise), timely (preventing irreversible harm), thorough (genuine rather than perfunctory human review), and proportionate, with escalation pathways from internal review through independent ombudsman services to judicial challenge.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:redress-procedure", "labels": [ "Redress Procedure", "Consumer Redress", "Redress", "Redress Mechanisms", "redress" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Governance Principle" ], "wikilinks": [ "AI Governance Principle", "Algorithmic Accountability Act", "Alternative Dispute Resolution", "Complaint Mechanism", "Decision Appeal", "Explanation Provision", "GDPR Article 22", "Human Review Process", "ISO/IEC 42001", "Judicial Review", "OECD AI Principles", "Ombudsman Service", "Remedy Implementation", "AIEthicsDomain", "ConceptualLayer", "EU AI Act" ] }, { "id": "redshift", "title": "Redshift", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:redshift", "labels": [ "Redshift" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "redundancy", "title": "Redundancy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Redundancy is the deliberate duplication of components, data, or pathways in a system so that the failure of any single element does not cause overall failure. It is a foundational technique for fault tolerance, implemented through replicas, standby nodes, mirrored storage, and multiple network routes. By eliminating single points of failure, redundancy raises availability at the cost of additional resources.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:redundancy", "labels": [ "Redundancy", "Network Redundancy" ], "is_subclass_of": [ "System Architecture" ], "wikilinks": [] }, { "id": "reed-smith", "title": "Reed Smith", "domain": "governance", "domain_name": "Governance", "definition": "Reed Smith is an international law firm headquartered in the United States with offices across several regions. It provides legal services to corporate and institutional clients.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:reed-smith", "labels": [ "Reed Smith" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Governance", "owl:Thing", "https://www.reedsmith.com", "https://www.reedsmith.com/en/about" ] }, { "id": "reed-solomon-codes", "title": "Reed-Solomon Codes", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Reed-Solomon codes are a class of non-binary, cyclic, block error-correcting codes defined over finite fields (Galois fields), capable of correcting both erasures and symbol errors with provably optimal efficiency at the Singleton bound. Introduced by Irving Reed and Gustave Solomon in 1960, they treat data blocks as polynomials over a finite field and encode them by evaluating the polynomial at multiple distinct points, allowing the original polynomial to be reconstructed from any sufficient subset of evaluation points. Reed-Solomon codes underpin data reliability in storage media (CDs, DVDs, RAID), satellite communications, QR codes, and are foundational to erasure-coded distributed storage and modern polynomial commitment schemes used in zero-knowledge proofs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reed-solomon-codes", "labels": [ "Reed-Solomon Codes", "Reed-Solomon Code", "Reed-Solomon Coding" ], "is_subclass_of": [ "Error Correction" ], "wikilinks": [] }, { "id": "reed-solomon-erasure-coding", "title": "Reed-Solomon Erasure Coding", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Reed-Solomon erasure coding is a family of error-correcting codes that transform k data symbols into n encoded symbols such that any k of the n suffice to reconstruct the original data, tolerating up to n minus k erasures. Operating over finite (Galois) fields, it provides maximum-distance-separable efficiency, meaning no scheme can recover from more erasures for the same redundancy. It is foundational to RAID storage, optical media, QR codes, satellite communication, and distributed-storage and blockchain data-availability systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:reed-solomon-erasure-coding", "labels": [ "Reed-Solomon Erasure Coding" ], "is_subclass_of": [ "Erasure Coding" ], "wikilinks": [] }, { "id": "reentrancy-attack", "title": "Reentrancy Attack", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A reentrancy attack is a smart-contract exploit in which a malicious contract repeatedly re-enters a vulnerable function before its state is updated, draining funds or corrupting state. It arises when a contract makes an external call before completing its own bookkeeping, allowing the callee to recursively invoke the caller. The class of bug was made notorious by the DAO incident and is mitigated by the checks-effects-interactions pattern and reentrancy guards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reentrancy-attack", "labels": [ "Reentrancy Attack" ], "is_subclass_of": [ "Blockchain Security" ], "wikilinks": [] }, { "id": "refactoring", "title": "Refactoring", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Refactoring is the disciplined practice of restructuring existing source code to improve its internal design\u2014readability, modularity, and maintainability\u2014without changing its external behaviour. It proceeds through small, behaviour-preserving transformations, typically guarded by automated tests that confirm functionality remains intact. Refactoring is a core technique for managing technical debt and sustaining the long-term evolvability of software systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:refactoring", "labels": [ "Refactoring" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "reference-architecture", "title": "Reference Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A standardized architectural template that provides proven structural frameworks and design patterns for building scalable, resilient enterprise applications, establishing shared vocabulary and best practices across development teams.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:reference-architecture", "labels": [ "Reference Architecture", "AR Reference Architecture", "MEC Reference Architecture", "XR Reference Architecture" ], "is_subclass_of": [ "Platform and Environment", "Software Architecture" ], "wikilinks": [ "Scalable Systems", "metaverse", "Software Architecture" ] }, { "id": "reference-ellipsoid-flattening", "title": "Reference Ellipsoid Flattening", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reference-ellipsoid-flattening", "labels": [ "Reference Ellipsoid Flattening" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reference-ellipsoid", "title": "Reference Ellipsoid", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reference-ellipsoid", "labels": [ "Reference Ellipsoid" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reference-frames", "title": "Reference Frames", "domain": "robotics", "domain_name": "Robotics", "definition": "A reference frame is a coordinate system, anchored to a chosen origin and orientation, against which positions, velocities, and orientations are measured. In robotics each link, sensor, and the world itself has its own frame, and motion is described by transformations between them. Correct frame definitions are essential for sensor fusion, kinematics, and consistent spatial reasoning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reference-frames", "labels": [ "Reference Frames", "Reference Frame" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "reference-implementation", "title": "Reference Implementation", "domain": "standards", "domain_name": "Standards", "definition": "A reference implementation is an authoritative, openly available realisation of a specification or standard that demonstrates how it is meant to be built and used. It serves as a concrete benchmark against which other implementations can be compared for correctness and conformance. By turning a written specification into working software, it reduces ambiguity and accelerates adoption of the standard.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reference-implementation", "labels": [ "Reference Implementation" ], "is_subclass_of": [ "Open Standards" ], "wikilinks": [] }, { "id": "reference-measurement", "title": "Reference Measurement", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reference-measurement", "labels": [ "Reference Measurement" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "reference-model", "title": "Reference Model", "domain": "robotics", "domain_name": "Robotics", "definition": "A reference model is an abstract, authoritative description of the desired behaviour or structure of a system, used as the standard against which an actual implementation is measured or controlled. In control theory it specifies the ideal closed-loop response that a model-reference adaptive controller drives the plant to follow. In evaluation it serves as the baseline whose outputs define the expected target.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reference-model", "labels": [ "Reference Model", "OAIS Reference Model" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "reference-signal", "title": "Reference Signal", "domain": "robotics", "domain_name": "Robotics", "definition": "A reference signal is the commanded target value, or setpoint trajectory, that a control system attempts to make its output follow over time. The controller computes the error between the reference and the measured output and acts to drive that error to zero. The shape and feasibility of the reference signal directly determine achievable tracking performance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reference-signal", "labels": [ "Reference Signal" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "reference-standard", "title": "Reference Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Reference Standard is a technology infrastructure concept and a type of infrastructure.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:reference-standard", "labels": [ "Reference Standard" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "reflection-loop", "title": "Reflection Loop", "domain": "ai", "domain_name": "Ai", "definition": "A reflection loop is an agentic control pattern in which a model critiques its own intermediate output, identifies errors or gaps, and revises before continuing or finalising. By iterating between generation and self-evaluation, the agent improves quality on tasks where a single pass is unreliable. It is a building block of self-correcting LLM agents and tool-using workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reflection-loop", "labels": [ "Reflection Loop" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "reflection-probe", "title": "Reflection Probe", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Reflection Probe is a spatial data structure placed in a 3D scene that captures a 360-degree image of its surroundings into a cubemap texture, which is then sampled by shaders to produce environment reflections on nearby surfaces. Probes are fundamental to physically based rendering pipelines, providing local ambient lighting and specular reflections without the cost of real-time ray tracing. They are placed by artists or computed automatically at bake time and can be blended across zones to produce continuous, plausible illumination.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:reflection-probe", "labels": [ "Reflection Probe" ], "is_subclass_of": [ "Display and Rendering", "Render Pipeline" ], "wikilinks": [] }, { "id": "reflection", "title": "Reflection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An agent pattern in which a model examines its own prior output, judges it against the goal and any available evidence, and then revises the work in a further pass. Reflection turns generation into a loop rather than a single shot: the agent produces a draft, critiques that draft \u2014 spotting errors, gaps, unmet constraints, or weak reasoning \u2014 and feeds the critique back as input to an improved attempt. It is distinguished from ordinary multi-step prompting by the fact that the intermediate judgement is about the agent's own work, making the agent both author and reviewer within the same task.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:reflection", "labels": [ "Reflection" ], "is_subclass_of": [ "Agentic Workflow", "AgenticWorkflow" ], "wikilinks": [ "AgenticWorkflow", "ChainOfThought", "SelfConsistency", "GoldenSet" ] }, { "id": "reflexion", "title": "Reflexion", "domain": "ai", "domain_name": "Ai", "definition": "Reflexion is a framework for language-model agents that converts feedback from failed attempts into reflective verbal self-critiques stored in an episodic memory, which condition subsequent attempts. Rather than updating model weights, it reinforces behaviour through natural-language reflections, enabling rapid trial-and-error learning. It improves agent performance on decision-making, reasoning, and coding benchmarks across repeated episodes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:reflexion", "labels": [ "Reflexion", "Reflexion Pattern" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "reformatted-data-state", "title": "Reformatted Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:reformatted-data-state", "labels": [ "Reformatted Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "refresh-token", "title": "Refresh Token", "domain": "security", "domain_name": "Security", "definition": "A refresh token is a long-lived credential issued alongside a short-lived access token, used to obtain new access tokens without prompting the user to re-authenticate. By keeping access tokens short-lived and exchanging the refresh token at the authorisation server, systems limit the damage of a leaked access token while preserving a smooth user session. Refresh tokens are sensitive and typically bound, rotated, and revocable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:refresh-token", "labels": [ "Refresh Token" ], "is_subclass_of": [ "OAuth 2.0" ], "wikilinks": [] }, { "id": "regtech", "title": "RegTech", "domain": "governance", "domain_name": "Governance", "definition": "RegTech (regulatory technology) is the application of technologies such as data analytics, machine learning, cloud computing and automation to help regulated firms meet compliance obligations more efficiently and accurately. It streamlines tasks including identity verification, transaction monitoring, regulatory reporting and risk surveillance. RegTech reduces the manual cost of compliance while improving the timeliness and consistency of regulatory outcomes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:regtech", "labels": [ "RegTech" ], "is_subclass_of": [ "Regulatory Technology" ], "wikilinks": [] }, { "id": "regenerative-finance", "title": "Regenerative Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Regenerative finance (ReFi) is a movement that uses blockchain-based financial primitives to fund and incentivise ecological and social regeneration, such as carbon removal, biodiversity, and public goods. It tokenises natural assets and routes capital through mechanisms like retroactive funding and on-chain carbon markets. ReFi reframes DeFi tooling toward restoring rather than merely extracting value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:regenerative-finance", "labels": [ "Regenerative Finance" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "regenerative-payload", "title": "Regenerative Payload", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:regenerative-payload", "labels": [ "Regenerative Payload" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "regolith", "title": "Regolith", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:regolith", "labels": [ "Regolith" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "regression-testing", "title": "Regression Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Regression testing is the practice of re-executing previously passing tests after a code change to confirm that existing behaviour has not been broken. It guards against regressions introduced by new features, bug fixes or refactoring by maintaining a suite of repeatable checks that are run automatically as part of integration and delivery pipelines. The discipline trades upfront test authoring and maintenance for sustained confidence in evolving systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:regression-testing", "labels": [ "Regression Testing" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "regression", "title": "Regression", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Regression is a class of statistical and machine learning methods that model the relationship between input variables and a continuous output. It is used for prediction and for estimating effects.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:regression", "labels": [ "Regression", "Least Squares Regression", "Least-Squares Regression", "Linear Regression", "Performance Regression Detection", "Regression Analysis" ], "is_subclass_of": [ "Supervised Learning", "AI Technique" ], "wikilinks": [ "Statistics", "Machine Learning", "Deep Learning", "Supervised Learning", "https://en.wikipedia.org/wiki/Regression_analysis", "https://scikit-learn.org/stable/modules/linear_model.html" ] }, { "id": "regridded-data-state", "title": "Regridded Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:regridded-data-state", "labels": [ "Regridded Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "regridding", "title": "Regridding", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:regridding", "labels": [ "Regridding", "Regrid" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "regular-expression", "title": "Regular Expression", "domain": "standards", "domain_name": "Standards", "definition": "A regular expression is a formal notation for describing sets of strings using a concise pattern language built from literals, character classes, quantifiers, and grouping. Regular expressions describe exactly the class of regular languages and are typically implemented by compiling the pattern into a finite-state machine for efficient matching. They are a foundational tool for searching, validating, and transforming text.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:regular-expression", "labels": [ "Regular Expression" ], "is_subclass_of": [ "Formal Language" ], "wikilinks": [] }, { "id": "regularisation", "title": "Regularisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A family of techniques that constrain or penalise model complexity during training to prevent overfitting and improve generalisation to unseen data. Common methods include L1 (Lasso) and L2 (Ridge) weight penalties, dropout, early stopping, and data augmentation, each discouraging the model from memorising noise at the expense of learning underlying patterns.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:regularisation", "labels": [ "Regularisation", "Regularization" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "regulation-d", "title": "Regulation D", "domain": "finance", "domain_name": "Finance", "definition": "A United States Securities and Exchange Commission regulation providing exemptions from the registration requirements of the Securities Act of 1933, allowing companies to raise capital through private placements to accredited investors without a full public offering. Its Rules 504 and 506 are the dominant legal pathway for security token offerings and early-stage fundraising, trading reduced disclosure burdens for restrictions on general solicitation and resale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:regulation-d", "labels": [ "Regulation D" ], "is_subclass_of": [ "Securities Regulation" ], "wikilinks": [ "Securities Regulation", "Security Token", "Know Your Customer", "Crowdfunding" ] }, { "id": "regulation", "title": "Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Regulation is the systematic use of authoritative rules, standards, and enforcement mechanisms by governments, inter-governmental bodies, or designated agencies to direct, constrain, or oversee the conduct of individuals, organisations, and systems within a defined domain. It encompasses the full lifecycle from rule-making and publication through monitoring, inspection, and sanctioning of non-compliance. Regulation is distinct from voluntary standards in that it carries legal force and can result in penalties, licence revocation, or other coercive consequences for non-conformance. In technical domains such as AI, data systems, and critical infrastructure, regulation increasingly intersects with algorithmic accountability, risk classification, and mandatory transparency obligations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:regulation", "labels": [ "Regulation", "Biosafety Regulation", "Machinery Regulation", "Regulation Best Interest", "Statutory Regulation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Regulatory Frameworks", "Compliance", "Governance", "https://www.oecd.org/gov/regulatory-policy/", "https://en.wikipedia.org/wiki/Regulation" ] }, { "id": "regulatory-approval", "title": "Regulatory Approval", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory approval is the formal authorisation by a regulator that permits an entity, product or activity to proceed under defined conditions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:regulatory-approval", "labels": [ "Regulatory Approval", "Drug Approval" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Regulatory Requirements", "Compliance", "Regulatory Frameworks", "Regulation", "https://www.oecd.org/gov/regulatory-policy/", "https://en.wikipedia.org/wiki/Regulatory_agency" ] }, { "id": "regulatory-authorisation", "title": "Regulatory Authorisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Regulatory authorisation is the formal permission granted by a competent authority allowing an entity to conduct a regulated activity such as issuing tokens, operating a payment system, or holding client assets. It is obtained through an application demonstrating compliance with capital, governance, and conduct requirements. Without authorisation, issuance and operation of regulated financial instruments are unlawful in most jurisdictions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-authorisation", "labels": [ "Regulatory Authorisation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "regulatory-authority", "title": "Regulatory Authority", "domain": "governance", "domain_name": "Governance", "definition": "A regulatory authority is a government body, statutory agency, or delegated entity empowered by legislation to create, enforce, and adjudicate rules governing a specific industry, market, or domain of public concern. Such bodies possess the legal power to issue binding standards, licenses, and penalties; conduct inspections and investigations; and compel disclosures, all with the aim of protecting consumers, ensuring market integrity, managing systemic risk, or advancing public policy objectives. Regulatory authorities operate at national, supranational, and sub-national levels and are increasingly engaged with digital markets, artificial intelligence, financial technology, and emerging technology sectors.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:regulatory-authority", "labels": [ "Regulatory Authority", "Financial Regulatory Authority", "Regulatory Agency", "Regulatory Body" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "regulatory-capture", "title": "Regulatory Capture", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory capture is a situation in which a regulator advances the interests of the entities it oversees rather than the public interest.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:regulatory-capture", "labels": [ "Regulatory Capture" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Regulatory Frameworks", "Regulation", "https://en.wikipedia.org/wiki/Regulatory_capture", "https://www.oecd.org/gov/regulatory-policy/" ] }, { "id": "regulatory-clarity", "title": "Regulatory Clarity", "domain": "governance", "domain_name": "Governance", "definition": "The condition in which the rules applying to an activity or product are sufficiently clear and predictable for participants to assess their legal obligations with confidence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-clarity", "labels": [ "Regulatory Clarity" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [ "Regulatory Framework", "Transparency", "Compliance", "Financial Regulation" ] }, { "id": "regulatory-compliance", "title": "Regulatory Compliance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Regulatory Compliance in AI contexts refers to the adherence to legal requirements, statutory obligations, and regulatory standards governing the development, deployment, and operation of artificial intelligence systems within specific jurisdictions or sectors. Compliance requires implementing governance structures, conducting impact assessments, maintaining documentation and audit trails, providing transparency, and demonstrating ongoing monitoring for regulatory adherence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:regulatory-compliance", "labels": [ "Regulatory Compliance", "BC-0479-regulatory-compliance", "Full Regulatory Compliance", "Regulatory AI Compliance", "Regulatory Compliance Custody", "Regulatory Ecosystem", "Regulatory Systems", "RegulatoryCompliance" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Governance Principle" ], "wikilinks": [ "AI Governance Principle", "Algorithmic Impact Assessment", "Conformity Assessment", "EU AI Act Compliance", "GDPR", "GDPR Compliance", "Human Oversight Requirement", "IEEE 7000 Model Process", "ISO/IEC 42001", "NIST AI Risk Management Framework", "Sector-Specific Compliance", "Transparency Obligation", "AIEthicsDomain", "Audit Trail", "ConceptualLayer", "EU AI Act" ] }, { "id": "regulatory-conformance", "title": "Regulatory Conformance", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory conformance is the demonstrable state of an organisation's systems and processes meeting the legal and regulatory requirements applicable to its domain. It is evidenced through controls, documentation, and audits that map obligations to operational practice. For AI and data systems, conformance increasingly covers risk classification, transparency, and accountability mandates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-conformance", "labels": [ "Regulatory Conformance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "regulatory-enforcement", "title": "Regulatory Enforcement", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory enforcement is the set of actions through which a competent authority compels compliance with laws and rules and sanctions those who breach them. It encompasses investigation, supervision, market surveillance, and the imposition of remedies such as fines, restitution, licence revocation or criminal referral. In financial markets it deters misconduct, protects consumers and investors, and sustains confidence in the integrity of the system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:regulatory-enforcement", "labels": [ "Regulatory Enforcement" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [] }, { "id": "regulatory-framework", "title": "Regulatory Framework", "domain": "governance", "domain_name": "Governance", "definition": "A regulatory framework for blockchain and cryptoassets comprises the laws, regulations, guidelines, and supervisory structures established by governmental authorities to govern the issuance, trading, and custody of digital assets. These frameworks address consumer protection, market integrity, anti-money laundering compliance, and financial stability whilst balancing innovation enablement with risk mitigation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:regulatory-framework", "labels": [ "Regulatory Framework", "FSB Regulatory Framework", "RegulatoryFramework" ], "is_subclass_of": [ "Legal Framework" ], "wikilinks": [ "Market Integrity", "Blockchain", "EU MiCA Regulation" ] }, { "id": "regulatory-frameworks", "title": "Regulatory Frameworks", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory frameworks are structured, hierarchical systems comprising primary legislation, secondary rules, supervisory authorities, and enforcement mechanisms that collectively govern conduct within a defined sector or jurisdiction. They establish who is subject to regulation, what obligations and prohibitions apply, how compliance is monitored, and what sanctions are available for breach. Frameworks may be rules-based, principles-based, or risk-based in design, and typically span multiple tiers from international treaties and standards through to national law and sector-specific codes of practice. Modern frameworks increasingly address cross-border, technology-mediated activities, creating interoperability requirements between national and supranational regimes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-frameworks", "labels": [ "Regulatory Frameworks" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Regulatory Requirements", "Compliance", "Regulation", "https://www.oecd.org/gov/regulatory-policy/", "https://en.wikipedia.org/wiki/Regulation" ] }, { "id": "regulatory-layer", "title": "Regulatory Layer", "domain": "governance", "domain_name": "Governance", "definition": "The Regulatory Layer is the cross-cutting stratum that represents jurisdiction-specific legal requirements and the authorities that impose them. It sits above the Compliance Layer, supplying the obligations that compliance verifies, and informs governance and institutional structures. It contains regulations, licensing regimes, reporting duties, and the mapping of activities to applicable rules.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:regulatory-layer", "labels": [ "Regulatory Layer" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Compliance Layer", "Institutional Layer", "Governance Layer", "Financial Regulation", "Data Protection Law", "owl:Thing" ] }, { "id": "regulatory-licence", "title": "Regulatory Licence", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A regulatory licence is a specific, often named permission issued by a financial supervisor that authorises an entity to carry out a defined regulated activity such as issuing e-money or operating as a payment institution. It binds the holder to ongoing conditions including capital adequacy, reporting, and conduct rules. For stablecoin issuers, the appropriate licence determines whether the token may legally be offered to the public.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-licence", "labels": [ "Regulatory Licence" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "regulatory-reporting-automation", "title": "Regulatory Reporting Automation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Regulatory reporting automation is the use of software, and increasingly smart contracts, to generate and submit mandated regulatory reports directly from source transaction data without manual compilation. On enterprise and consortium blockchains, shared ledgers let supervisors receive standardised, near-real-time disclosures. Automating reporting reduces error, lowers compliance cost, and shortens the lag between activity and oversight.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:regulatory-reporting-automation", "labels": [ "Regulatory Reporting Automation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "regulatory-reporting-module", "title": "Regulatory Reporting Module", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A regulatory reporting module is a self-contained software component that collects, formats, and dispatches the disclosures an entity owes to its regulators. It encapsulates jurisdiction-specific schemas, validation rules, and submission channels so they can be reused across systems. As a discrete subsystem it plugs into compliance-monitoring and tax-compliance pipelines to produce audit-ready outputs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-reporting-module", "labels": [ "Regulatory Reporting Module" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "regulatory-reporting", "title": "Regulatory Reporting", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Regulatory Reporting is the structured, machine-readable submission of financial transaction data, prudential metrics, operational incidents, and suspicious-activity indicators to supervisory authorities under legally mandated frameworks, spanning the full spectrum from trade-level derivatives re...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:regulatory-reporting", "labels": [ "Regulatory Reporting", "BC-0486-regulatory-reporting", "Regulatory Compliance Reporting", "Tax Reporting" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Compliance", "Financial Services", "Regulatory Technology", "Data Governance", "Supervisory Technology" ], "wikilinks": [ "Anti-Money Laundering", "Basel IV", "BIS", "Blockchain Analytics", "Capital Adequacy", "ComplianceLayer", "Cross-Border Data Sharing", "DataGovernanceDomain", "Data Quality Management", "DORA", "EBA", "EMIR Refit", "ESMA", "FATF", "FATF Recommendations", "FCA", "Financial Crime Detection", "Financial Market Infrastructure", "Financial Services", "FinancialServicesDomain" ] }, { "id": "regulatory-requirements", "title": "Regulatory Requirements", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory requirements are the legally binding or supervisory obligations imposed on organisations and individuals by governmental bodies, statutory agencies, and international instruments, specifying the conditions under which they may lawfully operate, offer products, or provide services. They derive from primary legislation, secondary instruments, and supervisory rulebooks, and are operationalised through licensing, capital adequacy, conduct-of-business, reporting, record-keeping, and data-protection mandates. Entities must demonstrate adherence through formal compliance programmes, independent audits, and periodic regulatory filings, with failure exposing them to financial penalties, operational restrictions, or revocation of authorisation. Regulatory requirements are increasingly sector-specific and jurisdictionally layered, requiring cross-border harmonisation mechanisms to reconcile conflicting national regimes.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:regulatory-requirements", "labels": [ "Regulatory Requirements" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [ "Compliance", "Regulatory Frameworks", "Regulation", "https://www.oecd.org/gov/regulatory-policy/", "https://en.wikipedia.org/wiki/Regulatory_compliance" ] }, { "id": "regulatory-sandbox", "title": "Regulatory Sandbox", "domain": "governance", "domain_name": "Governance", "definition": "A regulatory sandbox is a supervised testing environment in which firms can trial innovative products, services, or business models with real customers under temporary regulatory relief and close oversight. It lets regulators observe emerging technology and firms validate compliance before full-scale launch. Sandboxes are a common policy tool for fintech, blockchain, and other convergent technologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:regulatory-sandbox", "labels": [ "Regulatory Sandbox", "Regulatory Sandboxes" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "regulatory-standards", "title": "Regulatory Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Legal and compliance frameworks governing metaverse and extended reality platforms, including data protection (GDPR), artificial intelligence (EU AI Act), digital services (DSA), and accessibility requirements that ensure user rights protection in immersive environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:regulatory-standards", "labels": [ "Regulatory Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Compliance Framework" ], "wikilinks": [ "User Protection", "Compliance Framework", "metaverse" ] }, { "id": "regulatory-technology", "title": "Regulatory Technology", "domain": "governance", "domain_name": "Governance", "definition": "Regulatory Technology (RegTech) is the application of digital tools\u2014encompassing artificial intelligence, machine learning, natural language processing, distributed ledger technology, and cloud computing\u2014to automate, streamline, and improve the accuracy of regulatory compliance processes across financial services and adjacent sectors. Core functions include know-your-customer (KYC) and anti-money-laundering (AML) screening, real-time transaction monitoring, regulatory reporting, risk management, and consent lifecycle management. RegTech reduces the cost and latency of compliance obligations while providing regulators with higher-quality, machine-readable data through standardised reporting formats such as XBRL and ISO 20022. It operates at the intersection of legal obligation, data governance, and algorithmic automation, and is increasingly deployed in decentralised finance, open banking, and AI governance contexts.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:regulatory-technology", "labels": [ "Regulatory Technology" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [ "owl:Thing" ] }, { "id": "rehabilitation-robotics", "title": "Rehabilitation Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Rehabilitation robotics is the field of robotic devices that assist therapy and recovery of motor function, including exoskeletons, end-effector trainers, and assistive manipulators. These systems deliver repeatable, intensity-controlled movement while measuring patient progress, and rely on compliant, force-controlled actuation to interact safely with the body. They aim to improve outcomes and scale access to physical therapy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:rehabilitation-robotics", "labels": [ "Rehabilitation Robotics", "Rehabilitation Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "reinforcement-learning-algorithm", "title": "Reinforcement Learning Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Reinforcement Learning Algorithms enable agents to learn optimal decision-making policies through interaction with environments, guided by reward signals. Core families include value-based methods (Q-learning, DQN), policy gradient methods (REINFORCE, PPO, TRPO), and actor-critic approaches (A3C, SAC), all of which balance exploration against exploitation to maximise cumulative expected reward.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:reinforcement-learning-algorithm", "labels": [ "Reinforcement Learning Algorithm" ], "is_subclass_of": [ "AI Technique", "Reinforcement Learning" ], "wikilinks": [ "Actor-Critic Methods", "Deep Q-Network", "Policy Gradient", "owl:Thing", "Reinforcement Learning" ] }, { "id": "reinforcement-learning-for-robotics", "title": "Reinforcement Learning for Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Reinforcement Learning for Robotics is the application of reinforcement learning algorithms to train autonomous robots to acquire motor skills, manipulation capabilities, and navigation behaviours through interaction with physical or simulated environments, optimising reward signals without explicit programming of motion primitives. It addresses the unique challenges of high-dimensional continuous action spaces, sparse reward landscapes, and the sim-to-real transfer gap.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:reinforcement-learning-for-robotics", "labels": [ "Reinforcement Learning for Robotics" ], "is_subclass_of": [ "Robot Learning" ], "wikilinks": [] }, { "id": "reinforcement-learning-from-human-feedback", "title": "Reinforcement Learning from Human Feedback", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A technique for aligning language models with human preferences by training a reward model from human rankings of outputs and using reinforcement learning (typically PPO) to optimise the policy towards maximising predicted human preference. RLHF enables models to learn complex alignment objectives that are difficult to specify explicitly, forming a three-stage pipeline of supervised fine-tuning, reward model training, and RL optimisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reinforcement-learning-from-human-feedback", "labels": [ "Reinforcement Learning from Human Feedback", "RLHF", "Reinforcement Learning From Human Feedback" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "reinforcement-learning", "title": "Reinforcement Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Reinforcement learning is a machine learning paradigm in which agents learn optimal policies through interaction with an environment, receiving reward signals for actions and iteratively improving their decision-making through trial and error, encompassing model-free and model-based methods, policy gradient techniques, and value-based approaches.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reinforcement-learning", "labels": [ "Reinforcement Learning", "Globally Distributed Reinforcement Learning", "Model-Free Reinforcement Learning", "Reinforcement Learning Training", "ReinforcementLearning" ], "is_subclass_of": [ "Machine Learning", "AI Technique" ], "wikilinks": [ "BC-0452-policy", "GameAI", "Policy", "RecommendationSystem", "RoboticControl", "TemporalDifference", "ValueFunction", "ArtificialIntelligence", "ArtificialIntelligenceDomain", "AutonomousRobot", "Autonomous Robot" ] }, { "id": "rejection-sampling", "title": "Rejection Sampling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Rejection sampling is a Monte Carlo technique for drawing samples from a target probability distribution by sampling from a simpler proposal distribution and accepting or rejecting each draw according to a ratio test. It requires a proposal that bounds the target up to a constant and yields exact samples from the target when accepted. In machine learning it also names a practical method of generating candidate model outputs, scoring them and keeping only those that pass a quality or reward threshold.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rejection-sampling", "labels": [ "Rejection Sampling" ], "is_subclass_of": [ "Sampling" ], "wikilinks": [] }, { "id": "relation-extraction", "title": "Relation Extraction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Relation extraction is the natural-language-processing task of identifying semantic relationships between entities mentioned in text and classifying them into predefined or open relation types. It typically operates on the output of named-entity recognition, determining whether and how two entities are connected, for example employer-of, located-in or part-of. Relation extraction is foundational to knowledge-graph construction, supplying the typed edges that link extracted entities into structured assertions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:relation-extraction", "labels": [ "Relation Extraction" ], "is_subclass_of": [ "Information Extraction" ], "wikilinks": [] }, { "id": "relational-algebra", "title": "Relational Algebra", "domain": "data", "domain_name": "Data", "definition": "Relational algebra is a formal, procedural query language consisting of operators that take relations as input and produce relations as output. Its core operators \u2014 selection, projection, union, set difference, Cartesian product, rename, and derived operators such as join \u2014 provide a closed algebra that gives precise semantics to relational queries. It is the theoretical foundation for SQL and for the query optimisation performed by relational database systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:relational-algebra", "labels": [ "Relational Algebra" ], "is_subclass_of": [ "Relational Database" ], "wikilinks": [] }, { "id": "relational-database", "title": "Relational Database", "domain": "data", "domain_name": "Data", "definition": "A relational database organises data into tables of rows and columns, where relationships between tables are expressed through shared key values rather than physical pointers, following the relational model. Data is queried and manipulated declaratively, conventionally through SQL, and integrity is enforced via constraints, keys and typed schemas. Relational systems typically provide ACID transactions, making them the default choice for consistent, structured operational data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:relational-database", "labels": [ "Relational Database", "Relational Databases" ], "is_subclass_of": [ "Database Management System" ], "wikilinks": [] }, { "id": "relationship-edge", "title": "Relationship Edge", "domain": "data", "domain_name": "Data", "definition": "A relationship edge is a directed or undirected connection in a graph that represents a typed association between two entity nodes, often carrying a label and properties. In knowledge and identity graphs, edges encode facts such as ownership, membership, or similarity that give the graph its semantic structure. Edge type and weight determine how the graph can be queried, traversed, and reasoned over.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:relationship-edge", "labels": [ "Relationship Edge" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "relative-spacecraft-navigation", "title": "Relative Spacecraft Navigation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:relative-spacecraft-navigation", "labels": [ "Relative Spacecraft Navigation" ], "is_subclass_of": [ "Spacecraft Navigation" ], "wikilinks": [] }, { "id": "relay-network", "title": "Relay Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Relay Network is a set of intermediary nodes that forward messages, transactions, or data between participants who are not directly connected, improving reach, latency, privacy, or censorship-resistance. Relays do not necessarily originate or consume the content they pass on; they propagate it across the topology so that information reaches its destination efficiently. In blockchain systems, relay networks accelerate block and transaction propagation and connect cross-chain messaging; in privacy systems such as Tor and Nostr, relays forward traffic to obscure origin or to disseminate events. Relay networks depend on robust routing, redundancy, and incentive or trust assumptions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:relay-network", "labels": [ "Relay Network" ], "is_subclass_of": [ "Overlay Network", "Network Component" ], "wikilinks": [] }, { "id": "relayer", "title": "Relayer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Infrastructure operator that facilitates cross-chain message passing by monitoring source chains for events, generating cryptographic proofs of state, and submitting verified transactions to destination chains. Relayers provide non-custodial connectivity for interoperability protocols such as IBC, LayerZero, and Chainlink CCIP without holding user assets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:relayer", "labels": [ "Relayer", "Hub Relayer", "Relayer Network" ], "is_subclass_of": [ "Network Component", "Blockchain Interoperability", "Blockchain" ], "wikilinks": [ "Interoperability Protocol", "Message Passing", "Blockchain", "Blockchain Oracle", "Cross-Chain Bridge", "Interoperability", "Light Client" ] }, { "id": "relevance-ranking", "title": "Relevance Ranking", "domain": "ai", "domain_name": "Ai", "definition": "Relevance Ranking is the process of ordering a set of candidate documents, passages or items by their estimated usefulness to a given query or context. It combines lexical, semantic and behavioural signals into a score that determines the sequence in which results are presented. Ranking quality directly governs the perceived effectiveness of search, recommendation and retrieval-augmented generation systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:relevance-ranking", "labels": [ "Relevance Ranking" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "reliability-engineering", "title": "Reliability Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Reliability engineering is the engineering discipline that applies probabilistic and statistical methods to design, analyse, and verify that systems perform their required functions for a specified period under stated operating conditions without failure. It encompasses systematic techniques including fault tree analysis, failure mode and effects analysis, and accelerated life testing to quantify failure probabilities and identify design weaknesses. The discipline establishes dependability metrics such as MTBF, MTTF, and availability, and prescribes design strategies including redundancy, derating, and fault-tolerant architectures to achieve reliability targets. It spans hardware, software, and socio-technical systems and is foundational to safety-critical engineering in aerospace, automotive, medical devices, nuclear, and large-scale cloud infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:reliability-engineering", "labels": [ "Reliability Engineering" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [] }, { "id": "reliability", "title": "Reliability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The degree to which an AI system performs its intended function consistently and accurately over time and across repeated operations, producing predictable and dependable results under specified conditions.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:reliability", "labels": [ "Reliability", "Network Reliability", "Service Reliability", "Software Reliability", "System Reliability" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Monitoring", "Testing", "Validation (AI-0095)", "MetaverseDomain" ] }, { "id": "relying-party", "title": "Relying Party", "domain": "security", "domain_name": "Security", "definition": "A relying party is an application or service that depends on an external identity provider or credential issuer to authenticate users and assert their attributes, rather than managing credentials itself. It consumes and validates assertions, tokens or verifiable credentials to make access-control decisions. As a core role in federated and decentralised identity, the relying party trusts issuers within a defined trust framework and enforces the resulting authorisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:relying-party", "labels": [ "Relying Party" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "remediation-plan", "title": "Remediation Plan", "domain": "governance", "domain_name": "Governance", "definition": "A remediation plan is a structured set of corrective actions, owners, and deadlines drawn up to address identified non-conformities, risks, or harms discovered through audit or monitoring. It links each finding to a specific fix and tracks progress to closure. In compliance and ethical-sourcing contexts it is the mechanism that turns a violation into demonstrable improvement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:remediation-plan", "labels": [ "Remediation Plan" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "remittances", "title": "Remittances", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Remittances are cross-border transfers of money, typically sent by migrant workers to family in their home country, representing a major global flow of funds to developing economies. Traditional rails impose high fees and slow settlement, which cryptocurrencies and the Lightning Network aim to reduce. As a use case, remittances anchor much of the argument for Bitcoin and stablecoins as practical money.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:remittances", "labels": [ "Remittances", "Remittance" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "remote-assistance", "title": "Remote Assistance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Remote assistance is a service modality in which an expert or system provides real-time guidance, intervention, or control to a person or machine at a physically separate location, mediated by communication and rendering technology. It spans technical IT support via screen-sharing, clinical guidance through telestration overlays on AR headsets, and safety-critical interventions in teleoperated robotics. Effective remote assistance requires low-latency bidirectional communication, shared situational awareness, and appropriate interaction modalities for the task domain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:remote-assistance", "labels": [ "Remote Assistance", "Remote Expert Guidance", "Remote Technical Support" ], "is_subclass_of": [ "Remote Collaboration", "Telepresence (Distributed Collaboration)" ], "wikilinks": [] }, { "id": "remote-attestation", "title": "Remote Attestation", "domain": "security", "domain_name": "Security", "definition": "Remote attestation is a security mechanism by which one system proves the integrity and identity of its software and hardware state to a remote verifier. The attesting platform produces cryptographically signed evidence, rooted in a hardware root of trust, that captures measurements of its boot sequence and running code. A verifier checks this evidence against expected values and a trusted signing key before granting access or releasing secrets, allowing trust decisions to be made about a machine that is not physically controlled.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:remote-attestation", "labels": [ "Remote Attestation" ], "is_subclass_of": [ "Confidential Computing" ], "wikilinks": [] }, { "id": "remote-collaboration", "title": "Remote Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Remote Collaboration encompasses the work practices, platforms, and coordination methodologies that enable geographically distributed teams to plan, create, and deliver work without physical co-location. It integrates synchronous channels such as video conferencing, real-time co-editing, and spatial audio with asynchronous mechanisms such as version control, persistent messaging, and project management systems, binding these into coherent workflows that maintain team cohesion and productivity across time zones. The discipline draws on computer-supported cooperative work (CSCW) research, network infrastructure, human-computer interaction, and organisational science to address shared challenges of presence, trust, latency, and equitable participation. Increasingly, AI-powered features such as automatic transcription, meeting summarisation, and action-item extraction, together with immersive spatial interfaces from XR and the metaverse, are extending the fidelity and richness of distributed work.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:remote-collaboration", "labels": [ "Remote Collaboration", "RemoteDesignReview" ], "is_subclass_of": [ "Distributed Collaboration", "Distributed Teams" ], "wikilinks": [] }, { "id": "remote-communication", "title": "Remote Communication", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Remote Communication encompasses the protocols, platforms, and practices that enable synchronous and asynchronous exchange of information between geographically distributed participants. It spans text, voice, video, and immersive modalities, and underpins distributed work, telecollaboration, and telepresence systems by abstracting physical distance through networked infrastructure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:remote-communication", "labels": [ "Remote Communication" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "remote-education", "title": "Remote Education", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Remote education is the delivery of teaching and learning to geographically dispersed participants over digital networks, spanning live virtual classrooms, recorded content, and immersive metaverse environments. It depends on collaboration tools such as breakout rooms, shared media, and presence to recreate interactive instruction. Immersive variants add spatial co-presence to improve engagement over flat video.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:remote-education", "labels": [ "Remote Education" ], "is_subclass_of": [ "Educational Technology" ], "wikilinks": [] }, { "id": "remote-pair-programming", "title": "Remote Pair Programming", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Remote Pair Programming is a synchronous collaborative software-development practice in which two developers\u2014one acting as driver (typing) and one as navigator (reviewing, guiding, strategising)\u2014share a single logical programming environment across geographically separated workstations, communica...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:remote-pair-programming", "labels": [ "Remote Pair Programming" ], "is_subclass_of": [ "Workspace Tools", "Communication Technology", "Pair Programming", "Collaborative Development", "Distributed Software Engineering", "Agile Software Development", "Synchronous Collaboration" ], "wikilinks": [ "Agile Manifesto (2001)", "Agile Manifesto 2001", "AgileMethodologyDomain", "Agile Software Development", "Amazon Q Developer", "Apple Vision Pro", "Arisholm et al. 2007 Evaluating Pair Programming", "Asynchronous Code Review", "Audio/Video Channel", "Barke et al. 2023 Grounded Copilot", "Beck 1999 Extreme Programming Explained", "Carnegie Mellon University", "Cloud Development Environment", "Code Quality Assurance", "Code Review", "Code Walkthrough", "Collaborative Development", "Collective Code Ownership", "Continuous Integration", "Continuous Learning" ] }, { "id": "remote-procedure-call", "title": "Remote Procedure Call", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A remote procedure call (RPC) is a communication paradigm in which a program invokes a procedure that executes on a different address space, typically another machine on a network, as if it were a local call. The runtime marshals arguments, transmits them over a transport, executes the procedure remotely, and returns the result, hiding the underlying network mechanics from the caller. RPC underpins distributed systems and service-to-service communication, with modern frameworks adding streaming, code generation, and efficient binary serialisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:remote-procedure-call", "labels": [ "Remote Procedure Call" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "remote-rendering", "title": "Remote Rendering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Remote Rendering offloads GPU-intensive 3D scene computation to a server or cloud node, streaming compressed video frames to a thin client such as an XR headset or mobile device. This approach decouples visual fidelity from device hardware constraints, enabling photorealistic graphics on low-power endpoints while centralising GPU resources. Latency and bandwidth are critical quality-of-service parameters; edge computing deployments minimise round-trip delay to support interactive frame rates.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:remote-rendering", "labels": [ "Remote Rendering" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "remote-sensing-image-segmentation", "title": "Remote Sensing Image Segmentation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:remote-sensing-image-segmentation", "labels": [ "Remote Sensing Image Segmentation" ], "is_subclass_of": [ "Image Segmentation" ], "wikilinks": [] }, { "id": "remote-sensing-image", "title": "Remote Sensing Image", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:remote-sensing-image", "labels": [ "Remote Sensing Image" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "remote-sensing-platform", "title": "Remote Sensing Platform", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:remote-sensing-platform", "labels": [ "Remote Sensing Platform" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "remote-sensing", "title": "Remote Sensing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Remote sensing is the acquisition of information about physical objects or phenomena from a distance, typically using sensors mounted on satellites, aircraft, drones, or ground-based platforms, without direct physical contact with the subject. It encompasses passive sensing (measuring reflected or emitted electromagnetic radiation across optical, infrared, and microwave bands) and active sensing (radar and LiDAR, which emit and measure return signals). Remote sensing data underpins earth observation, environmental monitoring, precision agriculture, disaster response, urban planning, and military reconnaissance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:remote-sensing", "labels": [ "Remote Sensing" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "remote-surgery", "title": "Remote Surgery", "domain": "robotics", "domain_name": "Robotics", "definition": "Remote surgery, or telesurgery, is the performance of a surgical procedure by a surgeon operating a robotic system from a location physically separate from the patient, with instrument motion relayed over a network connection. It depends on low-latency, high-reliability telepresence links and haptic feedback to preserve the surgeon's sense of touch and precision. It extends specialist surgical care to remote or underserved locations.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:remote-surgery", "labels": [ "Remote Surgery" ], "is_subclass_of": [ "Surgical Robotics" ], "wikilinks": [] }, { "id": "remote-work-infrastructure", "title": "Remote Work Infrastructure", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Remote work infrastructure is the combined set of networking, collaboration, and presence technologies that let distributed teams work together as if co-located. It includes real-time communication, shared workspaces, presence and status signalling, and adaptive transport that copes with variable connectivity. As workforces decentralise, this infrastructure underpins productivity, security, and a sense of shared presence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:remote-work-infrastructure", "labels": [ "Remote Work Infrastructure" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "remote-work", "title": "Remote Work", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Remote work is a work arrangement in which employees perform their professional duties from locations outside the traditional employer premises \u2014 including home offices, co-working spaces, or while travelling \u2014 using digital communication, collaboration, and productivity tools to maintain output and team cohesion. It encompasses fully remote models, hybrid arrangements that blend in-office and remote days, and asynchronous-first organisational cultures that decouple productive work from synchronous physical presence. The practice spans individual contributor roles through entire distributed organisations and intersects deeply with workforce management, digital infrastructure provisioning, cybersecurity policy, and organisational design. Remote work has become a core dimension of modern knowledge-work governance, influencing talent acquisition geographically, commercial real estate demand, and the architecture of digital workplace platforms.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:remote-work", "labels": [ "Remote Work", "Inclusive Remote Work" ], "is_subclass_of": [ "Distributed Work" ], "wikilinks": [] }, { "id": "remotely-operated-vehicle-rov", "title": "Remotely Operated Vehicle (ROV)", "domain": "robotics", "domain_name": "Robotics", "definition": "A Remotely Operated Vehicle (ROV) is a tethered underwater robot controlled by a human operator at the surface via a cable that supplies electrical power and bidirectional communications. ROVs are deployed for inspection, maintenance, and intervention tasks in environments too hazardous or deep for human divers, including oil and gas infrastructure, offshore wind farms, and scientific ocean exploration.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:remotely-operated-vehicle-rov", "labels": [ "Remotely Operated Vehicle (ROV)", "Remote Operated Vehicle" ], "is_subclass_of": [ "Robot Type", "Underwater Robot" ], "wikilinks": [ "Robotics", "Underwater Robot" ] }, { "id": "render-farm", "title": "Render Farm", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A render farm is a cluster of networked computers dedicated to performing the computationally intensive image-synthesis work of rendering, distributing frames or tiles of a scene across many machines to reduce total completion time. It is a standard component of film production and other content creation pipelines that require photorealistic output at scale. Render farms may run on dedicated on-premises hardware or on elastically provisioned cloud compute.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:render-farm", "labels": [ "Render Farm" ], "is_subclass_of": [ "Rendering" ], "wikilinks": [] }, { "id": "render-pipeline", "title": "Render Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Render Pipeline is the ordered sequence of programmable and fixed-function GPU stages that transforms 3D scene geometry, material data, and lighting information into a final rasterised or ray-traced 2D image frame, typically encompassing vertex processing, tessellation, geometry shading, rasterisation, fragment shading, depth-stencil testing, blending, and screen-space post-processing. Modern graphics APIs \u2014 including Vulkan, DirectX 12, Metal, and WebGPU \u2014 expose explicit, low-overhead control over pipeline state objects and synchronisation barriers, enabling advanced techniques such as deferred rendering, clustered shading, variable-rate shading, and hardware-accelerated ray tracing. The render pipeline is the core computational artefact of any real-time interactive graphics system and is equally foundational to offline path-traced production rendering, though the stage granularity and scheduling strategies differ considerably between the two contexts. In spatial computing and metaverse platforms, render pipeline design determines latency, fidelity, and energy consumption on resource-constrained XR devices.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:render-pipeline", "labels": [ "Render Pipeline", "Universal Render Pipeline" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "render-target", "title": "Render Target", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A memory buffer or texture surface to which a GPU writes the output of a rendering pass, including the default framebuffer displayed on screen and off-screen targets used for post-processing effects, shadow maps, reflections, and multi-pass rendering in real-time graphics pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:render-target", "labels": [ "Render Target" ], "is_subclass_of": [ "Display and Rendering", "Render Pipeline" ], "wikilinks": [] }, { "id": "renderer", "title": "Renderer", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A renderer is the software or hardware component that converts a scene description, such as geometry, materials, lights, or a markup specification, into a final image or visual output. Renderers range from real-time rasterisers and ray tracers to text-and-diagram engines that draw from declarative source. The choice of renderer determines visual fidelity, performance, and the formats a system can produce.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:renderer", "labels": [ "Renderer" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "rendering-engine", "title": "Rendering Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A rendering engine is a software system that generates 2D images or animations from 3D scene descriptions through rasterisation, ray tracing, or hybrid pipelines. Modern real-time engines such as Unreal Engine 5 and Unity combine global illumination, virtualised geometry, neural shaders, and AI-driven upscaling to produce cinematic-quality visuals at interactive frame rates. They serve as the foundational compute layer for games, virtual reality, film production, and spatial-computing applications.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:rendering-engine", "labels": [ "Rendering Engine" ], "is_subclass_of": [ "Display and Rendering", "Computer Graphics" ], "wikilinks": [ "Computer Graphics", "Game Development", "Film Production", "Gaussian Splatting", "Metaverse", "NVIDIA", "Ray Tracing", "Virtual Reality" ] }, { "id": "rendering-pipeline", "title": "Rendering Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Rendering Pipeline is the ordered computational sequence by which a GPU transforms three-dimensional scene representations \u2014 vertex buffers, index buffers, textures, uniform data, and acceleration structures \u2014 into a two-dimensional raster image suitable for display or further processing.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:rendering-pipeline", "labels": [ "Rendering Pipeline", "GPU Rendering Pipeline", "Rasterisation Pipeline", "Rendering Cluster", "RenderingPipeline" ], "is_subclass_of": [ "Display and Rendering", "Rasterization", "GPU Compute", "Parallel Computing", "Real-Time Systems", "Computer Graphics", "Rasterisation" ], "wikilinks": [ "Acceleration Structure", "APILayer", "Apple Metal", "Architectural Visualisation", "BVH Acceleration Structure", "Clustered Shading", "Computer Graphics", "Deferred Rendering", "Depth Buffer", "DirectX 12", "DLSS", "DriverLayer", "EngineLayer", "Film Visual Effects", "Forward Rendering", "Fragment Shader", "FSR", "G-Buffer", "Geometry Shader", "GLSL" ] }, { "id": "rendering-technique", "title": "Rendering Technique", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A rendering technique is a specific algorithm or computational methodology for transforming a three-dimensional geometric scene description into a two-dimensional pixel image, inherently balancing image fidelity against computational cost and latency. The fundamental paradigms are rasterization (projecting geometry to screen space via edge-walking and interpolation), ray tracing (physically simulating light transport by casting rays from the virtual camera), and path tracing (extending ray tracing with Monte Carlo sampling for unbiased global illumination). Modern systems layer shading models (forward, deferred, clustered), global illumination approximations, and screen-space post-processing into hybrid pipelines that target real-time or offline quality budgets across hardware ranging from mobile GPUs to dedicated ray-tracing silicon.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:rendering-technique", "labels": [ "Rendering Technique" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "rendering-technology", "title": "Rendering Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Rendering Technology comprises the algorithms, pipelines, and hardware interfaces that convert geometric scene descriptions into pixel images. It encompasses rasterisation, ray tracing, and hybrid approaches, executed on GPUs through APIs such as Vulkan, Metal, and WebGPU, and is foundational to real-time interactive graphics, spatial computing experiences, and digital twin visualisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rendering-technology", "labels": [ "Rendering Technology" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "rendering", "title": "Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Rendering is the computational process of generating a two-dimensional image or display output from a scene description, encompassing geometry, materials, lighting, and camera parameters. Techniques range from real-time rasterisation on GPUs to offline ray tracing and neural rendering methods used in film and spatial computing applications. High-quality, low-latency rendering is critical for immersive virtual and augmented reality experiences.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:rendering", "labels": [ "Rendering" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "renewable-energy-certificates", "title": "Renewable Energy Certificates", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Digital or blockchain-tokenized instruments that certify the generation of one megawatt-hour (MWh) of electricity from renewable energy sources, providing transparent, immutable proof of renewable energy consumption for carbon accounting, sustainability reporting, and regulatory compliance, with ...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:renewable-energy-certificates", "labels": [ "Renewable Energy Certificates", "Renewable Energy Certificate", "Renewable Energy Certification" ], "is_subclass_of": [ "Token and Asset", "DigitalAsset", "CarbonCredit", "SustainabilityInstrument", "EnvironmentalCommodity" ], "wikilinks": [ "CarbonCredit", "CarbonMarkets", "Center for Resource Solutions", "CertificateMetadata", "EnergySector", "EnvironmentalCommodity", "European Energy Certificate System (EECS)", "GenerationData", "Green-e Energy", "IEA Renewable Energy Markets", "International REC Standard (I-REC)", "MeteringData", "OwnershipRecord", "PeerToPeerEnergyTrading", "RegistryInfrastructure", "RenewableEnergyGeneration", "RenewableEnergyTracking", "RetirementProof", "SustainabilityDomain", "SustainabilityInstrument" ] }, { "id": "renewable-energy-integration", "title": "Renewable Energy Integration", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Renewable energy integration is the practice of powering computational or industrial loads with solar, wind, hydro, and other low-carbon sources, often using flexible demand to absorb intermittent or otherwise curtailed generation. In blockchain it refers to siting mining near renewable supply to lower emissions and monetise stranded energy. Effective integration reduces carbon intensity and can improve grid economics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:renewable-energy-integration", "labels": [ "Renewable Energy Integration" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "renewable-energy-investment", "title": "Renewable Energy Investment", "domain": "finance", "domain_name": "Finance", "definition": "Renewable energy investment is the allocation of capital toward generation assets and infrastructure that produce energy from renewable sources such as solar, wind and hydro, spanning project finance, equity investment and green bonds. It is a core mechanism within green finance for directing capital toward decarbonisation, and it has been linked to blockchain-based schemes that monetise otherwise stranded renewable energy. Investment volumes and returns in this sector are sensitive to policy incentives, grid infrastructure and financing cost.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:renewable-energy-investment", "labels": [ "Renewable Energy Investment" ], "is_subclass_of": [ "Green Finance" ], "wikilinks": [] }, { "id": "renewable-energy-tracking", "title": "Renewable Energy Tracking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Renewable energy tracking is the verifiable measurement and accounting of clean-energy generation and consumption, often recorded on a ledger to prevent double counting of attributes. It underpins renewable energy certificates by binding each megawatt-hour to a unique, auditable claim. Reliable tracking is essential for credible decarbonisation reporting and voluntary climate commitments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:renewable-energy-tracking", "labels": [ "Renewable Energy Tracking" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "renewable-energy", "title": "Renewable Energy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Renewable Energy encompasses energy derived from naturally replenishing sources \u2014 principally solar photovoltaic, wind, hydropower, geothermal, tidal, and biomass \u2014 that are not depleted by use and produce little or no direct greenhouse gas emissions during operation. It forms the primary technical basis for decarbonising electricity systems worldwide, with cost reductions in solar and wind having made these technologies the cheapest sources of new generation capacity in most markets. Deployment is enabled by complementary systems including grid-scale energy storage, smart grid management, and market instruments such as Renewable Energy Certificates that allow attribution of clean generation to specific consumers. The sector intersects directly with digital infrastructure through the growing energy demands of data centres and AI workloads, driving hyperscaler Power Purchase Agreements and shaping where compute infrastructure is sited globally.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:renewable-energy", "labels": [ "Renewable Energy", "Curtailed Renewable Energy", "Renewable Energy Curtailment", "Renewable Energy Development", "Renewable Energy Supply" ], "is_subclass_of": [ "Energy and Power" ], "wikilinks": [] }, { "id": "reparameterisation-trick", "title": "Reparameterisation Trick", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The reparameterisation trick is a technique for obtaining low-variance gradient estimates of an expectation over a random variable by expressing that variable as a deterministic, differentiable function of the distribution parameters and an independent noise source. By moving the stochasticity outside the computation graph, gradients can flow through a sampling step via backpropagation, enabling end-to-end training of models with latent random variables. It is foundational to variational autoencoders and to many stochastic optimisation methods in deep learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reparameterisation-trick", "labels": [ "Reparameterisation Trick" ], "is_subclass_of": [ "Variational Autoencoder" ], "wikilinks": [] }, { "id": "repeat-ground-track", "title": "Repeat Ground Track", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:repeat-ground-track", "labels": [ "Repeat Ground Track" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "replication-protocol", "title": "Replication Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A replication protocol is the set of rules and message exchanges by which a distributed system maintains multiple copies of data across nodes so that they remain consistent according to a chosen consistency model despite failures and concurrency. It governs how updates are propagated, ordered, and acknowledged, balancing availability, latency, and durability. Replication protocols range from synchronous primary-backup and quorum schemes to asynchronous gossip and conflict-free replicated approaches.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:replication-protocol", "labels": [ "Replication Protocol" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "replication-system", "title": "Replication System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A replication system maintains synchronised copies of data or state across multiple nodes in a distributed environment, ensuring consistency, availability, and fault tolerance. It coordinates state propagation through protocols such as leader-based or leaderless replication, supporting both synchronous and asynchronous update strategies in blockchain, database, and metaverse infrastructure contexts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:replication-system", "labels": [ "Replication System" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "replication", "title": "Replication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Replication is the practice of maintaining multiple copies of data or services across different machines or locations to improve availability, durability, fault tolerance, and read performance. Strategies range from synchronous replication, which guarantees copies are identical before acknowledging a write, to asynchronous replication, which favours latency at the risk of temporary divergence. Replication is foundational to distributed databases, content-delivery networks, and high-availability systems, and its design forces explicit choices among consistency, availability, and partition tolerance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:replication", "labels": [ "Replication", "Full Replication", "Multi-datacenter Replication", "Optimistic Replication", "Primary-Backup Replication", "Replication Factor", "Replication Mechanism", "Replication Mechanisms", "Replication Service", "State Replication", "Storage Replication" ], "is_subclass_of": [ "Fault Tolerance" ], "wikilinks": [] }, { "id": "reporting-mechanisms", "title": "Reporting Mechanisms", "domain": "governance", "domain_name": "Governance", "definition": "Reporting mechanisms are the user-facing tools and processes through which people flag illegal content, abuse, or policy violations to a platform or authority. They are a mandated component of content-moderation regimes, requiring accessible channels, acknowledgement, and traceable handling of notices. Well-designed mechanisms balance ease of reporting against abuse of the reporting system itself.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reporting-mechanisms", "labels": [ "Reporting Mechanisms", "Reporting Tools" ], "is_subclass_of": [ "Metaverse governance and safeguarding" ], "wikilinks": [] }, { "id": "representation-learning", "title": "Representation Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Representation learning is a subfield of machine learning concerned with automatically discovering the data transformations and feature spaces \u2014 representations \u2014 that make subsequent learning tasks easier, more accurate, or more data-efficient. Rather than relying on hand-crafted features, representation learning systems learn to encode raw inputs such as images, text, or sensor readings into dense, structured latent vectors that capture semantically meaningful variation. Deep neural networks, particularly transformers and convolutional architectures, have made learned representations ubiquitous across computer vision, natural language processing, and multi-modal AI. The quality of learned representations directly determines the performance ceiling of downstream models trained on top of them.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:representation-learning", "labels": [ "Representation Learning", "Contextual Representation Learning", "Matryoshka Representation Learning", "Unsupervised Representation Learning" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "reproducibility", "title": "Reproducibility", "domain": "data", "domain_name": "Data", "definition": "Reproducibility is the property of a scientific or computational study whereby independent researchers can obtain the same or statistically equivalent results by applying the same methods and analysis procedures to the same dataset. It is a foundational criterion for the credibility and cumulative progress of empirical disciplines, distinguishing it from replicability, which extends the criterion to new data or new samples. Reproducibility failures arise from incomplete method documentation, software environment drift, undisclosed analytical flexibility, or data inaccessibility. Achieving it systematically requires version-controlled code, containerised execution environments, open datasets, and pre-registered analysis plans.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:reproducibility", "labels": [ "Reproducibility", "Model Reproducibility", "Research Reproducibility", "Scientific Reproducibility" ], "is_subclass_of": [ "Quality Assurance" ], "wikilinks": [ "Replication", "Robustness", "Sensitivity", "Quality Assurance" ] }, { "id": "reproducible-builds", "title": "Reproducible Builds", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Reproducible builds are a software build practice in which compiling the same source code under the same recorded conditions always yields bit-for-bit identical artifacts. By removing sources of non-determinism such as timestamps, build paths, and ordering, independent parties can verify that a published binary corresponds exactly to its claimed source. This independent verifiability strengthens software supply-chain security and trust in distributed binaries.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reproducible-builds", "labels": [ "Reproducible Builds" ], "is_subclass_of": [ "Supply Chain Security" ], "wikilinks": [] }, { "id": "reproducible-research", "title": "Reproducible Research", "domain": "ai", "domain_name": "Ai", "definition": "Reproducible research is the practice of conducting and reporting scientific work so that an independent party can obtain the same results from the same data and code. It encompasses sharing datasets, source code, computational environments, and detailed methodology alongside published findings. In machine learning it is the discipline of making experiments verifiable through fixed seeds, versioned artefacts, recorded hyperparameters, and documented hardware, distinguishing genuine effects from artefacts of a particular run.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reproducible-research", "labels": [ "Reproducible Research" ], "is_subclass_of": [ "Artificial Intelligence", "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "reprojection-error", "title": "Reprojection Error", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Reprojection error is the geometric distance, in image pixels, between an observed feature point and the position predicted by projecting its estimated three-dimensional point back through the estimated camera model. It serves as the primary residual minimised in camera calibration, pose estimation and bundle adjustment, where the sum of squared reprojection errors quantifies how well the reconstructed geometry explains the measurements. Low reprojection error indicates a consistent, well-calibrated reconstruction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reprojection-error", "labels": [ "Reprojection Error" ], "is_subclass_of": [ "Camera Calibration" ], "wikilinks": [] }, { "id": "reputation-data", "title": "Reputation Data", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured dataset containing historical records of user behavior, transaction outcomes, peer feedback, and trust metrics used to calculate reputation scores in peer-to-peer systems and virtual communities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:reputation-data", "labels": [ "Reputation Data" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Behavioral Pattern", "Community Moderation", "Cryptographic Signature", "Feedback Score", "Fraud Detection", "OpenReputation Protocol", "Social Graph", "Timestamp Service", "Transaction History", "Trust Indicator", "Trust Scoring", "W3C Verifiable Credentials", "Access Control", "Audit Trail", "Data Storage", "Identity Provider", "MiddlewareLayer", "Reputation System", "TrustAndGovernanceDomain", "Trust Framework" ] }, { "id": "reputation-scoring-model", "title": "Reputation Scoring Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An algorithmic process that computes quantitative reputation scores by aggregating behavioral data, applying weighted scoring functions, implementing temporal decay, and evaluating threshold conditions to generate trust indicators for entities in virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:reputation-scoring-model", "labels": [ "Reputation Scoring Model" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Access Control Decisions", "Behavioral Data Aggregator", "Behavioral Models", "Data Collection Pipeline", "ETSI GS MEC", "Governance Voting Weight", "Metric Computation", "Scoring Algorithms", "Statistical Analysis", "Temporal Decay Engine", "Threshold Evaluator", "Validation Rules", "Weighted Scoring Function", "MiddlewareLayer", "Reputation Data", "Risk Assessment", "TrustAndGovernanceDomain", "Trust Infrastructure", "Trust Score Metric", "VirtualSocietyDomain" ] }, { "id": "reputation-system", "title": "Reputation System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Reputation System is a decentralised computational infrastructure for aggregating verifiable behavioural signals about network participants into quantified trust scores that enable social coordination in trustless Blockchain environments, spanning algorithmic trust propagation architect...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:reputation-system", "labels": [ "Reputation System", "Brand Reputation Management", "Reputation" ], "is_subclass_of": [ "Governance and Regulation", "Decentralized Governance", "Trust Networks", "Social Coordination", "Decentralised Governance", "Identity Systems", "Feedback System" ], "wikilinks": [ "Aggregation Algorithm", "Attestation Schema", "Attestation Service", "Centralised Rating Systems", "Credential Registry", "Credit Scoring", "Cryptographic Proof", "Decentralised Governance", "Decentralised Identifiers", "Decentralised Identity", "EigenTrust Algorithm", "EIP-4804", "ENS", "ERC-721", "Ethereum Attestation Service", "Farcaster", "Feedback System", "Gitcoin", "Gitcoin Passport", "GovernanceDomain" ] }, { "id": "reputation-based-bft", "title": "Reputation-Based BFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Reputation-Based BFT is a variant of Byzantine Fault Tolerance in which each consensus participant is assigned a dynamic reputation score derived from its historical behaviour\u2014vote accuracy, uptime, and honest message propagation\u2014so that nodes with strong track records carry greater weight in the consensus outcome. This design allows the protocol to tolerate Byzantine actors more efficiently than pure stake-weighted schemes by penalising misbehaving validators through reputation decay rather than requiring slashing of locked collateral. It is commonly deployed in permissioned or consortium blockchain networks where participants are identified and accountable.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:reputation-based-bft", "labels": [ "Reputation-Based BFT" ], "is_subclass_of": [ "Protocol and Consensus", "Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Byzantine Fault Tolerance" ] }, { "id": "request-for-comments", "title": "Request For Comments", "domain": "standards", "domain_name": "Standards", "definition": "A Request for Comments (RFC) is a formally numbered document in a series used to define, describe and discuss internet technologies, protocols, procedures and conventions. Originating with the early ARPANET community and now stewarded primarily through the IETF, RFCs range from formal standards to informational and experimental memos. The series provides a durable, citable record of how core internet infrastructure is specified and evolved.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:request-for-comments", "labels": [ "Request For Comments", "Request for Comments" ], "is_subclass_of": [ "Open Standards" ], "wikilinks": [] }, { "id": "request-response-pattern", "title": "Request-Response Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The request-response pattern is a synchronous message-exchange model in which a client sends a request to a server and blocks, or awaits, until a corresponding response is returned. It establishes a one-to-one, correlated interaction where each request expects exactly one reply, forming the basis of most client-server communication. The pattern contrasts with asynchronous, event-driven, and publish-subscribe styles where senders do not wait for an immediate reply.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:request-response-pattern", "labels": [ "Request-Response Pattern" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "requirements-engineering", "title": "Requirements Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Requirements engineering is the systematic process of eliciting, analysing, specifying, validating and managing the needs and constraints that a system must satisfy. It transforms stakeholder goals into clear, verifiable and traceable requirements that guide design, implementation and testing. As a foundational software and systems engineering discipline, it reduces project risk by surfacing ambiguity, conflict and infeasibility early in the lifecycle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:requirements-engineering", "labels": [ "Requirements Engineering" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "reranker", "title": "Reranker", "domain": "ai", "domain_name": "Ai", "definition": "A reranker is a second-stage scoring model in information retrieval and retrieval-augmented generation pipelines that reorders a set of candidate documents or passages \u2014 initially retrieved by a fast first-stage retriever such as a dense vector index \u2014 using a more computationally intensive cross-encoder or LLM-based relevance model that jointly encodes the query and each candidate together, producing a higher-precision relevance score than bi-encoder similarity alone. Rerankers trade retrieval speed for ranking quality, operating on a reduced candidate set (typically 50-200 passages) rather than the full corpus, thereby making deep transformer inference tractable at query time.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:reranker", "labels": [ "Reranker" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "reranking", "title": "Reranking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Reranking is a second-stage retrieval step in which an initial, computationally cheap set of candidate documents or passages is reordered by a more expensive, higher-precision model that scores each candidate's relevance to the query more accurately. It is a standard component of retrieval-augmented generation and search pipelines, where a fast retriever (such as a bi-encoder or lexical index) first narrows a large corpus down to a manageable candidate set, and a cross-encoder or learned ranker then refines the ordering. Reranking improves precision at the cost of additional latency, so candidate set sizes are tuned to balance quality and speed.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:reranking", "labels": [ "Reranking" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "resampling", "title": "Resampling", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Resampling is a family of statistical and machine-learning techniques that repeatedly draw samples from observed data to estimate the variability of a statistic, validate a model, or rebalance a dataset. It includes methods such as bootstrapping, cross-validation, permutation testing, and over- and under-sampling for class imbalance. By substituting computation for restrictive distributional assumptions, resampling provides robust estimates of error, confidence and generalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:resampling", "labels": [ "Resampling" ], "is_subclass_of": [ "Statistical Inference" ], "wikilinks": [] }, { "id": "rescue-robot", "title": "Rescue Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Rescue Robot is a specialised mobile robot engineered to operate in hazardous, unstructured environments\u2014including collapsed structures, disaster zones, and nuclear incidents\u2014where direct human presence is unsafe. Such systems integrate multimodal locomotion (tracked, wheeled, or legged), sensor suites (IR, LIDAR, acoustic, tactile), and teleoperation or autonomous navigation to locate, assess, and extract casualties or gather situational data. Conformance with safety standards such as ISO 8373 and participation in benchmarks such as the RoboCup Rescue Robot League drives continuous capability development.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:rescue-robot", "labels": [ "Rescue Robot" ], "is_subclass_of": [ "Robot Type", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "research-agents", "title": "Research Agents", "domain": "ai", "domain_name": "Ai", "definition": "Research agents are autonomous LLM-based systems that plan and execute multi-step information-gathering tasks, issuing searches, reading sources, and synthesising cited findings into reports. They combine tool use, memory, and reasoning to pursue an objective across many web or document queries with minimal supervision. They represent a leading application of agentic AI for knowledge work.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:research-agents", "labels": [ "Research Agents" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "research-dissemination", "title": "Research Dissemination", "domain": "ai", "domain_name": "Ai", "definition": "Research dissemination is the systematic communication of research findings to academic peers, practitioners, and the public through conferences, journals, preprints, and open repositories. It encompasses peer review, publication, and presentation practices that allow new results to be scrutinised, replicated, and built upon. In artificial intelligence research it is closely tied to conference proceedings, workshops, and open-access archives that accelerate the field's collective progress.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:research-dissemination", "labels": [ "Research Dissemination" ], "is_subclass_of": [ "Knowledge Sharing" ], "wikilinks": [] }, { "id": "research-funding", "title": "Research Funding", "domain": "ai", "domain_name": "Ai", "definition": "Research funding is the provision of financial resources to support scientific, scholarly and technological investigation, supplied through mechanisms such as competitive grants, institutional block funding, philanthropy and industry sponsorship. It typically flows through proposal submission, peer review and award management, and shapes which questions are pursued and by whom. Research funding is a primary determinant of the direction, scale and independence of academic research and innovation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:research-funding", "labels": [ "Research Funding" ], "is_subclass_of": [ "Academic Research" ], "wikilinks": [] }, { "id": "research-institution", "title": "Research Institution", "domain": "ai", "domain_name": "Ai", "definition": "A research institution is an organisation whose primary mission is the systematic production of new knowledge through investigation, experimentation and scholarship. It encompasses universities, dedicated research laboratories, government agencies and independent institutes that employ researchers, secure funding, and disseminate findings via peer-reviewed publication. Research institutions provide the infrastructure, governance and intellectual community within which scientific and technical advances are pursued and validated.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:research-institution", "labels": [ "Research Institution" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "research-layer", "title": "Research Layer", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The Research Layer is the cross-cutting stratum where new methods, models, and understanding are generated before adoption into production strata. It sits above evaluation and experimentation concerns and feeds the algorithm and model layers with validated advances. It contains experiments, hypotheses, prototypes, and the findings that justify changes elsewhere.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:research-layer", "labels": [ "Research Layer" ], "is_subclass_of": [ "Workspace Tools", "owl:Thing" ], "wikilinks": [ "Evaluation Layer", "Simulation Layer", "Algorithm Layer", "Model Architecture Layer", "Scientific Method", "Reproducibility", "owl:Thing" ] }, { "id": "research-methods", "title": "Research Methods", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Research Methods are the systematic procedures and techniques used to collect, analyse, and interpret data in order to generate reliable knowledge. They encompass qualitative approaches (interviews, ethnography), quantitative approaches (controlled experiments, statistical analysis), and mixed-methods designs, with evaluation benchmarks and user studies being particularly relevant to AI and spatial computing research.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:research-methods", "labels": [ "Research Methods", "Empirical Research Methodology", "Observational Methods", "Research Protocol" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "research-university", "title": "Research University", "domain": "ai", "domain_name": "Ai", "definition": "A research university is a higher-education institution whose mission combines teaching with the production of original scholarship, conducting basic and applied research across disciplines and training postgraduate researchers. Such universities operate laboratories, secure competitive grant funding, publish peer-reviewed scholarship, and frequently anchor regional innovation ecosystems through technology transfer. They are central nodes in the pipeline that converts fundamental discovery into trained talent and commercialisable knowledge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:research-university", "labels": [ "Research University" ], "is_subclass_of": [ "Research Institution" ], "wikilinks": [] }, { "id": "research-and-development", "title": "Research and Development", "domain": "ai", "domain_name": "Ai", "definition": "Research and development (R&D) is the systematic, investigative activity undertaken to create new knowledge and to apply it in novel products, processes and services. It spans basic research that expands understanding, applied research that targets specific problems, and experimental development that turns findings into working technology. R&D is the principal engine of innovation and long-term competitive advantage in technology-intensive fields.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:research-and-development", "labels": [ "Research and Development" ], "is_subclass_of": [ "Innovation" ], "wikilinks": [] }, { "id": "reserve-asset", "title": "Reserve Asset", "domain": "finance", "domain_name": "Finance", "definition": "An asset held to back a liability or to provide a store of value, such as the holdings that back a stablecoin or the foreign currency reserves held by a central bank. Its purpose is to ensure that claims can be met.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:reserve-asset", "labels": [ "Reserve Asset", "Reserve Token" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [ "Stablecoin", "Monetary Policy" ] }, { "id": "reserve-requirements", "title": "Reserve Requirements", "domain": "finance", "domain_name": "Finance", "definition": "Reserve requirements are central-bank rules obliging commercial banks to hold a minimum fraction of their deposit liabilities as reserves, either as vault cash or as balances at the central bank. By constraining the proportion of deposits banks may lend, they influence credit creation, money supply and bank liquidity. They are a classic monetary-policy instrument, though many modern central banks now rely more on interest-rate tools.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reserve-requirements", "labels": [ "Reserve Requirements" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "residual-connection", "title": "Residual Connection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network connection that adds the input of a layer directly to its output, forming a skip connection that enables stable gradient flow during backpropagation in very deep architectures. Residual connections mitigate the vanishing gradient problem and allow networks of hundreds of layers to be trained effectively, forming a foundational component of ResNet and Transformer architectures.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:residual-connection", "labels": [ "Residual Connection", "Skip Connection" ], "is_subclass_of": [ "Neural Network Component" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "residual-network", "title": "Residual Network", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A deep neural network architecture introduced by He et al. (2016) that employs skip connections to allow gradients to propagate directly through layers, expressed as H(x) = F(x) + x. This residual formulation resolves the vanishing gradient problem in very deep networks, enabling training of architectures with hundreds of layers and achieving state-of-the-art performance on image recognition benchmarks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:residual-network", "labels": [ "Residual Network" ], "is_subclass_of": [ "AI Technique", "Deep Learning" ], "wikilinks": [ "MetaverseDomain", "Update Cycle" ] }, { "id": "resilience-engineering", "title": "Resilience Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Resilience engineering is the discipline concerned with how complex socio-technical systems sustain required operation under expected and unexpected disturbances, and how they adapt, degrade gracefully, and recover. Rather than treating failure solely as a deviation to be eliminated, it studies the adaptive capacities that let systems anticipate, monitor, respond to, and learn from disruption. In computing it informs architectures and operational practices that bend rather than break under load spikes, partial outages, and cascading faults.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:resilience-engineering", "labels": [ "Resilience Engineering" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [] }, { "id": "resilience-metric", "title": "Resilience Metric", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Resilience metrics are quantitative and qualitative measurements that assess the robustness, fault tolerance, and recovery capabilities of digital and physical infrastructure. Core indicators include availability percentage, Recovery Time Objective (RTO), Recovery Point Objective (RPO), Mean Time Between Failures (MTBF), and Mean Time to Repair (MTTR). Effective resilience measurement combines real-time monitoring, predictive analytics, and automated remediation to maintain service quality under adverse conditions and verify SLA compliance.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:resilience-metric", "labels": [ "Resilience Metric" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "BlockchainNetworks", "IncidentResponse", "MetaversePlatforms", "RecoveryTimeObjective", "SLAVerification", "CloudInfrastructure", "MetaverseDomain" ] }, { "id": "resilience", "title": "Resilience", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The capacity of a system, network, or organisation to anticipate, withstand, recover from, and adapt to adverse conditions, attacks, or failures whilst maintaining essential functions. In distributed and blockchain contexts, resilience is achieved through decentralisation, redundancy, byzantine fault tolerance, and adaptive response mechanisms.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:resilience", "labels": [ "Resilience" ], "is_subclass_of": [ "Software Engineering", "Blockchain" ], "wikilinks": [ "Blockchain" ] }, { "id": "resistojet", "title": "Resistojet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:resistojet", "labels": [ "Resistojet" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "resolution-bucketing", "title": "Resolution Bucketing", "domain": "ai", "domain_name": "Ai", "definition": "Resolution bucketing is a training-data preparation technique for diffusion and other image-generation models that groups training images into a fixed set of aspect-ratio and size buckets rather than forcing every image to a single square resolution. Each batch is drawn from a single bucket so images share dimensions, avoiding distortion from cropping or stretching while preserving GPU batching efficiency. It improves fidelity for non-square data and is a standard step in fine-tuning pipelines such as Kohya-based DreamBooth and LoRA training.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:resolution-bucketing", "labels": [ "Resolution Bucketing" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "resolution-test-chart", "title": "Resolution Test Chart", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Standardized calibration patterns such as ISO 12233 charts used to measure and validate the optical resolution, colour accuracy, and image quality of displays in virtual reality and augmented reality headsets through computational analysis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:resolution-test-chart", "labels": [ "Resolution Test Chart" ], "is_subclass_of": [ "Display and Rendering", "Calibration Tools" ], "wikilinks": [ "Display Calibration", "Calibration Tools", "metaverse" ] }, { "id": "resolver", "title": "Resolver", "domain": "standards", "domain_name": "Standards", "definition": "A resolver is a component that takes an identifier and returns the resource, address, or value it stands for. In identification infrastructure such as GS1 Digital Link and the Domain Name System, a resolver service receives a request containing a structured identifier and redirects the requester to one of potentially many linked resources \u2014 product information, traceability records, or network addresses \u2014 chosen by link type and context. The term also names the electromechanical rotary transformer used in robotics and motor control to resolve a shaft's absolute angular position from induced analogue signals.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:resolver", "labels": [ "Resolver" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [ "Middleware", "Gs1 Digital Link", "Proprioceptive Sensor", "Domain Name System", "Interoperability", "Rotary Encoder" ] }, { "id": "resource-allocation", "title": "Resource Allocation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Resource allocation is the economic and computational mechanism by which scarce resources, such as compute, capital, bandwidth, or stake, are distributed among competing uses or participants. In market and protocol contexts it determines who receives what and on what terms, often via pricing, auctions, or scheduling policies. Efficient allocation underpins value transfer and incentive alignment across economic and distributed systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:resource-allocation", "labels": [ "Resource Allocation", "Dynamic Resource Allocation", "Resource Allocation Unit" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "resource-efficiency", "title": "Resource Efficiency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Resource efficiency is the practice of producing the same output or value while consuming fewer material, energy, or computational inputs. It encompasses minimising waste, reusing by-products, and optimising processes so that resources deliver maximum useful work. In sustainability and industrial contexts it is a primary lever for reducing cost and environmental footprint.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:resource-efficiency", "labels": [ "Resource Efficiency", "Optimized Resource Utilization", "Resource Utilisation", "Resource-Efficient Training" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "resource-management", "title": "Resource Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Resource Management is the systematic discipline of allocating, scheduling, monitoring, and optimising computational and physical resources \u2014 including CPU, GPU, memory, storage, and network bandwidth \u2014 across applications, services, and infrastructure to ensure efficient utilisation, quality-of-service guarantees, and graceful degradation under load. It encompasses the full asset lifecycle from provisioning and pooling through dynamic scaling to decommissioning, balancing competing workload demands against capacity constraints. In distributed and cloud-native environments, resource management extends to orchestration of containerised workloads, quota enforcement, cost attribution, and autoscaling policies that respond to real-time demand signals. Effective resource management is a foundational prerequisite for reliable, cost-efficient, and performant infrastructure at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:resource-management", "labels": [ "Resource Management", "Resource Manager" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [] }, { "id": "response-time-prediction", "title": "Response Time Prediction", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Response Time Prediction applies statistical and machine learning models to forecast the end-to-end latency of requests in networked systems, enabling proactive quality-of-service management and resource scheduling. Inputs typically include historical latency distributions, network conditions, server load, and request characteristics; outputs drive adaptive scheduling, pre-emptive caching, and SLA alerting. Accurate prediction is critical for latency-sensitive applications such as real-time XR streaming, interactive robotics teleoperation, and cloud-gaming platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:response-time-prediction", "labels": [ "Response Time Prediction" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "responsible-ai-deployment", "title": "Responsible AI Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Responsible AI Deployment is the practice of releasing artificial-intelligence systems into production in ways that manage safety, fairness, transparency, and accountability risks throughout the system's operational lifecycle. It includes staged rollouts, monitoring for drift and harm, documentation such as model cards, and mechanisms for human oversight and rollback. It operationalises broader responsible-AI principles at the point where a model meets real users.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:responsible-ai-deployment", "labels": [ "Responsible AI Deployment" ], "is_subclass_of": [ "Responsible AI" ], "wikilinks": [] }, { "id": "responsible-ai-principles", "title": "Responsible AI Principles", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Foundational ethical commitments and normative guidelines governing AI system design, development, deployment, and monitoring to ensure beneficial, fair, and rights-respecting outcomes. Core principles\u2014including fairness, transparency, accountability, privacy, safety, and human oversight\u2014are operationalised through frameworks such as the OECD AI Principles, the EU Ethics Guidelines for Trustworthy AI, and UNESCO's Recommendation on the Ethics of AI.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:responsible-ai-principles", "labels": [ "Responsible AI Principles" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "EU HLEG AI", "OECD AI Principles", "UNESCO Recommendation on AI Ethics", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "responsible-ai", "title": "Responsible AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The practice of designing, developing, deploying, and operating artificial intelligence systems with explicit attention to their societal impacts, ethical implications, and potential harms, incorporating accountability mechanisms, stakeholder engagement, risk management, transparency, and governance throughout the AI lifecycle to ensure that AI systems are developed and used in ways that benefit individuals and society whilst minimising negative consequences, respecting human rights and democratic values, and maintaining clear lines of responsibility for AI-driven outcomes.", "entityType": "Class", "qualityScore": 0.65, "maturity": "draft", "iri": "urn:ngm:class:responsible-ai", "labels": [ "Responsible AI", "Responsible AI (AI-0033)" ], "is_subclass_of": [ "AI Governance" ], "wikilinks": [ "risk management processes", "social licence for AI", "Trustworthy AI", "Accountability", "AI Audit", "AI Governance", "AI Impact Assessment", "AI Risk Management", "Ethical AI", "Fairness", "Human Oversight", "MetaverseDomain", "Transparency" ] }, { "id": "responsible-deployment", "title": "Responsible Deployment", "domain": "ai", "domain_name": "Ai", "definition": "Responsible deployment is the disciplined release of AI systems into production with safeguards that manage risk to users and society throughout the system lifecycle. It involves staged rollout, monitoring, impact assessment, access controls, and the ability to roll back or restrict capabilities when harms emerge. It operationalises governance principles at the point where models meet real users.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:responsible-deployment", "labels": [ "Responsible Deployment" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "responsible-scaling-policy", "title": "responsible scaling policy", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Responsible Scaling Policy (RSP) is a voluntary self-regulatory commitment by a frontier AI developer that ties further capability scaling and deployment to measurable safety milestones. It specifies: the capability evaluation thresholds (often called AI Safety Levels) at which enhanced assessments must be completed before training or deploying a more capable model; the specific risk domains\u2014particularly uplift to chemical, biological, radiological, nuclear, and cyberweapon development\u2014that would trigger deployment restrictions or halts; and the technical and organisational safeguards required at each level. RSPs translate aspirational safety intentions into concrete, time-bound, falsifiable commitments, creating accountability mechanisms that operate ahead of formal government regulation.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:responsible-scaling-policy", "labels": [ "Responsible Scaling Policy", "Responsible Scaling Policies" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "restricted-three-body-problem", "title": "Restricted Three-body Problem", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:restricted-three-body-problem", "labels": [ "Restricted Three-body Problem" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "retail-cbdc", "title": "Retail CBDC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A retail central bank digital currency (CBDC) is a digital form of sovereign money issued directly by a central bank for use by the general public in everyday payments. Unlike wholesale CBDCs, which serve interbank settlement, retail CBDCs target consumers and merchants and may use account-based or token-based designs over distributed or centralised ledgers. They aim to provide a public digital payment instrument with central-bank credit safety.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:retail-cbdc", "labels": [ "Retail CBDC", "Retail CBDC Issuance" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "retention-policy", "title": "Retention Policy", "domain": "data", "domain_name": "Data", "definition": "A retention policy is a set of governance rules that specify how long data records are kept and when they are archived or permanently deleted. It encodes legal, regulatory, and operational requirements such as minimum retention for audit and maximum retention for privacy compliance. Automated enforcement ensures consistent lifecycle management across storage systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:retention-policy", "labels": [ "Retention Policy", "Data Retention Policy" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "retraining", "title": "Retraining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Retraining is the process of updating a deployed machine learning model's parameters using new or additional data, typically triggered when monitoring detects data drift or concept drift that degrades predictive performance. It may involve fine-tuning the existing model on recent data or training a fresh model from scratch on an updated dataset. Retraining cadence and triggers are core concerns of MLOps pipelines that keep production models aligned with the current data distribution.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:retraining", "labels": [ "Retraining" ], "is_subclass_of": [ "Model Training" ], "wikilinks": [] }, { "id": "retrieval-augmented-generation-rag", "title": "Retrieval Augmented Generation - RAG", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Retrieval Augmented Generation (RAG) is a neural architecture paradigm that augments large language model generation with dynamic retrieval of non-parametric external knowledge at inference time, enabling factually grounded, up-to-date, and citation-traceable responses without retraining model we...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:retrieval-augmented-generation-rag", "labels": [ "Retrieval Augmented Generation - RAG" ], "is_subclass_of": [ "AI Application", "Natural Language Processing", "Neural Information Retrieval", "Generative AI", "Knowledge-Intensive NLP", "Open Domain Question Answering", "Large-Scale Pretrained Foundation Model" ], "wikilinks": [ "Agentic RAG", "AlgorithmLayer", "Approximate Nearest Neighbour Search", "ARES Framework", "BEIR Benchmark", "BM25", "Chunking Strategy", "Citation Generation", "Code Navigation", "ColBERTv2", "Contextual Retrieval", "Corpus Preprocessing", "Customer Support Automation", "Dense Passage Retrieval", "Document Store", "Domain Adaptation Without Fine-Tuning", "Embedding Model", "Enterprise Search", "Factual Grounding", "Generator" ] }, { "id": "retrieval-augmented-generation", "title": "retrieval-augmented generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Retrieval-Augmented Generation (RAG) is an AI inference architecture that augments large language model generation by dynamically retrieving semantically relevant passages from an external knowledge store at query time, concatenating them into the model context window before the response is produced. A retriever component\u2014typically a dense bi-encoder backed by a vector database\u2014embeds both the query and document corpus into a shared latent space and selects the top-k most similar chunks via approximate nearest-neighbour search. This non-parametric memory mechanism decouples factual knowledge from frozen model weights, dramatically reducing hallucination rates, enabling post-deployment knowledge updates without retraining, and providing fine-grained source attribution for generated claims. RAG has become the dominant architectural pattern for enterprise knowledge-intensive natural language processing applications, including question answering, customer support, legal research, and medical information retrieval.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:retrieval-augmented-generation", "labels": [ "Retrieval-Augmented Generation", "RAG", "Retrieval Augmented Generation", "Retrieval-Augmented LLM" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "retroactive-public-goods-funding", "title": "retroactive public goods funding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Retroactive Public Goods Funding (RetroPGF) is a capital allocation mechanism that rewards contributors to open-source software, protocol infrastructure, or public-benefit projects after they have demonstrably created value, rather than distributing speculative upfront grants based on predicted future impact. By funding proven outcomes rather than promises, RetroPGF reduces misallocation risk and aligns incentives so that contributors can invest effort in public goods with confidence that eventual recognition is possible. Evaluation is typically performed by a trusted body \u2014 such as a DAO committee, a citizen house of badge holders, or an expert panel \u2014 who assess demonstrated impact and distribute treasury funds accordingly. The mechanism was pioneered by Vitalik Buterin and operationalised at scale by the Optimism Collective, making it one of the most influential experiments in decentralised public goods economics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:retroactive-public-goods-funding", "labels": [ "Retroactive Public Goods Funding", "Optimism Retroactive Public Goods Funding" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "retrospective", "title": "Retrospective", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A retrospective is a recurring, facilitated meeting in which a team reflects on a recent period of work to identify what went well, what did not, and concrete improvements to try next. It is a cornerstone agile practice that institutionalises continuous improvement by turning shared experience into actionable changes to process, tooling and collaboration. Retrospectives are increasingly run on collaborative whiteboards so distributed teams can participate equally.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:retrospective", "labels": [ "Retrospective" ], "is_subclass_of": [ "Agile Software Development" ], "wikilinks": [] }, { "id": "retry-with-backoff", "title": "Retry with Backoff", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A resilience strategy for handling transient failures in which a failed operation is attempted again after a deliberately growing delay, rather than immediately or at a fixed interval. Each successive retry waits longer \u2014 commonly the delay doubles \u2014 so that a system experiencing a temporary fault or overload is given increasing time to recover and is not hammered by a tight loop of identical requests. Randomised jitter is usually added to the delay so that many clients failing at once do not synchronise their retries into repeated coordinated bursts, and a cap on attempts or total elapsed time prevents the strategy from waiting forever on a failure that is not in fact transient.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:retry-with-backoff", "labels": [ "Retry with Backoff" ], "is_subclass_of": [ "Fallback" ], "wikilinks": [ "Fallback", "Rollback", "Guardrail", "AgenticWorkflow" ] }, { "id": "reusable-content-template-scaffolds", "title": "Reusable Content Template Scaffolds", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Templates are reusable structural scaffolds in software engineering and knowledge management that define a fixed format or pattern into which variable content is inserted. They reduce repetition, enforce consistency, and accelerate authoring across code generation, documentation, and prompt engineering workflows.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:reusable-content-template-scaffolds", "labels": [ "Reusable Content Template Scaffolds", "Templates" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "revenue-distribution", "title": "Revenue Distribution", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Revenue Distribution is the mechanism by which income generated by a platform, protocol, or creative ecosystem is allocated among contributors, token holders, and operational costs. In Web3 and spatial computing contexts, it is commonly encoded in smart contracts via tokenomics rules, protocol fee splits, and DAO-governed treasury allocations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:revenue-distribution", "labels": [ "Revenue Distribution", "Protocol Revenue Distribution", "Revenue Distribution Logic" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "revenue-sharing", "title": "Revenue Sharing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Blockchain-enabled distribution mechanisms that automatically allocate earnings between creators, platforms, and stakeholders through smart contracts, with NFT royalties providing perpetual income streams for digital asset creators in metaverse economies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:revenue-sharing", "labels": [ "Revenue Sharing" ], "is_subclass_of": [ "Content and Assets", "Creator Economy" ], "wikilinks": [ "Creator Monetization", "Creator Economy", "metaverse" ] }, { "id": "revenue-threshold", "title": "Revenue Threshold", "domain": "governance", "domain_name": "Governance", "definition": "A revenue threshold is a regulatory trigger that applies obligations only to entities whose annual revenue exceeds a defined monetary level. In AI legislation such thresholds scope rules to large developers, exempting smaller firms to reduce compliance burden while targeting actors with the greatest capacity and impact. The threshold value is a key parameter determining who falls within a law's reach.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:revenue-threshold", "labels": [ "Revenue Threshold" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "reverse-engineering", "title": "Reverse Engineering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Reverse engineering is the process of analysing an existing physical object, system, or software artefact to infer its design, structure, or manufacturing intent without access to original specifications. In spatial computing it commonly means recovering a parametric or mesh-based CAD model from scanned point-cloud or structured-light capture, enabling downstream editing and manufacture. It is also applied to software and hardware to recover behaviour or protocol specifications from observed operation.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:reverse-engineering", "labels": [ "Reverse Engineering" ], "is_subclass_of": [ "3D Reconstruction" ], "wikilinks": [] }, { "id": "reverse-logistics", "title": "Reverse Logistics", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Reverse logistics encompasses all operations related to the reuse, processing, and management of products and materials flowing backward through the supply chain from the consumer or end-of-life point toward the manufacturer or recycler. It includes product returns, refurbishment, remanufacturing, recycling, and responsible disposal, and is a key operational component of circular economy strategies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reverse-logistics", "labels": [ "Reverse Logistics" ], "is_subclass_of": [ "Logistics" ], "wikilinks": [] }, { "id": "reverse-proxy", "title": "Reverse Proxy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A reverse proxy is a server that sits in front of one or more backend servers and forwards client requests to them, presenting a single entry point to the outside world. It terminates client connections, can offload TLS, cache responses, compress payloads and apply access controls before relaying traffic. By decoupling clients from backends it improves security, scalability and operational flexibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reverse-proxy", "labels": [ "Reverse Proxy" ], "is_subclass_of": [ "Network Architecture" ], "wikilinks": [] }, { "id": "revisit-time", "title": "Revisit Time", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:revisit-time", "labels": [ "Revisit Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "revocation-mechanism", "title": "Revocation Mechanism", "domain": "security", "domain_name": "Security", "definition": "A revocation mechanism is a cryptographic or registry-based procedure for invalidating a previously issued credential, certificate, or access token before its natural expiry, allowing issuers to withdraw trust following compromise, policy change, or holder misconduct. In digital identity systems, revocation mechanisms range from Certificate Revocation Lists and OCSP for X.509 certificates to Bitstring Status Lists and cryptographic accumulators for verifiable credentials. The privacy characteristics and scalability of each approach differ substantially, making mechanism selection a critical design decision.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:revocation-mechanism", "labels": [ "Revocation Mechanism", "Instant Revocation" ], "is_subclass_of": [ "Cryptographic Infrastructure" ], "wikilinks": [] }, { "id": "revocation-registry", "title": "Revocation Registry", "domain": "security", "domain_name": "Security", "definition": "A revocation registry is a data structure or service that maintains the validity status of issued verifiable credentials or digital certificates, allowing verifiers to check whether a credential has been revoked by its issuer before accepting a presentation. In traditional PKI systems this role is fulfilled by Certificate Revocation Lists (CRLs) and OCSP responders; in self-sovereign identity ecosystems, revocation registries are implemented as privacy-preserving mechanisms including W3C Status List 2021 (bitstring-based), Hyperledger AnonCreds revocation with cryptographic accumulators, and on-chain smart contract registries. A well-designed revocation registry balances timely status updates, verifier privacy (preventing issuers from tracking when credentials are checked), and scalability to large credential populations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:revocation-registry", "labels": [ "Revocation Registry" ], "is_subclass_of": [ "Credential Verification" ], "wikilinks": [] }, { "id": "revolut", "title": "Revolut", "domain": "finance", "domain_name": "Finance", "definition": "Revolut is a British financial technology company offering banking, payments, currency exchange and investment services through a mobile app. It holds banking licences in several jurisdictions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:revolut", "labels": [ "Revolut" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [ "Payment System", "Cryptocurrency", "Financial Services" ] }, { "id": "reward-distribution", "title": "Reward Distribution", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Tokenomic mechanisms that calculate and allocate staking rewards, validator incentives, and participation benefits within blockchain networks through mathematically designed emission schedules and fee distribution protocols.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:reward-distribution", "labels": [ "Reward Distribution", "Reward Distribution Mechanism" ], "is_subclass_of": [ "Platform and Environment", "Tokenomics" ], "wikilinks": [ "Network Participation", "metaverse", "Tokenomics" ] }, { "id": "reward-function", "title": "Reward Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A mathematical formulation in reinforcement learning that maps state-action pairs to scalar values, guiding AI agent behaviour toward desired outcomes through feedback signals; central to policy optimisation, agent training, and objective specification in machine learning systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:reward-function", "labels": [ "Reward Function" ], "is_subclass_of": [ "Content and Assets", "Reinforcement Learning" ], "wikilinks": [ "Autonomous Agent Learning", "metaverse", "Reinforcement Learning" ] }, { "id": "reward-hacking", "title": "Reward Hacking", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Reward hacking is a failure mode in which a reinforcement-learning agent achieves high reward by exploiting flaws, loopholes or proxies in its reward function rather than accomplishing the intended task. Because the reward is only an imperfect proxy for the designer's true objective, an optimiser may discover unintended behaviours that maximise the measured reward while violating the spirit of the goal. It is a central concern in AI safety and alignment and is closely related to specification gaming.", "entityType": "Class", "qualityScore": 0.78, "maturity": "emerging", "iri": "urn:ngm:class:reward-hacking", "labels": [ "Reward Hacking" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "reward-model", "title": "Reward Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A neural network trained to predict scalar rewards for model outputs based on human feedback, used to provide learning signals in reinforcement learning from human feedback (RLHF). The reward model serves as a proxy for human preferences, enabling efficient optimisation without constant human evaluation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:reward-model", "labels": [ "Reward Model", "Outcome Reward Model" ], "is_subclass_of": [ "Reinforcement Learning", "Reward Modelling" ], "wikilinks": [ "Death of the Internet", "Direct Preference Optimization", "Google", "MetaverseDomain", "Proprietary Large Language Models" ] }, { "id": "reward-modelling", "title": "Reward Modelling", "domain": "ai", "domain_name": "Ai", "definition": "Reward modelling is a machine-learning technique that trains a separate model to predict human preferences and use its scores as a reward signal for optimising another agent. It underpins reinforcement learning from human feedback, where the reward model ranks candidate outputs so a policy can be tuned toward preferred behaviour. It is central to aligning large language models with human values and intent.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:reward-modelling", "labels": [ "Reward Modelling", "Rule-Based Reward Modelling" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "reward-shaping", "title": "Reward Shaping", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Reward shaping is a reinforcement-learning technique that augments an environment's native reward signal with additional intermediate rewards to guide and accelerate learning. Potential-based reward shaping provides theoretical guarantees that the optimal policy is preserved, avoiding the introduction of unintended behaviours. It is commonly used to address sparse-reward problems where useful feedback is rare.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reward-shaping", "labels": [ "Reward Shaping" ], "is_subclass_of": [ "Reinforcement Learning", "Reinforcement Learning Algorithm" ], "wikilinks": [] }, { "id": "reward-signal", "title": "Reward Signal", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A reward signal is the scalar feedback an agent receives from its environment in reinforcement learning, indicating the immediate desirability of the state-action pair just experienced. It is the primary mechanism by which goals are communicated to a learning agent, which seeks to maximise the cumulative reward it accumulates over time rather than any single immediate value. The design of the reward signal strongly shapes learned behaviour, and poorly specified rewards can lead to unintended or degenerate strategies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:reward-signal", "labels": [ "Reward Signal" ], "is_subclass_of": [ "Reinforcement Learning", "AI Technique" ], "wikilinks": [] }, { "id": "rfc-8032", "title": "Rfc 8032", "domain": "security", "domain_name": "Security", "definition": "RFC 8032 is the IETF specification that standardises the Edwards-curve Digital Signature Algorithm (EdDSA), defining the Ed25519 and Ed448 signature schemes. It specifies deterministic signature generation, key formats, and verification procedures over twisted Edwards curves, eliminating the dependence on a per-signature random number that has historically led to catastrophic key-recovery failures in other schemes. The standard is widely adopted across TLS, SSH, DNSSEC, and cryptocurrency systems for its strong security, high performance, and resistance to common implementation pitfalls.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rfc-8032", "labels": [ "Rfc 8032", "RFC 8032" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "right-ascension-of-the-ascending-node", "title": "Right Ascension of the Ascending Node", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:right-ascension-of-the-ascending-node", "labels": [ "Right Ascension of the Ascending Node" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "right-ascension", "title": "Right Ascension", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:right-ascension", "labels": [ "Right Ascension" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "right-to-erasure", "title": "Right To Erasure", "domain": "governance", "domain_name": "Governance", "definition": "The right to erasure, also known as the right to be forgotten, is a data-subject right under the UK GDPR and EU GDPR that allows individuals to require a controller to delete their personal data in defined circumstances. It applies where data is no longer necessary, consent is withdrawn, processing is unlawful, or the data subject objects without overriding legitimate grounds. The right is qualified by exemptions such as freedom of expression, legal obligations, and the public interest.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:right-to-erasure", "labels": [ "Right To Erasure", "Right to Erasure" ], "is_subclass_of": [ "Data Subject Rights" ], "wikilinks": [] }, { "id": "right-to-be-forgotten", "title": "Right to Be Forgotten", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A privacy right framework enabling individuals to request deletion or removal of personal data from online platforms and databases, with verification and audit mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:right-to-be-forgotten", "labels": [ "Right to Be Forgotten", "Right to Delete", "Right to be Forgotten" ], "is_subclass_of": [ "Legal and Regulatory", "Privacy Framework" ], "wikilinks": [ "CCPA", "Compliance Reporting", "Content Removal", "Data Controller", "Data Erasure", "Data Inventory", "Data Processor", "Data Protection Framework", "Deletion Request", "Erasure Verification", "GDPR Article 17", "Privacy Policy", "Privacy Rights System", "ApplicationLayer", "Audit Trail", "Consent Management", "Identity Verification", "Regulatory Compliance", "TrustAndGovernanceDomain", "User Privacy Control" ] }, { "id": "rights-protection", "title": "Rights Protection", "domain": "governance", "domain_name": "Governance", "definition": "Rights protection encompasses the institutional, legal, and technical mechanisms through which individual and collective entitlements\u2014civil liberties, digital rights, property rights, privacy rights, and human rights\u2014are identified, defended, and remedied when violated. It spans constitutional and statutory safeguards, judicial and administrative enforcement, technical privacy-by-design measures, and advocacy infrastructure. In digital and AI contexts, rights protection addresses algorithmic discrimination, surveillance overreach, automated decision-making without human review, and the erosion of informational self-determination. Effective rights protection requires both upstream prevention and downstream accountability pathways.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rights-protection", "labels": [ "Rights Protection" ], "is_subclass_of": [ "Human Rights" ], "wikilinks": [] }, { "id": "rigid-body-dynamics", "title": "Rigid Body Dynamics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Rigid Body Dynamics is the branch of classical mechanics that models solid objects as perfectly non-deformable, computing their translational and rotational motion under applied forces and torques using Newton-Euler equations or Lagrangian formulations. It addresses the full six-degree-of-freedom motion state \u2014 position, orientation, linear velocity, and angular velocity \u2014 and resolves contact constraints through collision detection, impulse resolution, and constraint solvers. The field underpins real-time simulation in game engines, robot motion planning, spacecraft attitude control, and extended-reality environments. Key numerical methods include symplectic Euler integration, Runge-Kutta schemes, and position-based dynamics for stable, interactive-rate simulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:rigid-body-dynamics", "labels": [ "Rigid Body Dynamics", "Rigid Body Mechanics", "Rigid Body Robotics" ], "is_subclass_of": [ "Physics Simulation" ], "wikilinks": [] }, { "id": "rigid-body", "title": "Rigid Body", "domain": "robotics", "domain_name": "Robotics", "definition": "A Rigid Body is an idealised mechanical object whose internal distances between all constituent points remain constant regardless of applied forces, making it the foundational abstraction for classical mechanics, robotic kinematics, and physics simulation. In robotics, links of a kinematic chain are modelled as rigid bodies connected by joints; their mass, centre of gravity, and inertia tensor parameterise the dynamics equations used for motion planning, control, and simulation. Rigid body assumptions break down for flexible or soft-robotic systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:rigid-body", "labels": [ "Rigid Body", "RigidBody" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "ring-signature", "title": "Ring Signature", "domain": "security", "domain_name": "Security", "definition": "A ring signature is a digital signature produced by one member of a group such that verifiers learn the signature came from the group but cannot identify which member signed. It provides signer anonymity within an ad hoc set.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ring-signature", "labels": [ "Ring Signature", "RingCT Protocol" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "Cryptography", "Pseudonymity", "Zero-Knowledge Proof" ] }, { "id": "riot-platforms", "title": "Riot Platforms", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Riot Platforms is a United States company that operates large-scale Bitcoin mining facilities. It is listed on a public stock exchange.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:riot-platforms", "labels": [ "Riot Platforms" ], "is_subclass_of": [ "Bitcoin Mining" ], "wikilinks": [ "ASIC", "Transaction Validation", "Bitcoin Network", "Bitcoin Mining", "https://www.riotplatforms.com", "https://www.riotplatforms.com/investors" ] }, { "id": "risk-analysis", "title": "Risk Analysis", "domain": "governance", "domain_name": "Governance", "definition": "Risk analysis is the systematic process of identifying, characterising and estimating the likelihood and impact of events that could threaten objectives, in order to inform decisions about how to treat them. It combines qualitative judgement with quantitative techniques such as scenario modelling and probabilistic simulation to express uncertainty in actionable terms. As a core component of risk management, it produces the evidence base for prioritising controls, allocating capital and setting tolerances.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:risk-analysis", "labels": [ "Risk Analysis" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "risk-assessment-engine", "title": "Risk Assessment Engine", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "A risk assessment engine is a software component that automatically evaluates the risk associated with transactions, shipments, or entities by applying rules, statistical models, or machine learning to available data. In customs and trade systems it scores consignments to prioritise inspection while expediting low-risk flows. It enables consistent, scalable, and auditable risk-based decision making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-assessment-engine", "labels": [ "Risk Assessment Engine" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "risk-assessment-matrix", "title": "Risk Assessment Matrix", "domain": "security", "domain_name": "Security", "definition": "A risk assessment matrix is a structured grid that ranks risks by cross-tabulating the likelihood of an event against the severity of its impact. Each cell maps to a qualitative or quantitative risk level, allowing teams to prioritise mitigation and allocate resources to the highest-rated risks. It is a standard visual tool in security and safety risk management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-assessment-matrix", "labels": [ "Risk Assessment Matrix" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "risk-assessment-methodology", "title": "Risk Assessment Methodology", "domain": "governance", "domain_name": "Governance", "definition": "A Risk Assessment Methodology is a structured, repeatable set of procedures for identifying, analysing, and evaluating potential hazards or threats to an organisation's objectives, assets, or stakeholders, producing a prioritised risk register that informs mitigation decisions. Methodologies range from qualitative (likelihood-impact matrices) to quantitative (Monte Carlo simulation, fault tree analysis) approaches, and are codified in standards such as ISO 31000, NIST SP 800-30, and FAIR for cybersecurity contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:risk-assessment-methodology", "labels": [ "Risk Assessment Methodology" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "risk-assessment", "title": "Risk Assessment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Risk Assessment is the systematic, evidence-based process of identifying, analysing, and evaluating hazards, vulnerabilities, and adverse-outcome chains across complex sociotechnical systems, forming the core analytical activity within broader Risk Management programmes aligned to", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:risk-assessment", "labels": [ "Risk Assessment", "Ethical Risk Assessment", "Hazard and Risk Assessment", "Risk Assessment (AI-0079)", "Risk Assessment Module", "Risk Assessment Procedure" ], "is_subclass_of": [ "Legal and Regulatory", "Software Engineering", "Risk Management", "Systems Analysis", "Compliance Framework", "Safety Engineering", "Audit" ], "wikilinks": [ "AI Incident Database", "AI Safety Institute", "AISafetyDomain", "Anthropic", "Attack Trees", "Audit", "Basel III", "Bayesian Inference", "Behavioural Science", "Bow-Tie Analysis", "ComplianceLayer", "DeFi Risk", "Decentralised Finance", "Fault Tree Analysis", "FCA PS21/3", "FinancialRegulatoryDomain", "Financial Stability", "FMEA", "FMEA", "FSB" ] }, { "id": "risk-based-regulation", "title": "Risk Based Regulation", "domain": "governance", "domain_name": "Governance", "definition": "Risk-Based Regulation is a regulatory methodology that calibrates the intensity of oversight, compliance requirements, and enforcement action to the assessed level of risk posed by regulated entities or activities. Rather than applying uniform rules to all actors, risk-based approaches tier obligations by factors such as likelihood of harm, severity of potential impact, and the capacity of regulated parties to manage risk. It is the foundational approach of the EU AI Act, financial services regulation, and many modern safety frameworks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:risk-based-regulation", "labels": [ "Risk Based Regulation", "Risk-Based Regulation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "risk-classification", "title": "Risk Classification", "domain": "governance", "domain_name": "Governance", "definition": "The systematic assignment of activities, systems, or products into ordered categories of risk severity so that regulatory obligations, controls, and oversight can be proportioned to potential harm. It combines criteria such as likelihood, impact, affected populations, and context of use into defined tiers, exemplified by the EU AI Act's four-level scheme running from prohibited unacceptable-risk practices through high-risk conformity requirements to limited-risk transparency duties and minimal-risk freedom.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:risk-classification", "labels": [ "Risk Classification" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [ "Risk Assessment", "Risk Based Regulation", "Policy Framework", "Regulatory Framework", "AI Governance Law and Privacy" ] }, { "id": "risk-intensity-heatmap", "title": "Risk Intensity Heatmap", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A heatmap is a data visualisation technique that encodes numerical values as colour intensities in a two-dimensional matrix, allowing rapid visual comparison of magnitudes across categorical dimensions. In the context of AI risk and impact analysis, heatmaps are used to compare urgency, impact, and composite severity scores across AI risk categories (such as security risks, algorithmic bias, job automation, and AI ethics), guiding prioritisation for governance and regulation. Heatmaps are typically rendered with colour-scheme gradients (e.g., Vega-Lite's blues scheme) and can be embedded directly in knowledge-graph pages for interactive analysis.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:risk-intensity-heatmap", "labels": [ "Risk Intensity Heatmap", "heatmap" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "risk-management-framework", "title": "Risk Management Framework", "domain": "governance", "domain_name": "Governance", "definition": "A Risk Management Framework is a structured methodology for identifying, assessing, treating, and monitoring risks to an organisation's objectives, assets, or systems. It provides a repeatable process and governance structure that links risk appetite, control selection, and assurance activities into a coherent programme aligned with recognised standards such as ISO 31000, NIST RMF, or the NIST AI RMF.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:risk-management-framework", "labels": [ "Risk Management Framework", "Risk Framework" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "risk-management", "title": "Risk Management", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The coordinated activities to direct and control an AI system with regard to risk, encompassing risk identification, assessment, treatment, monitoring, and communication throughout the AI lifecycle to minimize potential adverse effects while maximizing benefits.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:risk-management", "labels": [ "Risk Management", "Conduct Risk Management", "Counterparty Risk Management", "Risk Management (AI-0062)", "Risk Management (AI-0078)", "RiskManagement", "risk management processes" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "risk-mitigation", "title": "Risk Mitigation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Actions taken to reduce the likelihood or impact of identified AI risks through technical, organizational, or procedural controls, implemented throughout the AI lifecycle to achieve acceptable risk levels.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:risk-mitigation", "labels": [ "Risk Mitigation", "AI Risk Mitigation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "risk-register", "title": "Risk Register", "domain": "governance", "domain_name": "Governance", "definition": "A risk register is a structured record that captures identified risks together with their description, likelihood, impact, ownership, mitigation actions and current status. It serves as the central artefact of a risk management process, enabling organisations to track, prioritise and report on risks over time. The register supports governance by making risk exposure visible and accountable to decision-makers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:risk-register", "labels": [ "Risk Register", "Risk Registers" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "risk-scoring-engine", "title": "Risk Scoring Engine", "domain": "security", "domain_name": "Security", "definition": "A risk scoring engine is a software component that computes a numerical risk score for an entity, transaction, or session by combining multiple weighted signals through rules or models. In compliance and identity systems it quantifies how likely an action is fraudulent or non-compliant, feeding thresholds that trigger review, blocking, or escalation. The score provides a consistent, tunable basis for automated decisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-scoring-engine", "labels": [ "Risk Scoring Engine", "Risk Scoring Algorithm" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "risk-scoring", "title": "Risk Scoring", "domain": "finance", "domain_name": "Finance", "definition": "Risk scoring is the quantitative practice of assigning a numerical value to an entity, transaction or event to express the likelihood and potential severity of an adverse outcome such as credit default, fraud or money laundering. Scores are derived from statistical and machine-learning models trained on historical data and behavioural features, producing a calibrated probability or rank that informs automated and human decisions. In finance it underpins lending, underwriting, transaction monitoring and regulatory compliance workflows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:risk-scoring", "labels": [ "Risk Scoring" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "risk-treatment", "title": "Risk Treatment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of selecting and implementing measures to modify AI risk, including risk avoidance, risk reduction (mitigation), risk sharing (transfer), and risk retention, based on risk assessment outcomes and organisational risk appetite. Grounded in ISO/IEC 23894:2023 and the NIST AI RMF MANAGE function, risk treatment produces a treatment plan, residual risk documentation, and an ongoing monitoring regime. Options are evaluated for feasibility, cost, effectiveness, and stakeholder acceptability before selection.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:risk-treatment", "labels": [ "Risk Treatment", "Risk Treatment Plan" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "risk-weighted-assets", "title": "Risk Weighted Assets", "domain": "finance", "domain_name": "Finance", "definition": "Risk-weighted assets (RWA) are a bank's assets and off-balance-sheet exposures weighted according to their credit, market, and operational risk, forming the denominator of regulatory capital ratios. Under the Basel framework, riskier exposures attract higher weights, so banks must hold proportionally more capital against them. RWA links the composition of a bank's portfolio to its minimum capital requirements and overall solvency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:risk-weighted-assets", "labels": [ "Risk Weighted Assets", "Risk-Weighted Assets" ], "is_subclass_of": [ "Capital Adequacy" ], "wikilinks": [] }, { "id": "risk-based-approach", "title": "Risk-Based Approach", "domain": "governance", "domain_name": "Governance", "definition": "A risk-based approach is a regulatory and compliance methodology that allocates scrutiny and controls in proportion to assessed risk, rather than applying uniform requirements to all cases. In anti-money-laundering and cross-border compliance it directs enhanced due diligence at higher-risk customers and transactions while streamlining low-risk ones. It improves effectiveness and efficiency by focusing limited resources where harm is most likely.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-based-approach", "labels": [ "Risk-Based Approach", "Risk-Based Approach Methodology" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "risk-based-authentication", "title": "Risk-Based Authentication", "domain": "security", "domain_name": "Security", "definition": "Risk-based authentication (RBA) is an adaptive security method that adjusts the strength of identity verification based on the assessed risk of a login attempt. It evaluates contextual signals such as device, location, network reputation, and behaviour to decide whether to allow access, require step-up factors, or block the request. RBA balances security and usability by escalating friction only when risk is elevated.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:risk-based-authentication", "labels": [ "Risk-Based Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "risk", "title": "Risk", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The potential for an AI system or associated process to cause harm, produce unintended outcomes, or fail to achieve intended goals. Risk in AI encompasses technical failure modes, bias, adversarial vulnerabilities, misuse, and broader societal harms; risk management frameworks characterise likelihood and severity to prioritise mitigations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:risk", "labels": [ "Risk", "Risk Estimation", "Risk Prevention", "Technical Risk" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "ritual-artifact", "title": "Ritual Artifact", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital objects within virtual worlds that hold cultural, spiritual, or ceremonial significance, often represented as NFTs enabling communities to preserve and share cultural heritage through blockchain-verified authenticity and ownership.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ritual-artifact", "labels": [ "Ritual Artifact" ], "is_subclass_of": [ "Content and Assets", "Digital Heritage" ], "wikilinks": [ "Digital Cultural Heritage", "Digital Heritage", "metaverse" ] }, { "id": "ro-berta", "title": "RoBERTa", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "RoBERTa (Robustly Optimised BERT Pretraining Approach) is a transformer-based language model that refines BERT by removing the next-sentence-prediction objective, adopting dynamic masking, training on substantially larger corpora (160 GB vs 16 GB), and using larger batch sizes. These training-procedure improvements yield consistent performance gains on NLP benchmarks without altering the underlying transformer encoder architecture.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:ro-berta", "labels": [ "RoBERTa" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "robin-hanson-ai-sceptic-economist", "title": "Robin Hanson AI Sceptic Economist", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Robin Hanson is an economist, futurist, and author (George Mason University) known for contrarian analysis of AI progress timelines, the economics of prediction markets, and his book Age of Em which models a future dominated by brain emulations. He maintains a measured scepticism about near-term AGI, arguing that inertia, demographic decline, and adoption barriers make transformative AI further off than mainstream discourse suggests. His blog Overcoming Bias applies economic and evolutionary reasoning to cognition, signalling, and technology adoption.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:robin-hanson-ai-sceptic-economist", "labels": [ "Robin Hanson AI Sceptic Economist", "Robin Hanson" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Lead Poisoning Hypothesis" ] }, { "id": "roblox", "title": "Roblox", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Roblox is an online platform and storefront that lets users create, publish and play user-generated 3D experiences. It combines a game engine, an authoring environment and a social and economic layer, with creators building experiences that other users can enter as customisable avatars. An in-platform virtual currency underpins a creator economy in which developers monetise their experiences. Roblox is frequently cited as a consumer-scale realisation of metaverse concepts, blending gaming, social interaction and user-generated content.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:roblox", "labels": [ "Roblox" ], "is_subclass_of": [ "Metaverse Platform" ], "wikilinks": [] }, { "id": "robot-actuator", "title": "Robot Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot actuator is an electromechanical, pneumatic, hydraulic, or shape-memory device that converts stored or supplied energy into controlled mechanical motion, forming the effector substrate through which a robotic system exerts forces and displacements on its environment. Actuators are the physical implementation layer between a robot's control system and its mechanical structure, determining the speed, force, precision, and compliance characteristics achievable by the overall system. Modern robot actuators range from high-torque servo motors and linear voice-coil drives to soft pneumatic bellows and piezoelectric micro-actuators, each presenting distinct trade-offs in power density, bandwidth, back-drivability, and safety. The selection and design of actuators is a primary determinant of robot morphology, task capability, and energy efficiency across manipulation, locomotion, and human-collaborative applications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:robot-actuator", "labels": [ "Robot Actuator" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "robot-autonomy", "title": "Robot Autonomy", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot autonomy is the degree to which a robotic system can perceive its environment, make decisions, and execute tasks to achieve goals without ongoing human direction. It spans a spectrum from teleoperation, through supervised autonomy where a human approves key decisions, to full autonomy where the robot plans and acts independently within its operating envelope. Autonomy level is typically determined by the sophistication of a robot's perception, planning, and task-execution stack working together under uncertainty.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:robot-autonomy", "labels": [ "Robot Autonomy" ], "is_subclass_of": [ "Autonomous System" ], "wikilinks": [] }, { "id": "robot-component", "title": "Robot Component", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robot Component is any modular element forming part of a robotic system, encompassing hardware modules (motors, sensors, links, joints), electrical components (power supplies, motor drivers, embedded computers), software components (perception modules, planners, controllers), and interface components (communication protocols, connectors, mounting systems). Standardised component interfaces\u2014ISO 9409 tool flanges, EtherCAT, ROS 2\u2014enable plug-and-play integration, reducing development time and supporting modular system design, component reuse, and hierarchical decomposition.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:robot-component", "labels": [ "Robot Component" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "robot-control", "title": "Robot Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot Control encompasses the systems, algorithms, and methodologies that enable robots to execute tasks autonomously or semi-autonomously through sensing, decision-making, and actuation. It integrates perception, planning, and actuation into closed-loop systems operating in dynamic environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robot-control", "labels": [ "Robot Control", "Bimanual Robot Control", "Robot Control Architecture", "Robot Control Pipeline", "Robot Controllers" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "actuation", "Actuation Systems", "Adaptive Control Algorithms", "Autonomous Systems", "behavior trees", "control algorithms", "Control Algorithms", "Convolutional Neural Networks", "Decision Making", "IEC 61508", "IEEE 1872", "ISO 10218", "ISO 13849", "ISO 8373", "Machine Vision", "model predictive control", "Model Predictive Control", "Multi-Agent Robotics", "neural control", "Neural Control Methods" ] }, { "id": "robot-controller", "title": "Robot Controller", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robot Controller is the computational subsystem that governs a robot's behaviour by reading sensor data, computing control commands and driving actuators to achieve desired motion or tasks. It runs control loops in real time, coordinating kinematics, trajectory execution and safety logic while interfacing with higher-level planning software. Controllers range from embedded microcontrollers on a single joint to industrial cabinets coordinating an entire articulated arm.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robot-controller", "labels": [ "Robot Controller" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "robot-dynamics", "title": "Robot Dynamics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot Dynamics is the branch of classical and computational mechanics that characterises the forces, torques, inertias, and energy flows governing the motion of robotic mechanisms. It distinguishes between forward dynamics\u2014determining joint accelerations and Cartesian trajectories from applied actuator forces and torques\u2014and inverse dynamics\u2014computing the actuator effort required to produce a prescribed motion trajectory. The discipline provides the theoretical foundation for model-based control laws (such as computed-torque and feedback-linearisation controllers), physics-based simulation, and the design of energy-efficient manipulators and legged robots. Accurate dynamic models are indispensable for high-speed manipulation, safe human-robot collaboration, compliant actuation, and whole-body control of mobile robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:robot-dynamics", "labels": [ "Robot Dynamics", "Robot Dynamics Model" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RobotMechanics" ] }, { "id": "robot-hardware", "title": "Robot Hardware", "domain": "robotics", "domain_name": "Robotics", "definition": "RobotHardware is the integrated electromechanical substrate of robotic systems encompassing all physical components required for autonomous or semi-autonomous operation: compute platforms (NVIDIA Jetson Orin NX/AGX, Raspberry Pi 5, BeagleBone AI-64), real-time microcontrollers (STM32H7, Teensy 4....", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:robot-hardware", "labels": [ "Robot Hardware" ], "is_subclass_of": [ "Actuation and Control", "Mobile Robot Platform", "Robotics Platform", "Embedded Systems", "Mechatronics", "Industrial Robot" ], "wikilinks": [ "Actuator", "ARM Cortex-M", "Autonomous Operation", "Battery Management System", "BSI PAS 1085 Collaborative Robots 2021", "CAN-FD Bus", "CAN-FD Protocol", "CiA 301 CANopen Standard", "Communication Bus", "CommunicationLayer", "Craig Introduction to Robotics 2017", "DDS Middleware", "ElectromechanicalDomain", "Embedded Controller", "Embedded Systems", "EmbeddedSystemsDomain", "EtherCAT", "EtherCAT Protocol", "EtherCAT Technology Group", "EtherCAT Technology Group Specification v1.0.5" ] }, { "id": "robot-joint", "title": "Robot Joint", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robot Joint is a mechanical articulation between two robot links that permits controlled relative motion \u2014 rotational, translational, or compound \u2014 enabling the full kinematic range of a robotic arm or manipulator. Joint types include revolute, prismatic, and spherical, each characterised by degrees of freedom and load capacity.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robot-joint", "labels": [ "Robot Joint", "Joint" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "robot-kinematics", "title": "Robot Kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot Kinematics is the mathematical study of the geometry of robot motion\u2014comprising forward kinematics (mapping joint parameters to end-effector pose) and inverse kinematics (computing joint configurations that achieve a desired pose)\u2014without regard to the forces or torques that produce that motion. It is foundational to robot programming, trajectory planning, and the design of manipulation systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:robot-kinematics", "labels": [ "Robot Kinematics" ], "is_subclass_of": [ "Actuation and Control", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "robot-learning", "title": "Robot Learning", "domain": "robotics", "domain_name": "Robotics", "definition": "The application of machine learning algorithms and artificial intelligence techniques to enable robots to acquire new skills, adapt to changing environments, and improve performance through experience, encompassing supervised learning (imitation learning, learning from demonstration), reinforceme...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:robot-learning", "labels": [ "Robot Learning", "RobotLearning" ], "is_subclass_of": [ "Actuation and Control", "MachineLearning", "ArtificialIntelligence", "Robotics" ], "wikilinks": [ "Association for the Advancement of Artificial Intelligence (AAAI)", "ComputeResources", "Conference on Robot Learning (CoRL)", "IEEE Robotics and Automation", "International Conference on Learning Representations (ICLR)", "SelfSupervisedLearning", "SensorData", "SimulationEnvironment", "SkillAcquisition", "TransferLearning", "AdaptiveControl", "AIDomain", "ArtificialIntelligence", "AutonomousNavigation", "ImitationLearning", "MachineLearning", "Manipulation", "ReinforcementLearning", "Robotics", "RoboticsDomain" ] }, { "id": "robot-localisation", "title": "Robot Localisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot localisation is the problem of estimating a robot's pose, its position and orientation, within a known map from noisy sensor measurements and motion commands. It is typically solved with probabilistic filters that maintain a belief over possible poses and update it as new observations arrive. Accurate localisation is a prerequisite for reliable navigation and planning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:robot-localisation", "labels": [ "Robot Localisation" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "robot-locomotion", "title": "Robot Locomotion", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot locomotion is the study and engineering of how robots move through their environment, encompassing legged walking, wheeled rolling, crawling, swimming and flight. It integrates mechanical design, control theory and sensing to generate stable, efficient and adaptive movement over varied terrain. Locomotion distinguishes itself from manipulation by focusing on whole-body displacement and dynamic balance rather than interaction with objects.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robot-locomotion", "labels": [ "Robot Locomotion" ], "is_subclass_of": [ "Robotics", "Actuation and Control" ], "wikilinks": [] }, { "id": "robot-manipulation", "title": "Robot Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot manipulation is the sub-field of robotics concerned with the planning and execution of purposeful physical interactions between robotic systems and objects in the world, encompassing grasping, assembly, in-hand manipulation, and tool use. It integrates kinematics, dynamics, perception, and motion planning to move objects from one configuration to another while adapting to uncertainty in object shape, pose, surface properties, and environmental dynamics. Robust manipulation requires coordinating end-effectors, force-torque sensing, and real-time control loops to achieve reliable contact-rich behaviour. The field bridges classical planning and modern machine learning, increasingly leveraging deep visuomotor policies and foundation models trained on large-scale demonstration data.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:robot-manipulation", "labels": [ "Robot Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "robot-navigation", "title": "Robot Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot navigation is the capability of a mobile robot to determine its position, plan a route, and move safely through an environment toward a goal while avoiding obstacles. It integrates localisation, mapping, path planning, and motion control into a continuous perception-action loop. Robot navigation combines sensor fusion with algorithms such as SLAM to operate in unknown or dynamic surroundings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robot-navigation", "labels": [ "Robot Navigation" ], "is_subclass_of": [ "Robotics", "Navigation and Planning" ], "wikilinks": [] }, { "id": "robot-operating-system", "title": "Robot Operating System", "domain": "robotics", "domain_name": "Robotics", "definition": "The Robot Operating System (ROS / ROS 2) is an open-source middleware framework providing a structured communication layer, tool ecosystem, and package repository for robotic software development, enabling modular composition of perception, planning, and actuation subsystems through a publish-sub...", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:robot-operating-system", "labels": [ "Robot Operating System", "ROS (Robot Operating System)" ], "is_subclass_of": [ "Actuation and Control", "Robotics", "Robotic Middleware", "Publish-Subscribe Framework" ], "wikilinks": [ "AI Domain", "Autonomous Systems Platform", "C++ rclcpp", "CMake Build System", "colcon", "colcon Build System", "CommunicationLayer", "CycloneDDS", "Data Distribution Service", "DDS RTPS Protocol", "eProsima Fast DDS", "Large Language Model", "micro-ROS", "micro-ROS Project", "OMG DDS Specification", "ONNX Runtime", "OpenCV DNN", "Publish-Subscribe Framework", "Python 3", "Python rclpy" ] }, { "id": "robot-perception", "title": "Robot Perception", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot perception is the process by which a robot interprets sensor data from cameras, lidar, depth sensors and inertial units to build an actionable understanding of its environment, objects and its own pose.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:robot-perception", "labels": [ "Robot Perception", "Robot Learning", "Robot Perception System" ], "is_subclass_of": [ "Robotics", "Perception and Sensing" ], "wikilinks": [ "Sensors", "Sensor Fusion", "SLAM", "Machine Learning", "Robotics", "https://en.wikipedia.org/wiki/Machine_perception", "https://www.ros.org/" ] }, { "id": "robot-programming", "title": "Robot Programming", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot programming is the discipline of specifying the tasks, motions, and decision logic that a robot executes, ranging from low-level joint commands to high-level behavioural goals. It encompasses textual programming languages, graphical and teach-by-demonstration interfaces, and middleware frameworks that connect perception, planning, and actuation. The aim is to translate human intent into reliable, repeatable robot behaviour while respecting safety, timing, and hardware constraints.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robot-programming", "labels": [ "Robot Programming" ], "is_subclass_of": [ "Robotics", "Robot Operating System" ], "wikilinks": [] }, { "id": "robot-rb-0001", "title": "Robot RB-0001", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot RB-0001 is an identifier-class for a reference mobile-manipulator robot platform that combines a mobile base with an articulated manipulator arm. As a composite robot, it integrates locomotion and manipulation subsystems so the system can both navigate an environment and perform physical tasks within it. It serves as a canonical example linking mobile-robot and manipulator capabilities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:robot-rb-0001", "labels": [ "Robot RB-0001" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "robot-safety", "title": "Robot Safety", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot Safety encompasses the engineering principles, risk assessment methods, and regulatory standards that ensure robotic systems operate without causing harm to humans, other machines, or the environment. It includes functional safety standards (ISO 10218, ISO 13849), collaborative robot (cobot) application requirements, dynamic risk assessment, and the emerging challenges of AI-integrated and autonomous robot deployment in shared workspaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:robot-safety", "labels": [ "Robot Safety", "Human-Robot Safety" ], "is_subclass_of": [ "Safety and Standards", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "robot-sensor", "title": "Robot Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robot Sensor is a transducer or measurement device integrated into a robotic system to acquire data about the robot's internal state (proprioception: joint angles, torques, currents) or external environment (exteroception: proximity, force, vision, lidar). Sensor data drives closed-loop control, obstacle avoidance, and higher-level perception pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robot-sensor", "labels": [ "Robot Sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "robot-simulation", "title": "Robot Simulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot simulation is the use of physics-based virtual environments to model a robot's kinematics, dynamics, sensors, and interactions before or instead of deploying on hardware. It enables developers to test control software, validate motion plans, and generate training data safely and cheaply. Tight integration with robot middleware lets the same code run in simulation and on the physical robot.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:robot-simulation", "labels": [ "Robot Simulation", "Robot Simulation Environment" ], "is_subclass_of": [ "Robotics", "Actuation and Control" ], "wikilinks": [] }, { "id": "robot-singularity", "title": "Robot Singularity", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robot Singularity is a configuration of a serial or parallel robot manipulator at which the Jacobian matrix loses rank, causing the determinant to approach zero. At a singularity, the robot loses one or more degrees of freedom in Cartesian space: it cannot produce force or motion in certain directions, and inverse kinematics solutions become ill-conditioned or non-unique. Singularity avoidance and detection are central problems in motion planning, trajectory generation, and real-time control of robotic arms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:robot-singularity", "labels": [ "Robot Singularity" ], "is_subclass_of": [ "Actuation and Control", "RoboticsDomain" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "robot-standard", "title": "Robot Standard", "domain": "robotics", "domain_name": "Robotics", "definition": "RobotStandard is a normative technical instrument \u2014 a specification, guideline, or regulation \u2014 issued by a recognised standards development organisation (SDO) or regulatory body to define mandatory or voluntary safety requirements, performance criteria, interoperability protocols, and terminolog...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:robot-standard", "labels": [ "Robot Standard" ], "is_subclass_of": [ "Safety and Standards", "Technical Standard", "Safety Regulation", "Normative Framework", "Robotics", "Industrial Safety", "Product Regulation" ], "wikilinks": [ "AMRC", "ANSI", "Autonomous Mobile Robot Safety", "Autonomous Mobile Robots", "BARA", "BSI", "CE Marking", "Cobot Deployment", "Collaborative Robots", "ComplianceLayer", "Conformity Assessment", "Conformity Assessment Body", "De Facto Standard", "Declaration of Conformity", "Digital Product Passport", "DIN", "Domain-Specific Framework", "Driverless Industrial Trucks", "EHSR Framework", "EU Machinery Regulation" ] }, { "id": "robot", "title": "Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot is an actuated mechanism, programmable in two or more axes with a degree of autonomy, that moves within its environment to perform intended tasks without direct human intervention at the moment of task execution. Robots integrate mechanical structure, actuation, sensing, and control software to perceive their environment, reason about goals, and execute physical or digital actions. The concept spans a wide spectrum from fixed industrial manipulators to mobile autonomous agents, collaborative cobots, and software robots (robotic process automation). Standardised by ISO 8373:2021, the definition distinguishes robots from simple automated machines by requiring reprogrammability and environmental interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:robot", "labels": [ "Robot", "Elder Care Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "ISO 8373:2021", "MechatronicSystem", "AutonomousAgent", "Robotics" ] }, { "id": "robot-link", "title": "RobotLink", "domain": "robotics", "domain_name": "Robotics", "definition": "A rigid mechanical structural component in a robotic manipulator or kinematic chain that connects two consecutive joints and transmits motion and forces between them, characterized by fixed geometric and inertial properties including length, mass, center of mass location, and inertia tensor, desc...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:robot-link", "labels": [ "RobotLink", "Link" ], "is_subclass_of": [ "Actuation and Control", "RigidBody", "MechanicalComponent", "KinematicElement", "StructuralMember" ], "wikilinks": [ "Craig, J.J. Introduction to Robotics (2005)", "Denavit & Hartenberg (1955) Kinematic Notation", "DynamicBalancing", "ForceTransmission", "InertialFrame", "ISO 8373 Robotics Vocabulary", "ISO 9283 Manipulating Industrial Robots Performance Criteria", "KinematicChain", "LinkGeometry", "MassProperties", "MaterialComposition", "MaterialSelection", "MotionPropagation", "MountingInterface", "PayloadSupport", "Shabana, A.A. Dynamics of Multibody Systems (2013)", "StructuralAnalysis", "SurfaceFinish", "ThermalManagement", "VibrationDamping" ] }, { "id": "robotaxi", "title": "Robotaxi", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotaxi is a self-driving vehicle operated as an on-demand ride-hailing service without a human safety driver, combining autonomous vehicle technology with fleet dispatch, mapping, and remote-assistance infrastructure. Passengers summon a robotaxi through a mobile application in the same manner as a conventional ride-hailing service, but the vehicle navigates, avoids obstacles, and completes the trip using onboard perception and planning systems. Commercial robotaxi deployments require regulatory approval and typically operate within geofenced service areas where the vehicle's autonomy stack has been extensively validated.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:robotaxi", "labels": [ "Robotaxi" ], "is_subclass_of": [ "Autonomous Vehicle" ], "wikilinks": [] }, { "id": "robotic-arm", "title": "Robotic Arm", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotic arm is a programmable mechanical manipulator, typically composed of rigid links connected by actuated joints, that positions and orients an end-effector within a workspace. Its degrees of freedom allow it to reach and manipulate objects under the control of motion-planning and kinematics algorithms. Robotic arms range from industrial units performing repetitive high-precision tasks to collaborative arms designed to work safely alongside people. They are a foundational platform across manufacturing, surgery, logistics and research.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robotic-arm", "labels": [ "Robotic Arm" ], "is_subclass_of": [ "Manipulator" ], "wikilinks": [] }, { "id": "robotic-camera-virtual-production-integration", "title": "Robotic Camera Virtual Production Integration", "domain": "robotics", "domain_name": "Robotics", "definition": "The VP Robotics Project (VisionFlow) is a feasibility study investigating the integration of robotic camera control with machine learning and virtual production workflows. It inverts conventional pre-visualisation pipelines by deriving scene-driven camera motion from AI-generated content, combining open-source robotics software with parallax-plates-as-a-service delivery for the film and television industry.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robotic-camera-virtual-production-integration", "labels": [ "Robotic Camera Virtual Production Integration", "VP robotics project" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "robotic-control", "title": "Robotic Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotic control is the discipline of designing algorithms and hardware that regulate the motion, force, and behaviour of robotic systems to achieve desired trajectories or task objectives in the presence of dynamic uncertainty, environmental disturbances, and physical constraints. It spans classical feedback control strategies such as PID, computed-torque, and impedance control, through to model-predictive and learning-based controllers that adapt online. Robotic control integrates kinematics, dynamics, estimation, and optimisation to translate high-level task plans into actuator commands. It is a foundational competency enabling manipulation, locomotion, and human-robot interaction across industrial and service robot applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robotic-control", "labels": [ "Robotic Control", "Remote Robotic Control", "RoboticControl" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "robotic-grasping", "title": "Robotic Grasping", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotic Grasping is the subfield of robotics concerned with planning and executing stable, task-appropriate physical contact between a robotic end-effector and a target object. It integrates perception, grasp quality estimation, motion planning, and control to enable reliable object acquisition across varied geometries, materials, and environmental conditions.", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:robotic-grasping", "labels": [ "Robotic Grasping", "Compliant Grasping", "Soft Robotic Grasping" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "robotic-manipulation", "title": "Robotic Manipulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotic manipulation is the field concerned with how robots physically interact with and change the state of objects in their environment \u2014 grasping, moving, assembling, and reorienting items using arms, hands, and end-effectors. It integrates perception, motion planning, control, and contact reasoning so that a robot can compute and execute the forces and trajectories needed to handle objects reliably under uncertainty. Manipulation spans rigid pick-and-place in structured factories through to dexterous, contact-rich handling of deformable or unfamiliar objects in unstructured human environments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robotic-manipulation", "labels": [ "Robotic Manipulation" ], "is_subclass_of": [ "Manipulation" ], "wikilinks": [] }, { "id": "robotic-process-automation", "title": "Robotic Process Automation", "domain": "ai", "domain_name": "Ai", "definition": "Robotic Process Automation (RPA) is a software technology that deploys configurable software robots to emulate human interactions with digital systems \u2014 clicking, typing, copy-pasting data, and navigating graphical user interfaces \u2014 in order to execute repetitive, rule-based business processes without modifying underlying applications. RPA robots operate at the presentation layer, integrating with legacy systems through their UI rather than through APIs, and are managed by an orchestration platform that schedules work items, monitors execution, handles exceptions, and maintains audit trails. Modern RPA platforms increasingly incorporate AI capabilities such as optical character recognition, natural language processing, and machine learning to extend automation to semi-structured and unstructured data, giving rise to the broader discipline of intelligent automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:robotic-process-automation", "labels": [ "Robotic Process Automation" ], "is_subclass_of": [ "Process Automation" ], "wikilinks": [] }, { "id": "robotic-system", "title": "Robotic System", "domain": "robotics", "domain_name": "Robotics", "definition": "An integrated assembly of mechanical, electronic, and computational subsystems \u2014 including actuators, sensors, a control architecture, and software \u2014 that perceives its environment and executes physical tasks autonomously or under remote human direction, spanning industrial, collaborative, and telepresence applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robotic-system", "labels": [ "Robotic System", "Robotic Manipulation System", "Robotic Systems", "RoboticSystem" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "owl:Thing" ] }, { "id": "robotic-telepresence", "title": "Robotic Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The projection of human presence into remote physical locations through mobile robotic platforms equipped with cameras, displays, microphones, and speakers, enabling remote operators to navigate environments, interact with people, and manipulate objects as if physically present, bridging virtual...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:robotic-telepresence", "labels": [ "Robotic Telepresence", "TELE-200-robotic-telepresence" ], "is_subclass_of": [ "Telepresence", "TELE 001 telepresence" ], "wikilinks": [ "RemoteOfficeAttendance", "TELE-020-virtual-reality-telepresence", "TELE-150-webrtc", "TELE-201-teleoperation-systems", "TELE-202-remote-manipulation", "TELE-203-haptic-feedback-telepresence", "AutonomousNavigation", "HumanRobotInteraction", "TELE-001-telepresence" ] }, { "id": "robotics-application", "title": "Robotics Application", "domain": "robotics", "domain_name": "Robotics", "definition": "A deployed use-case or task domain in which robotic systems perform physical or cyber-physical work, spanning industrial automation, telepresence, telemedicine, logistics, hazardous environment inspection, and collaborative human-robot interaction. Robotics applications integrate perception, planning, actuation, and communication subsystems to accomplish domain-specific objectives with varying degrees of autonomy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robotics-application", "labels": [ "Robotics Application" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "robotics-control", "title": "Robotics Control", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The application of artificial intelligence and control theory to robotic systems to enable autonomous navigation, manipulation, perception, and task execution. AI-driven robotics control integrates reinforcement learning for policy optimisation, computer vision for scene perception, motion planning for collision-free trajectory generation, and sensor fusion for robust state estimation \u2014 operating in real time under uncertainty and safety constraints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robotics-control", "labels": [ "Robotics Control" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Computer Vision", "Motion Planning", "Reinforcement Learning", "Sensor Fusion" ] }, { "id": "robotics-perception", "title": "Robotics Perception", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Robotics perception is the set of capabilities by which a robot senses, interprets, and builds an internal representation of its physical environment from sensor data. It fuses inputs from cameras, lidar, depth sensors, and inertial units to perform object detection, scene understanding, localisation, and mapping. Robust perception is the foundation for autonomous navigation and manipulation, transforming raw, noisy measurements into actionable spatial knowledge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:robotics-perception", "labels": [ "Robotics Perception", "Robotic Perception" ], "is_subclass_of": [ "Perception", "Computer Vision" ], "wikilinks": [] }, { "id": "robotics-platform", "title": "Robotics Platform", "domain": "robotics", "domain_name": "Robotics", "definition": "RoboticsPlatform is an integrated hardware\u2013software\u2013middleware ecosystem providing standardised communication layers, hardware abstraction, simulation environments, motion-planning stacks, real-time control loops, and cloud-edge orchestration that collectively reduce engineering effort for robot ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:robotics-platform", "labels": [ "Robotics Platform", "Robotics Software Ecosystem" ], "is_subclass_of": [ "Actuation and Control", "Software Framework", "Control System", "Embedded System", "Middleware", "Simulation Environment", "Digital Twin" ], "wikilinks": [ "5G", "Agricultural Robotics", "Assistive Robotics", "Autonomous Vehicles", "Behaviour Tree", "C++", "CAN Bus", "ControlSystemsDomain", "CUDA", "Custom RTOS Firmware", "DDS", "DDS Middleware", "DDS QoS", "Digital Manufacturing", "Eigen", "Embedded System", "EmbeddedSystemsDomain", "EtherCAT", "Ethernet", "Field Robotics" ] }, { "id": "robotics-process", "title": "Robotics Process", "domain": "robotics", "domain_name": "Robotics", "definition": "A Robotics Process encompasses the systematic procedures, methodologies, and workflows for robot development, deployment, operation, and maintenance, spanning requirements engineering, system design, hardware-software integration, verification and validation, commissioning, and lifecycle management. Agile and iterative development methods accommodate uncertainty and enable rapid prototyping; simulation-driven approaches using digital twins support virtual commissioning. Compliance with process standards such as ISO 9001 and IEC 61508 is required for safety-critical robotic applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:robotics-process", "labels": [ "Robotics Process" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "robotics-simulation", "title": "Robotics Simulation", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotics simulation is the discipline of creating high-fidelity virtual worlds in which robots and their environments are modelled for development, testing, and synthetic-data generation. It emphasises scene description, physically based rendering, and interoperable asset pipelines so simulated environments transfer faithfully to deployment. Standards such as Universal Scene Description enable shared, composable simulation scenes across tools.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:robotics-simulation", "labels": [ "Robotics Simulation" ], "is_subclass_of": [ "Robotics", "Simulation" ], "wikilinks": [] }, { "id": "robotics-systems", "title": "Robotics Systems", "domain": "robotics", "domain_name": "Robotics", "definition": "The foundational technology domain encompassing autonomous and semi-autonomous mechanical systems, including manipulators, mobile robots, humanoids, and collaborative robots (cobots), along with their control systems, sensors, actuators, kinematics, and the integration of perception, planning, and action for physical world interaction.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robotics-systems", "labels": [ "Robotics Systems" ], "is_subclass_of": [ "Actuation and Control", "Technology Domain" ], "wikilinks": [ "Artificial Intelligence", "Collaborative Robot", "HumanRobotInteraction", "Humanoid Robot", "Manipulator", "Metaverse Technology", "Mobile Robot", "Motion Planning", "Robot Control", "Robot Kinematics", "Robot Sensor", "Robotics Systems", "SwarmRobotics", "Technology Domain", "Telecollaboration" ] }, { "id": "robotics-telepresence-bridge", "title": "Robotics-Telepresence Bridge", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The conceptual and technical integration between robotics systems and telepresence technologies, where remote operators experience physical presence in distant real-world locations through robot-mediated perception and action, combining robotic manipulation capabilities with telepresence social ...", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:robotics-telepresence-bridge", "labels": [ "Robotics-Telepresence Bridge" ], "is_subclass_of": [ "Telepresence", "ConvergenceConcept" ], "wikilinks": [ "HazardousEnvironmentAccess", "TELE-020-virtual-reality-telepresence", "TELE-150-webrtc", "TELE-154-edge-computing-telepresence", "TELE-157-predictive-tracking", "TELE-200-robotic-telepresence", "TELE-201-teleoperation-systems", "TELE-203-haptic-feedback-telepresence", "TELE-205-surgical-telepresence", "CollaborativeRobot", "ConvergenceConcept", "HumanRobotInteraction", "RoboticSystem", "TELE-001-telepresence" ] }, { "id": "robotics", "title": "Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotics is the interdisciplinary field encompassing the design, construction, operation, and application of robots and automated systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:robotics", "labels": [ "Robotics", "Advanced Robotics", "Autonomous Robotics", "Embedded Robotics" ], "is_subclass_of": [ "Thing", "Automation" ], "wikilinks": [ "Actuators", "Automation", "Autonomous Manufacturing", "Collaborative Robotics", "Control Systems", "Humanoid Robotics", "Humanoid Robots", "Industrial Automation", "Industrial Robots", "Machine Vision", "Precision Engineering", "Robotic Arm", "Sensor System", "Smart Manufacturing", "Warehouse Automation", "Artificial Intelligence", "Computer Vision", "Digital Twin", "IoT Sensors", "Kinematics" ] }, { "id": "robotics-core-concepts", "title": "RoboticsCoreConcepts", "domain": "robotics", "domain_name": "Robotics", "definition": "A hierarchical taxonomic framework organizing the fundamental conceptual domains within robotics science and engineering, comprising ten core subdomains structured into four primary categories\u2014autonomous systems (encompassing independent robotic platforms implementing path planning algorithms suc...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:robotics-core-concepts", "labels": [ "RoboticsCoreConcepts", "Robotics Core Concepts" ], "is_subclass_of": [ "Actuation and Control", "TaxonomicFramework", "KnowledgeOrganizationSystem", "DomainOntology", "ConceptualHierarchy" ], "wikilinks": [ "AutonomousVehicles", "ComputationalMethods", "Craig J.J. Introduction to Robotics Mechanics and Control", "HardwareDesign", "IEEE Robotics and Automation Society RAS", "IndustrialAutomation", "International Federation of Robotics IFR", "ISO 8373 Robotics Vocabulary", "MathematicalFoundations", "PhysicsModeling", "RoboticsEducation", "RoboticsResearch", "ServiceRobotics", "Siciliano et al. Robotics Modelling Planning and Control", "Springer Handbook of Robotics", "Thrun et al. Probabilistic Robotics", "AutonomousRobot", "ConceptualHierarchy", "DomainOntology", "HumanRobotInteraction" ] }, { "id": "robust-control", "title": "Robust Control", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Robust Control is a branch of control theory that designs controllers guaranteeing stable and acceptable performance across a bounded set of model uncertainties and disturbances. It formalises worst-case design requirements through H-infinity and H2 optimisation frameworks, ensuring actuated systems remain within specification even when plant parameters deviate from nominal values.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robust-control", "labels": [ "Robust Control" ], "is_subclass_of": [ "Content and Assets", "ControlAlgorithms" ], "wikilinks": [ "ControlAlgorithms" ] }, { "id": "robustness-oecd", "title": "Robustness (OECD)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The OECD Robustness principle (Principle 1.4, revised 2024) mandates that AI systems function reliably and securely throughout their lifecycle, demonstrating resilience against errors, faults, distributional inconsistencies, and adversarial attempts to alter system use or performance. It requires continuous risk assessment covering statistical robustness, fault tolerance, and adversarial resistance, and connects to the EU AI Act Article 15 technical requirements. Providers are responsible for design and testing; deployers are responsible for ongoing monitoring.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:robustness-oecd", "labels": [ "Robustness (OECD)" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "robustness-testing", "title": "Robustness Testing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Robustness Testing is the systematic evaluation of a system's behaviour under adversarial, noisy, or out-of-distribution inputs to determine whether its outputs remain reliable and safe outside normal operating conditions. For AI models this includes adversarial perturbation testing, distribution-shift evaluation, and stress testing of edge cases. It is a required input to conformity assessment and broader trustworthy-AI evaluation frameworks.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:robustness-testing", "labels": [ "Robustness Testing" ], "is_subclass_of": [ "Model Evaluation" ], "wikilinks": [] }, { "id": "robustness", "title": "Robustness", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The ability of an AI system to maintain consistent, correct, and safe performance across diverse operating conditions, including unexpected inputs, environmental variations, and adversarial perturbations, without catastrophic failure or significant degradation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:robustness", "labels": [ "Robustness" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "rocket-equation", "title": "Rocket Equation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:rocket-equation", "labels": [ "Rocket Equation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "rocks-db", "title": "RocksDB", "domain": "data", "domain_name": "Data", "definition": "RocksDB is an embedded, high-performance key-value store built on a log-structured merge-tree and optimised for fast storage such as SSDs. It provides ordered keys, atomic batch writes, snapshots, and tunable compaction, and is embedded as a library rather than run as a server. It is widely used as the local state backend in databases and blockchain platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rocks-db", "labels": [ "RocksDB" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "role-based-access-control", "title": "Role-Based Access Control", "domain": "security", "domain_name": "Security", "definition": "Role-Based Access Control (RBAC) is an access control paradigm in which permissions to perform operations on system resources are assigned to roles rather than to individual users, and users acquire those permissions by being assigned to one or more roles that reflect their organisational function. It enforces the principle of least privilege and separation of duties by decoupling user identity from resource authorisation through an intermediate role abstraction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:role-based-access-control", "labels": [ "Role-Based Access Control" ], "is_subclass_of": [ "Access Control" ], "wikilinks": [] }, { "id": "rollback", "title": "Rollback", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Rollback is the operation of restoring a system, workspace, or agent to a previously captured good state, discarding the changes made since that point after an error, failed action, or unwanted outcome. It depends on the earlier capture of restorable state \u2014 a checkpoint, a snapshot, or a version-control commit \u2014 and on the changes since being either reversible or discardable. In autonomous-agent orchestration, rollback lets a supervisor undo a subagent's destructive or incorrect edits and retry from a known-safe baseline, turning risky irreversible automation into a recoverable, bounded-blast-radius process.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:rollback", "labels": [ "Rollback" ], "is_subclass_of": [ "Error Recovery", "ErrorRecovery" ], "wikilinks": [ "ErrorRecovery", "Checkpointing", "VersionControl", "FaultTolerance" ] }, { "id": "rollup", "title": "Rollup", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Rollup is a blockchain scaling technique that executes transactions off a base layer such as Ethereum while posting compressed transaction data and state commitments back to it for settlement and data availability. By batching many transactions and proving or asserting their validity on the main chain, rollups increase throughput and reduce fees while inheriting much of the base layer's security. Two main types exist: optimistic rollups, which assume validity and rely on fraud proofs during a challenge window, and zero-knowledge rollups, which post validity proofs verifying each batch. Rollups are central to Ethereum's layer-2 scaling strategy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:rollup", "labels": [ "Rollup", "Layer 2 Rollup", "Layer-2 Rollup", "Rollup Scaling", "Validity Rollup" ], "is_subclass_of": [ "Layer 2 Scaling", "Blockchain Domain" ], "wikilinks": [ "Fraud Proof", "Zero Knowledge Proof", "Ethereum", "Optimism", "Arbitrum", "zkSync", "Polygon", "Blockchain Domain" ] }, { "id": "room-scale", "title": "Room Scale", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A VR interaction paradigm in which the user can physically walk within a tracked play area \u2014 typically 2m \u00d7 2m or larger \u2014 with inside-out or lighthouse-based positional tracking translating real-world movement into the virtual environment. Room-scale VR requires boundary (chaperone) systems to prevent collisions with physical obstacles, and is contrasted with seated or standing-only VR experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:room-scale", "labels": [ "Room Scale" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "room-state-tracking", "title": "Room State Tracking", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Room state tracking is the mechanism in real-time collaboration and conferencing systems that maintains an authoritative, synchronised record of a virtual room's membership, media status, and shared state. It tracks who is present, their roles, mute and stream states, and assignments such as breakout-room placement. Reliable state tracking ensures all participants and the server share a consistent view of the session.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:room-state-tracking", "labels": [ "Room State Tracking" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "root-cause-analysis", "title": "Root Cause Analysis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Root cause analysis (RCA) is a structured problem-solving discipline that identifies the underlying origin of a fault, failure, or incident rather than merely treating its visible symptoms. In infrastructure and reliability engineering it traces a chain of contributing causes back to the conditions that, if corrected, would have prevented the event. RCA produces durable, systemic fixes and feeds learning back into operational practice.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:root-cause-analysis", "labels": [ "Root Cause Analysis", "Root-Cause Analysis" ], "is_subclass_of": [ "Incident Management" ], "wikilinks": [] }, { "id": "root-certificate", "title": "Root Certificate", "domain": "security", "domain_name": "Security", "definition": "A root certificate is a self-signed X.509 certificate that identifies a root certificate authority and serves as the trust anchor at the top of a certificate chain. Relying parties pre-install trusted root certificates in trust stores, and any certificate that chains back to a trusted root is accepted as authentic. Because a compromised root undermines all certificates beneath it, root keys are protected with the highest assurance and kept offline.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:root-certificate", "labels": [ "Root Certificate" ], "is_subclass_of": [ "Digital Certificate" ], "wikilinks": [] }, { "id": "root-mean-square-error", "title": "Root Mean Square Error", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A regression performance metric representing the square root of the average squared differences between predicted and actual values, calculated by taking the mean of squared prediction errors and then applying the square root, providing a measure of prediction accuracy in the same units as the ta...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:root-mean-square-error", "labels": [ "Root Mean Square Error", "Root Mean Squared Error" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Bias-Variance Trade-off", "Error Analysis", "Mean Squared Error", "Model Evaluation", "Outlier", "Regression", "RMSLE", "Standard Deviation", "Mean Absolute Error", "MetaverseDomain", "Model Performance" ] }, { "id": "rootstock", "title": "Rootstock", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bitcoin sidechain that supports Ethereum-compatible smart contracts, secured by merged mining with Bitcoin and connected through a two-way peg.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:rootstock", "labels": [ "Rootstock" ], "is_subclass_of": [ "Sidechain" ], "wikilinks": [ "Sidechain", "Bitcoin", "Smart Contract", "Ethereum Virtual Machine" ] }, { "id": "rotary-encoder", "title": "Rotary Encoder", "domain": "robotics", "domain_name": "Robotics", "definition": "A rotary encoder is an electromechanical transducer that converts shaft angular position or velocity into digital pulses or absolute position codes. Incremental encoders output A/B quadrature pulses enabling direction and relative position measurement; absolute encoders output a unique binary code at every shaft position. Used ubiquitously in robot joints, servo motors, and motion-control systems for precise closed-loop feedback.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rotary-encoder", "labels": [ "Rotary Encoder" ], "is_subclass_of": [ "Perception and Sensing", "Encoder" ], "wikilinks": [ "Encoder", "Robotics" ] }, { "id": "rotary-position-embedding", "title": "Rotary Position Embedding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Rotary Position Embedding (RoPE) is a method for injecting positional information into transformer attention by rotating the query and key vectors by an angle proportional to each token's absolute position, so that their dot product depends only on relative position. Because the rotation is applied multiplicatively in feature pairs rather than added to the embeddings, RoPE unifies absolute and relative positional encoding while preserving the inner-product structure that attention relies on. It is the dominant positional scheme in modern large language models such as Llama, and its frequency basis can be rescaled to extrapolate context windows far beyond the training length.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rotary-position-embedding", "labels": [ "Rotary Position Embedding" ], "is_subclass_of": [ "Positional Encoding" ], "wikilinks": [] }, { "id": "rotation-matrix", "title": "Rotation Matrix", "domain": "robotics", "domain_name": "Robotics", "definition": "A rotation matrix is a square orthogonal matrix with determinant one that represents a rigid rotation of vectors in Euclidean space about a fixed origin. In robotics and computer graphics it encodes the orientation of one coordinate frame relative to another, mapping direction vectors from one frame into another without altering their length. Rotation matrices compose by matrix multiplication, enabling chains of rotations to be combined into a single transformation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rotation-matrix", "labels": [ "Rotation Matrix" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "rough-consensus", "title": "Rough Consensus", "domain": "standards", "domain_name": "Standards", "definition": "Rough consensus is a decision-making principle, central to internet standards development, in which a working group adopts a position when the dominant view has been established and substantive objections have been addressed, rather than requiring unanimity or a formal majority vote. Famously paired with running code in the IETF, it favours technically sound, broadly acceptable outcomes over precise headcounts. The chair judges the sense of the group, ensuring dissent is engaged with rather than merely outnumbered.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:rough-consensus", "labels": [ "Rough Consensus" ], "is_subclass_of": [ "Consensus" ], "wikilinks": [] }, { "id": "round-trip-time", "title": "Round-Trip Time", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Round-trip time is the duration measured from when a signal is sent to when its corresponding acknowledgement is received back at the origin, capturing the combined effect of propagation delay, processing delay, and queuing along the path. It is a standard diagnostic for network latency and is used to estimate achievable throughput, retransmission timers, and interactive responsiveness. Round-trip time grows with physical distance because of propagation delay but is also affected by congestion and routing.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:round-trip-time", "labels": [ "Round-Trip Time", "Round Trip Time" ], "is_subclass_of": [ "Network Latency" ], "wikilinks": [] }, { "id": "routing-algorithm", "title": "Routing Algorithm", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A routing algorithm is a procedure that determines the path along which data, messages, or value should travel from a source to a destination across a network of interconnected nodes. It typically models the network as a graph and selects routes by optimising criteria such as shortest path, lowest cost, available capacity, or reliability, while adapting to changing topology and link state. In blockchain payment networks such as the Lightning Network, routing algorithms find viable multi-hop paths across payment channels, balancing fees, liquidity, and privacy through techniques like source routing and onion encryption.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:routing-algorithm", "labels": [ "Routing Algorithm" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "routing-protocol", "title": "Routing Protocol", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A protocol that determines how packets are forwarded between nodes in a network by computing and distributing routing information among routers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:routing-protocol", "labels": [ "Routing Protocol", "AODV Routing", "Routing" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "Communication Protocol", "Network Architecture", "Interoperability", "Network Topology" ] }, { "id": "routing-table", "title": "Routing Table", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A routing table is a data structure, held in a router, switch or networked host, that stores the mappings between destination network addresses and the next hop or outgoing interface used to forward packets toward them. Each entry typically records a destination prefix, a next-hop address, an interface, and a metric or administrative cost used to select among competing routes. Routing tables are populated by static configuration and by dynamic routing protocols, and are consulted on every forwarding decision.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:routing-table", "labels": [ "Routing Table" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "routing", "title": "Routing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Routing is the process of selecting paths in a network along which data packets or messages are forwarded from source to destination. Routing algorithms and protocols such as BGP, OSPF, and segment routing determine optimal or policy-compliant paths based on metrics including latency, bandwidth, and cost. It is a foundational function of IP networks, software-defined networking, and overlay networks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:routing", "labels": [ "Routing" ], "is_subclass_of": [ "Network and Communications" ], "wikilinks": [] }, { "id": "royalty-distribution", "title": "Royalty Distribution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Automated, trustless distribution of royalty payments to creators and rights-holders via blockchain smart contracts, triggered on every secondary-market sale or licensing event, enforcing on-chain revenue sharing without intermediaries and enabling programmable intellectual property economics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:royalty-distribution", "labels": [ "Royalty Distribution", "Automated Royalty Distribution", "Royalty Splitting" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "Blockchain" ] }, { "id": "royalty-mechanism", "title": "Royalty Mechanism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Automated process ensuring creators receive compensation when their assets are resold or used in secondary markets.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:royalty-mechanism", "labels": [ "Royalty Mechanism" ], "is_subclass_of": [ "Governance and Safety", "Digital Rights Management" ], "wikilinks": [ "Creator Compensation", "Creator Wallet", "MSF Use Cases", "Payment Distribution", "Perpetual Revenue", "Rights Enforcement", "Rights Registry", "Royalty Calculation", "Transaction Tracking", "Blockchain Infrastructure", "Digital Rights Management", "Marketplace Integration", "MiddlewareLayer", "NFT Standard", "Smart Contract", "VirtualEconomyDomain" ] }, { "id": "rsa-algorithm", "title": "Rsa Algorithm", "domain": "security", "domain_name": "Security", "definition": "The RSA algorithm is a public-key cryptosystem whose security rests on the practical difficulty of factoring the product of two large prime numbers. A public key derived from this product encrypts data or verifies signatures, while the corresponding private key, recoverable only with knowledge of the prime factors, decrypts or signs. RSA was among the first practical asymmetric schemes and remains widely used for key exchange, digital signatures, and certificate-based authentication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:rsa-algorithm", "labels": [ "Rsa Algorithm", "RSA Algorithm" ], "is_subclass_of": [ "Public-Key Cryptography" ], "wikilinks": [] }, { "id": "rsa-signature", "title": "Rsa Signature", "domain": "security", "domain_name": "Security", "definition": "An RSA signature is a digital signature scheme built on the RSA public-key cryptosystem, where a message digest is signed with the holder's private exponent and verified with the corresponding public exponent and modulus. Security rests on the difficulty of factoring large composite integers, and practical deployments use padding schemes such as PKCS#1 v1.5 or PSS together with a cryptographic hash function. RSA signatures are widely used for certificate authorities, code signing, and transport-layer authentication, though they produce larger keys and signatures than elliptic-curve alternatives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:rsa-signature", "labels": [ "RSA Signature" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "rsa", "title": "Rsa", "domain": "security", "domain_name": "Security", "definition": "RSA is a public-key cryptosystem, named after Rivest, Shamir and Adleman, whose security rests on the computational difficulty of factoring the product of two large prime numbers. It supports both encryption \u2014 where a message encrypted with a public key can only be decrypted with the corresponding private key \u2014 and digital signatures, where a private key signs data that anyone can verify with the public key. As one of the earliest and most widely deployed asymmetric algorithms, RSA underpins much of the legacy public-key infrastructure, though it is gradually being supplemented by faster elliptic-curve schemes and, prospectively, post-quantum alternatives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:rsa", "labels": [ "Rsa", "RSA" ], "is_subclass_of": [ "Public Key Cryptography" ], "wikilinks": [] }, { "id": "rule-engine", "title": "Rule Engine", "domain": "governance", "domain_name": "Governance", "definition": "A rule engine is a software component that evaluates a set of declarative if-then rules against input data and executes the associated actions when conditions are met, separating business or policy logic from application code. It is a specialised form of inference engine that applies forward- or backward-chaining evaluation over a rule base rather than general-purpose reasoning. Rule engines underpin automated compliance and policy engine systems, where regulatory or organisational rules must be applied consistently and updated without redeploying software.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:rule-engine", "labels": [ "Rule Engine" ], "is_subclass_of": [ "Inference Engine" ], "wikilinks": [] }, { "id": "rule-of-law", "title": "Rule of Law", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The constitutional and institutional principle that all persons, institutions, and entities\u2014including AI systems and their operators\u2014are accountable to laws that are publicly promulgated, equally enforced, independently adjudicated, and consistent with international human rights norms. Applied to AI governance, the rule of law requires legal certainty, non-arbitrary decision-making, procedural fairness, and effective remedies for AI-related harms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rule-of-law", "labels": [ "Rule of Law", "Rule Of Law" ], "is_subclass_of": [ "Legal and Regulatory", "Legal Framework" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "rule-based-systems", "title": "Rule-Based Systems", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Rule-based systems are a class of knowledge-based AI systems that encode domain expertise as a collection of condition-action rules (IF THEN ) stored in a knowledge base, and employ an inference engine to match those rules against a working memory of current facts to derive new facts, trigger actions, or reach conclusions. The inference engine operates through either forward chaining (data-driven, propagating from known facts to goals) or backward chaining (goal-driven, working from desired conclusions back to supporting facts), enabling transparent, explainable reasoning where every inference can be traced to the specific rules that fired. Rule-based systems are the foundational architecture underlying classical expert systems, modern business rules engines, production systems, and many decision-automation pipelines across medicine, finance, manufacturing, and legal compliance. Their central tension lies between the interpretability of explicit symbolic rules and the scalability challenges that arise as rule sets grow large and rule interactions become complex.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:rule-based-systems", "labels": [ "Rule-Based Systems", "Rule Based System", "Rule-Based System" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Inference Engine", "Expert Systems", "Artificial Intelligence", "Knowledge Representation" ] }, { "id": "runbook", "title": "Runbook", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A runbook is a documented set of procedures for operating, maintaining and recovering a system, giving operators a repeatable sequence of steps for routine tasks and known failure scenarios. Modern runbooks range from human-readable checklists to executable automations that orchestrate remediation directly. They are a core artefact of site reliability engineering, reducing reliance on individual expertise during incidents.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:runbook", "labels": [ "Runbook" ], "is_subclass_of": [ "Documentation" ], "wikilinks": [] }, { "id": "runes-protocol", "title": "Runes Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Runes Protocol is a fungible token standard for the Bitcoin blockchain, introduced by Casey Rodarmor in 2024, that encodes token creation and transfer instructions directly into transaction outputs using the UTXO model. Unlike account-based token systems, Runes attaches a OP_RETURN-stored protocol message (a Runestone) to each transaction, assigning token balances to specific outputs so that ownership is tracked through spendable transaction outputs rather than a separate ledger. The protocol is designed to be more on-chain efficient than earlier Bitcoin-native token conventions such as BRC-20 and Ordinals-based token schemes, minimising UTXO proliferation while remaining fully compatible with the base-layer Bitcoin settlement mechanism. It reached significant ecosystem traction following its mainnet launch at the Bitcoin halving block in April 2024.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:runes-protocol", "labels": [ "Runes Protocol" ], "is_subclass_of": [ "Token Standard" ], "wikilinks": [ "UTXO", "Fungible Token", "Bitcoin", "Token Standard" ] }, { "id": "runes", "title": "Runes", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Runes is a protocol for issuing fungible tokens directly on the Bitcoin blockchain, using the transaction model and recorded in the witness data. It launched at the Bitcoin halving in April 2024.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:runes", "labels": [ "Runes" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Bitcoin Network", "Token", "Bitcoin Ordinals", "Bitcoin" ] }, { "id": "runtime-environment", "title": "Runtime Environment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Runtime Environment is the integrated set of software components, libraries, and managed services that provide the execution context in which application code operates at run time, encompassing memory management, instruction dispatch, system-call mediation, and API surface exposure. It abstracts the underlying hardware and operating system to deliver a consistent, portable execution substrate \u2014 whether a bytecode interpreter, a just-in-time compiler, a managed virtual machine, or a native process sandbox. Runtime environments coordinate lifecycle events (initialisation, garbage collection, signal handling, graceful shutdown) and expose introspection facilities such as profilers, debuggers, and telemetry hooks. They are foundational to cloud-native, edge, and immersive computing stacks, underpinning containerised microservices, WebAssembly workloads, game engine scripting layers, and on-device AI inference pipelines.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:runtime-environment", "labels": [ "Runtime Environment", "Host Runtime", "Runtime Engine" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "runtime-inspection", "title": "Runtime Inspection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Runtime Inspection is the practice of dynamically examining the internal state, activations, attention patterns, and behavioural properties of an AI model while it is executing inference, as opposed to static analysis of weights or architecture prior to deployment. It enables detection of unexpected or dangerous reasoning paths, verification of safety constraints during live operation, and post-hoc explanation of individual predictions by observing intermediate computations at the moment they occur.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:runtime-inspection", "labels": [ "Runtime Inspection" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Ai Governance", "Artificial Intelligence" ] }, { "id": "runtime-layer", "title": "Runtime Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Runtime Layer is the stratum that provides the execution environment in which higher-level code and models actually run. It sits above the Compute Layer that allocates resources and below the application and inference strata it hosts. It contains process and memory management, schedulers, sandboxes, and the libraries that support running programmes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:runtime-layer", "labels": [ "Runtime Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Compute Layer", "Inference Layer", "Application Layer", "Virtual Machine", "Garbage Collection", "owl:Thing" ] }, { "id": "russell-group", "title": "Russell Group", "domain": "ai", "domain_name": "Ai", "definition": "The Russell Group is a self-selecting association of leading research-intensive universities in the United Kingdom that advocates collectively on research funding, policy, and higher-education strategy. Its members are characterised by substantial research output, strong postgraduate provision, and extensive industry and international links. Membership is frequently used as a shorthand for institutions with high research standing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:russell-group", "labels": [ "Russell Group" ], "is_subclass_of": [ "Education" ], "wikilinks": [] }, { "id": "rust-systems-programming-language", "title": "Rust Systems Programming Language", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Rust is a compiled, statically typed systems programming language created at Mozilla Research and first released in 2015, designed to provide C-like performance and low-level memory control without the memory safety vulnerabilities that plague C and C++. Its defining innovation is the ownership-and-borrowing type system, which enforces at compile time that each value has exactly one owner, references obey strict lifetime rules, and data races are structurally impossible \u2014 all without a garbage collector. Rust has rapidly become the preferred language for writing safe, high-performance systems software, blockchain runtimes, WebAssembly modules, embedded firmware, and operating system kernels.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:rust-systems-programming-language", "labels": [ "Rust Systems Programming Language", "Rust", "Rust Programming Language", "Rust Runtime" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "s3-api", "title": "S3 Api", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The S3 API is a RESTful, HTTP-based application programming interface for object storage that exposes buckets and objects through operations such as PUT, GET, DELETE and LIST. Originating with Amazon Simple Storage Service, it has become a de facto standard implemented by many cloud and on-premises storage systems, enabling portable, vendor-neutral object access. Clients authenticate with signed requests and address resources by bucket and key, allowing scalable, durable storage of unstructured data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:s3-api", "labels": [ "S3 Api", "S3 API" ], "is_subclass_of": [ "Object Storage" ], "wikilinks": [] }, { "id": "sa8000", "title": "SA8000", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "SA8000 is a voluntary, auditable social-accountability standard developed by Social Accountability International that certifies organisations against decent-work criteria. It covers child and forced labour, health and safety, freedom of association, discrimination, working hours, and management systems, drawing on ILO conventions and human-rights instruments. Certification is used to demonstrate ethical labour practices across supply chains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sa8000", "labels": [ "SA8000", "SA8000 Labour Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "sae-j-3016", "title": "SAE J3016", "domain": "standards", "domain_name": "Standards", "definition": "SAE J3016 is a foundational technical standard published by SAE International that defines a six-level taxonomy of driving automation, ranging from Level 0 (no automation) through Level 5 (full automation), by specifying the allocation of the dynamic driving task between the human driver and the automated driving system. First issued in 2014 and subsequently revised in 2016, 2018, and 2021, the standard establishes precise vocabulary \u2014 including terms such as dynamic driving task, operational design domain, and minimal risk condition \u2014 that underpins regulatory frameworks, vehicle development programmes, and public communication about autonomous and semi-autonomous vehicles worldwide. The taxonomy distinguishes Levels 1\u20132, where a human driver must supervise and remain ready to intervene, from Levels 3\u20135, where the automated system assumes full authority over the driving task within a defined operational design domain.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:sae-j-3016", "labels": [ "SAE J3016", "SAE Levels of Driving Automation" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Standards Body", "Autonomous Vehicle" ] }, { "id": "saml-2-0", "title": "SAML 2.0", "domain": "security", "domain_name": "Security", "definition": "Security Assertion Markup Language 2.0 (SAML 2.0) is an OASIS open standard ratified in 2005 that defines an XML-based framework for exchanging authentication and authorisation data between an Identity Provider and a Service Provider. It enables web-based federated single sign-on by allowing an IdP to issue digitally signed XML assertions attesting a user's identity and attributes, which the SP accepts without requiring the user to re-authenticate. SAML 2.0 consolidates and supersedes SAML 1.0 and 1.1, Liberty Alliance ID-FF 1.2, and Shibboleth 1.3, incorporating bindings for HTTP Redirect, POST, Artifact, and SOAP transports. The standard is governed by OASIS and remains the dominant federation protocol in enterprise, higher-education, and government identity ecosystems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:saml-2-0", "labels": [ "SAML 2.0", "SAML 2.0 Federation" ], "is_subclass_of": [ "Federated Identity" ], "wikilinks": [ "Authentication", "Single Sign-On", "Identity Management" ] }, { "id": "saml", "title": "SAML", "domain": "security", "domain_name": "Security", "definition": "Security Assertion Markup Language (SAML) is an XML-based open standard developed by OASIS that defines formats and protocols for exchanging authentication and authorisation data between an identity provider (IdP) and a service provider (SP). SAML 2.0, published in 2005, is the dominant version in enterprise deployments and enables web-based single sign-on by allowing the IdP to issue digitally signed XML assertions that service providers trust without requiring the user to re-authenticate. It decouples the identity layer from application services, enabling cross-domain federated identity across organisational and cloud boundaries.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:saml", "labels": [ "SAML" ], "is_subclass_of": [ "Federated Identity" ], "wikilinks": [ "Authentication", "Single Sign-On", "Identity Management" ] }, { "id": "sar-coherence", "title": "SAR Coherence", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sar-coherence", "labels": [ "SAR Coherence" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sar-terrain-correction", "title": "SAR Terrain Correction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sar-terrain-correction", "labels": [ "SAR Terrain Correction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sbti", "title": "SBTi", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The Science Based Targets initiative (SBTi) is a body that defines and validates corporate greenhouse-gas reduction targets against climate science, specifically the goal of limiting warming to 1.5C. It publishes sector-specific methodologies and independently certifies that company targets are consistent with the Paris Agreement. SBTi validation has become a widely referenced benchmark in ESG reporting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sbti", "labels": [ "SBTi" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "scada", "title": "SCADA", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SCADA (Supervisory Control and Data Acquisition) is a control-system architecture that uses computers, networked data communications and graphical interfaces to monitor and supervise industrial processes across geographically distributed sites. It gathers real-time telemetry from field sensors and controllers, presents it to operators, and issues supervisory commands back to actuators and programmable controllers. SCADA underpins critical infrastructure such as power grids, water treatment, manufacturing and pipeline operations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scada", "labels": [ "SCADA" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "scara-robot", "title": "SCARA Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "Selective Compliance Articulated Robot Arm \u2014 a four-axis industrial robot with two rotary joints in the horizontal plane that give high compliance laterally whilst providing rigid vertical stiffness along the Z-axis. SCARA robots excel at high-speed pick-and-place, assembly, and packaging tasks that require precise horizontal positioning but resist vertical forces.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:scara-robot", "labels": [ "SCARA Robot" ], "is_subclass_of": [ "Robot Type", "Industrial Robot" ], "wikilinks": [ "Industrial Robot", "Robotics" ] }, { "id": "scim", "title": "SCIM", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SCIM (System for Cross-domain Identity Management) is an open standard that defines a common schema and a RESTful protocol for automating the exchange of user and group identity information between identity providers and service providers. It lets organisations create, update, deactivate and deprovision accounts across applications without bespoke integrations. SCIM is widely used to automate user lifecycle management in enterprise single sign-on and cloud deployments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:scim", "labels": [ "SCIM" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "sd-jwt-vc", "title": "SD-JWT VC", "domain": "security", "domain_name": "Security", "definition": "A verifiable credential format that uses Selective Disclosure JSON Web Tokens, allowing a holder to reveal only chosen claims to a verifier. It combines the JWT structure with salted hashes that support selective disclosure.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:sd-jwt-vc", "labels": [ "SD-JWT VC" ], "is_subclass_of": [ "JSON Web Token" ], "wikilinks": [ "JSON Web Token", "Selective Disclosure", "Digital Identity", "Privacy" ] }, { "id": "sd-jwt", "title": "SD-JWT", "domain": "security", "domain_name": "Security", "definition": "SD-JWT (Selective Disclosure JSON Web Token) is an IETF-specified extension to the JSON Web Token standard that allows an Issuer to create a signed token containing hashed claim values, from which the Holder can selectively reveal only the specific claims needed for a given presentation \u2014 without exposing other claims or enabling correlation across presentations. The mechanism uses SHA-256 salted disclosure objects appended to the base JWT; verifiers can validate revealed disclosures against the issuer signature whilst remaining blind to undisclosed claims. SD-JWT forms the primary credential format for the European Union's eIDAS 2.0 digital identity wallet system.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sd-jwt", "labels": [ "SD-JWT", "IETF SD-JWT" ], "is_subclass_of": [ "Selective Disclosure" ], "wikilinks": [] }, { "id": "sdf", "title": "SDF", "domain": "robotics", "domain_name": "Robotics", "definition": "SDF (Simulation Description Format) is an XML-based format for describing robots, sensors, and environments for simulation and control. Developed alongside the Gazebo simulator, it specifies kinematic and dynamic properties, links, joints, visual and collision geometry, and world contents. It is widely used in robotics for modelling platforms and scenes consistently across simulation and tooling.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sdf", "labels": [ "SDF" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "sdxl", "title": "SDXL", "domain": "ai", "domain_name": "Ai", "definition": "SDXL (Stable Diffusion XL) is a large-scale latent diffusion model released by Stability AI in 2023, comprising a 3.5-billion-parameter UNet and an ensemble of two CLIP text encoders that condition image generation at native 1024\u00d71024 resolution. It introduces a two-stage architecture \u2014 a base model followed by a refinement model \u2014 and supports advanced conditioning mechanisms including aesthetic scoring and crop coordinates.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sdxl", "labels": [ "SDXL", "SDXL Refiner Handoff" ], "is_subclass_of": [ "Latent Diffusion", "Diffusion Models" ], "wikilinks": [] }, { "id": "sec", "title": "SEC", "domain": "governance", "domain_name": "Governance", "definition": "The Securities and Exchange Commission (SEC) is the principal US federal regulatory agency responsible for enforcing federal securities laws, overseeing securities markets, and protecting investors. Established by the Securities Exchange Act of 1934 in the wake of the 1929 stock market crash, the SEC requires public companies to disclose material financial information and polices against fraud, insider trading, and market manipulation. In the digital asset domain, the SEC has asserted jurisdiction over many cryptocurrencies and token offerings on the grounds that they constitute investment contracts under the Howey Test, leading to high-profile enforcement actions against exchanges, issuers, and DeFi protocols. The SEC's regulatory posture towards crypto assets is a defining factor in the shape of the US digital asset industry.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sec", "labels": [ "SEC", "SEC Crypto Enforcement" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "sha-256-hashing", "title": "SHA-256 Hashing", "domain": "security", "domain_name": "Security", "definition": "SHA-256 Hashing refers to the application of the SHA-256 (Secure Hash Algorithm 256-bit) cryptographic hash function \u2014 a member of the SHA-2 family standardised by NIST in 2001 \u2014 to produce a fixed-length 256-bit (32-byte) message digest from an arbitrary-length input, with the properties of determinism, pre-image resistance, second pre-image resistance, and collision resistance. The algorithm operates via 64 rounds of bit manipulation, modular addition, and nonlinear functions over a 512-bit block schedule. SHA-256 is the foundational hash function of Bitcoin's proof-of-work mining, block header commitments, and Merkle tree construction, as well as a critical primitive in TLS, code signing, and certificate transparency.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sha-256-hashing", "labels": [ "SHA-256 Hashing" ], "is_subclass_of": [ "Cryptographic Hash Function" ], "wikilinks": [] }, { "id": "sha-256", "title": "SHA-256", "domain": "security", "domain_name": "Security", "definition": "Secure Hash Algorithm producing a deterministic 256-bit digest from arbitrary input, forming the cryptographic foundation of Bitcoin's proof-of-work, Merkle tree construction, block header chaining, and address derivation in most major blockchain systems. Produces avalanche-effect collision resistance with 2^128 preimage security. Standardised in FIPS 180-4 by NIST as part of the SHA-2 family.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:sha-256", "labels": [ "SHA-256", "SHA-256 Hash Function", "SHA256" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "sha-3", "title": "SHA-3", "domain": "security", "domain_name": "Security", "definition": "SHA-3 is a family of cryptographic hash functions standardised by NIST in FIPS 202, based on the Keccak sponge construction rather than the Merkle-Damgaard design of SHA-2. Its sponge structure provides strong resistance to length-extension attacks and offers configurable output lengths and extendable-output functions. It is used as a collision-resistant primitive in digital signatures and integrity verification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sha-3", "labels": [ "SHA-3" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "shacl", "title": "SHACL", "domain": "data", "domain_name": "Data", "definition": "SHACL (Shapes Constraint Language) is a W3C standard for validating RDF graphs against a set of declarative conditions expressed as shapes. It defines constraints on the structure, datatypes, cardinality and value ranges that nodes in a graph must satisfy, and reports conformance results identifying any violations. SHACL enables data quality assurance and interface contracts for linked data, complementing OWL by focusing on validation rather than inference.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:shacl", "labels": [ "SHACL" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "shap", "title": "SHAP", "domain": "ai", "domain_name": "Ai", "definition": "SHAP, short for SHapley Additive exPlanations, is a method for explaining the output of machine-learning models by attributing each prediction to its input features. It is grounded in Shapley values from cooperative game theory, which fairly distribute a payoff among contributors, treating each feature as a player and the prediction as the payoff. SHAP provides locally accurate, consistent feature attributions and unifies several earlier explanation techniques under a common framework.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:shap", "labels": [ "SHAP", "SHAP (Shapley Additive Explanations)", "SHAP Values" ], "is_subclass_of": [ "Explainable AI", "Machine Learning Domain" ], "wikilinks": [ "Shapley Value", "Machine Learning Model", "Model Interpretability", "Feature Attribution", "Explainable AI", "LIME", "Machine Learning Domain", "Lundberg and Lee, A Unified Approach to Interpreting Model Predictions (2017)" ] }, { "id": "siem", "title": "SIEM", "domain": "security", "domain_name": "Security", "definition": "SIEM (Security Information and Event Management) is a platform that aggregates, normalises, and correlates log and event data from across an organisation's IT estate to detect threats and support compliance. It combines real-time alerting, anomaly detection, and historical search with dashboards and audit reporting. It is a foundational tool of security operations centres and regulatory compliance programmes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:siem", "labels": [ "SIEM" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "siggraph-pipeline-wg", "title": "SIGGRAPH Pipeline WG", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The SIGGRAPH Pipeline Working Group is an industry effort associated with the SIGGRAPH community that addresses interoperability and best practice across the stages of 3D content production pipelines.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:siggraph-pipeline-wg", "labels": [ "SIGGRAPH Pipeline WG" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [ "3D Asset Standard", "Asset Interoperability", "3D Content Pipeline", "MaterialX", "SIGGRAPH", "Standards Body" ] }, { "id": "siggraph", "title": "SIGGRAPH", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "SIGGRAPH (Special Interest Group on Computer Graphics and Interactive Techniques) is the ACM's premier annual conference on computer graphics and interactive technology, considered the world's most prestigious venue for publishing advances in rendering, animation, simulation, geometry processing, and human-computer interaction. First held in 1974, SIGGRAPH has been the introduction point for transformative techniques including ray tracing, physically based rendering, GPU shading languages, neural rendering, and real-time global illumination. Its technical papers programme is among the most selective in computer science, with acceptance rates typically below 25 per cent.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:siggraph", "labels": [ "SIGGRAPH", "ACM SIGGRAPH" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "simd", "title": "SIMD", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SIMD (Single Instruction, Multiple Data) is a class of parallel computing hardware and instruction sets in which one instruction operates simultaneously on multiple data elements packed into a wide register, rather than processing them one at a time. It is implemented in CPU vector extensions and underlies the lock-step execution model of GPU cores. Numerical libraries such as NumPy and GPU programming frameworks rely on SIMD execution to accelerate array and tensor operations.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:simd", "labels": [ "SIMD" ], "is_subclass_of": [ "Parallel Computing" ], "wikilinks": [] }, { "id": "siopv2", "title": "SIOPv2", "domain": "security", "domain_name": "Security", "definition": "SIOPv2 (Self-Issued OpenID Provider v2) is an OpenID Foundation specification that extends OpenID Connect so a user's own wallet acts as the identity provider, authenticating with self-issued and decentralised identifiers rather than a third-party server. It lets relying parties verify a subject controls a DID and, paired with OpenID for Verifiable Presentations, accept verifiable credentials. It is a key protocol for decentralised, wallet-centric digital identity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:siopv2", "labels": [ "SIOPv2" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "slam-toolbox", "title": "SLAM Toolbox", "domain": "robotics", "domain_name": "Robotics", "definition": "SLAM Toolbox is an open-source ROS package for 2D simultaneous localisation and mapping using laser scan matching and pose-graph optimisation. It supports online mapping, lifelong mapping with map serialisation and continuation, and localisation against pre-built maps, making it a default mapping stack for indoor mobile robots. It is widely deployed on ground robots running the ROS navigation stack.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:slam-toolbox", "labels": [ "SLAM Toolbox" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "slam", "title": "SLAM", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Simultaneous Localization and Mapping (SLAM) is a robotics and computer vision technique enabling devices to build maps of unknown environments whilst simultaneously tracking their own position within those environments, combining localisation and map construction in real-time using probabilistic state estimation and sensor fusion.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:slam", "labels": [ "SLAM", "Cartographer SLAM", "FastSLAM", "SLAM Algorithm", "SLAM Technology", "SLAM Tracking" ], "is_subclass_of": [ "Spatial Mapping" ], "wikilinks": [ "AutonomousVehicles", "buildsMap", "dt:enablesFor", "dt:enhancedBy", "dt:optimizedBy", "dt:usedIn", "dt:validatedIn", "estimatesState", "GraphOptimization", "implementsAlgorithm", "LiDAR", "LocalizationTechnique", "LoopClosure", "MapConstruction", "Mapping", "OccupancyGrid", "RGBDCamera", "usesSensor", "VisualOdometry", "AugmentedReality" ] }, { "id": "sme-ai-productivity-toolkit", "title": "SME AI Productivity Toolkit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Consumer Tools for SMEs encompasses the accessible AI-powered productivity applications, creative tools, and automation platforms used by small and medium enterprises. These tools lower the barrier to adopting generative AI and workflow automation without requiring technical expertise or enterprise-scale infrastructure investment.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:sme-ai-productivity-toolkit", "labels": [ "SME AI Productivity Toolkit", "Consumer Tools for SMEs" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Blockchain", "Gemini" ] }, { "id": "smpl-body-model", "title": "SMPL Body Model", "domain": "ai", "domain_name": "Ai", "definition": "SMPL (Skinned Multi-Person Linear model) is a parametric 3D human body model that represents body shape and pose with a low-dimensional set of parameters driving a deformable triangle mesh. It separates identity-dependent shape blend shapes from pose-dependent corrective deformations, enabling realistic articulated bodies that fit motion capture and image data. It is a foundational representation in human pose and shape estimation pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:smpl-body-model", "labels": [ "SMPL Body Model", "Body Model", "SMPL" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "smpte-st-2128", "title": "SMPTE ST 2128", "domain": "standards", "domain_name": "Standards", "definition": "SMPTE ST 2128 is a standard from the Society of Motion Picture and Television Engineers in the area of professional media and video signalling. It forms part of the SMPTE family of broadcast and production standards.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:smpte-st-2128", "labels": [ "SMPTE ST 2128", "SMPTE NDI Standard" ], "is_subclass_of": [ "Standards and Interoperability", "owl:Thing" ], "wikilinks": [ "Content Delivery", "Video Compression", "owl:Thing" ] }, { "id": "smart-bft", "title": "SMaRT-BFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "SMaRT-BFT is a Byzantine fault tolerant consensus approach built on the SMaRt (State Machine Replication) library, used to order transactions deterministically across a fixed set of validating nodes. It tolerates up to f faulty replicas out of 3f+1 total while guaranteeing safety and liveness under partial synchrony. It is commonly deployed in permissioned ledgers where validator identity is known and high throughput with finality is required.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:smart-bft", "labels": [ "SMaRT-BFT" ], "is_subclass_of": [ "Consensus Protocol" ], "wikilinks": [] }, { "id": "soap", "title": "SOAP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SOAP (Simple Object Access Protocol) is a standardised, XML-based messaging protocol for exchanging structured information between applications, typically over HTTP but also over other transports. It defines an envelope structure with a header and body, supports remote procedure calls and document-style messaging, and is described by WSDL contracts and extended by the WS-* specifications for security, transactions, and reliable messaging. SOAP underpinned the first generation of enterprise web services and remains common in regulated and legacy integration scenarios, contrasting with the lighter, resource-oriented REST style.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:soap", "labels": [ "SOAP" ], "is_subclass_of": [ "Web Services" ], "wikilinks": [] }, { "id": "soar", "title": "SOAR", "domain": "security", "domain_name": "Security", "definition": "SOAR (Security Orchestration, Automation and Response) is a class of platforms that coordinate detection tools, automate repetitive incident-handling workflows, and orchestrate response actions across a security stack through codified playbooks. It ingests alerts from disparate sources, enriches them with context, and executes graded responses to reduce analyst workload and mean time to respond. The acronym also denotes Soar, a cognitive architecture for modelling general intelligence, but in security contexts the orchestration sense dominates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:soar", "labels": [ "SOAR", "Soar Workshop" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "sparql-examples", "title": "SPARQL EXAMPLES", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "SPARQL (SPARQL Protocol and RDF Query Language) is the W3C-standardised query language for RDF-based knowledge graphs and linked data endpoints, supporting SELECT, CONSTRUCT, ASK, and DESCRIBE query forms, as well as federated queries across distributed SPARQL endpoints and SPARQL Update for graph mutation. SPARQL 1.1 (2013) added aggregates, subqueries, and property paths; SPARQL 1.2 (2023) further refined the protocol and service description vocabulary. It is the primary access interface for semantic knowledge graph exploration, ontology navigation, and cross-domain data integration.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:sparql-examples", "labels": [ "SPARQL EXAMPLES" ], "is_subclass_of": [ "Standards and Interoperability", "Semantic Web Standards" ], "wikilinks": [ "ASKQuery", "CONSTRUCTQuery", "DESCRIBEQuery", "KnowledgeGraphQuerying", "LinkedData", "RDF", "SELECTQuery", "W3C", "MetaverseDomain" ] }, { "id": "sparql-endpoint", "title": "SPARQL Endpoint", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A SPARQL Endpoint is a network-accessible service that accepts SPARQL Protocol and RDF Query Language queries and returns structured results over HTTP, enabling federated access to RDF knowledge graphs and linked data stores. It acts as the primary interface between client applications and triple stores, supporting SELECT, CONSTRUCT, ASK, and UPDATE operations. SPARQL Endpoints are foundational to the Semantic Web stack and to provenance-aware knowledge graph systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:sparql-endpoint", "labels": [ "SPARQL Endpoint" ], "is_subclass_of": [ "Standards and Interoperability", "Data Access Interface" ], "wikilinks": [ "Data Access Interface" ] }, { "id": "sparql", "title": "sparql", "domain": "data", "domain_name": "Data", "definition": "SPARQL (SPARQL Protocol and RDF Query Language) is a W3C Recommendation defining a query language, update language, and HTTP-based protocol for retrieving and modifying data stored in RDF graph databases and Linked Data endpoints. SPARQL 1.1 (2013) introduced federated queries across multiple remote endpoints via the SERVICE keyword, property paths for graph traversal, aggregation functions, and the SPARQL Update (SPARUL) sublanguage for graph mutation. It serves as the standard access layer for the Semantic Web and knowledge graph ecosystems, occupying a role analogous to SQL for relational databases. The four primary query forms \u2014 SELECT, CONSTRUCT, ASK, and DESCRIBE \u2014 provide flexible mechanisms for retrieving variable bindings, materialising derived graphs, testing graph patterns, and describing resources.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:sparql", "labels": [ "SPARQL", "W3C SPARQL Specification" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "spir-v", "title": "SPIR-V", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SPIR-V is a binary intermediate representation for shaders and parallel-compute kernels standardised by the Khronos Group, used as the portable bytecode target for Vulkan, OpenCL, and OpenGL. High-level shading languages such as GLSL and HLSL compile to SPIR-V, which drivers then translate to native GPU instructions, decoupling source languages from hardware. It enables offline compilation, validation, and optimisation of graphics and compute programs across vendors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:spir-v", "labels": [ "SPIR-V" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "sql", "title": "SQL", "domain": "data", "domain_name": "Data", "definition": "SQL (Structured Query Language) is a declarative, domain-specific language for defining, manipulating, and querying data held in relational database management systems. A user expresses the desired result set through clauses such as SELECT, FROM, WHERE, JOIN, and GROUP BY, while the database query optimiser determines the physical execution plan. Standardised by ANSI and ISO since 1986, SQL spans data definition (DDL), data manipulation (DML), and transaction control, and remains the predominant interface for structured data despite the rise of NoSQL alternatives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:sql", "labels": [ "SQL" ], "is_subclass_of": [ "Database System" ], "wikilinks": [] }, { "id": "ssim-loss", "title": "SSIM Loss", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "SSIM loss is a training objective derived from the Structural Similarity Index that penalises differences in luminance, contrast, and structure between a rendered or reconstructed image and its target, typically computed as one minus the SSIM score. It is used alongside or instead of pixel-wise losses such as L1 or L2 because it better reflects perceptual image quality. SSIM loss is a common component of the optimisation objective in 3D Gaussian Splatting and other differentiable rendering pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ssim-loss", "labels": [ "SSIM Loss" ], "is_subclass_of": [ "Loss Function" ], "wikilinks": [] }, { "id": "ssim-metric", "title": "SSIM Metric", "domain": "ai", "domain_name": "Ai", "definition": "The Structural Similarity Index (SSIM) is a perceptual image-quality metric that compares two images by modelling luminance, contrast, and structural correlation over local windows, rather than measuring pixel-wise error alone. It correlates better with human judgement of quality than mean squared error or PSNR, producing a score between -1 and 1 where 1 indicates identical structure. It is widely used to evaluate compression, restoration, and generative reconstruction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ssim-metric", "labels": [ "SSIM Metric" ], "is_subclass_of": [ "Evaluation Metric", "Model Performance" ], "wikilinks": [] }, { "id": "stark", "title": "STARK", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A STARK is a scalable transparent argument of knowledge, a cryptographic proof system that lets a verifier check a computation was performed correctly without a trusted setup and with proof verification far cheaper than re-execution.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:stark", "labels": [ "STARK", "STARK Proof", "STARK Proofs" ], "is_subclass_of": [ "Cryptography", "Cryptography Domain" ], "wikilinks": [ "Zero-Knowledge Proof", "Verifiable Computation", "Cryptographic Hash Function", "Blockchain", "Cryptography Domain" ] }, { "id": "strips", "title": "STRIPS", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "STRIPS (Stanford Research Institute Problem Solver) is a classical automated planning formalism that represents world states as conjunctions of propositions and defines actions via precondition, add-list, and delete-list operators. It introduced the core abstraction underlying modern planning languages such as PDDL and remains foundational to symbolic AI planning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:strips", "labels": [ "STRIPS" ], "is_subclass_of": [ "AI Technique", "Knowledge Representation" ], "wikilinks": [ "Knowledge Representation" ] }, { "id": "stun-and-turn", "title": "STUN and TURN", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "STUN (Session Traversal Utilities for NAT) and TURN (Traversal Using Relays around NAT) are complementary protocols used to establish peer-to-peer connectivity across Network Address Translators and firewalls. STUN allows a client to discover its public IP address and port mapping, while TURN provides a relay server as a fallback when direct connectivity cannot be achieved. Together they form the ICE (Interactive Connectivity Establishment) framework used by WebRTC.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:stun-and-turn", "labels": [ "STUN and TURN", "STUN Server", "STUN TURN Servers" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "summary", "title": "SUMMARY", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The metaverse represents a convergent technological paradigm combining virtual reality, augmented reality, and extended reality to create persistent, 3D virtual shared environments enabling user interaction, economic transactions, and social collaboration. This digital infrastructure integrates blockchain-based assets, smart contracts, and decentralised identity systems to support genuine digital ownership, interoperability, and governance across platforms.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:summary", "labels": [ "SUMMARY" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "BlockchainTechnology", "DecentralizedIdentity", "ExtendedReality", "AugmentedReality", "MetaverseDomain", "SmartContracts", "VirtualReality" ] }, { "id": "swift-messaging", "title": "SWIFT Messaging", "domain": "finance", "domain_name": "Finance", "definition": "SWIFT Messaging is the standardised, structured financial message exchange system operated over the Society for Worldwide Interbank Financial Telecommunication (SWIFT) network, enabling banks, brokers, custodians, and other financial institutions to communicate payment instructions, securities transactions, treasury operations, and trade finance data. Messages are governed by two principal format families: the legacy MT (Message Type) standards encoded in SWIFT's own syntax, and the modern ISO 20022 MX messages encoded in XML, which carry richer, more granular data and support global harmonisation. SWIFT itself functions as a secure messaging infrastructure rather than a settlement system; actual fund movement occurs via correspondent banking relationships, real-time gross settlement (RTGS) systems, and central bank facilities. The network is subject to multi-jurisdictional regulatory oversight and is a critical node in the global financial stability architecture.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:swift-messaging", "labels": [ "SWIFT Messaging", "SWIFT MT", "SWIFT Messaging Network" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [ "SWIFT", "Payment System", "CPMI-IOSCO PFMI" ] }, { "id": "swift", "title": "SWIFT", "domain": "finance", "domain_name": "Finance", "definition": "SWIFT, the Society for Worldwide Interbank Financial Telecommunication, is a cooperative that operates a secure messaging network used by banks and other financial institutions to exchange payment and securities instructions. It does not move money itself but standardises the messages that instruct transfers between institutions, which then settle through correspondent banking relationships or payment systems. Founded in 1973 and based in Belgium, it connects thousands of institutions across most countries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:swift", "labels": [ "SWIFT", "SWIFT Network", "Swift" ], "is_subclass_of": [ "Financial Infrastructure", "Financial Infrastructure Domain" ], "wikilinks": [ "Correspondent Banking", "Cross-Border Payments", "Interbank Settlement", "Payment Systems Domain", "ISO 20022", "Financial Infrastructure Domain" ] }, { "id": "saas-pricing-models", "title": "SaaS Pricing Models", "domain": "economics", "domain_name": "Economics", "definition": "The various commercial frameworks, such as per-seat, usage-based, or outcome-based, used by Software-as-a-Service providers to monetize access to their applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:saas-pricing-models", "labels": [ "SaaS Pricing Models" ], "is_subclass_of": [ "Software As A Service" ], "wikilinks": [] }, { "id": "sablier", "title": "Sablier", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Ethereum protocol for token streaming that pays a recipient continuously over time by releasing funds from a smart contract at a defined rate.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:sablier", "labels": [ "Sablier" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Smart Contract", "Ethereum", "DeFi", "Token Economics" ] }, { "id": "safe-ai-deployment", "title": "Safe AI Deployment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Safe AI deployment is the set of engineering and governance practices applied when releasing an AI system into production so that it operates within intended bounds, degrades gracefully, and remains subject to human oversight. It draws on techniques from AI safety and AI alignment, including staged rollouts, monitoring, kill switches, and red-teaming, to reduce the risk of harmful or unintended behaviour once a model is exposed to real users. It is the operational counterpart to alignment research, translating safety properties validated offline into guarantees that hold in live systems.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:safe-ai-deployment", "labels": [ "Safe AI Deployment", "Safe Deployment" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "safe-human-robot-interaction", "title": "Safe Human-Robot Interaction", "domain": "robotics", "domain_name": "Robotics", "definition": "Safe human-robot interaction is the body of methods and controls that allow robots to operate in close proximity to people without causing harm. It combines compliant control, force and contact limiting, collision avoidance, and speed-and-separation monitoring so that physical contact, when it occurs, stays within safe limits. It is essential for collaborative robots that share workspaces with humans.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:safe-human-robot-interaction", "labels": [ "Safe Human-Robot Interaction", "Physical Human Robot Interaction Safety", "Safe Human Interaction", "Safe Physical Human-Robot Interaction", "Safe Physical Interaction" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [] }, { "id": "safe-immersive-experience", "title": "Safe Immersive Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Design principles and technical safeguards that protect users from physical harm, psychological distress, and privacy violations in virtual and augmented reality environments through content moderation, comfort settings, and safety boundaries.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:safe-immersive-experience", "labels": [ "Safe Immersive Experience" ], "is_subclass_of": [ "Governance and Safety", "User Safety" ], "wikilinks": [ "Responsible XR Design", "metaverse", "User Safety" ] }, { "id": "safe", "title": "Safe", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart contract wallet platform on Ethereum and compatible chains, formerly Gnosis Safe, that provides multi-signature account management and programmable transaction approval.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:safe", "labels": [ "Safe" ], "is_subclass_of": [ "Multi-Signature Wallet" ], "wikilinks": [ "Smart Contract", "Ethereum", "Account Abstraction", "Gnosis Safe", "Wallet", "Multi-Signature Wallet" ] }, { "id": "safe-snap", "title": "SafeSnap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "SafeSnap is a tool that connects off-chain Snapshot governance votes to on-chain execution through a Gnosis Safe and the Reality.eth oracle. It lets DAOs enact decisions without paying gas to vote.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:safe-snap", "labels": [ "SafeSnap", "On-Chain Execution via SafeSnap", "SafeSnap Module" ], "is_subclass_of": [ "DAO Governance" ], "wikilinks": [ "Snapshot", "Reality.eth", "Decentralized Autonomous Organization", "Smart Contract", "DAO Governance" ] }, { "id": "safetensors-format", "title": "Safetensors Format", "domain": "ai", "domain_name": "Ai", "definition": "Safetensors is a file format for storing tensors \u2014 the weight matrices of machine-learning models \u2014 designed to be safe, fast, and simple. Unlike Python pickle-based formats, safetensors stores only raw tensor data and a JSON header describing shapes, dtypes, and offsets, so loading a file cannot execute arbitrary code. The layout supports zero-copy and memory-mapped loading, enabling rapid model initialisation and lazy access to individual tensors. Developed by Hugging Face, it has become a de facto standard for distributing open model weights.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:safetensors-format", "labels": [ "Safetensors Format" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "safetensors", "title": "Safetensors", "domain": "ai", "domain_name": "Ai", "definition": "Safetensors is a secure, simple, and fast file format for storing and loading neural network tensor weights, developed by Hugging Face as a safe alternative to Python's pickle-based serialisation. The format stores tensor metadata in a JSON header followed by raw binary data, enabling zero-copy memory-mapped loading without executing arbitrary code during deserialisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:safetensors", "labels": [ "Safetensors" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [] }, { "id": "safety-oecd", "title": "Safety (OECD)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Under the OECD AI Principles (Principle 1.4, updated 2024), AI systems must operate without posing unacceptable risks of physical, psychological, or environmental harm. This requires lifecycle-spanning hazard analysis, fail-safe design, continuous monitoring, and incident response mechanisms, with responsibilities shared between providers who design safety in and deployers who maintain operational safety.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:safety-oecd", "labels": [ "Safety (OECD)" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "safety-assessment", "title": "Safety Assessment", "domain": "governance", "domain_name": "Governance", "definition": "Safety assessment is a systematic analytical process that identifies, evaluates, and mitigates hazards associated with a system, product, or operational process to demonstrate that residual risk is tolerable within accepted standards and regulatory frameworks. It encompasses methods such as Failure Modes and Effects Analysis (FMEA), Fault Tree Analysis (FTA), Hazard and Operability Studies (HAZOP), and probabilistic safety analysis, applied across domains including aerospace, automotive, nuclear, medical devices, and increasingly AI systems. Safety assessments form the evidentiary basis for certification against safety standards such as IEC 61508 and ISO 26262.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:safety-assessment", "labels": [ "Safety Assessment", "Independent Safety Assessment" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "safety-assurance", "title": "Safety Assurance", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety assurance is the systematic process of building and demonstrating justified confidence that a system will operate safely throughout its lifecycle. It draws on hazard analysis, structured safety cases, verification evidence, and continuous monitoring to argue that residual risk is acceptable. For autonomous and industrial systems it provides the evidentiary basis for deployment and regulatory acceptance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:safety-assurance", "labels": [ "Safety Assurance" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "safety-case", "title": "Safety Case", "domain": "robotics", "domain_name": "Robotics", "definition": "A safety case is a structured, evidence-backed argument that a system is acceptably safe to operate in a defined context and operational environment. It comprises explicit safety claims, the reasoning that connects them, and the supporting evidence drawn from hazard analysis, design measures, verification, and operational data. Commonly documented using notations such as the Goal Structuring Notation or Claims-Arguments-Evidence, the safety case is reviewed by stakeholders and regulators and maintained throughout the system lifecycle. It is the central assurance artefact in safety-critical and autonomous-systems engineering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:safety-case", "labels": [ "Safety Case" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [] }, { "id": "safety-certification", "title": "Safety Certification", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety certification is the formal attestation by an accredited body that a product, system, or process conforms to applicable safety standards and regulations. It involves assessment, testing, and audit against criteria such as functional-safety integrity levels, after which a mark or certificate authorises deployment. For robots it verifies that protective measures like collision detection meet required safety levels.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:safety-certification", "labels": [ "Safety Certification", "UL Certification" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "safety-engineering", "title": "Safety Engineering", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety engineering is the engineering discipline concerned with designing systems that avoid causing harm to people, property, and the environment under both normal and fault conditions. It applies hazard analysis, redundancy, fail-safe design, and quantified risk assessment across the system lifecycle. Functional safety and reliability engineering are specialised branches that address fault behaviour and dependable operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:safety-engineering", "labels": [ "Safety Engineering" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "safety-evaluation", "title": "Safety Evaluation", "domain": "ai", "domain_name": "Ai", "definition": "Safety evaluation is the systematic assessment of whether an AI system behaves acceptably under a defined threat and risk model, measuring propensities for harmful outputs, susceptibility to misuse, robustness under adversarial pressure, and the presence of dangerous capabilities. It combines automated benchmarks, red-teaming, and structured human review to produce evidence used in deployment decisions and governance reporting. Distinct from capability evaluation, safety evaluation asks not only what a model can do but how reliably it refrains from causing harm.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:safety-evaluation", "labels": [ "Safety Evaluation", "AI Safety Evaluation", "Model Safety Evaluation" ], "is_subclass_of": [ "Model Evaluation", "AI Safety" ], "wikilinks": [] }, { "id": "safety-filter", "title": "Safety Filter", "domain": "ai", "domain_name": "Ai", "definition": "A safety filter is a content-moderation component placed around a generative AI model that screens inputs and outputs to block disallowed, harmful, or policy-violating content. It typically combines classifiers, keyword and pattern rules, and policy thresholds to detect unsafe prompts or generations and refuse, redact, or regenerate them. It is a core safeguard in deployed image and conversational AI products.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:safety-filter", "labels": [ "Safety Filter" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "safety-fine-tuning", "title": "Safety Fine Tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Safety Fine Tuning is a specialised training stage applied after general capability training to reduce harmful outputs and align model behaviour with safety principles. It employs safety-curated datasets, RLHF, and constitutional AI objectives to harden models against adversarial misuse and emergent misalignment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:safety-fine-tuning", "labels": [ "Safety Fine Tuning" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Alignment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "safety-function", "title": "Safety Function", "domain": "robotics", "domain_name": "Robotics", "definition": "A safety function is a function implemented by a control system or safeguard whose failure results in an immediate increase of risk to people, equipment, or the environment. In robotics and machine safety, examples include emergency stop, speed and separation monitoring, and safe torque off. Safety functions are specified, designed, and validated to meet a required performance level so that residual risk remains tolerable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:safety-function", "labels": [ "Safety Function" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "safety-instrumented-system", "title": "Safety Instrumented System", "domain": "robotics", "domain_name": "Robotics", "definition": "An engineered system of sensors, logic solvers, and final elements dedicated to automatically bringing an industrial process to a safe state when hazardous conditions are detected \u2014 for example shutting an emergency valve on high pressure; each of its safety instrumented functions is assigned a safety integrity level (SIL 1-4) quantifying required risk reduction, and the system is specified, designed, validated, and maintained under the functional safety lifecycle of IEC 61508 and IEC 61511, independently of the basic process control system.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:safety-instrumented-system", "labels": [ "Safety Instrumented System" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [ "Functional Safety", "Safety Lifecycle", "Failure Mode And Effects Analysis", "IEC 62061" ] }, { "id": "safety-integrity-level", "title": "Safety Integrity Level", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety Integrity Level (SIL) is a discrete measure of the risk-reduction provided by a safety function, defined by the IEC 61508 family of standards on four levels (SIL 1 to SIL 4). Each level corresponds to a target probability of dangerous failure on demand or per hour. SIL is used to specify, design, and verify that safety-related control systems achieve a tolerable level of risk.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:safety-integrity-level", "labels": [ "Safety Integrity Level" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [] }, { "id": "safety-lifecycle", "title": "Safety Lifecycle", "domain": "standards", "domain_name": "Standards", "definition": "The safety lifecycle is a structured, end-to-end process defined by functional-safety standards for managing safety from concept through decommissioning. It organises activities such as hazard and risk analysis, safety requirement allocation, design, implementation, verification, validation, operation, and modification into defined phases with documented inputs, outputs, and review gates. Standards such as IEC 61508, IEC 62061, and ISO 26262 prescribe safety lifecycles to ensure that safety-related systems achieve and sustain their required integrity levels.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:safety-lifecycle", "labels": [ "Safety Lifecycle" ], "is_subclass_of": [ "Functional Safety" ], "wikilinks": [] }, { "id": "safety-measure", "title": "Safety Measure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Safety Measure is a technical, procedural, or organisational control applied to an AI system to prevent, detect, or mitigate harm arising from system failures, misuse, or unintended behaviour. Safety measures span the full AI lifecycle\u2014from architecture choices that constrain dangerous capabilities, to testing regimes that surface failure modes, to runtime guardrails and human oversight mechanisms that limit impact when systems operate outside intended boundaries. Effective safety measures are proportionate to the risk profile of the system and are validated against adversarial conditions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:safety-measure", "labels": [ "Safety Measure", "Protective Measure" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Governance" ], "wikilinks": [ "Ai Governance", "Artificial Intelligence" ] }, { "id": "safety-metrics", "title": "Safety Metrics", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety Metrics are quantitative measures used to assess how safely a robotic or autonomous system behaves around humans, such as minimum separation distance, collision frequency, time-to-contact, and force limits during physical contact. They provide an objective basis for comparing systems, setting regulatory thresholds, and validating that a human-robot interaction design meets acceptable risk levels. They are typically measured through instrumented testing rather than self-reported system logs alone.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:safety-metrics", "labels": [ "Safety Metrics" ], "is_subclass_of": [ "Robot Safety" ], "wikilinks": [] }, { "id": "safety-monitoring", "title": "Safety Monitoring", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety monitoring is the continuous supervision of a robotic or automated system to detect hazardous states, component degradation or human proximity and to trigger protective responses. It fuses proprioceptive and exteroceptive sensing with state estimation, anomaly detection and decision logic to enforce safety constraints independently of the nominal task controller. It underpins certified human-robot collaboration and autonomous operation in shared workspaces.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:safety-monitoring", "labels": [ "Safety Monitoring" ], "is_subclass_of": [ "Robot Safety" ], "wikilinks": [] }, { "id": "safety-plc", "title": "Safety PLC", "domain": "robotics", "domain_name": "Robotics", "definition": "A Safety Programmable Logic Controller (Safety PLC) is a specialised industrial controller certified to IEC 61511 or IEC 62061 safety integrity levels, designed to execute safety instrumented functions that bring a process to a safe state upon detecting hazardous conditions. Unlike standard PLCs, Safety PLCs implement redundant processing, self-diagnostics, and rigorous failure-mode analysis to achieve the high diagnostic coverage required for safety-critical robotics and industrial automation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:safety-plc", "labels": [ "Safety PLC" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "safety-standard", "title": "Safety Standard", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A safety standard is a normative document \u2014 published by a standards body, industry consortium, or regulatory authority \u2014 that defines requirements, processes, and verification methods for ensuring that a system, product, or environment achieves an acceptable level of risk to human life, health, and property. Safety standards range from generic functional safety frameworks such as IEC 61508 and ISO 26262 to domain-specific codes covering robotics, medical devices, aviation, and industrial machinery, and they typically specify hazard analysis methods, safety integrity levels, design constraints, and evidence of compliance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:safety-standard", "labels": [ "Safety Standard", "Machinery Safety Standards", "Product Safety Standards", "Voluntary AI Safety Standard" ], "is_subclass_of": [ "Compliance Standards" ], "wikilinks": [] }, { "id": "safety-system", "title": "Safety System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A safety system is an engineered arrangement of hardware, software, and procedural controls designed to prevent, detect, and mitigate hazards in industrial, transport, robotic, or cyber-physical environments, ensuring that a process or machine reaches a safe state when faults or dangerous conditions are detected. Safety systems are defined by their Safety Integrity Level (SIL) or Performance Level (PL) ratings under standards such as IEC 61508 and ISO 26262, which quantify the required probability of failure on demand. They encompass emergency shutdown systems, safety PLCs, watchdog timers, interlocks, redundant sensors, and AI-based anomaly detection.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:safety-system", "labels": [ "Safety System", "Safety Systems" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "safety-vulnerability", "title": "Safety Vulnerability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Safety Vulnerability is an identified weakness or flaw in an AI system's design, training data, architecture, or deployment context that could be exploited to cause harmful, unintended, or unsafe behaviours\u2014including adversarial attacks, data poisoning, prompt injection, or misalignment failures. Safety vulnerabilities are distinct from conventional software security vulnerabilities in that they may arise from statistical properties of learned models rather than explicit coding errors, making them harder to enumerate exhaustively and more sensitive to distribution shift. Systematic identification requires red-teaming, adversarial testing, and formal verification where feasible.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:safety-vulnerability", "labels": [ "Safety Vulnerability" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "safety-and-alignment", "title": "Safety and alignment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI safety and alignment is the interdisciplinary research programme and engineering practice concerned with ensuring that artificial intelligence systems \u2014 particularly large language models and future general AI \u2014 behave in accordance with human intentions, values, and oversight mechanisms acros...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:safety-and-alignment", "labels": [ "Safety and alignment", "Safety Alignment", "Safety and Alignment" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Safety", "AI Ethics", "Machine Learning Discipline", "AI Governance", "Human-Computer Interaction", "Existential Risk Research" ], "wikilinks": [ "Activation Patching", "Adversarial Red Teaming", "AI Capabilities Research", "AI Safety Institute", "AlgorithmLayer", "Bayesian Inference", "Bletchley Declaration", "Catastrophic Risk Reduction", "Causal Tracing", "Chain of Thought", "Dangerous Capability Evaluations", "Debate", "Debate Protocol", "Decision Theory", "EthicsAndGovernanceDomain", "Evaluation Benchmarks", "EvaluationLayer", "Existential Risk Research", "Governance Frameworks", "Interpretability Tools" ] }, { "id": "safety-critical-systems", "title": "Safety-Critical Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Safety-critical systems are hardware and software systems whose failure or malfunction could result in death, serious injury, significant property damage, or severe environmental harm. They are subject to rigorous engineering processes, formal standards, and independent certification to demonstrate that residual risk is reduced to an acceptable level.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:safety-critical-systems", "labels": [ "Safety-Critical Systems", "Automotive Safety-Critical Systems", "Safety Critical Computing", "Safety Critical Systems", "Safety-Critical System" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [] }, { "id": "safety", "title": "Safety", "domain": "governance", "domain_name": "Governance", "definition": "The condition whereby an AI system operates without causing unacceptable risk of physical injury, harm to human health or well-being, damage to property, or harm to the environment, achieved through hazard identification, risk assessment, and implementation of appropriate safeguards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:safety", "labels": [ "Safety", "Brand Safety", "Safety Requirement" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Hazard Analysis", "Risk Assessment (AI-0079)", "Bitcoin", "Death of the Internet", "Decentralised Web", "Infrastructure", "Large Language Models", "MetaverseDomain", "Politics, Law, Privacy", "Safety and alignment", "Social contract and jobs", "Solid", "Trust and Safety", "webid" ] }, { "id": "saga-pattern", "title": "Saga Pattern", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The saga pattern is a distributed-systems design for managing long-lived business transactions that span multiple services without a global lock or two-phase commit. A saga is a sequence of local transactions, each of which has an associated compensating transaction that semantically undoes its effect if a later step fails. Coordination is achieved either through orchestration, where a central coordinator drives the steps, or choreography, where services react to events, trading strong atomicity for eventual consistency and availability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:saga-pattern", "labels": [ "Saga Pattern" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "salt", "title": "Salt", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Salt is a randomly generated value appended or prepended to input data before it is processed by a cryptographic hash function, ensuring that two identical inputs produce distinct hash outputs. Salts defeat precomputed dictionary and rainbow-table attacks on hashed credentials and commitments by making each hash unique even when the underlying plaintext is shared. In blockchain contexts, salts appear in commitment schemes, zero-knowledge proofs, and password-based key derivation functions, where they guarantee that a hash reveals no information about the original value to an observer who does not know the salt.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:salt", "labels": [ "Salt" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "sam-hammond-ai-policy-economist", "title": "Sam Hammond AI Policy Economist", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Sam Hammond is a senior economist and policy analyst specialising in the intersection of transformative technology and governance, known for work at the Foundation for American Innovation. His analysis explores AI-driven economic disruption, techno-feudalism, the co-evolution of state institutions with autonomous systems, and the geopolitical implications of superintelligence, offering a libertarian-leaning but empirically grounded perspective on long-horizon AI futures.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:sam-hammond-ai-policy-economist", "labels": [ "Sam Hammond AI Policy Economist", "Sam Hammond" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Social contract and jobs" ] }, { "id": "saml-assertion", "title": "Saml Assertion", "domain": "security", "domain_name": "Security", "definition": "A SAML assertion is an XML-based security token, issued by an identity provider, that conveys statements about an authenticated subject to a relying service provider. Assertions carry authentication, attribute and authorisation-decision statements, are bound to a subject and validity window, and are protected by XML digital signatures to ensure integrity and origin. They are the core data structure exchanged in SAML single sign-on and federated identity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:saml-assertion", "labels": [ "Saml Assertion", "SAML Assertion" ], "is_subclass_of": [ "SAML" ], "wikilinks": [] }, { "id": "sample-efficiency", "title": "Sample Efficiency", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Sample efficiency is a measure of how much task performance a learning algorithm achieves per unit of training data or environment interaction it consumes. In reinforcement learning it is particularly critical because real-world or simulated interactions can be expensive or slow to collect, making algorithms that learn from fewer trials more practical to deploy. Techniques such as curriculum learning, model-based planning, and off-policy replay are used specifically to improve sample efficiency.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:sample-efficiency", "labels": [ "Sample Efficiency" ], "is_subclass_of": [ "Reinforcement Learning" ], "wikilinks": [] }, { "id": "sample-return", "title": "Sample Return", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sample-return", "labels": [ "Sample Return" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sampling-based-planning", "title": "Sampling Based Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Sampling-based planning is a family of motion-planning methods that find feasible paths by randomly sampling the configuration space and connecting collision-free samples into a graph or tree, rather than constructing an explicit representation of the free space. By trading completeness for probabilistic completeness, these methods scale to the high-dimensional spaces typical of articulated robots. Representative algorithms include the Probabilistic Roadmap and the Rapidly Exploring Random Tree.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sampling-based-planning", "labels": [ "Sampling Based Planning", "Sampling-Based Planning" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [] }, { "id": "sampling-bias", "title": "Sampling Bias", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sampling-bias", "labels": [ "Sampling Bias" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sampling-procedure", "title": "Sampling Procedure", "domain": "ai", "domain_name": "Ai", "definition": "A sampling procedure is the method by which a generative model draws output samples from its learned probability distribution. It governs how latent or noise variables are mapped to concrete outputs, including techniques such as ancestral sampling, temperature scaling, top-k and nucleus sampling, and iterative denoising in diffusion models. The choice of procedure controls the diversity, fidelity, and computational cost of generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sampling-procedure", "labels": [ "Sampling Procedure" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "sampling-theory", "title": "Sampling Theory", "domain": "data", "domain_name": "Data", "definition": "Sampling theory is the body of statistical and signal-processing principles governing how a subset of observations is selected from a larger population or continuous signal so that valid inferences can be drawn about the whole. In statistics it formalises how representative samples are drawn, how estimator variance and bias behave, and how confidence in conclusions scales with sample size. In signal processing it specifies the conditions under which a continuous signal can be reconstructed without loss from discrete samples. The discipline underpins survey design, experimental design, digital signal acquisition, and Monte Carlo estimation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:sampling-theory", "labels": [ "Sampling Theory" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "sampling", "title": "Sampling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Sampling is the family of computational and statistical procedures that draw representative or informative values from probability distributions \u2014 ranging from classical Markov chain Monte Carlo mods (Metropolis-Hastings, Gibbs, Hamiltonian Monte Carlo, NUTS) and sequential mods (Sequential Monte...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:sampling", "labels": [ "Sampling", "Data Sampling", "Hemisphere Sampling", "KLD-Sampling", "Sampling Algorithm", "Sampling Strategies", "Statistical Sampling", "Stratified Sampling", "Uncertainty Sampling" ], "is_subclass_of": [ "AI Technique", "Numerical Methods", "Stochastic Processes", "Probabilistic Model", "Bayesian Inference", "Statistical Computing" ], "wikilinks": [ "AlgorithmLayer", "Autoregressive Sampling", "Bayesian Inference", "Bayesian Neural Networks", "Convergence Criterion", "DDIM", "Deterministic Integration", "DPM-Solver", "Entropy", "Ergodic Theory", "GenerationLayer", "Gibbs Sampling", "Gradient Information", "Grid Approximation", "Hamiltonian Monte Carlo", "Hugging Face Diffusers", "Importance Sampling", "InferenceLayer", "Information Theory", "Kullback-Leibler Divergence" ] }, { "id": "sanctions-compliance", "title": "Sanctions Compliance", "domain": "governance", "domain_name": "Governance", "definition": "Sanctions compliance is the set of controls organisations implement to avoid transacting with individuals, entities, jurisdictions, or assets prohibited by sanctions regimes such as those administered by the US OFAC, the EU, the UK OFSI, and the UN. It involves screening counterparties against sanctions lists, monitoring transactions for prohibited patterns, blocking or freezing prohibited dealings, and maintaining auditable records. In digital-asset markets it extends to wallet-address screening and analysis of fund provenance, posing distinctive challenges for permissionless and privacy-preserving systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sanctions-compliance", "labels": [ "Sanctions Compliance", "Financial Sanctions Compliance" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "sanctions-enforcement", "title": "Sanctions Enforcement", "domain": "governance", "domain_name": "Governance", "definition": "Sanctions enforcement is the regulatory and operational process by which financial institutions, governments, and technology platforms identify, screen, and block transactions, assets, and relationships involving sanctioned individuals, entities, or jurisdictions as designated by authorities such as OFAC, the UN Security Council, the European Union, or HM Treasury. It encompasses real-time screening of payment flows against consolidated sanctions lists, asset-freezing procedures, reporting obligations, and the maintenance of audit trails demonstrating compliance. In digital asset and blockchain contexts, sanctions enforcement extends to on-chain transaction monitoring and the blocking of wallet addresses associated with designated parties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sanctions-enforcement", "labels": [ "Sanctions Enforcement" ], "is_subclass_of": [ "Compliance Framework" ], "wikilinks": [] }, { "id": "sanctions-screening", "title": "Sanctions Screening", "domain": "finance", "domain_name": "Finance", "definition": "Sanctions Screening is the regulatory compliance process by which financial institutions, fintech operators, and other regulated entities check customers, counterparties, and transactions against government-issued sanctions lists\u2014such as those maintained by OFAC, the UN Security Council, HM Treasury, and the EU\u2014to prevent prohibited dealings with designated individuals, entities, and jurisdictions. The process involves name-matching algorithms, fuzzy logic to handle transliterations and aliases, and risk-based escalation procedures for potential matches. In blockchain and digital asset contexts, screening extends to wallet addresses and on-chain transaction flows. Failure to screen adequately exposes institutions to severe civil and criminal penalties.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sanctions-screening", "labels": [ "Sanctions Screening", "PEP Screening", "Politically Exposed Person Screening", "Portfolio Screening", "Sanctions Screening Engine", "Screening" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "sandbox-environment", "title": "Sandbox Environment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A sandbox environment is an isolated execution context that constrains a program's access to the host system, filesystem, network, and resources. It allows untrusted or autonomous code, such as AI-generated programs or agent actions, to run with controlled side effects and contained failure. Sandboxes are implemented through containers, virtual machines, OS-level namespaces, or language-level interpreters.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sandbox-environment", "labels": [ "Sandbox Environment" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "sandboxed-code-execution", "title": "Sandboxed Code Execution", "domain": "ai", "domain_name": "Ai", "definition": "Sandboxed code execution is the practice of running AI-agent-generated code inside an isolated runtime \u2014 a container, microVM, or restricted interpreter \u2014 that constrains filesystem, network, and system-call access. It lets an agent execute arbitrary code to compute results or verify hypotheses without risking the host environment or leaking credentials. It matters for autonomous coding agents and terminal-based assistants, where generated code is untrusted by default and must be contained until its effects are reviewed or accepted.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:sandboxed-code-execution", "labels": [ "Sandboxed Code Execution" ], "is_subclass_of": [ "Code Execution" ], "wikilinks": [] }, { "id": "sandboxed-execution", "title": "Sandboxed Execution", "domain": "security", "domain_name": "Security", "definition": "Sandboxed execution is the practice of running untrusted or partially trusted code inside an isolated environment with restricted access to the host filesystem, network and system resources. It contains the effects of faulty or malicious code by enforcing resource and permission boundaries at the process, container or virtual-machine level. Sandboxed execution underpins both security-sensitive contexts, such as running third-party plugins, and autonomous AI agents that must be prevented from taking unintended actions on a host system.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:sandboxed-execution", "labels": [ "Sandboxed Execution" ], "is_subclass_of": [ "Sandbox Environment" ], "wikilinks": [] }, { "id": "sat-solving", "title": "Sat Solving", "domain": "ai", "domain_name": "Ai", "definition": "SAT solving is the algorithmic process of determining whether a propositional logic formula, typically in conjunctive normal form, has a satisfying truth assignment \u2014 the Boolean satisfiability problem. Although SAT is the canonical NP-complete problem, modern conflict-driven clause-learning solvers routinely decide instances with millions of variables and clauses, making SAT a practical engine for many computational tasks. SAT solving uses systematic search with unit propagation, clause learning, and intelligent backtracking. It underpins formal verification, automated planning, and constraint solving, and is closely related to satisfiability-modulo-theories reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:sat-solving", "labels": [ "Sat Solving", "SAT Solving" ], "is_subclass_of": [ "Automated Reasoning" ], "wikilinks": [] }, { "id": "satellite-antenna", "title": "Satellite Antenna", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-antenna", "labels": [ "Satellite Antenna" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-beam", "title": "Satellite Beam", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-beam", "labels": [ "Satellite Beam" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-channel-coding", "title": "Satellite Channel Coding", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-channel-coding", "labels": [ "Satellite Channel Coding" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-clock-error", "title": "Satellite Clock Error", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-clock-error", "labels": [ "Satellite Clock Error" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-communication", "title": "Satellite Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Satellite communication is the use of artificial Earth-orbiting satellites as relay stations to transmit signals \u2014 including voice, data, video, and telemetry \u2014 between ground terminals separated by large distances or challenging terrain. Signals are uplinked from a ground station to a satellite transponder, which amplifies and retransmits them on a different frequency to one or more receiving terminals. Different orbital regimes (GEO, MEO, LEO) offer distinct trade-offs between coverage footprint, latency, and capacity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:satellite-communication", "labels": [ "Satellite Communication", "Satellite Communications" ], "is_subclass_of": [ "Infrastructure", "Telecommunications" ], "wikilinks": [] }, { "id": "satellite-constellation", "title": "Satellite Constellation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A satellite constellation is a fleet of spacecraft designed and operated as one service or observing system. The satellites occupy coordinated orbits so their coverage, revisit, geometry or capacity is more useful than that of any one member. A constellation also includes operational infrastructure: ground stations, control centres, user terminals, network routing and fleet-management software.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-constellation", "labels": [ "Satellite Constellation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-doppler-shift", "title": "Satellite Doppler Shift", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-doppler-shift", "labels": [ "Satellite Doppler Shift" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-downlink", "title": "Satellite Downlink", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-downlink", "labels": [ "Satellite Downlink" ], "is_subclass_of": [ "Space-to-ground Link" ], "wikilinks": [] }, { "id": "satellite-gateway", "title": "Satellite Gateway", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-gateway", "labels": [ "Satellite Gateway" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-handover", "title": "Satellite Handover", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-handover", "labels": [ "Satellite Handover" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-imagery", "title": "Satellite Imagery", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Imagery of the Earth's surface captured by sensors aboard orbiting satellites, spanning optical, multispectral, hyperspectral, thermal, and synthetic-aperture radar modalities at spatial resolutions from kilometres to tens of centimetres. Georeferenced and revisited on regular orbits, satellite imagery is the primary raster substrate of geospatial analysis, feeding mapping, agriculture, climate monitoring, disaster response, and defence, and \u2014 via open programmes such as Landsat and Sentinel \u2014 machine learning at planetary scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:satellite-imagery", "labels": [ "Satellite Imagery" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [ "Remote Sensing", "Geospatial Data", "Image Processing", "Computer Vision" ] }, { "id": "satellite-link-budget", "title": "Satellite Link Budget", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-link-budget", "labels": [ "Satellite Link Budget" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-modulation", "title": "Satellite Modulation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-modulation", "labels": [ "Satellite Modulation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-monitoring", "title": "Satellite Monitoring", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Satellite monitoring is the use of orbital remote-sensing platforms to observe and measure conditions on the Earth's surface over time. In sustainability and supply-chain contexts it supports verification of deforestation, land use, emissions plumes, and facility activity using multispectral, radar, and thermal imagery. The resulting geospatial data feeds carbon accounting, compliance, and ethical-sourcing assessments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:satellite-monitoring", "labels": [ "Satellite Monitoring", "Satellite Earth Observation" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "satellite-multiple-access", "title": "Satellite Multiple Access", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-multiple-access", "labels": [ "Satellite Multiple Access" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-navigation-solution", "title": "Satellite Navigation Solution", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-navigation-solution", "labels": [ "Satellite Navigation Solution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-remote-sensing", "title": "Satellite Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-remote-sensing", "labels": [ "Satellite Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "satellite-service-coverage", "title": "Satellite Service Coverage", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-service-coverage", "labels": [ "Satellite Service Coverage" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-spectrum-allocation", "title": "Satellite Spectrum Allocation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-spectrum-allocation", "labels": [ "Satellite Spectrum Allocation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-terminal", "title": "Satellite Terminal", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-terminal", "labels": [ "Satellite Terminal" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-transceiver", "title": "Satellite Transceiver", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-transceiver", "labels": [ "Satellite Transceiver" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satellite-uplink", "title": "Satellite Uplink", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-uplink", "labels": [ "Satellite Uplink" ], "is_subclass_of": [ "Space-to-ground Link" ], "wikilinks": [] }, { "id": "satellite-based-augmentation-system", "title": "Satellite-based Augmentation System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:satellite-based-augmentation-system", "labels": [ "Satellite-based Augmentation System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "satisfiability", "title": "Satisfiability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Satisfiability is the problem of determining whether there exists an assignment of values to variables that makes a logical formula true, most prominently the Boolean satisfiability problem (SAT). SAT is the canonical NP-complete problem, and modern SAT solvers can decide formulas with millions of variables despite this worst-case hardness. Satisfiability provides a unifying computational substrate for verification, planning, scheduling and many forms of automated reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:satisfiability", "labels": [ "Satisfiability" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "satoshi-nakamoto", "title": "Satoshi Nakamoto", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Satoshi Nakamoto is the pseudonym used by the unidentified person or group who designed Bitcoin, authored its 2008 white paper and released the first reference implementation in 2009. Nakamoto introduced a practical solution to double-spending in a peer-to-peer electronic cash system through proof-of-work and a public, append-only ledger. The true identity behind the name remains unknown, and Nakamoto withdrew from public involvement in the project around 2010 to 2011.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:satoshi-nakamoto", "labels": [ "Satoshi Nakamoto" ], "is_subclass_of": [ "Protocol and Consensus", "owl:Thing" ], "wikilinks": [ "Bitcoin", "Proof of Work", "Blockchain", "Cryptographic Domain", "Distributed Systems Domain", "owl:Thing", "Bitcoin white paper (2008)" ] }, { "id": "satoshi", "title": "Satoshi", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A satoshi is the smallest indivisible unit of bitcoin, equal to one hundred-millionth of a single bitcoin, named after Bitcoin's pseudonymous creator Satoshi Nakamoto. It is the base unit against which on-chain protocols such as Ordinals and Inscriptions operate, since each satoshi can be individually tracked and inscribed with arbitrary data. Fee markets, token standards such as BRC-20, and wallet accounting are all ultimately denominated in satoshis.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:satoshi", "labels": [ "Satoshi" ], "is_subclass_of": [ "Bitcoin" ], "wikilinks": [] }, { "id": "saturated-zone", "title": "Saturated Zone", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:saturated-zone", "labels": [ "Saturated Zone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "scalability-pattern", "title": "Scalability Pattern", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Scalability Pattern is a reusable architectural strategy that enables a system to handle increasing load by adding resources or restructuring components without degrading performance or reliability. Patterns include horizontal scaling (adding parallel instances), vertical scaling (increasing instance capacity), sharding, caching, and event-driven decomposition, each suited to different bottleneck profiles.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scalability-pattern", "labels": [ "Scalability Pattern" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "scalability-solutions", "title": "Scalability Solutions", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Scalability solutions are a family of techniques, architectures, and protocols designed to increase the throughput, reduce latency, and lower transaction costs of distributed systems\u2014especially blockchain networks\u2014without sacrificing security or decentralisation. They span both on-chain approaches (sharding, improved consensus algorithms, data availability sampling) and off-chain or layer-2 approaches (payment channels, rollups, sidechains) that defer computation or data storage away from the base layer. The discipline also extends beyond blockchains into general distributed-systems engineering, encompassing horizontal scaling, caching strategies, and peer-to-peer load distribution. Together these mechanisms address the fundamental tension between throughput, security, and decentralisation commonly formalised as the Blockchain Trilemma.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scalability-solutions", "labels": [ "Scalability Solutions", "Scaling Solutions" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Blockchain", "Layer 2 Scaling", "Sharding" ] }, { "id": "scalability", "title": "Scalability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scalability is the fundamental system property describing the capacity to handle increasing workloads\u2014in throughput, data volume, concurrency, or geographic reach\u2014by adding resources without proportionate degradation in performance, cost efficiency, or reliability. It encompasses both vertical scaling (augmenting the capacity of existing nodes) and horizontal scaling (adding more nodes to a distributed cluster), each with distinct architectural implications governed by coordination overhead, data partitioning strategies, and consistency trade-offs. In distributed systems, achieving linear or near-linear scalability requires careful attention to the CAP theorem and its practical refinement, PACELC, which explicitly models the latency\u2013consistency tension even in partition-free conditions. Scalability is a first-order design concern in cloud-native architectures, blockchain networks, edge computing platforms, and AI inference pipelines serving large or rapidly growing user populations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:scalability", "labels": [ "Scalability", "Read Scalability", "System Scalability", "Throughput Scalability" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "scalable-architecture", "title": "Scalable Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "System designs that maintain performance and availability as user demand grows, employing horizontal scaling, load balancing, distributed computing, and cloud-native patterns to support expanding metaverse populations and concurrent interactions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:scalable-architecture", "labels": [ "Scalable Architecture" ], "is_subclass_of": [ "Platform and Environment", "System Architecture" ], "wikilinks": [ "High Availability", "metaverse", "System Architecture" ] }, { "id": "scalable-deployment", "title": "Scalable Deployment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scalable deployment is the practice of releasing and operating software so that capacity can grow or shrink with demand without redesign. It relies on stateless services, horizontal scaling, load balancing, container orchestration, and infrastructure-as-code to add resources elastically. It is essential for systems such as digital twins that must serve fluctuating workloads reliably.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scalable-deployment", "labels": [ "Scalable Deployment" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "scalable-oversight", "title": "Scalable Oversight", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Scalable oversight is the AI-safety research problem of reliably supervising, evaluating, and steering AI systems whose capabilities approach or exceed human ability on the tasks being judged. It seeks mechanisms that let limited human supervisors provide accurate training signal and verification even when they cannot directly check a model's outputs. Approaches decompose hard judgements, amplify human judgement with AI assistance, and use adversarial or recursive structures to surface errors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:scalable-oversight", "labels": [ "Scalable Oversight" ], "is_subclass_of": [ "AI Alignment" ], "wikilinks": [] }, { "id": "scaled-data-state", "title": "Scaled Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:scaled-data-state", "labels": [ "Scaled Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "scaled-dot-product-attention", "title": "Scaled Dot Product Attention", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An attention mechanism that computes attention weights using the dot product of queries and keys, scaled by the square root of the key dimension, followed by a softmax normalisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scaled-dot-product-attention", "labels": [ "Scaled Dot Product Attention", "Scaled Dot Product", "Scaled Dot-Product Attention" ], "is_subclass_of": [ "AI Model Architecture", "Attention Mechanism" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "scaling-laws", "title": "scaling laws", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Scaling Laws are empirical power-law relationships describing how the performance of neural networks \u2014 typically measured as held-out cross-entropy loss \u2014 varies predictably as a function of model parameters (N), training data volume (D), and total compute budget (C). Foundational Kaplan et al. (2020) work established smooth, predictable loss curves across many orders of magnitude, while Hoffmann et al. (2022) Chinchilla analyses refined optimal compute allocation to roughly equal scaling of model size and training tokens. These relationships guide architectural decisions, compute budgeting, and capability forecasting for large foundation models. Scaling laws have since been extended beyond language modelling to vision transformers, multimodal architectures, and reinforcement learning from human feedback.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:scaling-laws", "labels": [ "Scaling Laws", "Chinchilla Scaling Laws", "Neural Scaling Laws", "Scaling Law" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "scarcity", "title": "Scarcity", "domain": "finance", "domain_name": "Finance", "definition": "Scarcity is the economic condition in which a resource is limited relative to demand for it, giving rise to value, price and the need for allocation choices. In monetary and token economics scarcity is often engineered through fixed or capped supply, issuance schedules or burning mechanisms to support a store-of-value proposition. It is the foundational premise from which supply-and-demand pricing dynamics emerge.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scarcity", "labels": [ "Scarcity" ], "is_subclass_of": [ "Monetary Policy" ], "wikilinks": [] }, { "id": "scattergram", "title": "Scattergram", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:scattergram", "labels": [ "Scattergram" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "scatterometer", "title": "Scatterometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:scatterometer", "labels": [ "Scatterometer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "scenario-analysis", "title": "Scenario Analysis", "domain": "data", "domain_name": "Data", "definition": "Scenario analysis is a structured method for exploring how a system behaves under a set of plausible alternative futures rather than a single forecast. It defines coherent scenarios over key drivers and evaluates outcomes, risks, and decisions across each. It is widely used in financial planning, climate disclosure, and operational risk to stress-test strategies against uncertainty.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scenario-analysis", "labels": [ "Scenario Analysis" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "scene-capture-and-reconstruction", "title": "Scene Capture and Reconstruction", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Scene Capture and Reconstruction is a fundamental computer-vision and spatial-computing discipline concerned with recovering complete geometric, radiometric, and semantic representations of physical environments from sensor observations \u2014 primarily calibrated multi-view photographs, RGB-D frames,...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:scene-capture-and-reconstruction", "labels": [ "Scene Capture and Reconstruction" ], "is_subclass_of": [ "AI Technique", "Computer Vision", "3D Reconstruction", "Neural Rendering", "Spatial Computing Paradigm", "Photogrammetry" ], "wikilinks": [ "Agisoft Metashape", "AlgorithmLayer", "Barron et al. 2022 Mip-NeRF-360 CVPR", "Barron et al. 2023 Zip-NeRF ICCV", "Bundle Adjustment", "Camera Calibration", "Chen et al. 2022 TensoRF ECCV", "COLMAP", "Cultural Heritage Preservation", "CVPR", "Differentiable Renderer", "Differentiable Rendering", "ECCV", "Feature Matching", "Fei et al. 2024 3DGS Survey IEEE TVCG", "Gao et al. 2024 NeRF Survey IEEE TPAMI", "Gaussian Splatting Rasterisation", "Generative AI 3D", "GPU Compute", "ICCV" ] }, { "id": "scene-design", "title": "Scene Design", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Scene design is the process of composing and arranging the spatial, visual, and behavioural elements of a virtual or augmented environment. It encompasses layout of geometry, lighting, materials, cameras, and interactive objects within a scene graph. It is a foundational creative activity in metaverse, game, and immersive-media authoring.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scene-design", "labels": [ "Scene Design" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "scene-geometry", "title": "Scene Geometry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scene geometry is the structured description of the spatial shape, surfaces, and arrangement of objects within a three-dimensional environment. It captures positions, depths, surface orientations, and connectivity needed to render, simulate, or reason about a scene. Accurate scene geometry underpins realistic lighting, occlusion, collision, and interaction in spatial computing and computer graphics applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:scene-geometry", "labels": [ "Scene Geometry" ], "is_subclass_of": [ "Scene Representation" ], "wikilinks": [] }, { "id": "scene-graph-format", "title": "Scene Graph Format", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Scene Graph Format is a standardised serialisation schema for directed acyclic graphs that represent 3D scene hierarchies, encoding spatial transforms, geometry, materials, lights, cameras, and behaviours. Major formats include glTF (optimised for runtime delivery), USD (layered composition for production pipelines), X3D (ISO web standard), and Collada (interchange for DCC tools).", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scene-graph-format", "labels": [ "Scene Graph Format" ], "is_subclass_of": [ "Standards and Interoperability", "Scene Graph" ], "wikilinks": [] }, { "id": "scene-graph", "title": "Scene Graph", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A hierarchical tree-based data structure organizing and describing the spatial, logical, and rendering relationships among objects in a 3D scene, enabling efficient traversal, culling, and rendering operations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:scene-graph", "labels": [ "Scene Graph", "3D Scene Graph", "Game Engine Scene Graph" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "3D Engine", "Bounding Volume", "Camera Node", "Frustum Culling", "Geometry Node", "Group Node", "ISO/IEC 19775-2", "Scene Node", "Scene Rendering", "Transform Matrix", "Transform Node", "Web3D", "Collision Detection", "Coordinate System", "CreativeMediaDomain", "DataLayer", "Game Engine", "glTF (3D File Format)", "Graphics API", "InteractionDomain" ] }, { "id": "scene-management", "title": "Scene Management", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scene Management encompasses the runtime systems and data structures responsible for organising, loading, and unloading 3D scene content in real-time environments. It coordinates scene graph traversal, hierarchical object relationships, asset streaming, and spatial partitioning to ensure that only geometrically and logically relevant content is active at any moment, enabling scalable and performant virtual worlds.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:scene-management", "labels": [ "Scene Management" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "scene-optimization", "title": "Scene Optimization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scene Optimization is the set of techniques applied to 3D environments to reduce computational and bandwidth overhead while preserving perceptual fidelity, encompassing polygon reduction, texture compression and atlasing, draw call batching, occlusion culling, and level-of-detail management. These techniques are essential for achieving real-time frame rates on resource-constrained XR hardware and for supporting large concurrent user counts in metaverse platforms.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:scene-optimization", "labels": [ "Scene Optimization" ], "is_subclass_of": [ "Display and Rendering", "Render Pipeline" ], "wikilinks": [] }, { "id": "scene-representation", "title": "Scene Representation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scene Representation is the set of data structures and mathematical models used to encode the geometry, appearance, and spatial relationships of objects within a 3D environment. Representations include polygonal meshes, voxel grids, point clouds, implicit surfaces, and neural radiance fields, each with different trade-offs in accuracy, compactness, and rendering speed. The choice of scene representation directly influences the quality and performance of spatial computing and computer vision systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:scene-representation", "labels": [ "Scene Representation" ], "is_subclass_of": [ "Spatial Computing Content and Assets" ], "wikilinks": [] }, { "id": "scene-understanding", "title": "Scene Understanding", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Scene Understanding is the high-level semantic interpretation of visual and sensor data to comprehend the structure, context, objects, relationships, and dynamics of an environment. It encompasses object detection and recognition, spatial layout inference, activity recognition, contextual reasoning, and semantic scene categorisation, enabling autonomous and interactive systems to make contextually appropriate decisions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:scene-understanding", "labels": [ "Scene Understanding", "3D Scene Understanding" ], "is_subclass_of": [ "Perception System" ], "wikilinks": [ "Computer Vision", "Knowhere", "MetaverseDomain", "Perception System", "Semantic Segmentation", "software engineering" ] }, { "id": "scheduled-ai-tasks", "title": "Scheduled AI Tasks", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A feature in AI assistants that allows users to define recurring or one-time prompts to be executed automatically at specified times, extending the model's utility from reactive chat to proactive automation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:scheduled-ai-tasks", "labels": [ "Scheduled AI Tasks" ], "is_subclass_of": [ "Agentic Workflow" ], "wikilinks": [] }, { "id": "scheduler", "title": "Scheduler", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A scheduler is a system component that decides which units of work run, where they run and in what order, allocating finite computing resources among competing tasks over time. Schedulers exist at many layers, from operating-system process and thread scheduling to cluster and container orchestration that places workloads across machines. Their policies trade off throughput, latency, fairness and resource utilisation according to the goals of the platform.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scheduler", "labels": [ "Scheduler" ], "is_subclass_of": [ "Resource Management" ], "wikilinks": [] }, { "id": "schema-definition", "title": "Schema Definition", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Schema Definition is a formal, machine-readable specification of the structure, data types, constraints, and relationships that govern a dataset, message format, document, or knowledge representation, serving as a contract between data producers and consumers. Schema languages include W3C XML Schema Definition (XSD), JSON Schema (for JSON documents), OWL/RDFS (for ontologies over RDF graphs), SHACL and ShEx (for RDF graph shape constraints), Protocol Buffers and Apache Avro (for binary-serialised messages), OpenAPI (for REST API request/response bodies), and GraphQL SDL (for graph API types). A schema definition enables automated validation, code generation, documentation, and inter-system interoperability; in knowledge-graph contexts, the schema defines classes, properties, cardinality constraints, and axioms that allow OWL-based reasoning over the graph.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:schema-definition", "labels": [ "Schema Definition" ], "is_subclass_of": [ "Data Format Standard" ], "wikilinks": [] }, { "id": "schema-evolution", "title": "Schema Evolution", "domain": "data", "domain_name": "Data", "definition": "Schema evolution is the practice of modifying a data schema over time, such as adding, removing, or renaming fields, while preserving the ability to read data written under earlier versions. It is a core concern for serialisation formats such as Avro, Protobuf, and Parquet, which define explicit compatibility rules governing which changes are backward, forward, or fully compatible. Systems that manage schema evolution typically rely on a schema registry to version schemas centrally and validate that producers and consumers remain compatible before a change is deployed. Poorly managed schema evolution is a common source of pipeline breakage in streaming and data lake architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:schema-evolution", "labels": [ "Schema Evolution" ], "is_subclass_of": [ "Data Schema" ], "wikilinks": [] }, { "id": "schema-management", "title": "Schema Management", "domain": "data", "domain_name": "Data", "definition": "Schema management is the practice of defining, versioning, evolving, and enforcing the structure of data across systems and over time. It governs how data schemas change without breaking producers and consumers, using compatibility rules, schema registries, and validation to keep pipelines reliable. As a discipline within data governance, it ensures that interfaces between services and analytical stores remain consistent and trustworthy as requirements evolve.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:schema-management", "labels": [ "Schema Management" ], "is_subclass_of": [ "Data Governance" ], "wikilinks": [] }, { "id": "schema-mapping", "title": "Schema Mapping", "domain": "data", "domain_name": "Data", "definition": "Schema mapping is the specification of correspondences between the elements of two or more data schemas so that data structured under one can be translated to another. It captures how fields, types and relationships relate, including transformations, defaults and conflict resolution. Schema mapping is a core enabler of data integration, migration and semantic interoperability across heterogeneous systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:schema-mapping", "labels": [ "Schema Mapping" ], "is_subclass_of": [ "Data Integration" ], "wikilinks": [] }, { "id": "schema-registry", "title": "Schema Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A centralized service that stores, validates, and manages data schemas for event streaming and API contracts, ensuring compatibility and controlled evolution of data formats across distributed microservices and event-driven architectures.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:schema-registry", "labels": [ "Schema Registry" ], "is_subclass_of": [ "Standards and Interoperability", "Data Management" ], "wikilinks": [ "Schema Evolution", "Data Management", "metaverse" ] }, { "id": "schema-validation", "title": "Schema Validation", "domain": "standards", "domain_name": "Standards", "definition": "Schema validation is the process of checking that a data instance conforms to a formally declared structure, asserting required fields, types, value constraints and relationships before the data is accepted or processed. It uses a schema language such as JSON Schema or XML Schema to define the contract and a validator to report conformance and errors. Schema validation enforces data integrity at system boundaries and underpins reliable interoperability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:schema-validation", "labels": [ "Schema Validation" ], "is_subclass_of": [ "Data Validation", "JSON Schema" ], "wikilinks": [] }, { "id": "schema-versioning", "title": "Schema Versioning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Schema versioning is the discipline of managing changes to a data schema over time so that producers and consumers can evolve independently without breaking interoperability. It defines policies and mechanisms \u2014 version identifiers, compatibility rules, deprecation windows and migration paths \u2014 that govern how additions, removals and modifications to fields and types are introduced. Robust schema versioning underpins reliable data interchange in distributed systems, event streams and long-lived APIs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:schema-versioning", "labels": [ "Schema Versioning" ], "is_subclass_of": [ "Data Versioning" ], "wikilinks": [] }, { "id": "schema", "title": "Schema", "domain": "data", "domain_name": "Data", "definition": "A schema is a formal specification of the structure, types and constraints that valid data must satisfy within a given system. It defines entities, their attributes, relationships and permissible values, acting as a contract between data producers and consumers. Schemas underpin validation, interoperability and the reliable exchange of structured data across applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:schema", "labels": [ "Schema", "Database Schema" ], "is_subclass_of": [ "Data Model" ], "wikilinks": [] }, { "id": "schema-org", "title": "Schema.org", "domain": "standards", "domain_name": "Standards", "definition": "Schema.org is a collaborative, community-maintained vocabulary project founded in 2011 by Google, Microsoft, Yahoo, and Yandex to define a shared set of structured data markup schemas for web pages, enabling search engines and other consumers to understand the semantic content of web resources. The vocabulary defines types and properties for entities such as persons, organisations, events, products, reviews, and creative works, expressed using JSON-LD, Microdata, or RDFa. Schema.org markup embedded in web pages allows search engines to generate rich snippets, knowledge panels, and structured results. The vocabulary is extensible and hosted at schema.org, with governance managed by a W3C community group.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:schema-org", "labels": [ "Schema.org", "schema.org Vocabulary" ], "is_subclass_of": [ "Linked Data" ], "wikilinks": [] }, { "id": "schnorr-signature", "title": "Schnorr Signature", "domain": "security", "domain_name": "Security", "definition": "A provably secure digital signature scheme based on the discrete logarithm problem over elliptic curves, offering linearity properties that enable key and signature aggregation. Within blockchain systems Schnorr signatures underpin features such as Taproot, MuSig multi-signature protocols, and threshold signing, providing smaller signature sizes and stronger privacy compared to ECDSA.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:schnorr-signature", "labels": [ "Schnorr Signature", "Schnorr Protocol" ], "is_subclass_of": [ "Digital Signature", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "schnorr-signatures", "title": "Schnorr Signatures", "domain": "security", "domain_name": "Security", "definition": "A digital signature scheme based on the discrete logarithm problem over elliptic curve groups, producing compact fixed-size signatures with linear aggregation properties that enable efficient multi-signature and threshold constructions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:schnorr-signatures", "labels": [ "Schnorr Signatures" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [ "Elliptic Curve Cryptography", "MuSig2", "Taproot", "Schnorr Signature", "Digital Signature" ] }, { "id": "scholarly-manuscript-composition-process", "title": "Scholarly Manuscript Composition Process", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The structured process of composing and submitting scholarly manuscripts, including literature review, argument construction, citation management, and formatting to venue-specific standards. In AI-adjacent contexts, paper writing increasingly involves AI-assisted drafting, ontology-grounded claims, and tool-augmented workflows (LaTeX, reference managers, large language models) that must be balanced against academic integrity requirements.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scholarly-manuscript-composition-process", "labels": [ "Scholarly Manuscript Composition Process", "Paper Writing" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "science-based-targets", "title": "Science Based Targets", "domain": "governance", "domain_name": "Governance", "definition": "Science Based Targets (SBTs) are greenhouse gas emission reduction targets set by companies in alignment with the level of decarbonisation required by climate science to limit global warming to 1.5\u00b0C above pre-industrial levels, as defined by the Science Based Targets initiative (SBTi). Targets are validated by the SBTi against approved methods including absolute contraction, sectoral decarbonisation approach, and economy-wide linear regression, and must cover Scope 1 and Scope 2 emissions with a growing requirement to include Scope 3.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:science-based-targets", "labels": [ "Science Based Targets", "Science Based Targets Initiative", "Science-Based Targets" ], "is_subclass_of": [ "Sustainability Framework" ], "wikilinks": [] }, { "id": "science-operations-phase", "title": "Science Operations Phase", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:science-operations-phase", "labels": [ "Science Operations Phase" ], "is_subclass_of": [ "Mission Phase" ], "wikilinks": [] }, { "id": "scientific-computing", "title": "Scientific Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scientific computing is the discipline concerned with the development and application of computational methods and software to solve mathematical and scientific problems that are analytically intractable or whose scale demands automation. It encompasses numerical analysis, algorithm design, software engineering for high-performance systems, and the management of large-scale simulation workflows. Key application domains include climate modelling, computational fluid dynamics, molecular dynamics, finite element analysis, and astronomical simulation. Scientific computing increasingly overlaps with machine learning as data-driven models augment or replace first-principles simulations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:scientific-computing", "labels": [ "Scientific Computing" ], "is_subclass_of": [ "High-Performance Computing" ], "wikilinks": [] }, { "id": "scientific-discovery-acceleration", "title": "Scientific Discovery Acceleration", "domain": "ai", "domain_name": "Ai", "definition": "Scientific discovery acceleration is the application of artificial intelligence to compress the cycle of hypothesis generation, experimentation, and analysis across scientific disciplines. It includes AI-driven literature synthesis, simulation surrogates, automated experiment design, and prediction of structures or materials. It is often cited as a transformative potential outcome of advanced and general-purpose AI systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:scientific-discovery-acceleration", "labels": [ "Scientific Discovery Acceleration" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "scientific-discovery", "title": "scientific discovery", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Scientific discovery in AI refers to the application of machine learning, knowledge-graph reasoning, and autonomous experimentation systems to accelerate the identification of novel scientific findings, generate and test hypotheses, and interpret complex experimental data at scales beyond human cognitive capacity. It encompasses AI-driven approaches across domains including drug discovery, materials science, genomics, and climate modelling, where the goal is to augment or automate stages of the scientific method from hypothesis generation through experimental design to result interpretation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scientific-discovery", "labels": [ "Scientific Discovery", "AI-Enabled Scientific Discovery", "Scientific Discovery AI", "Scientific Discovery Agents", "Scientific Exploration" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "scientific-machine-learning", "title": "Scientific Machine Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Scientific machine learning is the discipline that fuses data-driven learning with the governing equations, conservation laws and mechanistic structure of the physical sciences. It embeds domain knowledge such as differential equations and symmetries directly into model architectures and loss functions so that learned models remain physically consistent, data-efficient and extrapolative. The field spans surrogate modelling of expensive simulators, discovery of governing equations from data, and hybrid models that blend numerical solvers with neural components.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:scientific-machine-learning", "labels": [ "Scientific Machine Learning" ], "is_subclass_of": [ "Machine Learning", "Scientific Discovery" ], "wikilinks": [] }, { "id": "scientific-method", "title": "Scientific Method", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The scientific method is a systematic process for acquiring knowledge through observation, formulation of testable hypotheses, controlled experimentation, and analysis of results, with conclusions subject to revision in light of new evidence. It emphasises empirical testing, reproducibility, and peer scrutiny to distinguish well-supported claims from conjecture. In artificial intelligence it underpins rigorous experimentation, benchmarking, and the validation of models and theories.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scientific-method", "labels": [ "Scientific Method" ], "is_subclass_of": [ "Artificial Intelligence", "Scientific Discovery" ], "wikilinks": [] }, { "id": "scientific-research", "title": "Scientific Research", "domain": "ai", "domain_name": "Ai", "definition": "Scientific research is the systematic investigation of phenomena through observation, hypothesis, experiment, and analysis to produce reproducible knowledge. As an application domain for AI, it spans literature review, data analysis, modelling, and experiment automation. Modern AI systems and autonomous agents increasingly support researchers across the discovery, verification, and dissemination stages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scientific-research", "labels": [ "Scientific Research" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "scientific-simulation", "title": "Scientific Simulation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scientific simulation is the numerical modelling of physical, chemical, or biological systems to predict behaviour that is impractical to observe directly. It solves governing equations across discretised domains, as in fluid dynamics, molecular dynamics, and climate modelling, typically on high-performance computing infrastructure. Machine learning increasingly augments simulation through learned surrogates and data-driven generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scientific-simulation", "labels": [ "Scientific Simulation", "Molecular Simulation" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "scientific-visualisation", "title": "Scientific Visualisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scientific visualisation is the rendering of multidimensional scientific data into visual form to support analysis and communication. It includes volume rendering, isosurface extraction, flow and vector-field visualisation, and time-varying simulation playback. It depends on graphics rendering pipelines and techniques such as ray tracing to convey spatial and quantitative structure accurately.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scientific-visualisation", "labels": [ "Scientific Visualisation", "Scientific Visualization" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "scikit-learn", "title": "Scikit Learn", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Scikit-learn is an open-source Python library providing a unified, consistent API for classical machine learning algorithms covering classification, regression, clustering, dimensionality reduction and model selection. Built on NumPy and SciPy, it emphasises clean estimator interfaces, reproducible pipelines and rigorous evaluation tooling rather than deep learning. It is one of the most widely used libraries for non-neural machine learning, data science education and rapid prototyping.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scikit-learn", "labels": [ "Scikit Learn", "Scikit-Learn", "Scikit-learn", "scikit-learn" ], "is_subclass_of": [ "Machine Learning", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "scope-1-emissions", "title": "Scope 1 Emissions", "domain": "governance", "domain_name": "Governance", "definition": "Scope 1 emissions are direct greenhouse gas (GHG) emissions from sources that are owned or controlled by a reporting organisation, as defined by the GHG Protocol Corporate Standard. They include combustion of fuels in owned vehicles, boilers, and furnaces; process emissions from chemical or physical reactions in production; and fugitive emissions from refrigerant leaks, methane from waste operations, and similar inadvertent releases. Scope 1 emissions are denominated in CO2-equivalent (CO2e) tonnes and represent the most controllable category of an organisation's carbon footprint.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:scope-1-emissions", "labels": [ "Scope 1 Emissions" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "scope-2-emissions", "title": "Scope 2 Emissions", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Scope 2 Emissions are the indirect greenhouse gas emissions attributable to an organisation arising from the generation of purchased or acquired electricity, steam, heat, or cooling consumed in its operations, as defined by the GHG Protocol Corporate Standard. Although the physical emissions occur at the power plant or thermal facility, they are accounted for by the purchasing organisation because its demand drives that generation, making Scope 2 the primary decarbonisation lever for energy-intensive organisations such as data centres and manufacturers.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:scope-2-emissions", "labels": [ "Scope 2 Emissions" ], "is_subclass_of": [ "Environmental Sustainability" ], "wikilinks": [] }, { "id": "scope-3-emissions", "title": "Scope 3 Emissions", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Scope 3 emissions are the indirect greenhouse gas (GHG) emissions that occur across an organisation's value chain as a consequence of its activities but outside its operational boundary, encompassing both upstream sources such as purchased goods and services, capital goods, and business travel, and downstream sources such as the use and end-of-life treatment of sold products. Defined under Category 3 of the GHG Protocol Corporate Standard, Scope 3 emissions typically constitute the largest share of a company's total carbon footprint \u2014 often exceeding 70 percent \u2014 making their measurement, reporting, and reduction critical to credible corporate climate strategies. Accurate Scope 3 accounting requires collaboration across supply chains and the use of spend-based, activity-based, or supplier-specific emission factors.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:scope-3-emissions", "labels": [ "Scope 3 Emissions", "Scope 3 Emissions Accounting", "Scope 3 Emissions Reporting" ], "is_subclass_of": [ "Carbon Accounting" ], "wikilinks": [] }, { "id": "scope-definition", "title": "Scope Definition", "domain": "governance", "domain_name": "Governance", "definition": "Scope definition is the act of establishing the precise boundaries, entities, and conditions to which an assessment, consent, or measurement applies. In governance and consent systems it specifies which permissions, data, or activities are covered, while in environmental assessment it delineates the system boundary for impact accounting. Clear scope is a prerequisite for auditable and comparable results.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:scope-definition", "labels": [ "Scope Definition", "Goal and Scope Definition" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "score-function", "title": "Score Function", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In score-based generative modelling, the Score Function is the gradient of the log probability density of the data with respect to the input, indicating the direction of increasing data likelihood. Diffusion models learn to estimate this score across noise levels, then use it to iteratively denoise samples drawn from a simple prior. The score function connects diffusion models to Langevin-style sampling and energy-based formulations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:score-function", "labels": [ "Score Function" ], "is_subclass_of": [ "Diffusion Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "score-matching", "title": "Score Matching", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Score matching is a method for fitting probability models by matching the gradient of the log-density, the score, of the model to that of the data, avoiding the intractable normalising constant. It underpins score-based generative models and diffusion models.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:score-matching", "labels": [ "Score Matching" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Technique" ], "wikilinks": [ "Probability Theory", "Statistics", "Diffusion Model", "Generative Model", "Machine Learning" ] }, { "id": "score-based-generative-model", "title": "Score-Based Generative Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A score-based generative model is a class of generative model that learns the gradient of the log probability density of data \u2014 the score function \u2014 and samples by reversing a noising process using that learned score. Training perturbs data with noise at multiple scales and fits a neural network to estimate the score at each scale, after which Langevin-style or reverse stochastic-differential-equation dynamics transform noise into samples. The framework provides a unifying continuous-time view that subsumes denoising diffusion models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:score-based-generative-model", "labels": [ "Score-Based Generative Model", "Score-Based Generative Models", "Score-based Generative Model" ], "is_subclass_of": [ "Generative Model", "AI Model Architecture" ], "wikilinks": [] }, { "id": "scorm", "title": "Scorm", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "SCORM (Sharable Content Object Reference Model) is a set of technical standards for packaging and exchanging e-learning content so that courses run consistently across compliant learning management systems. It defines content packaging, run-time communication and sequencing so that a course can report completion, scores and time to the host platform. SCORM established interoperability for digital learning content and is a predecessor to xAPI.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:scorm", "labels": [ "Scorm", "SCORM" ], "is_subclass_of": [ "Education Technology" ], "wikilinks": [] }, { "id": "scratchpad-reasoning", "title": "Scratchpad Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Scratchpad reasoning is a prompting and inference technique in which a language model generates intermediate working steps into an explicit textual workspace before committing to a final answer. The scratchpad externalises latent computation, letting the model decompose a problem, track partial results, and self-correct, which improves performance on multi-step arithmetic, logic, and code tasks. It is a foundational mechanism behind chain-of-thought and related deliberate-reasoning methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:scratchpad-reasoning", "labels": [ "Scratchpad Reasoning" ], "is_subclass_of": [ "Chain-of-Thought Reasoning" ], "wikilinks": [] }, { "id": "screen-capture-api", "title": "Screen Capture API", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Screen Capture API is a W3C web platform interface, centred on getDisplayMedia, that lets web applications obtain a live media stream of a user-selected screen, window, or browser tab. It exposes the captured display as a MediaStream usable for recording, sharing, or processing, gated by an explicit user-selection prompt. It is the standard browser foundation for screen sharing and recording features.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:screen-capture-api", "labels": [ "Screen Capture API", "W3C Screen Capture API" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "screen-capture", "title": "Screen Capture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Screen capture is the acquisition of the visual contents of a display, window, or application as still images or a stream of frames. It underpins screenshots, screen recording, and remote presentation, and is increasingly used to provide perceptual input to AI agents that operate graphical interfaces. Capture is mediated by operating-system or browser APIs subject to user permission.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:screen-capture", "labels": [ "Screen Capture" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "screen-recording", "title": "Screen Recording", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Screen recording is the capture of pixel-level output from a computer display \u2014 including cursor movement, application windows, and system UI \u2014 encoded into a video stream for later playback, streaming, or analysis. It combines display capture with optional audio recording and may include region selection, frame-rate control, hardware-accelerated encoding, and metadata tagging.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:screen-recording", "labels": [ "Screen Recording" ], "is_subclass_of": [ "Screen Capture" ], "wikilinks": [] }, { "id": "screen-sharing", "title": "Screen Sharing", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Screen sharing is the real-time capture and transmission of a computing device's visual display output \u2014 encompassing full-desktop, application-window, browser-tab, and mobile-device viewports \u2014 to one or more remote participants over a network, enabling synchronous visual communication of interf...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:screen-sharing", "labels": [ "Screen Sharing" ], "is_subclass_of": [ "Communication Technology", "Visual Communication", "Synchronous Collaboration Tools", "Remote Desktop Technology", "Distributed Collaboration", "Real-Time Communication" ], "wikilinks": [ "Alliance for Open Media AV1", "Annotation Overlay", "Application Window Capture", "AR Screen Projection", "AV1 Codec", "Clipboard Synchronisation", "Design Collaboration", "Display Capture", "Distributed Collaboration", "DistributedCollaborationDomain", "GPU Hardware Encoding", "H.264", "H.264 Screen Content Coding", "HumanComputerInteractionDomain", "Hybrid Working", "IETF WebRTC", "Intraframe Refresh", "ITU-T H.264", "Live Code Demonstration", "NDI Protocol" ] }, { "id": "script", "title": "Script", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Bitcoin's stack-based, non-Turing-complete scripting language used to encode spending conditions on transaction outputs and to provide the corresponding unlocking data in transaction inputs. Script programs define locking and unlocking conditions\u2014such as pay-to-public-key-hash\u2014that a validator executes to authorise fund transfer, making transaction validation a programmable process without full smart-contract complexity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:script", "labels": [ "Script", "Redeem Script", "Script Language" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "scripting-language", "title": "Scripting Language", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A scripting language is an interpreted or dynamically compiled programming language used to automate tasks, define interactive behaviours, and extend the capabilities of host applications or game engines without modifying core engine code. In metaverse contexts, scripting languages such as JavaScript, Lua, Python, and C# enable content creators to author custom gameplay mechanics, NPC behaviour, and dynamic environment logic with rapid iteration cycles.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:scripting-language", "labels": [ "Scripting Language" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "scroll", "title": "Scroll", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Scroll is an Ethereum layer-two network that uses zero-knowledge rollup technology to scale transactions while remaining compatible with the Ethereum Virtual Machine. It posts validity proofs to Ethereum.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:scroll", "labels": [ "Scroll" ], "is_subclass_of": [ "Layer 2 Scaling" ], "wikilinks": [ "Ethereum", "PLONK", "DeFi", "Rollup", "Layer 2 Scaling", "https://scroll.io", "https://docs.scroll.io" ] }, { "id": "sea-ice-floe", "title": "Sea Ice Floe", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-ice-floe", "labels": [ "Sea Ice Floe", "IceFloe" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sea-ice-lead", "title": "Sea Ice Lead", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-ice-lead", "labels": [ "Sea Ice Lead" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sea-ice-monitoring", "title": "Sea Ice Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-ice-monitoring", "labels": [ "Sea Ice Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sea-ice", "title": "Sea Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-ice", "labels": [ "Sea Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sea-surface-height-measurement", "title": "Sea Surface Height Measurement", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-surface-height-measurement", "labels": [ "Sea Surface Height Measurement" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sea-surface-temperature-retrieval", "title": "Sea Surface Temperature Retrieval", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sea-surface-temperature-retrieval", "labels": [ "Sea Surface Temperature Retrieval" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "search-algorithm", "title": "Search Algorithm", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Search Algorithms in AI systematically explore solution spaces to find optimal or satisfactory solutions to problems. Classical algorithms include uninformed search (breadth-first, depth-first, uniform-cost) and informed search (A*, greedy best-first, beam search). Advanced techniques incorporate heuristics, pruning, bidirectional search, and iterative deepening. Modern AI integrates learning-based search (Monte Carlo Tree Search with neural networks, learned heuristics) and continuous optimisation methods.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:search-algorithm", "labels": [ "Search Algorithm", "A* Search", "Best-First Search", "Tree Search Algorithm" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Constraint Satisfaction", "Planning", "Heuristic Methods", "Monte Carlo Tree Search" ] }, { "id": "search-algorithms", "title": "Search Algorithms", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Computational methods for systematically navigating problem spaces to find solutions, optimal paths, or goal states, employing strategies such as breadth-first, depth-first, heuristic-guided, or adversarial search to efficiently discover answers to complex problems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:search-algorithms", "labels": [ "Search Algorithms" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "Graph Theory", "Optimization", "Pathfinding", "Artificial Intelligence", "Planning and Scheduling" ] }, { "id": "search-discovery", "title": "Search Discovery", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systems and interfaces that enable users to find relevant content, assets, experiences, and other users within metaverse platforms through keyword search, semantic queries, recommendations, and spatial exploration mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:search-discovery", "labels": [ "Search Discovery", "Search & Discovery" ], "is_subclass_of": [ "Standards and Interoperability", "Information Retrieval" ], "wikilinks": [ "User Navigation", "Information Retrieval", "metaverse" ] }, { "id": "search-engine-optimisation", "title": "Search Engine Optimisation", "domain": "data", "domain_name": "Data", "definition": "Search engine optimisation (SEO) is the discipline of improving a website's visibility and ranking in organic search results. It combines technical optimisation, content relevance, structured data, and authority signals to align pages with how search engines crawl, index, and rank content. SEO bridges marketing intent with the indexing and ranking machinery of search engines to attract qualified, unpaid traffic.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:search-engine-optimisation", "labels": [ "Search Engine Optimisation" ], "is_subclass_of": [ "Digital Marketing" ], "wikilinks": [] }, { "id": "search-engine", "title": "Search Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A search engine is a software system that systematically crawls, indexes, and ranks digital content to retrieve relevant results in response to user queries. It combines web crawling, inverted-index construction, relevance ranking (including PageRank-style link analysis and learning-to-rank models), and query understanding (tokenisation, stemming, NLP) into an end-to-end pipeline. Modern search engines increasingly integrate semantic embeddings, dense retrieval, and large language model components to handle natural-language and multimodal queries. They constitute foundational information infrastructure for the open web, enterprise knowledge bases, e-commerce catalogues, and emerging spatial and metaverse content layers.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:search-engine", "labels": [ "Search Engine" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [ "Content Discovery", "Information Retrieval", "metaverse" ] }, { "id": "search-index", "title": "Search Index", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A data structure that maps terms, embeddings, or attributes to document locations, enabling rapid retrieval of relevant content from large metaverse asset catalogues through inverted indexes, vector indexes, or hybrid approaches.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:search-index", "labels": [ "Search Index", "Full-text Search Index" ], "is_subclass_of": [ "Content and Assets", "Data Structure" ], "wikilinks": [ "Fast Query Response", "Data Structure", "metaverse" ] }, { "id": "search-interface", "title": "Search Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User-facing components that enable query input, result presentation, and navigation through search results in metaverse platforms, including text fields, voice input, spatial gestures, and augmented reality overlays for asset and experience discovery.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:search-interface", "labels": [ "Search Interface" ], "is_subclass_of": [ "Interaction Technology", "User Interface" ], "wikilinks": [ "User Query Experience", "metaverse", "User Interface" ] }, { "id": "search-space-definition", "title": "Search Space Definition", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Search space definition is the process of specifying the set of candidate configurations, such as model architectures, hyperparameters, or feature transformations, that an automated search or optimisation algorithm is permitted to explore. A well-formed search space bounds each parameter's type and range and captures dependencies between parameters, directly shaping the efficiency and quality of the resulting search. It is a foundational step in AutoML and neural architecture search, where a poorly defined space can make the optimisation problem intractable or exclude high-performing solutions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:search-space-definition", "labels": [ "Search Space Definition" ], "is_subclass_of": [ "AutoML" ], "wikilinks": [] }, { "id": "search-space", "title": "Search Space", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Search Space is the full set of candidate configurations, such as hyperparameter values, architectures, or solutions, that an optimisation or search algorithm is permitted to explore in pursuit of an objective. Its size and structure directly determine how tractable a search problem is: a poorly bounded or high-dimensional search space can make exhaustive search infeasible. Algorithms such as grid search and random search differ chiefly in how they sample points from this space.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:search-space", "labels": [ "Search Space" ], "is_subclass_of": [ "Hyperparameter Optimisation" ], "wikilinks": [] }, { "id": "search-technology", "title": "Search Technology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Search technology encompasses the algorithms, indexing systems, and retrieval mechanisms that enable efficient discovery of relevant information across structured and unstructured data sources. Modern search systems combine inverted indices, vector embeddings, and machine learning ranking models to support keyword, semantic, and hybrid queries at scale.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:search-technology", "labels": [ "Search Technology" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "owl:Thing" ] }, { "id": "search-and-rescue-robotics", "title": "Search and Rescue Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Search and rescue robotics is the application of robotic systems to locate, assess and assist victims in disaster and emergency environments that are dangerous or inaccessible to human responders. These robots traverse rubble, water and confined spaces, carrying sensors to detect survivors and relay situational awareness to rescue teams. By extending responders' reach into hazardous zones, they reduce risk to human life while accelerating victim location.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:search-and-rescue-robotics", "labels": [ "Search and Rescue Robotics" ], "is_subclass_of": [ "Robotics", "Robot Type" ], "wikilinks": [] }, { "id": "seasonal-ice", "title": "Seasonal Ice", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:seasonal-ice", "labels": [ "Seasonal Ice" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "second-order-optimisation", "title": "Second Order Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Second-order optimisation is a family of optimisation methods that use second-derivative (curvature) information, typically the Hessian matrix or its approximations, to determine search directions and step sizes. By accounting for the curvature of the objective, these methods can converge in far fewer iterations than first-order methods near a minimum, at the cost of higher per-iteration computation and memory. Examples include Newton's method, quasi-Newton methods such as L-BFGS, and conjugate-gradient approaches.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:second-order-optimisation", "labels": [ "Second Order Optimisation", "Second-Order Optimisation" ], "is_subclass_of": [ "Convex Optimisation" ], "wikilinks": [] }, { "id": "secondary-market-integration", "title": "Secondary Market Integration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical and business infrastructure connecting primary asset creation with resale marketplaces, enabling liquidity for digital collectibles, virtual real estate, and NFTs through automated royalty distribution and cross-platform trading protocols.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:secondary-market-integration", "labels": [ "Secondary Market Integration", "Secondary Market Trading" ], "is_subclass_of": [ "Platform and Environment", "Digital Asset Trading" ], "wikilinks": [ "Asset Liquidity", "Digital Asset Trading", "metaverse" ] }, { "id": "secondary-market", "title": "Secondary Market", "domain": "finance", "domain_name": "Finance", "definition": "A secondary market is a venue in which existing financial assets are bought and sold among investors after their initial issuance, rather than purchased directly from the issuer. By allowing holders to exit positions and new buyers to enter, it provides liquidity, continuous price discovery, and a mechanism for valuing assets through ongoing trading. Secondary markets are essential to the functioning of capital markets because the ability to resell makes initial investment far more attractive.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:secondary-market", "labels": [ "Secondary Market" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "secret-sharing", "title": "Secret Sharing", "domain": "security", "domain_name": "Security", "definition": "Secret sharing is a cryptographic technique that splits a secret into multiple shares distributed among participants such that only an authorised subset, meeting a defined threshold, can reconstruct the secret, while any smaller subset learns nothing about it. The canonical construction, Shamir's secret sharing, uses polynomial interpolation over a finite field. Secret sharing underpins threshold cryptography, distributed key management and secure multi-party computation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:secret-sharing", "labels": [ "Secret Sharing" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "section-508", "title": "Section 508", "domain": "governance", "domain_name": "Governance", "definition": "Section 508 is an amendment to the US Rehabilitation Act requiring federal agencies to make their electronic and information technology accessible to people with disabilities. Its 2017 refresh harmonised the technical requirements with the WCAG 2.0 Level AA success criteria. It functions as an enforceable procurement and compliance standard for accessibility in US government IT.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:section-508", "labels": [ "Section 508" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "secure-aggregation", "title": "Secure Aggregation", "domain": "security", "domain_name": "Security", "definition": "Secure aggregation is a cryptographic protocol that computes the sum of inputs held by many parties without revealing any individual input to the aggregator or other participants. It is most prominent in federated learning, where a server combines model updates from clients while learning only the aggregate. Typical constructions use pairwise masking, secret sharing, or additively homomorphic encryption, with dropout-resilient designs so the protocol completes even when some clients disconnect.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:secure-aggregation", "labels": [ "Secure Aggregation" ], "is_subclass_of": [ "Secure Multi-Party Computation" ], "wikilinks": [] }, { "id": "secure-boot", "title": "Secure Boot", "domain": "security", "domain_name": "Security", "definition": "Secure Boot is a platform security mechanism that verifies the cryptographic signature of each component loaded during system start-up, allowing only software trusted by an established chain of keys to execute. By validating firmware, bootloaders, and the operating system loader before handing over control, it prevents persistent low-level malware such as bootkits from running. Secure Boot establishes a hardware-anchored chain of trust from power-on through to the operating system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:secure-boot", "labels": [ "Secure Boot" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "secure-channel", "title": "Secure Channel", "domain": "security", "domain_name": "Security", "definition": "A Secure Channel is a communication path between two parties that protects the confidentiality, integrity and authenticity of exchanged data against eavesdropping and tampering. It is typically established by a key-exchange and authentication handshake that derives session keys, then protects subsequent traffic with authenticated encryption. Secure channels underpin protocols such as TLS and are foundational to trustworthy communication over untrusted networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:secure-channel", "labels": [ "Secure Channel" ], "is_subclass_of": [ "Secure Communication" ], "wikilinks": [] }, { "id": "secure-communication", "title": "Secure Communication", "domain": "security", "domain_name": "Security", "definition": "Secure communication is the practice and set of technologies that protect the confidentiality, integrity, and authenticity of information exchanged between parties over potentially adversarial networks, ensuring that eavesdroppers cannot read message content, tamperers cannot alter it undetected, and impersonators cannot forge the identity of legitimate participants. The discipline encompasses transport-layer security protocols (TLS, DTLS, QUIC with TLS), end-to-end encryption protocols (Signal Protocol, MLS, Matrix), secure messaging standards (S/MIME, PGP), VPN tunnelling (IPsec, WireGuard), and the cryptographic primitives\u2014asymmetric key exchange, symmetric cipher suites, authenticated encryption, and digital signatures\u2014that underpin them. Security properties are formally analysed through cryptographic protocol proofs and verified implementations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:secure-communication", "labels": [ "Secure Communication", "Encrypted Communication", "Secure Communication Channel", "Secure Communication Protocols", "Ultra-Secure Communication" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "secure-data-sharing", "title": "Secure Data Sharing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Cryptographic and access control mechanisms that enable controlled exchange of sensitive information between parties in metaverse environments while maintaining confidentiality, integrity, and regulatory compliance through encryption and permissioned access.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:secure-data-sharing", "labels": [ "Secure Data Sharing" ], "is_subclass_of": [ "Security and Identity", "Data Security" ], "wikilinks": [ "Confidential Collaboration", "Data Security", "metaverse" ] }, { "id": "secure-element", "title": "Secure Element", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Secure Element (SE) is a tamper-resistant hardware component, typically a dedicated microcontroller, that securely stores cryptographic keys and executes sensitive operations such as signing and authentication in physical isolation from the host system. It enforces hardware-backed access controls so that secret material never leaves the chip in plaintext, resisting both software extraction and many physical attacks. Secure Elements underpin hardware wallets, payment cards, SIMs and mobile secure payment systems, and are commonly certified against standards such as Common Criteria EAL levels.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:secure-element", "labels": [ "Secure Element" ], "is_subclass_of": [ "Hardware Security Module" ], "wikilinks": [] }, { "id": "secure-email", "title": "Secure Email", "domain": "security", "domain_name": "Security", "definition": "Secure email is the set of technologies and practices that protect the confidentiality, integrity, and authenticity of electronic mail in transit and at rest. It combines transport encryption between mail servers, end-to-end message encryption and signing schemes, and sender-authentication and anti-spoofing controls, so that recipients can trust who sent a message and that its contents were not read or altered along the way.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:secure-email", "labels": [ "Secure Email" ], "is_subclass_of": [ "Secure Communication" ], "wikilinks": [] }, { "id": "secure-enclave", "title": "secure enclave", "domain": "security", "domain_name": "Security", "definition": "A secure enclave is a hardware-isolated execution environment embedded within a processor that maintains confidentiality and integrity guarantees for code and data even when the host operating system, hypervisor, or other privileged software is compromised. Implemented through technologies such as Intel SGX, ARM TrustZone, Apple Secure Enclave Processor, and AMD SEV-SNP, these environments use hardware memory encryption and access-control mechanisms enforced within the CPU package itself. Remote attestation allows a verifier to cryptographically confirm that specific code runs inside a genuine, unmodified enclave before transmitting sensitive data to it, without trusting the surrounding software stack. Secure enclaves are foundational to confidential computing, privacy-preserving machine learning, decentralised identity, and secure multi-party computation.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:secure-enclave", "labels": [ "Secure Enclave", "Enclave", "Secure Element Chip", "Secure Enclaves" ], "is_subclass_of": [ "Trusted Execution Environment" ], "wikilinks": [] }, { "id": "secure-messaging", "title": "Secure Messaging", "domain": "security", "domain_name": "Security", "definition": "Secure messaging is the exchange of messages protected by cryptography so that only intended participants can read or authenticate them. It typically provides end-to-end encryption, forward secrecy, and integrity, often through ratcheting key-agreement protocols. It underpins privacy-preserving communication in consumer apps, enterprise systems, and decentralised networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:secure-messaging", "labels": [ "Secure Messaging", "Private Messaging", "Secure Messaging Protocol" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "secure-multi-party-computation", "title": "Secure Multi-Party Computation", "domain": "security", "domain_name": "Security", "definition": "Secure Multi-Party Computation (MPC) is a cryptographic protocol enabling multiple parties to jointly compute functions over their combined private inputs without revealing individual inputs to other participants or third parties. Only the final output is disclosed; intermediate computations remain confidential. Key implementation techniques include secret sharing (Shamir), garbled circuits (Yao), and oblivious transfer primitives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:secure-multi-party-computation", "labels": [ "Secure Multi-Party Computation", "Multiparty Computation", "Secure Computation", "Secure Multiparty Computation" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [ "MP-SPDZ", "Shamir Secret Sharing", "Yao's Garbled Circuits", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "secure-storage", "title": "Secure Storage", "domain": "security", "domain_name": "Security", "definition": "Secure storage is the persistence of data with protections for confidentiality, integrity, and controlled access throughout its lifecycle. It combines encryption at rest, access control, tamper-evidence, and key management, sometimes anchored in hardware security modules or trusted execution environments. It is essential where stored records must remain trustworthy, such as audit trails and chains of custody.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:secure-storage", "labels": [ "Secure Storage", "Secure Key Storage", "Secure Token Storage" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "securities-act-1933", "title": "Securities Act 1933", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Securities Act of 1933 is the foundational United States federal statute governing the offer and sale of securities. It requires that securities offered to the public be registered with regulators and accompanied by truthful disclosure, with the aim of preventing fraud and protecting investors. In digital-asset contexts it is central to determining whether a token sale constitutes an unregistered securities offering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:securities-act-1933", "labels": [ "Securities Act 1933" ], "is_subclass_of": [ "Securities Law" ], "wikilinks": [] }, { "id": "securities-exchange-act-1934", "title": "Securities Exchange Act 1934", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Securities Exchange Act of 1934 is the US federal law that governs secondary trading of securities and created the Securities and Exchange Commission. It establishes registration, disclosure, and anti-fraud requirements for exchanges, brokers, and listed companies. It is a foundational reference for determining how digital assets and investment products are regulated in the United States.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:securities-exchange-act-1934", "labels": [ "Securities Exchange Act 1934", "Securities Exchange Act" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "securities-law", "title": "Securities Law", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Securities law is the body of statutory and regulatory rules governing the issuance, sale, and trading of financial instruments classified as securities, with the aim of protecting investors and ensuring fair, transparent markets. It defines disclosure obligations, registration requirements, and anti-fraud provisions, and applies tests to determine when an instrument such as a token constitutes a security. In the digital-asset context it shapes whether token offerings must register and comply with investor-protection regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:securities-law", "labels": [ "Securities Law" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "securities-regulation", "title": "Securities Regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The legal and regulatory frameworks determining wher digital assets \u2014 cryptocurrency tokens, stablecoins, security tokens, DeFi protocol interests, NFTs, and AI-generated financial instruments \u2014 qualify as securities requiring registration, disclosure, and ongoing compliance obligations across mu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:securities-regulation", "labels": [ "Securities Regulation", "Principles of Securities Regulation", "Securities Registration", "crypto-securities-law, digital-asset-regulation, token-securities-compliance, investment-contract-analysis" ], "is_subclass_of": [ "Governance and Regulation", "DeFi and Economics", "Financial Regulation", "Investment Law", "Compliance Framework", "Digital Asset Law", "Capital Markets Regulation" ], "wikilinks": [ "Accredited Investor Verification", "Alternative Trading System", "Bank Secrecy Act", "Banking Regulation", "Basel Committee on Banking Supervision", "BIS", "Broker-Dealer Registration", "Capital Formation", "Capital Markets Regulation", "CFTC", "Commodity Regulation", "ComplianceLayer", "Contract Law Frameworks", "CPMI", "DeFi", "Decentralisation Analysis", "Digital Asset Custody", "Digital Asset Custody Innovation", "Digital Asset Law", "Disclosure Obligations" ] }, { "id": "securities-settlement", "title": "Securities Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Securities settlement is the process by which the buyer of a security receives the asset and the seller receives payment, completing a trade through the final transfer of legal ownership and funds between counterparties, typically via a central securities depository (CSD) or clearing house. Traditional settlement occurs on a T+1 or T+2 basis after trade execution, involves complex chains of custodians, CSDs, and central counterparty clearing houses (CCPs), and carries counterparty risk during the settlement window. Distributed ledger technology and tokenised securities are driving a shift toward atomic and real-time settlement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:securities-settlement", "labels": [ "Securities Settlement", "24-7 Securities Settlement", "Digital Securities Settlement", "Securities Transfer" ], "is_subclass_of": [ "Financial System" ], "wikilinks": [] }, { "id": "securitize", "title": "Securitize", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A financial technology company providing infrastructure for issuing, managing and trading tokenised securities on blockchain networks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:securitize", "labels": [ "Securitize" ], "is_subclass_of": [ "Asset Tokenisation" ], "wikilinks": [ "Blockchain", "Securities Regulation", "Security Token", "Tokenisation", "Asset Tokenisation" ] }, { "id": "security-architecture", "title": "Security Architecture", "domain": "security", "domain_name": "Security", "definition": "A comprehensive framework defining security controls, policies, and technologies that protect systems, infrastructure, user data, and digital assets through defence-in-depth strategies including authentication, encryption, access control, threat monitoring, and zero-trust principles.", "entityType": "Class", "qualityScore": 0.7, "maturity": "draft", "iri": "urn:ngm:class:security-architecture", "labels": [ "Security Architecture", "SecurityArchitecture" ], "is_subclass_of": [ "Governance and Safety", "System Architecture" ], "wikilinks": [ "Secure Systems", "metaverse", "System Architecture" ] }, { "id": "security-audit", "title": "Security Audit", "domain": "security", "domain_name": "Security", "definition": "A security audit is a systematic evaluation of a system's controls, configurations, and code against security requirements and threats. It combines techniques such as code review, configuration assessment, penetration testing, and control verification to identify vulnerabilities and compliance gaps. Audits produce evidence and remediation guidance used to reduce risk and demonstrate assurance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:security-audit", "labels": [ "Security Audit", "Security Auditing" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "security-by-design", "title": "Security By Design", "domain": "security", "domain_name": "Security", "definition": "Security by design is an engineering approach in which security is treated as a foundational requirement and built into systems from the earliest stages rather than bolted on retrospectively. It favours secure defaults, minimal attack surface, defence in depth and least-privilege access, supported by threat modelling and continuous verification across the development lifecycle. The principle is increasingly mandated by regulators and procurement frameworks for connected products and critical infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:security-by-design", "labels": [ "Security By Design", "Security by Design" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "security-engineering", "title": "Security Engineering", "domain": "security", "domain_name": "Security", "definition": "Security engineering is the discipline of designing, building, and maintaining systems that remain dependable and trustworthy in the face of malice, error, and accident. It applies systematic engineering practices to protect the confidentiality, integrity, and availability of information and assets, integrating threat modelling, secure architecture, and verifiable controls across the development lifecycle. Security engineers reason about adversaries, attack surfaces, and trust boundaries to make risk decisions defensible and economically rational.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:security-engineering", "labels": [ "Security Engineering" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "security-framework", "title": "Security Framework", "domain": "security", "domain_name": "Security", "definition": "A Security Framework is a structured, normative system of policies, controls, standards, and procedural guidelines that an organisation applies to protect digital assets, information systems, and users from adversarial threats and accidental harm. It integrates risk management principles with technical controls across identity, access, cryptography, monitoring, and incident response to establish a cohesive defence posture. Widely adopted frameworks such as NIST CSF, ISO/IEC 27001, and CIS Controls provide systematic vocabularies for assessing security maturity and aligning investment with threat landscape. In emergent domains such as spatial computing, decentralised infrastructure, and AI-mediated platforms, security frameworks extend to cover smart-contract auditing, decentralised identity, privacy-preserving computation, and supply-chain integrity.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:security-framework", "labels": [ "Security Framework", "DLT Security Framework" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "security-information-and-event-management", "title": "Security Information and Event Management", "domain": "security", "domain_name": "Security", "definition": "Security Information and Event Management (SIEM) is a security discipline and platform category that aggregates, normalises and correlates log and event data from across an estate to detect, investigate and respond to threats. It combines real-time monitoring and alerting with longer-term storage for forensics and compliance reporting. SIEM is the analytical core of most security operations, turning raw telemetry into actionable detections.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:security-information-and-event-management", "labels": [ "Security Information and Event Management", "Security Information And Event Management" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "security-infrastructure", "title": "Security Infrastructure", "domain": "security", "domain_name": "Security", "definition": "Security infrastructure encompasses the integrated set of hardware, software, processes, and policies that protect an organisation's information systems, networks, and data assets from unauthorised access, disruption, and exploitation. It includes perimeter defences, identity and access management systems, cryptographic key management, endpoint protection, monitoring and detection platforms, and incident response capabilities. Effective security infrastructure is layered\u2014applying defence-in-depth principles so that compromise of any single control does not expose the entire system. Modern security infrastructure increasingly adopts zero-trust architectural principles, eliminating implicit trust based on network location and requiring continuous verification of every access request. Security infrastructure must align with regulatory requirements and evolve continuously in response to the changing threat landscape.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:security-infrastructure", "labels": [ "Security Infrastructure", "SecurityInfrastructure" ], "is_subclass_of": [ "Security Framework" ], "wikilinks": [] }, { "id": "security-layer", "title": "Security Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Security Layer is the implementation-focused abstraction level that encapsulates cryptographic mechanisms, security protocols, threat models, and defensive architectures protecting blockchain and distributed systems from attacks whilst ensuring data integrity, authenticity, and confidentiality. It encompasses concrete implementations such as hash functions, digital signatures, zero-knowledge proofs, access control, and formal verification, distinct from higher-level conceptual security properties.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:security-layer", "labels": [ "Security Layer", "SecurityLayer" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "security-module", "title": "Security Module", "domain": "security", "domain_name": "Security", "definition": "A security module is a dedicated hardware or software component that provides cryptographic and protection services such as key generation, secure storage, signing, and access enforcement. Hardware security modules and secure elements isolate sensitive operations from the general computing environment to resist extraction and tampering. Such modules are required wherever keys and credentials must be safeguarded, as in custody and IoT systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:security-module", "labels": [ "Security Module" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "security-monitoring", "title": "Security Monitoring", "domain": "security", "domain_name": "Security", "definition": "Security monitoring is the continuous collection, correlation and analysis of telemetry from systems, networks and applications to detect indicators of compromise and policy violations. It feeds detection rules, baselines and analytics that surface suspicious behaviour for investigation and response. As a discipline it spans log aggregation, intrusion detection, threat intelligence enrichment and alerting, and is a core function of a security operations centre.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:security-monitoring", "labels": [ "Security Monitoring" ], "is_subclass_of": [ "Threat Detection" ], "wikilinks": [] }, { "id": "security-operations-centre", "title": "Security Operations Centre", "domain": "security", "domain_name": "Security", "definition": "A Security Operations Centre (SOC) is a centralised function, combining people, processes and technology, that continuously monitors, detects, analyses and responds to cybersecurity threats across an organisation. Analysts triage alerts from telemetry sources, investigate incidents, and coordinate containment and remediation, typically aided by SIEM and SOAR platforms. The SOC is the operational hub of an enterprise security programme, providing the situational awareness and response capability needed to limit the impact of attacks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:security-operations-centre", "labels": [ "Security Operations Centre" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "security-operations", "title": "Security Operations", "domain": "security", "domain_name": "Security", "definition": "Security operations is the ongoing practice of monitoring, detecting, investigating, and responding to security threats across an organisation's systems. Centred on a security operations centre, it integrates log collection, SIEM correlation, alert triage, threat intelligence, and incident response. Its goal is to reduce dwell time and limit the impact of attacks through continuous vigilance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:security-operations", "labels": [ "Security Operations", "Security Operations Center" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "security-policy", "title": "Security Policy", "domain": "security", "domain_name": "Security", "definition": "A security policy is a formalised set of rules, principles, and procedures that govern how an organisation protects its information assets, systems, and personnel. It defines acceptable use, access control objectives, incident response obligations, and compliance requirements. Security policies serve as the authoritative reference for all subordinate security controls, technical configurations, and procedural guidelines within an enterprise.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:security-policy", "labels": [ "Security Policy" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "security-protocol", "title": "Security Protocol", "domain": "standards", "domain_name": "Standards", "definition": "A security protocol is a defined sequence of message exchanges and cryptographic operations that lets parties achieve security goals such as confidentiality, integrity, authentication or key establishment over an untrusted channel. Protocols specify message formats, ordering, cryptographic primitives and state transitions, and are designed to resist defined adversaries. Examples include TLS for transport security and authentication protocols for identity verification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:security-protocol", "labels": [ "Security Protocol" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "security-scanning", "title": "Security Scanning", "domain": "security", "domain_name": "Security", "definition": "The automated process of inspecting software artifacts, code, or data for known vulnerabilities, malware, or policy violations before deployment or execution.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:security-scanning", "labels": [ "Security Scanning" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "security-services", "title": "Security Services", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Security Services are the technical and organisational mechanisms that protect spatial computing platforms from unauthorised access, data breaches, and malicious activity. They encompass authentication, encryption, access control, and identity management components that collectively enforce security policies across metaverse, cloud, and immersive technology deployments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:security-services", "labels": [ "Security Services" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "owl:Thing" ] }, { "id": "security-standards", "title": "Security Standards", "domain": "security", "domain_name": "Security", "definition": "Security standards are published specifications that define requirements, controls, and best practices for protecting information systems. They span management frameworks such as ISO/IEC 27001, control catalogues such as NIST SP 800-53, and authentication standards. They provide a common basis for designing, assessing, and certifying the security posture of organisations and products.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:security-standards", "labels": [ "Security Standards", "Security Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "security-technology", "title": "Security Technology", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Security Technology refers to the hardware, software, and protocol mechanisms deployed to protect systems, data, and communications from unauthorised access, tampering, or disruption. In the blockchain and distributed systems context this encompasses cryptographic primitives, authentication schemes, digital signatures, zero-knowledge proofs, and secure enclaves that collectively enforce integrity, confidentiality, and non-repudiation across decentralised networks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:security-technology", "labels": [ "Security Technology" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "owl:Thing" ] }, { "id": "security-testing", "title": "Security Testing", "domain": "security", "domain_name": "Security", "definition": "Security testing is the evaluation of software to discover vulnerabilities and verify that security controls behave as intended. It includes static application security testing, dynamic testing, dependency and secret scanning, fuzzing, and penetration testing. Integrated into development pipelines, it shifts vulnerability discovery earlier and continuously throughout the software lifecycle.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:security-testing", "labels": [ "Security Testing" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "security-token-offering", "title": "Security Token Offering", "domain": "finance", "domain_name": "Finance", "definition": "A Security Token Offering (STO) is a regulated capital-raising mechanism in which blockchain-based digital tokens representing ownership rights, equity interests, debt obligations, revenue participation, or other financial entitlements are issued and sold to investors in compliance with applicable securities law. STOs require issuers to satisfy jurisdictional regulatory frameworks \u2014 such as the SEC's Regulation D or Regulation S exemptions in the United States, or MiFID II and EU prospectus rules in Europe \u2014 including investor accreditation verification, Know Your Customer and Anti-Money Laundering checks, and mandatory disclosure obligations. Unlike Initial Coin Offerings, which frequently issued utility tokens, STOs embed compliance logic directly into programmable smart contracts on distributed ledger platforms, automating transfer restrictions, cap-table management, and dividend or interest distributions. The STO model bridges traditional capital markets infrastructure with blockchain-based asset tokenisation, enabling fractional ownership of previously illiquid assets such as real estate, private equity, and infrastructure funds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:security-token-offering", "labels": [ "Security Token Offering" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "owl:Thing" ] }, { "id": "security-token", "title": "Security Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain-based token representing complete or fractional ownership interests in real-world assets or entities, subject to securities regulations that may restrict transfer based on investor identity, jurisdiction, or asset category. Security tokens provide holders with entitlements such as dividends, profit-sharing, or voting rights, and are typically issued through security token offerings (STOs) under regulatory compliance frameworks.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:security-token", "labels": [ "Security Token" ], "is_subclass_of": [ "Fungible Token" ], "wikilinks": [ "Blockchain", "Fungible Token" ] }, { "id": "security", "title": "Security", "domain": "security", "domain_name": "Security", "definition": "The protection of AI systems and their components against unauthorized access, manipulation, disruption, or exploitation, encompassing confidentiality, integrity, and availability of data, models, and infrastructure throughout the AI lifecycle.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:security", "labels": [ "Security", "Cyber Security", "Cybersecurity", "IT Security", "Security Mechanism", "Security Obligations", "System Security" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "apostolaki2016hijacking; @apostolaki2017hijacking; @johnson2014game; @stinner2022proof", "dymydiuk2020rubicon", "Eurodollar", "Anthropic Claude", "CBDCs", "Cyber Security and Cryptography", "Distributed Identity", "Gaussian Splatting", "Hardware and Edge", "machine learning", "MetaverseDomain", "Politics, Law, Privacy", "robotics" ] }, { "id": "seed-phrase", "title": "Seed Phrase", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A seed phrase, also called a recovery or mnemonic phrase, is an ordered list of words, typically twelve or twenty-four, that encodes the master secret from which a cryptocurrency wallet derives all of its private keys. Generated from random entropy and mapped to words via a standard wordlist, it provides a human-readable backup that can fully restore a wallet on any compatible device. Because anyone holding the phrase controls the funds, its secrecy and secure storage are paramount to self-custody.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:seed-phrase", "labels": [ "Seed Phrase" ], "is_subclass_of": [ "Digital Wallet" ], "wikilinks": [] }, { "id": "seepage-face", "title": "Seepage Face", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:seepage-face", "labels": [ "Seepage Face" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "seg-wit", "title": "SegWit", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Segregated Witness (SegWit) is a Bitcoin protocol upgrade (BIP141/BIP143/BIP144) that moves witness data (signatures and scripts) into a separate structure outside the traditional transaction serialisation, eliminating transaction malleability, introducing a weight-based block-size accounting system, and enabling second-layer protocols such as the Lightning Network and subsequent upgrades such as Taproot.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:seg-wit", "labels": [ "SegWit" ], "is_subclass_of": [ "Bitcoin Protocol" ], "wikilinks": [ "Bitcoin Script", "Lightning Network", "Taproot", "Bitcoin", "Bitcoin Protocol" ] }, { "id": "segmentation-and-identification", "title": "Segmentation and Identification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Segmentation and Identification is a foundational computer vision subdomain encompassing the computational processes of partitioning digital images and video frames into semantically meaningful regions and assigning categorical or instance-level identities to those regions.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:segmentation-and-identification", "labels": [ "Segmentation and Identification" ], "is_subclass_of": [ "AI Application", "Computer Vision", "Object Recognition", "Scene Understanding", "Image Analysis", "Deep Learning" ], "wikilinks": [ "ADE20K", "AIPerceptionDomain", "AlgorithmLayer", "Annotated Training Data", "AR Content Removal", "Autonomous Driving Perception", "Backbone Network", "BDD100K", "Cityscapes Benchmark", "Cityscapes Evaluation", "CLIP", "COCO Benchmark", "COCO Dataset", "ComputerVisionDomain", "Convolutional Neural Networks", "Cross-Entropy Loss", "DeepLab", "DETR", "Dice Loss", "Feature Pyramid Network" ] }, { "id": "segregated-witness", "title": "Segregated Witness", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Segregated Witness (SegWit) is a protocol upgrade, activated on Bitcoin as a backwards-compatible soft fork, that moves the witness data \u2014 the digital signatures authorising spends \u2014 out of the main transaction body into a separate structure. By segregating signatures, it fixes transaction malleability, since the transaction identifier no longer depends on mutable signature data, and effectively increases block capacity through a block-weight accounting scheme. SegWit also created the foundation for second-layer protocols such as the Lightning Network and enabled later script upgrades.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:segregated-witness", "labels": [ "Segregated Witness" ], "is_subclass_of": [ "Blockchain Scalability" ], "wikilinks": [] }, { "id": "segregation-of-duties", "title": "Segregation of Duties", "domain": "governance", "domain_name": "Governance", "definition": "Segregation of duties is an internal control principle that divides critical tasks among multiple people so that no single individual can both execute and conceal an error or fraud. It separates responsibilities such as authorisation, custody, recording, and reconciliation. It is a core requirement of financial, security, and compliance control frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:segregation-of-duties", "labels": [ "Segregation of Duties" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "selective-disclosure", "title": "selective disclosure", "domain": "security", "domain_name": "Security", "definition": "Selective disclosure is a cryptographic capability that allows the holder of a verifiable credential to present only a chosen subset of the credential's claims to a verifier, without revealing the undisclosed fields or requiring re-issuance by the issuer. It is implemented through specialised signature schemes such as BBS+ signatures and SD-JWT, as well as zero-knowledge proof systems, enabling fine-grained data minimisation in decentralised identity architectures. As a core privacy-engineering primitive, selective disclosure satisfies regulatory requirements such as GDPR's data minimisation principle and supports the eIDAS 2.0 digital wallet framework. The concept bridges cryptographic credential security with practical privacy-preserving authentication across identity, finance, and access-control domains.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:selective-disclosure", "labels": [ "Selective Disclosure", "Selective Disclosure Mechanism", "SelectiveDisclosure" ], "is_subclass_of": [ "Privacy Preserving Technology" ], "wikilinks": [] }, { "id": "selective-forwarding-unit", "title": "Selective Forwarding Unit", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Selective Forwarding Unit (SFU) is a media server architecture used in multi-party real-time communication that receives media streams from each participant and selectively forwards them to other participants without mixing or decoding the content. This approach reduces server-side computational cost compared to Multipoint Control Units (MCUs) while still enabling scalable group video and audio sessions. SFUs allow clients to subscribe to individual streams, supporting adaptive bitrate and simulcast strategies.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:selective-forwarding-unit", "labels": [ "Selective Forwarding Unit" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "self-attention", "title": "Self Attention", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An attention mechanism where every token in a sequence attends to every other token in the same sequence, computing query-key compatibility scores to produce context-aware weighted value representations that capture intra-sequence dependencies without recurrence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:self-attention", "labels": [ "Self Attention", "Masked Self Attention", "Masked Self-Attention", "Self-Attention", "Self-Attention Layer" ], "is_subclass_of": [ "Attention Mechanism" ], "wikilinks": [ "presentation", "Visionflow", "Human vs AI", "Knowhere", "MetaverseDomain", "Proprietary Large Language Models", "State Space and Other Approaches", "Transformers" ] }, { "id": "self-driving-car", "title": "Self Driving Car", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A self-driving car is an autonomous passenger vehicle capable of sensing its environment and operating with minimal or no human input, employing AI-driven perception, decision-making, and control systems to navigate roads, comply with traffic regulations, and transport occupants safely. Self-driving cars represent the consumer application of autonomous vehicle technology, typically targeting SAE Level 3\u20135 automation in urban and highway environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:self-driving-car", "labels": [ "Self Driving Car", "Self-Driving Car" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "ADAS", "Autonomous Vehicle", "MetaverseDomain", "Perception System" ] }, { "id": "self-improvement", "title": "Self Improvement", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The capacity of an AI system to autonomously modify its own architecture, weights, or learning algorithms to improve performance, generality, or efficiency without direct human intervention. Central to recursive self-improvement scenarios and AI safety research, raising fundamental questions about capability control, value alignment, and the conditions under which iterative self-modification remains aligned with intended objectives.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:self-improvement", "labels": [ "Self Improvement", "Self-Improvement" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "self-presence", "title": "Self Presence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The subjective sense that one's virtual or avatar representation in a mediated environment is a genuine extension of the physical self \u2014 encompassing body ownership, proprioceptive coherence, and self-location within virtual space. Self presence is a component of overall presence and a prerequisite for authentic immersive collaboration, as it governs how naturally users interact with and move through shared virtual environments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:self-presence", "labels": [ "Self Presence" ], "is_subclass_of": [ "Telepresence", "Presence" ], "wikilinks": [ "Avatar Psychology", "Presence Research", "Presence", "TelecollaborationDomain" ] }, { "id": "self-service-analytics", "title": "Self Service Analytics", "domain": "data", "domain_name": "Data", "definition": "Self-service analytics is an approach that empowers business users to explore data, build reports, and answer their own questions through governed, accessible tools without depending on specialist data teams for every request. It combines intuitive interfaces, curated and trustworthy data sources, and embedded governance so that broad access does not compromise consistency or security. The aim is to accelerate decision-making and foster data literacy across an organisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:self-service-analytics", "labels": [ "Self Service Analytics", "Self-Service Analytics" ], "is_subclass_of": [ "Data Analytics" ], "wikilinks": [] }, { "id": "self-sovereign-identity", "title": "Self Sovereign Identity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Self-Sovereign Identity (SSI) is an identity management paradigm \u2014 first articulated as a systematic framework by Christopher Allen in \"The Path to Self-Sovereign Identity\" (April 2016) \u2014 where individuals hold and control cryptographically verifiable credentials in their own Digital Wallet r...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:self-sovereign-identity", "labels": [ "Self Sovereign Identity", "BC-0188-self-sovereign-identity", "BC-0456-self-sovereign-identity", "Self-Sovereign Identity", "self-sovereign identity" ], "is_subclass_of": [ "Network Component", "Decentralised Identity", "Privacy-Enhancing Technology", "Digital Identity", "User-Centric Identity", "Cryptographic Protocol", "Trust Framework" ], "wikilinks": [ "AnonCreds", "BBS+ Signatures", "Bitstring Status List", "Blockchain Technology", "Camenisch-Lysyanskaya Signatures", "CBOR", "Centralised Identity", "Correlation Resistance", "Credential Definition", "Credential Schema", "Cross-Border Identity", "CryptographyDomain", "Decentralised Identity", "Decentralized Identity Foundation", "DID Document", "DIDComm", "DIDComm v2", "Digital Government", "eIDAS", "ECDH-1PU" ] }, { "id": "self-training", "title": "Self Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A semi-supervised learning technique where a model is iteratively improved by training on its own high-confidence predictions on unlabelled data. Self-training enables learning from large amounts of unlabelled data by using the model's own predictions as pseudo-labels.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:self-training", "labels": [ "Self Training", "Self-Training" ], "is_subclass_of": [ "AI Technique", "Semi-Supervised Learning" ], "wikilinks": [ "Education and AI", "MetaverseDomain" ] }, { "id": "self-consistency", "title": "self-consistency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Self-consistency is a decoding and prompting strategy for large language models in which multiple independent reasoning chains are sampled stochastically for a given problem and the final answer is selected by majority vote across those chains. The technique exploits the observation that correct reasoning paths, although varied in surface form, converge on the same answer whilst incorrect paths remain scattered across the answer space. Originally introduced alongside chain-of-thought prompting, self-consistency substantially improves accuracy on arithmetic, commonsense, and multi-step logical reasoning tasks without requiring additional model training or fine-tuning.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:self-consistency", "labels": [ "Self-Consistency", "Self-Consistency Decoding" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "self-custody", "title": "self-custody", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Self-custody is the security model and operational practice whereby an individual or entity retains exclusive, unmediated control of the private cryptographic keys that authorise transactions over their digital assets, entirely without delegating that custody to a centralised exchange, broker, or financial institution. The axiom underpinning self-custody \u2014 'not your keys, not your coins' \u2014 reflects the fact that whoever controls the private key controls the asset on-chain; counterparty risk from exchange insolvency, fraud, or regulatory asset freezes is therefore eliminated, but operational responsibility for key generation, backup, and signing security is transferred fully to the key holder. Self-custody is realised through software wallets (hot wallets), dedicated hardware security devices (cold wallets), multi-signature threshold schemes that distribute key material across several signatories, and advanced constructions such as threshold signature schemes (TSS) and social recovery wallets. Loss or compromise of the controlling private key or seed phrase results in permanent, irrecoverable loss of the associated assets with no recourse.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:self-custody", "labels": [ "Self-Custody", "Self Custody", "Self Custody Wallets", "Self-Custody Wallet" ], "is_subclass_of": [ "Key Management" ], "wikilinks": [] }, { "id": "self-liquidation", "title": "Self-Liquidation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Self-liquidation is a DeFi technique in which a borrower proactively closes their own under-collateralised or at-risk lending position \u2014 often using a flash loan to repay debt and withdraw collateral atomically \u2014 rather than waiting for a third-party liquidator to seize it at a penalty. It lets borrowers capture the collateral value that would otherwise be lost to liquidation fees. It is used on lending protocols such as Aave where liquidation penalties make self-initiated closure economically preferable.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:self-liquidation", "labels": [ "Self-Liquidation" ], "is_subclass_of": [ "Flash Loan" ], "wikilinks": [] }, { "id": "self-organisation", "title": "Self-Organisation", "domain": "ai", "domain_name": "Ai", "definition": "Self-organisation is the spontaneous emergence of order, structure, or pattern in a complex system without external direction, arising purely from local interactions among components. It operates through positive and negative feedback mechanisms that amplify or dampen perturbations, leading to globally coherent behaviour from decentralised, local rules.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:self-organisation", "labels": [ "Self-Organisation", "Self Organisation", "Self-Organization" ], "is_subclass_of": [ "Complex Systems" ], "wikilinks": [] }, { "id": "self-regulation", "title": "Self-Regulation", "domain": "governance", "domain_name": "Governance", "definition": "An arrangement in which an industry or professional community sets and enforces its own rules of conduct, typically through codes of practice, member oversight, and industry-led standards rather than direct state legislation; effectiveness depends on incentive alignment, credible enforcement, and public accountability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:self-regulation", "labels": [ "Self-Regulation", "Voluntary Self-Regulation" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [ "Accountability", "Compliance", "Transparency", "Regulatory Framework", "Governance Framework" ] }, { "id": "self-service-portal", "title": "Self-Service Portal", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A self-service portal is a customer-facing software interface that lets users resolve requests -- account changes, order tracking, troubleshooting -- without contacting a human agent, often backed by AI-driven search, chatbots or workflow automation. It reduces support-team load by handling routine, well-defined requests automatically while escalating complex cases to human agents. Self-service portals are a common deliverable of customer service and customer support automation initiatives.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:self-service-portal", "labels": [ "Self-Service Portal" ], "is_subclass_of": [ "User Interface" ], "wikilinks": [] }, { "id": "self-sovereign-identity-ssi", "title": "Self-Sovereign Identity (SSI)", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Self-Sovereign Identity (SSI) is an identity model in which individuals control their own digital identifiers and credentials without dependence on a centralised issuing authority, using Decentralised Identifiers (DIDs) anchored to distributed ledgers and Verifiable Credentials signed by trusted issuers to enable selective disclosure. SSI shifts identity management from siloed service-provider accounts to portable, user-held credentials that can be verified without querying the original issuer.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:self-sovereign-identity-ssi", "labels": [ "Self-Sovereign Identity (SSI)", "SSI Protocol" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [] }, { "id": "self-supervised-learning", "title": "Self-Supervised Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Self-Supervised Learning (SSL) is a machine learning paradigm in which a model learns rich representations of data by solving pretext tasks whose supervisory signal is derived automatically from the input data itself, requiring no human-provided labels. The model learns to predict masked or hidden portions of an input, to match different views of the same data, or to distinguish positive from negative data pairs, developing features that transfer effectively to downstream supervised tasks with limited labelled data. Self-supervised learning has become the dominant pre-training strategy for large language models, visual foundation models, and multimodal systems, enabling training at scales that would be infeasible with manually annotated datasets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:self-supervised-learning", "labels": [ "Self-Supervised Learning", "Self Supervised Learning", "SelfSupervisedLearning" ], "is_subclass_of": [ "Machine Learning Technique" ], "wikilinks": [] }, { "id": "selfish-mining", "title": "Selfish Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A strategic block-withholding attack in proof-of-work blockchains where a mining pool privately mines a secret chain and selectively publishes blocks to waste the computational work of honest miners, thereby earning a disproportionate share of block rewards relative to its hash-rate contribution. Selfish mining is profitable when the attacker controls more than ~33% of network hash-rate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:selfish-mining", "labels": [ "Selfish Mining" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Network Component", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "AI Agent System", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "semantic-html", "title": "Semantic HTML", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Semantic HTML is the practice of using HTML elements according to their intended meaning rather than for presentation alone. Elements such as header, nav, main, article, and button convey document structure and role to browsers and assistive technologies. It is a foundational requirement for accessible, machine-interpretable web content.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:semantic-html", "labels": [ "Semantic HTML" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "semantic-interoperability", "title": "semantic interoperability", "domain": "data", "domain_name": "Data", "definition": "Semantic interoperability is the capability of disparate information systems to exchange data with shared, unambiguous meaning by aligning their data models, ontologies, and controlled vocabularies at the level of semantics rather than syntax. Unlike syntactic interoperability, which concerns format agreement alone, semantic interoperability requires that the interpretation of exchanged values is consistent across system boundaries \u2014 typically enforced through shared formal ontologies expressed in OWL2, RDF-based linked data representations, and standardised schema mappings. It is a foundational property of federated knowledge graphs, cross-enterprise data sharing, healthcare information exchange, and machine-readable regulatory compliance. Achieving it demands governance as much as technology: communities must agree on canonical concept identifiers, term definitions, and relationship axioms before any technical implementation can succeed.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:semantic-interoperability", "labels": [ "Semantic Interoperability" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "semantic-mapping", "title": "Semantic Mapping", "domain": "ai", "domain_name": "Ai", "definition": "Semantic mapping is the process of establishing correspondences between concepts, terms, or schema elements across different knowledge representations so that their meanings align. It links source vocabularies to target ontologies, enabling data expressed under one model to be interpreted consistently under another. Semantic mapping underpins interoperability across heterogeneous knowledge graphs, databases, and linked-data sources.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:semantic-mapping", "labels": [ "Semantic Mapping" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "semantic-metadata-registry", "title": "Semantic Metadata Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A centralized catalog managing structured metadata schemas, controlled vocabularies, and semantic relationships to enable consistent asset description, cross-platform interoperability, and intelligent discovery.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:semantic-metadata-registry", "labels": [ "Semantic Metadata Registry" ], "is_subclass_of": [ "Data Management", "Metadata Registry" ], "wikilinks": [ "Asset Cataloging", "Linked Data Platform", "Metadata Management Infrastructure", "Metadata Schemas", "Namespace Management", "Ontology Repository", "RDF Store", "Relationship Mappings", "Schema Validator", "Schema Versioning", "Semantic Reasoning Engine", "Term Definitions", "W3C Semantic Web Standards", "Controlled Vocabularies", "CreativeMediaDomain", "Cross-Platform Interoperability", "Data Integration", "InfrastructureDomain", "MiddlewareLayer", "Semantic Search" ] }, { "id": "semantic-network", "title": "Semantic Network", "domain": "ai", "domain_name": "Ai", "definition": "A semantic network is a knowledge-representation structure that models concepts as nodes and their relationships as labelled edges. It captures meaning through typed links such as is-a and part-of, supporting inference, inheritance, and associative retrieval. It is a long-standing formalism underlying ontologies, knowledge graphs, and cognitive models of memory.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:semantic-network", "labels": [ "Semantic Network" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "semantic-parsing", "title": "Semantic Parsing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semantic parsing is the task of mapping natural-language utterances onto structured, machine-interpretable meaning representations such as logical forms, executable queries or programs. It converts ambiguous human language into precise formalisms that can be reasoned over or executed against a database or knowledge graph. Applications include question answering, text-to-SQL and instruction-to-code translation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:semantic-parsing", "labels": [ "Semantic Parsing" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "semantic-reasoning-engine", "title": "Semantic Reasoning Engine", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software that derives logically implied facts and checks consistency over a knowledge base expressed in a formal language such as OWL or RDF Schema. It applies inference rules of the underlying logic to make implicit information explicit.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:semantic-reasoning-engine", "labels": [ "Semantic Reasoning Engine", "Semantic Reasoning" ], "is_subclass_of": [ "Inference Engine" ], "wikilinks": [ "Description Logic", "OWL 2 Web Ontology Language", "Knowledge Graph", "Reasoning", "Knowledge Representation", "Inference Engine", "https://www.w3.org/TR/owl2-primer/" ] }, { "id": "semantic-reasoning", "title": "Semantic Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semantic reasoning is the process of deriving new facts or checking consistency by applying logical inference rules to formally represented knowledge, such as an ontology or knowledge graph, rather than to raw text or numeric data. It exploits explicit semantics, including class hierarchies and property definitions, to infer relationships not directly asserted, for example that an individual belongs to a class through transitive subsumption. Semantic reasoning underlies description-logic reasoners and is affected by whether a knowledge base assumes an open- or closed-world semantics.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:semantic-reasoning", "labels": [ "Semantic Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [] }, { "id": "semantic-scene-understanding", "title": "Semantic Scene Understanding", "domain": "ai", "domain_name": "Ai", "definition": "Semantic scene understanding is the computer-vision task of parsing a visual environment into labelled, structured representations of objects, surfaces, and their spatial and functional relationships. It goes beyond object detection by assigning meaning to regions, inferring affordances, and building a coherent model of the scene that downstream systems can reason over. It is foundational to spatial computing, where digital content must be anchored to real-world geometry and semantics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:semantic-scene-understanding", "labels": [ "Semantic Scene Understanding" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "semantic-search", "title": "Semantic Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semantic search is a retrieval paradigm that understands the meaning and intent of queries and documents rather than relying solely on lexical keyword overlap, deploying continuous vector representations of text\u2014produced by neural encoder models that compress sentences into dense embedding spaces...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:semantic-search", "labels": [ "Semantic Search" ], "is_subclass_of": [ "AI Application", "Search Technology", "Information Retrieval", "Natural Language Processing", "Neural Information Retrieval", "Knowledge Retrieval" ], "wikilinks": [ "AlgorithmLayer", "Approximate Nearest Neighbour", "BEIR Benchmark", "Bi-Encoder Architecture", "Biomedical Information Retrieval", "BM25", "Boolean Retrieval", "Chroma", "ColBERT Late Interaction", "Conversational Search", "Cosine Similarity", "Cross-Encoder Reranking", "Dense Retrieval", "Document Encoder", "Document Retrieval", "E-Commerce Search", "Embedding Model", "Enterprise Search", "Evaluation Benchmark", "Faiss" ] }, { "id": "semantic-segmentation", "title": "Semantic Segmentation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semantic Segmentation is the computer vision task of assigning a class label to every pixel in an image, partitioning the image into semantically meaningful regions without distinguishing between individual object instances. Architectures such as FCN, U-Net, and DeepLab produce dense pixel-wise predictions enabling scene understanding in autonomous driving, medical imaging, and satellite analysis.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:semantic-segmentation", "labels": [ "Semantic Segmentation", "SemanticSegmentation" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "costigan2018world", "Solid-Lite", "ArtificialIntelligenceDomain", "Computer Vision", "Instance Segmentation", "Nostr protocol", "Panoptic Segmentation", "Semantic Web", "semanticWeb", "Solid" ] }, { "id": "semantic-spatial-web-layer", "title": "Semantic Spatial Web Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Spatial Web is a computing paradigm that overlays machine-readable semantic information on physical locations and objects, enabling devices to understand and interact with the real world in context. It combines spatial computing, geospatial data, IoT sensors, and web standards to blur the boundary between physical and digital environments, and is regarded as a foundational layer for persistent augmented reality and metaverse experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:semantic-spatial-web-layer", "labels": [ "Semantic Spatial Web Layer", "Spatial Web" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "semantic-versioning", "title": "Semantic Versioning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Semantic Versioning (SemVer) is a versioning scheme that encodes compatibility information in a three-part MAJOR.MINOR.PATCH number: major increments signal breaking changes, minor increments add backward-compatible functionality, and patch increments fix bugs without changing the interface. It gives consumers of a package or API a predictable way to reason about upgrade risk. It is widely adopted across package registries, REST and gRPC API contracts, and machine learning model registries.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:semantic-versioning", "labels": [ "Semantic Versioning" ], "is_subclass_of": [ "Version Control" ], "wikilinks": [] }, { "id": "semantic-web-infrastructure", "title": "Semantic Web Infrastructure", "domain": "data", "domain_name": "Data", "definition": "Semantic web infrastructure is the stack of standards, vocabularies, and services that enable machine-readable, linked data on the web, including RDF, OWL, SPARQL endpoints, triple stores, and ontology registries. It provides the technical substrate for representing entities and relationships as interoperable graphs that agents can query and reason over. It is the backbone for knowledge graphs and linked-data ecosystems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:semantic-web-linked-data-standard-infrastructure", "labels": [ "Semantic Web Infrastructure" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "semantic-web-linked-data-standard", "title": "Semantic Web Linked Data Standard", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semantic Web extends the World Wide Web with machine-readable metadata, ontologies, and linked data to enable intelligent information discovery, integration, and reasoning. Core technologies include RDF, OWL, SPARQL, and knowledge graphs; AI techniques such as entity linking, relation extraction, ontology alignment, and automated reasoning further enhance these systems.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:semantic-web-linked-data-standard", "labels": [ "Semantic Web Linked Data Standard", "Semantic Web", "semanticWeb" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Information Extraction", "Linked Data", "Ontology", "Knowledge Graph" ] }, { "id": "semantic-web-standards", "title": "Semantic Web Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A set of W3C specifications including RDF (Resource Description Framework) and OWL (Web Ontology Language) that enable machine-readable data interchange, knowledge representation, and automated reasoning across distributed web applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:semantic-web-linked-data-standard-standards", "labels": [ "Semantic Web Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Web Standards" ], "wikilinks": [ "Data Interoperability", "metaverse", "Web Standards" ] }, { "id": "semantic-web", "title": "Semantic Web", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Semantic Web is an extension of the World Wide Web in which information is given well-defined, machine-readable meaning so that software agents can interpret, combine and reason over data published by different parties. It is realised through a stack of standards including RDF for data modelling, ontologies in OWL for shared vocabularies, and SPARQL for querying. The Semantic Web enables linked data, interoperability and automated inference across distributed sources.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:semantic-web", "labels": [ "Semantic Web" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "semi-fungible-token", "title": "Semi-Fungible Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain token standard that combines fungible and non-fungible properties, typically implemented under ERC-1155. Semi-fungible tokens represent classes of interchangeable assets (e.g. event tickets of the same tier) that may transition to uniquely non-fungible assets upon redemption or use, enabling a single contract to manage both fungible currencies and unique collectibles with reduced gas costs.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:semi-fungible-token", "labels": [ "Semi-Fungible Token" ], "is_subclass_of": [ "Token and Asset", "Token" ], "wikilinks": [ "Blockchain", "Token" ] }, { "id": "semi-supervised-learning", "title": "Semi-Supervised Learning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Semi-Supervised Learning is a artificial intelligence concept and a type of Machine Learning. that enables Data-Efficient Learning.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:semi-supervised-learning", "labels": [ "Semi-Supervised Learning" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "Data-Efficient Learning", "Machine Learning" ] }, { "id": "semi-major-axis", "title": "Semi-major Axis", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:semi-major-axis", "labels": [ "Semi-major Axis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "semiconductor-fabrication", "title": "Semiconductor Fabrication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Semiconductor fabrication is the industrial process by which integrated circuits and discrete semiconductor devices are constructed on crystalline silicon or compound semiconductor wafers through iterative cycles of deposition, photolithography, etching, doping, and planarisation. Each process generation, characterised by a technology node (e.g. 7 nm, 3 nm, 2 nm), defines achievable feature sizes and transistor densities, directly determining the computational density and power efficiency of the resulting chips. Fabrication occurs in ISO Class 1\u20135 cleanrooms to suppress particle contamination, with yield management and statistical process control being critical disciplines. The field bridges materials science, quantum mechanics, chemical engineering, and precision metrology to produce the foundational hardware layer for all digital computing, AI accelerators, communications infrastructure, and embedded systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:semiconductor-fabrication", "labels": [ "Semiconductor Fabrication", "Semiconductor Technology" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Computer Science", "Energy Consumption", "owl:Thing" ] }, { "id": "semiconductor-industry", "title": "Semiconductor Industry", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The global economic sector that designs, fabricates, packages, tests, and supplies integrated circuits and related components. It spans fabless design houses, foundries, integrated device manufacturers, equipment and materials suppliers, and EDA vendors in a deeply specialised international supply chain whose output underpins computing, communications, automotive, and defence systems, and whose leading-edge capacity has become a strategic chokepoint amid the accelerating hardware demands of artificial intelligence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:semiconductor-industry", "labels": [ "Semiconductor Industry" ], "is_subclass_of": [ "Technology Ecosystem" ], "wikilinks": [ "Technology Ecosystem", "Semiconductor", "Supply Chain", "AI Hardware", "ASIC", "Moore's Law" ] }, { "id": "semiconductor-manufacturing", "title": "Semiconductor Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Semiconductor manufacturing is the multi-stage industrial process that fabricates integrated circuits on silicon wafers through cycles of deposition, photolithographic patterning, etching, doping, and metallisation in cleanroom fabs. It transforms electronic designs into physical chips at nanometre feature sizes and is the foundation of all modern computing hardware. The capital-intensive supply chain spans wafer production, fabrication, packaging, and test.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:semiconductor-manufacturing", "labels": [ "Semiconductor Manufacturing" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "semiconductor-self-reliance", "title": "Semiconductor Self-Reliance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The strategic objective of a nation to develop domestic capabilities in semiconductor design and manufacturing to reduce or eliminate dependence on foreign supply chains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:semiconductor-self-reliance", "labels": [ "Semiconductor Self-Reliance" ], "is_subclass_of": [ "Semiconductor" ], "wikilinks": [] }, { "id": "semiconductor-supply-chain", "title": "Semiconductor Supply Chain", "domain": "governance", "domain_name": "Governance", "definition": "The semiconductor supply chain is the globally distributed network of design, fabrication, assembly, testing and distribution activities that produce integrated circuits. It is highly specialised and geographically concentrated, depending on leading-edge foundries, advanced lithography equipment, rare materials and complex logistics. Because chips underpin computing, AI and critical infrastructure, the semiconductor supply chain has become a focal point of geopolitics, export controls and national-security policy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:semiconductor-supply-chain", "labels": [ "Semiconductor Supply Chain" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [] }, { "id": "semiconductor", "title": "Semiconductor", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A semiconductor is a material, typically silicon, whose electrical conductivity lies between that of a conductor and an insulator and can be precisely controlled by doping, enabling the transistors that form the basis of modern electronics. Semiconductor devices are fabricated into integrated circuits through processes such as photolithography and etching, and underpin every class of computing hardware from CPUs to sensors. Global compute capacity, including GPU and CPU manufacturing, is bounded by semiconductor fabrication capacity and supply.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:semiconductor", "labels": [ "Semiconductor" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "sensitivity-analysis", "title": "Sensitivity Analysis", "domain": "data", "domain_name": "Data", "definition": "Sensitivity analysis is the study of how variation in the output of a model or calculation can be attributed to variation in its inputs and parameters. It quantifies which factors drive results, identifies fragile assumptions, and supports robustness checks across deterministic and probabilistic models. It is widely used in risk assessment, optimisation, and the calibration of analytics pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensitivity-analysis", "labels": [ "Sensitivity Analysis" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "sensitivity", "title": "Sensitivity", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Sensitivity quantifies the degree to which the output of a model, classifier, or system changes in response to variation in its inputs, parameters, or underlying assumptions. In binary classification, sensitivity (also called recall or true positive rate) measures the proportion of actual positive instances correctly identified by a model. In sensitivity analysis, the concept is generalised to any computational or physical system to identify which input factors most strongly drive output variance, thereby guiding uncertainty quantification and model validation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensitivity", "labels": [ "Sensitivity" ], "is_subclass_of": [ "Model Evaluation", "AI Evaluation" ], "wikilinks": [ "Parameter", "Robustness", "Reproducibility" ] }, { "id": "sensor-calibration", "title": "Sensor Calibration", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor calibration is the process of determining and correcting systematic measurement errors in sensing devices by comparing their outputs against known reference values or through geometric constraint solving, yielding intrinsic parameter models and extrinsic transformation matrices that map raw sensor readings to physically meaningful quantities. It is a prerequisite for sensor fusion, perception pipelines, and any application requiring quantitatively accurate environmental measurements.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:sensor-calibration", "labels": [ "Sensor Calibration", "Depth Sensor Calibration" ], "is_subclass_of": [ "Calibration" ], "wikilinks": [] }, { "id": "sensor-data", "title": "sensor data", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Sensor data comprises the raw or pre-processed digital measurements produced by physical transducers\u2014including cameras, LiDAR scanners, inertial measurement units (IMUs), ultrasonic rangers, and microphones\u2014that encode observable properties of the environment such as geometry, colour, acceleration, and sound. In robotic, autonomous, and spatial computing systems, sensor data forms the primary input to perception pipelines responsible for state estimation, object detection, and scene understanding. Data quality characteristics\u2014including frame rate, resolution, noise floor, and synchronisation latency\u2014directly constrain the capabilities of downstream algorithms such as SLAM, sensor fusion, and learned perception models. Sensor data is collected, timestamped, and transmitted through data acquisition systems before being processed through calibration, fusion, and inference stages to produce actionable world representations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:sensor-data", "labels": [ "Sensor Data", "IoT Sensor Data", "Raw Sensor Data", "Sensor Data Stream", "SensorData" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "sensor-feedback", "title": "Sensor Feedback", "domain": "robotics", "domain_name": "Robotics", "definition": "Measurement data returned from sensors into a control loop so that a controller can compare the actual state of a dynamic system against its desired state and compute corrective action. Sensor feedback closes the loop in control engineering and robotics: encoders, IMUs, force-torque sensors, thermocouples, and cameras report position, velocity, force, temperature, or pose, and the resulting error signal drives actuator commands, with latency, noise, and sampling rate directly bounding achievable control performance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:sensor-feedback", "labels": [ "Sensor Feedback" ], "is_subclass_of": [ "Sensor Data" ], "wikilinks": [ "Sensor Data", "Feedback Loop", "Control Algorithm" ] }, { "id": "sensor-footprint", "title": "Sensor Footprint", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sensor-footprint", "labels": [ "Sensor Footprint" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sensor-fusion-layer", "title": "Sensor Fusion Layer", "domain": "robotics", "domain_name": "Robotics", "definition": "The Sensor Fusion Layer is the stratum that combines data from multiple sensors into a single, more reliable estimate of state. It sits above the Hardware sensing devices and below the Perception Layer that interprets the fused result. It contains alignment, filtering, and fusion algorithms together with uncertainty models.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:sensor-fusion-layer", "labels": [ "Sensor Fusion Layer" ], "is_subclass_of": [ "Perception and Sensing", "owl:Thing" ], "wikilinks": [ "Hardware Layer", "Perception Layer", "Kalman Filter", "Bayesian Inference", "owl:Thing" ] }, { "id": "sensor-fusion", "title": "Sensor Fusion", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor Fusion is the process of combining data from multiple sensors (camera, lidar, radar, GPS, IMU) to produce more accurate, reliable, and complete information than could be obtained from any individual sensor. It employs algorithms including Kalman filtering, particle filtering, and deep learning-based fusion to integrate complementary sensor modalities whilst managing noise, uncertainties, and hardware failures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sensor-fusion", "labels": [ "Sensor Fusion", "IMU Sensor Fusion", "Sensor Data Fusion", "SensorFusion" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "Localisation", "3D and 4D", "AI Video", "copyright", "Microsoft Copilot", "Music and Audio", "Object Detection", "Perception System", "robotics", "RoboticsDomain", "Stable Diffusion Image Model", "Update Cycle" ] }, { "id": "sensor-housing", "title": "Sensor Housing", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor housing is the mechanical enclosure that protects a sensing element from environmental hazards such as moisture, dust, vibration, and electromagnetic interference while maintaining its measurement window. It defines mounting geometry, ingress protection rating, thermal management, and the optical or acoustic path to the sensed medium. It is a structural component of robotic and exteroceptive sensing assemblies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensor-housing", "labels": [ "Sensor Housing" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "sensor-input", "title": "Sensor Input", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor input refers to the raw or pre-processed signals acquired from physical or virtual transducers that convert observable environmental quantities\u2014position, force, temperature, light intensity, pressure, chemical concentration\u2014into electrical or digital representations suitable for computation. It constitutes the primary interface between an autonomous or intelligent system and its environment, providing the perceptual foundation for state estimation, feedback control, and situational awareness. Effective use of sensor input requires calibration, filtering, and often fusion with complementary modalities to produce reliable world-state estimates, and the quality and latency of sensor inputs directly constrain the operational envelope of any system that must perceive and respond to the physical world. Sensor inputs vary by modality, sampling rate, resolution, noise characteristics, and temporal precision, spanning domains from robotics and autonomous vehicles to spatial computing, industrial IoT, and embodied AI.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensor-input", "labels": [ "Sensor Input", "Input Sensors" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "sensor-intercalibration", "title": "Sensor Intercalibration", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sensor-intercalibration", "labels": [ "Sensor Intercalibration" ], "is_subclass_of": [ "Calibration" ], "wikilinks": [] }, { "id": "sensor-interface", "title": "Sensor Interface", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor interface is the hardware and protocol boundary through which a controller acquires data from a sensor, encompassing signal conditioning, analogue-to-digital conversion, and bus protocols such as I2C, SPI, CAN, or analogue voltage lines. It standardises timing, addressing, and electrical levels so that heterogeneous sensors can be integrated into a control loop. It is a core subsystem of embedded and robotic control architectures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensor-interface", "labels": [ "Sensor Interface" ], "is_subclass_of": [ "Control System" ], "wikilinks": [] }, { "id": "sensor-measurements", "title": "Sensor Measurements", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor measurements are the time-stamped observations produced by physical or virtual sensors, each carrying a value, a unit, and an associated uncertainty or noise model. In probabilistic robotics they form the observation stream that estimators condition on to infer latent state such as pose or velocity. Their statistical characterisation is essential for filtering, fusion, and localisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensor-measurements", "labels": [ "Sensor Measurements" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "sensor-model", "title": "Sensor Model", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor model, also called an observation or measurement model, is a probabilistic description of how a robot's sensor readings relate to the underlying state of the world, expressing the likelihood of an observation given a hypothesised state. It captures sensor characteristics such as noise, resolution, range limits, and failure modes, allowing a robot to weight evidence appropriately when fusing measurements. Sensor models are central to Bayesian state estimation, where they form the update step that corrects predictions using incoming data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sensor-model", "labels": [ "Sensor Model" ], "is_subclass_of": [ "Probabilistic Robotics", "Perception and Sensing" ], "wikilinks": [] }, { "id": "sensor-networks", "title": "Sensor Networks", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor networks are distributed collections of sensor nodes that collect, process and communicate measurements about their environment.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:sensor-networks", "labels": [ "Sensor Networks", "Sensor Network" ], "is_subclass_of": [ "Sensors", "Perception and Sensing" ], "wikilinks": [ "Sensor Fusion", "Distributed Systems", "Sensors", "https://en.wikipedia.org/wiki/Wireless_sensor_network", "https://www.nist.gov/" ] }, { "id": "sensor-saturation", "title": "Sensor Saturation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sensor-saturation", "labels": [ "Sensor Saturation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sensor-suite", "title": "Sensor Suite", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor suite is the integrated collection of complementary sensors mounted on a robotic platform, such as cameras, LiDAR, radar, IMUs, and encoders, chosen to provide redundant and complementary perception. The combination supports sensor fusion that compensates for the failure modes and blind spots of any single modality. It is a defining hardware subsystem of autonomous mobile robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensor-suite", "labels": [ "Sensor Suite", "Autonomous Vehicle Sensor Suite", "Robot Sensor Suite", "Robotic Sensor Suite" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "sensor-system", "title": "Sensor System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A sensor system is an integrated assembly of physical transducers, signal conditioning hardware, analogue-to-digital conversion, and data management software that captures physical-world phenomena and transforms them into structured digital representations suitable for processing, analysis, or control. Sensor systems range from single-chip inertial measurement units to multi-modal perception arrays combining cameras, LiDAR, RADAR, and acoustic sensors. Calibration, synchronisation, and fusion across heterogeneous modalities are core engineering concerns. Sensor systems are foundational to robotics, autonomous vehicles, industrial IoT, and environmental monitoring.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sensor-system", "labels": [ "Sensor System" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [] }, { "id": "sensor-technology", "title": "Sensor Technology", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensor Technology encompasses the diverse hardware devices and fusion algorithms that capture physical-world data for use in robotics, spatial computing, and XR systems. This includes inertial measurement units, LiDAR scanners, depth cameras, eye-tracking, haptic sensors, and biometric devices, combined with fusion techniques such as visual-inertial odometry and Kalman filtering to produce accurate, robust state estimates for autonomous systems and immersive experiences.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:sensor-technology", "labels": [ "Sensor Technology" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "sensor", "title": "Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor is a transducer or measurement device that detects, converts, and quantifies a physical, chemical, or environmental stimulus \u2014 such as light, pressure, temperature, motion, or electromagnetic fields \u2014 into an electrical signal suitable for processing, storage, or actuation. Sensors form the perceptual interface between computational systems and the physical world, enabling autonomous robots, IoT devices, spacecraft, and industrial machinery to react to real-world conditions. They are characterised by key metrology attributes including sensitivity, resolution, dynamic range, linearity, bandwidth, and noise floor. Modern sensor fusion architectures combine heterogeneous sensor streams \u2014 e.g. [[LiDAR]], [[Camera]], and [[IMU]] \u2014 using probabilistic filters to produce robust, high-fidelity world models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:sensor", "labels": [ "Sensor", "Multispectral Sensor", "Sensor Transducer", "Sensor Validation", "Soft Sensor" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [ "Robotics Systems" ] }, { "id": "sensors", "title": "Sensors", "domain": "robotics", "domain_name": "Robotics", "definition": "Sensors are devices that transduce physical quantities\u2014distance, force, temperature, orientation, light, electromagnetic fields\u2014into electrical or digital signals that a robot or computational system can process, providing the perceptual input for state estimation, control, and decision making.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensors", "labels": [ "Sensors", "Gz Sensors", "Motion Sensors", "Visual Sensors" ], "is_subclass_of": [ "Robotics", "Sensor Technology", "Robotics Domain" ], "wikilinks": [ "Perception System", "Sensor Fusion", "Lidar", "Radar", "Robotics Domain" ] }, { "id": "sensory-feedback", "title": "Sensory Feedback", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Sensory feedback is the delivery of stimuli to a user's senses, including visual, auditory, haptic, and proprioceptive channels, in response to their actions within a virtual or mixed environment. Timely and coherent feedback closes the perception-action loop and is a precondition for the user's sense of presence and emotional engagement. It is a core mechanism of immersive interaction design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sensory-feedback", "labels": [ "Sensory Feedback" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "sensory-immersion", "title": "Sensory Immersion", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Sensory immersion is the dimension of immersion produced by saturating a user's perceptual channels with synthetic stimuli, such that the virtual environment dominates their awareness over the physical surroundings. It is driven by display fidelity, spatial audio, haptics, and wide field of view, and is distinct from narrative or cognitive immersion. It is a constituent component of the broader experience of immersion.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sensory-immersion", "labels": [ "Sensory Immersion" ], "is_subclass_of": [ "Virtual Environment" ], "wikilinks": [] }, { "id": "sentence-piece", "title": "SentencePiece", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "SentencePiece is a language-independent subword tokenisation library that processes raw Unicode text without language-specific pre-tokenisation, learning vocabulary units via Byte-Pair Encoding or the Unigram Language Model directly from corpora. It produces fully reversible, fixed-vocabulary tokenisations widely used in multilingual large language models such as T5, mT5, and ALBERT, and is particularly valuable for languages lacking explicit word boundaries.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:sentence-piece", "labels": [ "SentencePiece" ], "is_subclass_of": [ "AI Technique", "Neural Network Text Tokenisation" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "sentiment-analysis", "title": "Sentiment Analysis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Sentiment Analysis is the NLP task of determining the emotional tone, attitude, or opinion expressed in text, classifying content as positive, negative, or neutral, and extracting fine-grained emotional attributes including aspect-level sentiment, emotion detection, and subjectivity classification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sentiment-analysis", "labels": [ "Sentiment Analysis", "Sentiment Classification Model", "Sentiment Tracking" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Opinion Mining", "Text Classification", "Landscape", "MetaverseDomain", "Natural Language Processing" ] }, { "id": "seoul-declaration", "title": "Seoul Declaration", "domain": "governance", "domain_name": "Governance", "definition": "The Seoul Declaration is an international AI-safety commitment adopted at the 2024 AI Seoul Summit, in which participating states and companies affirmed principles of safe, innovative, and inclusive AI and pledged cooperation on frontier-model risk. It built on the Bletchley Declaration and advanced commitments to risk thresholds and safety frameworks for advanced AI systems. It is a governance instrument shaping Asia-Pacific and global AI regulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:seoul-declaration", "labels": [ "Seoul Declaration" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "separation-of-duties", "title": "Separation Of Duties", "domain": "security", "domain_name": "Security", "definition": "Separation of duties is a control principle that divides a sensitive task among multiple people or roles so that no single individual can complete it alone, reducing the risk of fraud, error and abuse of privilege. It requires collusion to subvert controls and is foundational to access governance, financial controls and compliance regimes. It complements least-privilege and is operationalised through role-based access control.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:separation-of-duties", "labels": [ "Separation Of Duties", "Separation of Duties" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "separation-of-powers", "title": "Separation Of Powers", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Governance design patterns in decentralized autonomous organizations that distribute authority across distinct functional roles, preventing concentration of control through checks and balances between proposal creation, voting, execution, and oversight functions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:separation-of-powers", "labels": [ "Separation Of Powers" ], "is_subclass_of": [ "Governance and Safety", "DAO Governance" ], "wikilinks": [ "Decentralized Control", "Blockchain", "DAO Governance", "metaverse" ] }, { "id": "separation-of-concerns", "title": "Separation of Concerns", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A foundational design principle holding that a system should be decomposed so that each part addresses a single, distinct concern \u2014 one aspect of functionality or one axis of decision-making \u2014 with minimal overlap between parts. Articulated by Dijkstra in 1974, it underlies modularity, layering, encapsulation, and interface design: by isolating concerns behind boundaries, changes to one concern can be made, understood, tested, and reused without cascading through the rest of the system.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:separation-of-concerns", "labels": [ "Separation of Concerns" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [ "Software Architecture", "Component", "OSI Model", "Hardware Abstraction Layer", "Interoperability" ] }, { "id": "sequence-labelling", "title": "Sequence Labelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Sequence labelling is a class of natural-language-processing tasks in which each element of an input sequence is assigned a categorical label from a fixed tag set. It encompasses tasks such as named-entity recognition, part-of-speech tagging and slot filling, where contextual dependencies between adjacent tokens matter. Classical approaches use hidden Markov models and conditional random fields, while modern systems use neural encoders.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sequence-labelling", "labels": [ "Sequence Labelling" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "sequence-model", "title": "Sequence Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Any machine learning model that captures dependencies across ordered data \u2014 text, speech, video frames, genomic strings, time series \u2014 by assigning probabilities to sequences or mapping input sequences to outputs; the family spans n-gram models and hidden Markov models through recurrent networks (LSTM, GRU) to transformers and modern state space models, and underlies language modelling, speech recognition, machine translation and time-series forecasting.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:sequence-model", "labels": [ "Sequence Model" ], "is_subclass_of": [ "Machine Learning Model" ], "wikilinks": [ "Machine Learning Model", "Hidden Markov Model", "Recurrent Neural Network", "Transformer", "Beam Search" ] }, { "id": "sequence-to-sequence-learning", "title": "Sequence To Sequence Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Sequence-to-sequence learning is a neural modelling framework that maps an input sequence to an output sequence of possibly different length, using an encoder to compress the input into a context representation and a decoder to generate the output token by token. Originally built on recurrent networks, it now predominantly uses the attention-based transformer architecture. It is the dominant paradigm for machine translation, summarisation and other transduction tasks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sequence-to-sequence-learning", "labels": [ "Sequence To Sequence Learning", "Sequence-to-Sequence Learning" ], "is_subclass_of": [ "Supervised Learning", "AI Technique" ], "wikilinks": [] }, { "id": "sequence-to-sequence-model", "title": "Sequence To Sequence Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A sequence-to-sequence model is a neural architecture that maps an input sequence of arbitrary length to an output sequence of arbitrary length, classically using an encoder to compress the input into a context representation and a decoder to generate the output one element at a time. Originally built from recurrent networks such as LSTMs and GRUs and later augmented with attention to overcome the fixed-context bottleneck, the paradigm became the foundation for the transformer. Sequence-to-sequence models power machine translation, text summarisation, speech recognition, and other tasks where input and output structures differ.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sequence-to-sequence-model", "labels": [ "Sequence To Sequence Model", "Sequence-to-Sequence Model" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "sequencer", "title": "Sequencer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A sequencer is a component in a blockchain rollup or layer-2 system that receives user transactions, orders them into a canonical sequence, and produces blocks or batches for execution and settlement. By fixing transaction order off-chain before posting to the base layer, the sequencer enables fast confirmations and low fees. Sequencer design directly governs liveness, fairness, and the centralisation risk of a rollup.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:sequencer", "labels": [ "Sequencer" ], "is_subclass_of": [ "Layer 2 Scaling" ], "wikilinks": [] }, { "id": "sequential-monte-carlo", "title": "Sequential Monte Carlo", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Sequential Monte Carlo is a family of methods that approximate evolving probability distributions using a set of weighted samples updated recursively as new observations arrive.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sequential-monte-carlo", "labels": [ "Sequential Monte Carlo" ], "is_subclass_of": [ "Numerical Methods" ], "wikilinks": [ "Monte Carlo Integration", "Particle Filter", "Bayesian Inference", "Importance Sampling", "Numerical Methods" ] }, { "id": "serialisation-format", "title": "Serialisation Format", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A serialisation format is a specification for encoding in-memory data structures into a byte or text stream that can be persisted or transmitted and later reconstructed. Formats differ in schema rigidity, compactness, speed, and cross-language support, spanning text formats like JSON and YAML and binary formats like Protocol Buffers, Avro, and Parquet. It is fundamental to checkpointing, messaging, and distributed computation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:serialisation-format", "labels": [ "Serialisation Format", "Crate Format", "Data Serialisation Format", "Serialisation System", "Serialization Format" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "serialisation", "title": "Serialisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Serialisation is the process of converting in-memory data structures or object graphs into a linear byte sequence or text representation that can be stored, transmitted, and later reconstructed. The inverse operation, deserialisation, rebuilds the original structure from the encoded form. Serialisation underpins persistence, inter-process and network communication, and the interoperability of systems that exchange structured data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:serialisation", "labels": [ "Serialisation" ], "is_subclass_of": [ "Data Format Standard" ], "wikilinks": [] }, { "id": "series-elastic-actuation", "title": "series elastic actuation", "domain": "robotics", "domain_name": "Robotics", "definition": "Series Elastic Actuation (SEA) is a robotic actuation paradigm in which a calibrated compliant element \u2014 typically a torsional or linear spring \u2014 is interposed in series between a motor-gearbox drive train and the output link of a robot joint. The spring deflection under load provides an indirect torque measurement via Hooke's Law, enabling high-fidelity closed-loop torque control without the complexity of strain-gauge transducers. This intrinsic mechanical compliance attenuates shock loads, lowers the reflected inertia experienced during unintended contact, and passively limits peak interaction forces \u2014 properties that are essential for safe physical human\u2013robot interaction. SEA is a foundational technology in legged locomotion systems, rehabilitation exoskeletons, prosthetic limbs, and compliant collaborative robots, and stands as the seminal example of intentional mechanical compliance in actuation design.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:series-elastic-actuation", "labels": [ "Series Elastic Actuation" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "series-elastic-actuator", "title": "Series Elastic Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A Series Elastic Actuator (SEA) is a robotic joint mechanism that interposes a compliant spring element in series between the gearbox output and the load, deliberately introducing controlled elasticity into the drivetrain. By measuring the spring deflection with a position sensor, the SEA provides accurate, low-noise torque measurement without a dedicated force-torque sensor, while simultaneously providing passive mechanical compliance that absorbs impact energy and reduces stiffness at the point of contact. This architecture enables safe, backdrivable interactions between robots and humans or unstructured environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:series-elastic-actuator", "labels": [ "Series Elastic Actuator" ], "is_subclass_of": [ "Robot Actuator" ], "wikilinks": [] }, { "id": "serious-incident", "title": "Serious Incident", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Under EU AI Act Article 3(44), a Serious Incident is any incident or malfunctioning of an AI system that directly or indirectly causes death, serious health damage, serious disruption of critical infrastructure, or serious infringement of fundamental rights. Such incidents trigger mandatory reporting obligations for providers and deployers of high-risk AI systems and GPAI models with systemic risk, enabling rapid regulatory response and market surveillance under Article 73.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:serious-incident", "labels": [ "Serious Incident" ], "is_subclass_of": [ "Governance and Safety", "Spatial Embodiment Harm Taxonomy" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "server-push", "title": "Server Push", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Server push is an HTTP/2 mechanism that allows a server to proactively send resources to a client before the client explicitly requests them, anticipating what a page will need next. It reduces round-trip latency by eliminating the request phase for predictable secondary resources such as stylesheets or scripts. The related technique of Server-Sent Events uses a similar unidirectional push model over a persistent connection to stream updates to a client.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:server-push", "labels": [ "Server Push" ], "is_subclass_of": [ "Http2" ], "wikilinks": [] }, { "id": "server-sent-events", "title": "server-sent events", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Server-Sent Events (SSE) is a W3C and WHATWG-standardised unidirectional server-to-client push protocol layered atop HTTP/1.1 and HTTP/2, enabling servers to emit a continuous stream of newline-delimited text events to browser or API clients over a single persistent connection. Each event record may carry optional 'id', 'event', 'data', and 'retry' fields; clients reconnect automatically using the last-received event id as a cursor. SSE is consumed via the browser's EventSource API and is the dominant transport for streaming large-language-model inference outputs in APIs such as OpenAI and Anthropic, as well as in the Model Context Protocol (MCP) transport layer.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:server-sent-events", "labels": [ "Server-Sent Events", "Server-Sent Push" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "server", "title": "Server", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A computer system engineered to provide services \u2014 computation, storage, or content \u2014 to other machines over a network, typically built for sustained load, remote management, and high availability with error-correcting memory, redundant power supplies, and hot-swappable storage. Racked in their thousands inside data centres, servers are the physical substrate of cloud computing, and the term equally names the software process that answers client requests in the client-server model.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:server", "labels": [ "Server" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [ "Hardware", "Data Centre", "Computing Infrastructure", "Power Supply", "Operating System", "Cloud Computing", "Networking" ] }, { "id": "serverless-architecture", "title": "Serverless Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Serverless Architecture is a cloud execution model in which application logic is deployed as discrete, stateless functions that are provisioned and scaled automatically by the cloud provider in response to events, with billing proportional to actual execution time. By abstracting away server provisioning and capacity management, serverless architectures reduce operational overhead and enable fine-grained, event-driven compute patterns suited to metaverse backends, AI inference endpoints, and IoT data pipelines.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:serverless-architecture", "labels": [ "Serverless Architecture", "Serverless Computing", "Serverless Edge Compute" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "service-design", "title": "Service Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Service design is the activity of planning and orchestrating the people, infrastructure, communication, and touchpoints of a service to improve its quality and the interaction between a provider and its users. It takes a holistic, end-to-end view that spans front-stage customer experiences and the back-stage processes and systems that enable them. Service designers use tools such as journey maps and service blueprints to align organisational capabilities with user needs across every channel.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:service-design", "labels": [ "Service Design" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [] }, { "id": "service-discovery", "title": "Service Discovery", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Service discovery is the runtime mechanism by which distributed applications automatically locate the network endpoints of services they depend on, without relying on hardcoded addresses or manual configuration. It operates through a service registry that maintains a dynamic catalogue of available service instances alongside their health status, enabling clients to resolve service names to live endpoints at query time. Discovery patterns divide into client-side discovery, where the consumer queries the registry directly and selects from returned healthy instances, and server-side discovery, where a load balancer or API gateway mediates resolution transparently. It is a foundational pattern in microservices and cloud-native architectures, enabling dynamic scaling, zero-downtime rolling updates, and fault-tolerant inter-service communication.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:service-discovery", "labels": [ "Service Discovery" ], "is_subclass_of": [ "Distributed System Architecture" ], "wikilinks": [] }, { "id": "service-endpoint", "title": "Service Endpoint", "domain": "security", "domain_name": "Security", "definition": "A Service Endpoint is a network-addressable location, typically expressed as a URL or URI, at which a service exposes its functionality to clients. It defines where requests are sent and, together with a protocol and interface contract, how interactions are framed and authenticated. Service endpoints are central to APIs, microservices and decentralised-identity documents, where they advertise the reachable interfaces associated with an identity or capability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:service-endpoint", "labels": [ "Service Endpoint" ], "is_subclass_of": [ "API" ], "wikilinks": [] }, { "id": "service-integration", "title": "Service Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Service integration is the practice of connecting independently deployed software services so that they can exchange data and coordinate behaviour as a coherent system. It is commonly achieved through APIs, message brokers, and middleware layers that translate between differing protocols and data formats. Effective service integration allows organisations to compose new capabilities from existing systems without rebuilding them.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:service-integration", "labels": [ "Service Integration" ], "is_subclass_of": [ "System Integration" ], "wikilinks": [] }, { "id": "service-layer", "title": "Service Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Collection of reusable services exposed via APIs for identity, assets, physics, and analytics that enable application functionality and interoperability in virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:service-layer", "labels": [ "Service Layer", "ServiceLayer" ], "is_subclass_of": [ "Platform and Environment", "Spatial Computing" ], "wikilinks": [ "Analytics Service", "API Integration", "API Management", "Asset Service", "Data Models", "Database Systems", "EWG/MSF Taxonomy", "Identity Service", "Message Queue", "Physics Service", "Service Composition", "Service Discovery", "Service Orchestration", "API Gateway", "Data Layer", "InfrastructureDomain", "Microservices Architecture", "Middleware Layer" ] }, { "id": "service-level-agreement", "title": "Service Level Agreement", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A service level agreement (SLA) is a formal contract between a service provider and its customers that defines the measurable level of service to be delivered, including availability, performance, and response-time targets, together with the remedies or credits owed when those targets are breached. SLAs translate abstract reliability expectations into quantitative service level objectives and indicators that can be monitored and enforced. They are central to cloud computing, managed services, and outsourcing, providing the accountability framework around which capacity, support, and operational practices are organised.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:service-level-agreement", "labels": [ "Service Level Agreement" ], "is_subclass_of": [ "Infrastructure", "High Availability" ], "wikilinks": [] }, { "id": "service-level-objective", "title": "Service Level Objective", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A service level objective (SLO) is a target value or range for a measurable property of a service, such as availability or latency, that defines the acceptable level of reliability over a stated time window. SLOs are expressed against service level indicators and provide the quantitative basis for engineering decisions, error budgets and the contractual commitments of service level agreements. They are a cornerstone practice of site reliability engineering, balancing reliability against the pace of change.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:service-level-objective", "labels": [ "Service Level Objective", "Service Level Objectives" ], "is_subclass_of": [ "Site Reliability Engineering" ], "wikilinks": [] }, { "id": "service-mesh", "title": "Service Mesh", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A service mesh is a dedicated infrastructure layer for managing service-to-service communication within a microservices architecture, providing traffic management, mutual TLS encryption, observability, and policy enforcement transparently to application code through sidecar proxies or eBPF-based data planes. It decouples operational concerns\u2014load balancing, retries, circuit breaking, telemetry\u2014from business logic, enabling consistent reliability and security across heterogeneous services.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:service-mesh", "labels": [ "Service Mesh" ], "is_subclass_of": [ "Microservices Architecture" ], "wikilinks": [] }, { "id": "service-oriented-architecture", "title": "Service Oriented Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An architectural style organizing software as loosely coupled, interoperable services that communicate through standardized interfaces, enabling modular composition of metaverse functionality through reusable components and well-defined contracts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:service-oriented-architecture", "labels": [ "Service Oriented Architecture", "Service-Based Architecture", "Service-Oriented Architecture" ], "is_subclass_of": [ "Platform and Environment", "Software Architecture" ], "wikilinks": [ "metaverse", "Software Architecture", "System Interoperability" ] }, { "id": "service-registry", "title": "Service Registry", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A Service Registry is a database of available service instances and their network locations that underpins dynamic service discovery in distributed and microservice architectures. Instances register on startup and deregister on shutdown, while clients or load balancers query the registry to resolve a logical service name to a healthy endpoint. Health checks expire stale entries, keeping the registry an accurate view of live topology in elastic, frequently changing deployments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:service-registry", "labels": [ "Service Registry" ], "is_subclass_of": [ "Service Discovery" ], "wikilinks": [] }, { "id": "service-robot", "title": "Service Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Service Robot is an autonomous or semi-autonomous robotic system designed to perform tasks for humans in non-industrial settings such as healthcare, logistics, hospitality, and domestic environments. Unlike fixed industrial robots, service robots operate in dynamic, unstructured spaces and must navigate safely around people, relying on sensor fusion, motion planning, and human-robot interaction capabilities.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:service-robot", "labels": [ "Service Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "service-robotics", "title": "Service Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Service robotics is the branch of robotics concerned with robots that perform useful tasks for humans outside industrial manufacturing, including cleaning, logistics, hospitality, healthcare, and domestic assistance. These systems emphasise autonomous navigation in unstructured human environments, safe interaction, and task planning under uncertainty. It is a rapidly growing application domain of mobile and collaborative robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:service-robotics", "labels": [ "Service Robotics" ], "is_subclass_of": [ "Robotics", "Service Robot" ], "wikilinks": [] }, { "id": "service-now", "title": "ServiceNow", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "ServiceNow is a cloud platform for IT service management and workflow automation, delivering digital workflows across IT, security and business operations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:service-now", "labels": [ "ServiceNow" ], "is_subclass_of": [ "Workflow Automation" ], "wikilinks": [ "REST", "Threat Intelligence", "Database Systems", "Workflow Automation", "https://www.servicenow.com/", "https://docs.servicenow.com/" ] }, { "id": "servo-control", "title": "Servo Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Servo Control is a closed-loop control methodology that uses feedback signals \u2014 typically from encoders or resolvers \u2014 to precisely regulate the position, velocity, or torque of an actuator. A servo controller computes the error between a desired setpoint and the measured output, then drives a servo motor or hydraulic actuator to minimise that error, making servo control foundational to high-precision robotic joint control, CNC machining, and collaborative robot safety systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:servo-control", "labels": [ "Servo Control", "Servo Controller", "ServoControl" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "servo-drive", "title": "Servo Drive", "domain": "robotics", "domain_name": "Robotics", "definition": "A servo drive is the power-electronic controller that regulates the current, velocity, and position of a servomotor by closing feedback loops on encoder or resolver signals. It converts commanded setpoints from a motion controller into precisely modulated motor currents, enabling high-bandwidth, high-accuracy actuation. It is a core component of industrial robot joints and CNC motion systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:servo-drive", "labels": [ "Servo Drive" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "servo-motor", "title": "Servo Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "A servo motor is a closed-loop electromechanical actuator that couples an electric motor with a feedback sensor, typically a rotary encoder, and a controller to deliver precise position, velocity, and torque control. Servo motors drive robot joints, manipulator arms, and CNC machinery, and other applications demanding accurate, repeatable motion under varying load.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:servo-motor", "labels": [ "Servo Motor" ], "is_subclass_of": [ "Actuation and Control", "Electric Motor" ], "wikilinks": [ "Electric Motor", "Robotics" ] }, { "id": "servo-valve", "title": "Servo Valve", "domain": "robotics", "domain_name": "Robotics", "definition": "A servo valve is an electrohydraulic device that proportionally controls the flow and pressure of hydraulic fluid in response to a low-power electrical command, typically via a torque-motor-driven pilot stage. It provides high-bandwidth, closed-loop modulation of hydraulic actuators, delivering large forces with precise position control. It is a key element of high-power robotic and aerospace actuation systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:servo-valve", "labels": [ "Servo Valve", "Electrohydraulic Servo Valve" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "session-initiation-protocol", "title": "Session Initiation Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Session Initiation Protocol (SIP) is an application-layer signalling protocol used to establish, modify and terminate real-time communication sessions such as voice and video calls over IP networks. It handles user location, session setup negotiation and call control, delegating the actual media transport to companion protocols. SIP is text-based and request-response oriented, modelled on HTTP, and underpins much of modern internet telephony and unified communications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:session-initiation-protocol", "labels": [ "Session Initiation Protocol" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "session-key", "title": "Session Key", "domain": "security", "domain_name": "Security", "definition": "A session key is a temporary symmetric cryptographic key generated for the duration of a single communication session, used to encrypt the data exchanged between parties after an initial key exchange handshake. Session keys have bounded lifetimes and are discarded at the end of the session, limiting the window of vulnerability if the key is compromised. Ephemeral session keys derived from asymmetric key exchanges (such as Diffie-Hellman) provide forward secrecy, ensuring past sessions remain private even if long-term private keys are later exposed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:session-key", "labels": [ "Session Key" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "session-management", "title": "Session Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Session management is the set of mechanisms by which a networked system creates, maintains, secures, and terminates bounded interaction contexts \u2014 called sessions \u2014 between a client and a server or distributed service. It encompasses token issuance and validation, state serialisation and synchronisation, session expiry and renewal, concurrent-session policy, and secure revocation. Session management bridges authentication (establishing identity) and authorisation (enforcing capability) by preserving verified context across otherwise stateless request\u2013response cycles. It is a foundational layer for any multi-user application, from web platforms and APIs to real-time collaborative environments and spatial computing runtimes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:session-management", "labels": [ "Session Management", "Stateless Session Management" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "session-manager", "title": "Session Manager", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Session Manager is a software component responsible for creating, maintaining, tracking, and terminating user or application sessions within a computing system, ensuring that stateful context is preserved across multiple interactions or network requests. It issues session tokens or identifiers, enforces timeout and expiry policies, replicates session state for high-availability scenarios, and integrates with authentication services to verify that sessions remain bound to authenticated principals. Session managers are critical security components: misconfigurations can lead to session fixation, hijacking, or replay attacks. In distributed architectures they must handle session affinity, cross-node replication, and graceful failover without exposing stale state.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:session-manager", "labels": [ "Session Manager" ], "is_subclass_of": [ "Middleware" ], "wikilinks": [] }, { "id": "session-recording", "title": "Session Recording", "domain": "security", "domain_name": "Security", "definition": "Session Recording is the systematic capture of a user's interaction with a digital system during a defined session, including screen state, input events (mouse, keyboard, touch), audio, video, and network activity. In security contexts it provides an audit trail of privileged access for forensic and compliance purposes. In UX research it enables replay analysis of usability test sessions. In digital forensics it constitutes primary evidence of user actions on a system. Robust implementations mask sensitive data fields to balance observability with privacy requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:session-recording", "labels": [ "Session Recording" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "set-theory", "title": "Set Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The branch of mathematical logic that studies collections of objects called sets, providing a foundational language for most of modern mathematics including logic, topology, algebra, and knowledge representation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:set-theory", "labels": [ "Set Theory", "Axiomatic Set Theory" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Description Logic", "Probability Theory", "Graph Theory", "Ontology", "owl:Thing" ] }, { "id": "set-of-mark-prompting", "title": "Set-of-Mark Prompting", "domain": "ai", "domain_name": "Ai", "definition": "Set-of-mark prompting is a visual-prompting technique that overlays an image with numbered or coloured marks on segmented regions so a multimodal language model can refer to and reason about specific elements by label. By grounding the model's references in explicit visual tokens, it sharply improves spatial grounding, visual question answering, and GUI element selection. It is a key enabler of vision-driven computer-use agents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:set-of-mark-prompting", "labels": [ "Set-of-Mark Prompting" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "setpoint", "title": "Setpoint", "domain": "robotics", "domain_name": "Robotics", "definition": "A setpoint is the desired target value that a control system attempts to maintain for a measured process variable, serving as the reference against which the controller compares the actual measurement. The difference between setpoint and measured value is the error signal that drives corrective action in a feedback loop. Setpoints can be fixed, scheduled, or continuously varying, as when a controller tracks a moving reference trajectory.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:setpoint", "labels": [ "Setpoint" ], "is_subclass_of": [ "Closed-Loop Control" ], "wikilinks": [] }, { "id": "settlement-finality", "title": "Settlement Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Settlement finality is the point at which a transfer of value or assets becomes irrevocable and unconditional, such that it can no longer be reversed, unwound, or repudiated even in the event of a participant's insolvency. In traditional finance it is defined by legal frameworks governing payment and securities systems, while in blockchain systems it emerges from consensus guarantees that may be deterministic or probabilistic. Finality is essential to systemic stability because it eliminates settlement risk once the defined threshold is reached.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:settlement-finality", "labels": [ "Settlement Finality" ], "is_subclass_of": [ "Settlement" ], "wikilinks": [] }, { "id": "settlement-layer", "title": "Settlement Layer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Settlement Layer is the stratum that achieves final, irreversible transfer of value or state between parties. In layered ledger systems it sits beneath faster execution and netting strata and above the Consensus Layer that secures finality. It contains the records, accounts, and finality conditions under which obligations are discharged.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:settlement-layer", "labels": [ "Settlement Layer", "Cardano Settlement Layer", "On-Chain or Off-Chain Settlement Layer" ], "is_subclass_of": [ "Protocol and Consensus", "owl:Thing" ], "wikilinks": [ "Consensus Layer", "Smart Contract Layer", "Lightning Network Layer", "Atomic Settlement", "Double-Spending", "owl:Thing" ] }, { "id": "settlement", "title": "Settlement", "domain": "finance", "domain_name": "Finance", "definition": "Settlement is the final, legally binding transfer of an asset or funds from one party to another that extinguishes the obligation arising from a trade, payment, or contract. In traditional financial markets, settlement follows trade execution after a defined delay (e.g. T+2) during which counterparty risk persists; in distributed ledger systems, settlement occurs when a transaction achieves irreversible confirmation according to the network's consensus rules. Achieving settlement finality is the core objective of payment and securities clearing infrastructure, as it determines when legal title definitively passes and operational risk is eliminated. The shift towards real-time gross settlement (RTGS), delivery-versus-payment (DvP), and atomic settlement on distributed ledgers represents a decades-long effort to compress settlement latency and reduce systemic risk.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:settlement", "labels": [ "Settlement", "Settlement Contract", "Settlement Infrastructure", "Settlement Mechanism", "Settlement Protocol", "Settlement System", "Settlement Systems", "T+0 Settlement", "Tokenised Settlement" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [ "Atomic Settlement", "Payment System", "Cross-Border Settlement", "Financial Infrastructure Domain" ] }, { "id": "sh-ex", "title": "ShEx", "domain": "data", "domain_name": "Data", "definition": "ShEx (Shape Expressions) is a language for describing and validating the structure of RDF graphs, specifying which properties a node must have, their cardinalities, and value constraints. It serves a role for linked data analogous to schemas for XML or JSON, enabling data producers and consumers to agree on graph shapes. It is widely used to validate Solid pods and Wikidata-style knowledge graphs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sh-ex", "labels": [ "ShEx" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "shader-compiler", "title": "Shader Compiler", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A shader compiler is the toolchain component that translates high-level shading-language source code, such as GLSL or HLSL, into an intermediate representation or the native instruction set executed by a GPU. It performs parsing, optimisation, register allocation, and code generation so that programmable graphics and compute stages run efficiently on diverse hardware. Shader compilation may occur ahead of time or just in time, and its results are frequently cached to avoid costly recompilation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:shader-compiler", "labels": [ "Shader Compiler" ], "is_subclass_of": [ "Compiler" ], "wikilinks": [] }, { "id": "shader-language", "title": "Shader Language", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Shader Language is a domain-specific programming language designed to express GPU-executable programs that control one or more programmable stages of a graphics or compute pipeline. Shader languages provide specialised type systems encompassing vectors, matrices, samplers, and atomic types, while deliberately restricting features incompatible with massively parallel execution such as dynamic memory allocation and unbounded recursion. Major production languages include GLSL (OpenGL Shading Language) for OpenGL and WebGL, HLSL (High-Level Shading Language) for DirectX, Metal Shading Language for Apple silicon and macOS/iOS platforms, and the emerging WGSL (WebGPU Shading Language) for the web-native WebGPU API. Modern shader toolchains commonly cross-compile to the hardware-agnostic intermediate representation SPIR-V, enabling portability across vendors and runtime environments.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:shader-language", "labels": [ "Shader Language", "Shading Language" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "shader", "title": "Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A shader is a programmable GPU program that replaces fixed-function rendering pipeline stages, executing artist-authored or engine-generated algorithms in parallel across thousands of GPU threads to determine vertex positions, surface colour, lighting, and post-process effects. Shader types include vertex, tessellation control, tessellation evaluation, geometry, fragment (pixel), mesh, ray-generation, and compute shaders, each targeting a distinct stage of the GPU execution pipeline. Written in high-level shading languages such as GLSL, HLSL, Metal Shading Language, or WGSL, shaders are compiled to hardware-specific bytecode and scheduled by the GPU driver onto shader processor cores. In the context of spatial computing, shaders are the fundamental mechanism through which all real-time visual content is generated for head-mounted displays, AR overlays, and immersive 3D environments.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:shader", "labels": [ "Shader", "Shader Code", "Shader Core", "Shader Program", "Shader Programming", "Shader System" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "shading-model", "title": "Shading Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A shading model is a mathematical description of how a surface reflects and emits light, determining the colour and intensity seen at each point given the lighting and viewing geometry. It encapsulates a bidirectional reflectance distribution function together with parameters such as albedo, roughness and metalness. Shading models range from simple empirical formulations to physically based formulations grounded in energy conservation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:shading-model", "labels": [ "Shading Model" ], "is_subclass_of": [ "Rendering" ], "wikilinks": [] }, { "id": "shadow-banking", "title": "Shadow Banking", "domain": "finance", "domain_name": "Finance", "definition": "Shadow banking refers to credit intermediation and maturity transformation conducted by non-bank financial entities that operate outside the perimeter of conventional deposit-taking banking regulation. It encompasses activities such as money-market funds, securitisation vehicles, repo markets, and hedge-fund lending that perform bank-like functions without access to central-bank backstops or deposit insurance. Because it sits beyond standard prudential oversight, shadow banking can amplify systemic risk during periods of market stress.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:shadow-banking", "labels": [ "Shadow Banking" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "shadow-mapping", "title": "Shadow Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Shadow Mapping is a real-time computer graphics technique for rendering shadows by rendering the scene from the perspective of each light source into a depth texture (the shadow map), then comparing scene-point depth values against that map during the main render pass to determine visibility. It is the dominant method for dynamic shadows in games and real-time rendering engines due to its GPU efficiency and flexibility. Artefacts such as shadow acne and perspective aliasing are mitigated through techniques like bias adjustment, percentage-closer filtering, and cascaded shadow maps. The technique is fundamental to photorealistic rendering in spatial computing and XR applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:shadow-mapping", "labels": [ "Shadow Mapping" ], "is_subclass_of": [ "Spatial Computing", "Display and Rendering" ], "wikilinks": [] }, { "id": "shamir-secret-sharing", "title": "Shamir Secret Sharing", "domain": "security", "domain_name": "Security", "definition": "Shamir secret sharing splits a secret into shares so that any threshold number of them reconstructs it while fewer reveal nothing, using polynomial interpolation over a finite field.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:shamir-secret-sharing", "labels": [ "Shamir Secret Sharing" ], "is_subclass_of": [ "Cryptography", "Cryptography Domain" ], "wikilinks": [ "Modular Arithmetic", "Threshold Cryptography", "Cryptography", "STARK", "Cryptography Domain" ] }, { "id": "shape-memory-alloy-actuator", "title": "Shape Memory Alloy Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A shape memory alloy (SMA) actuator exploits the thermoelastic phase transformation of nickel-titanium (Nitinol) or similar alloys, which contract and generate force when thermally activated, then return to their original shape on cooling. SMA actuators are valued for their high force-to-weight ratio, silent operation, and inherent compliance, making them well-suited to soft robotics, minimally invasive surgical tools, and wearable exoskeletons. Control bandwidth is limited by thermal cycle times, which remains a key engineering challenge.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:shape-memory-alloy-actuator", "labels": [ "Shape Memory Alloy Actuator" ], "is_subclass_of": [ "Actuation and Control", "Robotics" ], "wikilinks": [ "Materials Science", "Robotics", "RoboticsDomain", "Soft Robotics" ] }, { "id": "shapley-value", "title": "Shapley Value", "domain": "ai", "domain_name": "Ai", "definition": "The Shapley Value is a solution concept from cooperative game theory that fairly allocates a total payoff among players according to their average marginal contribution across every possible coalition. In machine learning it is repurposed to attribute a model's prediction to individual input features, treating each feature as a player contributing to the model's output. This attribution use underlies explainability methods such as SHAP.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:shapley-value", "labels": [ "Shapley Value" ], "is_subclass_of": [ "Game Theory" ], "wikilinks": [] }, { "id": "sharding", "title": "Sharding", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sharding is a horizontal partitioning technique for distributed databases and blockchain networks in which a dataset or workload is divided into disjoint subsets called shards, each maintained by a distinct subset of nodes, so that the total system throughput scales with the number of shards rather than being bounded by the capacity of a single node. In databases, sharding routes queries to the appropriate shard by a sharding key. In blockchain, each shard processes its own subset of transactions and stores its own portion of the state, with cross-shard communication handled by a coordination layer. Sharding dramatically increases transaction throughput and reduces storage requirements per node at the cost of increased architectural complexity and cross-shard coordination overhead.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sharding", "labels": [ "Sharding", "Data Sharding", "ZeRO-3 Sharding" ], "is_subclass_of": [ "Blockchain Scalability" ], "wikilinks": [] }, { "id": "shared-ar-experiences", "title": "Shared Ar Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Collaborative augmented reality interactions enabling multiple users to simultaneously view and interact with digital content overlaid on shared physical environments, supporting real-time synchronization of virtual objects across devices for social connection and joint activities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:shared-ar-experiences", "labels": [ "Shared Ar Experiences", "Shared AR Experience" ], "is_subclass_of": [ "Platform and Environment", "Augmented Reality" ], "wikilinks": [ "Social AR Interaction", "Augmented Reality", "metaverse" ] }, { "id": "shared-cursors", "title": "Shared Cursors", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Shared Cursors are real-time visual indicators that show the pointer position and identity of each collaborator within a shared document or canvas. By rendering each participant's cursor with a distinct colour and name label, they communicate spatial presence and prevent conflicting edits without requiring verbal interruptions. They are a foundational multiplayer feature in live co-authoring environments such as Figma and Google Docs.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:shared-cursors", "labels": [ "Shared Cursors" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "shared-knowledge-base", "title": "Shared Knowledge Base", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A collaborative knowledge repository enabling multiple agents or users to exchange, integrate, and query structured information using semantic web technologies, providing a common understanding of domain concepts through ontologies and linked data standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:shared-knowledge-base", "labels": [ "Shared Knowledge Base" ], "is_subclass_of": [ "Content and Assets", "Knowledge Management System" ], "wikilinks": [ "Collaborative Knowledge Discovery", "Knowledge Management System", "metaverse" ] }, { "id": "shared-memory", "title": "Shared Memory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Shared memory is a memory region that multiple processes or threads can access concurrently, providing a low-latency mechanism for interprocess communication without the overhead of copying data through the kernel. It is exposed by operating systems through APIs such as POSIX shm_open or System V IPC and is heavily used in high-performance computing, GPU programming, and multi-process architectures. Because concurrent access requires explicit synchronisation, shared memory is typically paired with locks, semaphores, or atomic operations to avoid race conditions. It trades convenience for correctness risk, making it well-suited to performance-critical paths where the coordination cost is justified.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:shared-memory", "labels": [ "Shared Memory" ], "is_subclass_of": [ "Memory Management" ], "wikilinks": [] }, { "id": "shared-ownership-model", "title": "Shared Ownership Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A blockchain-enabled framework for fractional ownership of high-value assets through tokenization, allowing multiple parties to hold proportional stakes in real estate, digital art, intellectual property, or virtual assets with automated governance via smart contracts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:shared-ownership-model", "labels": [ "Shared Ownership Model" ], "is_subclass_of": [ "Governance and Safety", "Asset Tokenisation" ], "wikilinks": [ "Democratized Investment", "Asset Tokenization", "metaverse" ] }, { "id": "shared-spatial-anchors", "title": "Shared Spatial Anchors", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cloud-based reference points that lock virtual objects to specific physical locations, enabling multiple users across different devices to perceive digital content in the same position and orientation relative to the real-world environment for collaborative mixed reality experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:shared-spatial-anchors", "labels": [ "Shared Spatial Anchors" ], "is_subclass_of": [ "Platform and Environment", "Spatial Computing Paradigm" ], "wikilinks": [ "Collaborative AR Experiences", "metaverse", "Spatial Computing", "Telecollaboration" ] }, { "id": "shared-virtual-space", "title": "Shared Virtual Space", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Shared Virtual Space is a persistent, synchronised three-dimensional environment that multiple remote participants inhabit simultaneously through their avatars or representations. It provides a common spatial context for collaboration, enabling participants to co-locate, manipulate shared artefacts, and engage in spatial communication as if physically together. Such spaces underpin immersive remote collaboration by combining real-time synchronisation, access control, and spatial audio to create a convincing sense of co-presence.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:shared-virtual-space", "labels": [ "Shared Virtual Space", "Shared Virtual Environment" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [] }, { "id": "shared-virtual-world", "title": "Shared Virtual World", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A persistent, synchronous three-dimensional digital environment where unlimited users interact simultaneously through avatars, featuring continuous data persistence for identity, assets, and social relationships across sessions while supporting real-time rendering and cross-platform access.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:shared-virtual-world", "labels": [ "Shared Virtual World" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "Digital Social Presence", "metaverse", "Virtual Environment" ] }, { "id": "shared-whiteboards", "title": "Shared Whiteboards", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Digital canvases in telepresence and collaboration platforms enabling distributed team members to simultaneously draw, write, annotate, and manipulate visual content \u2014 including text, images, diagrams, and sticky notes \u2014 in real time. Shared whiteboards replicate physical whiteboard collaboration dynamics through multi-user synchronisation and persistent storage of collaborative artefacts, supporting brainstorming, diagramming, and visual sensemaking across geographic distances.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:shared-whiteboards", "labels": [ "Shared Whiteboards", "TELE-302-shared-whiteboards" ], "is_subclass_of": [ "Workspace Tools", "Telecollaboration" ], "wikilinks": [ "TELE-002-telecollaboration", "TELE-028-horizon-workrooms", "TELE-301-virtual-office-spaces" ] }, { "id": "shared-workspace", "title": "Shared Workspace", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A shared workspace is a common digital environment in which multiple people can view, edit and organise shared artefacts together, whether simultaneously or over time. It provides a persistent space holding documents, boards, tasks or models alongside cues about who is present and what they are doing. Shared workspaces are a core construct of collaborative and groupware systems, turning individual tools into venues for coordinated teamwork.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:shared-workspace", "labels": [ "Shared Workspace" ], "is_subclass_of": [ "Collaboration", "Workspace Tools" ], "wikilinks": [] }, { "id": "sheffield-advanced-manufacturing", "title": "Sheffield Advanced Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sheffield Advanced Manufacturing denotes the cluster of research institutions, industrial facilities, and technology parks centred on the Advanced Manufacturing Park (AMP) in South Yorkshire, UK, specialising in materials science, aerospace engineering, robotics, and Industry 4.0 technologies. Anchored by the University of Sheffield's Advanced Manufacturing Research Centre (AMRC) and hosting global firms including Boeing and Rolls-Royce, the cluster forms the core of the Advanced Manufacturing Innovation District (AMID) \u2014 the UK's largest research-led advanced manufacturing ecosystem. It bridges traditional industrial heritage with digital manufacturing, additive processes, and intelligent automation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:sheffield-advanced-manufacturing", "labels": [ "Sheffield Advanced Manufacturing" ], "is_subclass_of": [ "UK Tech Ecosystem" ], "wikilinks": [ "Northern Powerhouse", "North England Innovation Corridor", "UK Tech Ecosystem" ] }, { "id": "sheffield", "title": "Sheffield", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sheffield is a metropolitan city in South Yorkshire, England, historically renowned as the Steel City for its dominance in cutlery, tool-making, and crucible steel production from the 18th century onwards. Today it operates as a major centre for advanced manufacturing, materials science, robotics research, and digital innovation, anchored by the University of Sheffield and Sheffield Hallam University. The city hosts the Advanced Manufacturing Research Centre (AMRC) \u2014 a globally recognised aerospace and industrial manufacturing cluster \u2014 as well as the Nuclear AMRC and a growing technology and data economy. Sheffield participates in the Northern Powerhouse initiative and is developing smart city infrastructure and innovation-district strategies aligned with Catapult programmes and UK industrial strategy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sheffield", "labels": [ "Sheffield" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Material Science", "Northern Powerhouse", "Manchester", "Entity" ] }, { "id": "side-channel-attack", "title": "Side-Channel Attack", "domain": "security", "domain_name": "Security", "definition": "A side-channel attack recovers secret information by observing physical or behavioural artefacts of a cryptographic implementation rather than breaking its underlying mathematics. Exploitable channels include timing variation, power consumption, electromagnetic emissions, cache access patterns, and acoustic emanations. It is a critical implementation-level threat that motivates constant-time code and masking countermeasures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:side-channel-attack", "labels": [ "Side-Channel Attack", "Side-Channel Attacks" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "sidechain", "title": "Sidechain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Independent blockchain connected to a parent blockchain via a two-way peg mechanism that operates with its own consensus and validation rules while enabling asset transfers between chains, providing scalability and experimental capabilities without impacting the main chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:sidechain", "labels": [ "Sidechain", "Sidechains" ], "is_subclass_of": [ "Network Component", "Blockchain" ], "wikilinks": [ "Federated Byzantine Agreement", "Gnosis Chain", "Layer 1", "Liquid Network", "Polygon PoS", "Rollup", "Ronin", "Scalability Solutions", "Two-Way Peg", "Blockchain", "Cross-Chain Bridge", "State Channel" ] }, { "id": "sidetree-protocol", "title": "Sidetree Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Sidetree Protocol is a layer-two protocol for operating scalable decentralised identifier networks on top of any existing decentralised ledger without requiring trusted intermediaries or special-purpose consensus. It batches large numbers of DID create, update, recover, and deactivate operations, anchors a single compact commitment to the underlying chain, and stores the operation data in content-addressed storage so that any node can deterministically replay the operation log to compute current DID states. This separation of anchoring from data lets identifier throughput scale far beyond the base chain's transaction capacity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:sidetree-protocol", "labels": [ "Sidetree Protocol" ], "is_subclass_of": [ "DID Method" ], "wikilinks": [] }, { "id": "sigma-algebra", "title": "Sigma-Algebra", "domain": "data", "domain_name": "Data", "definition": "A sigma-algebra is a collection of subsets of a sample space that contains the whole space and is closed under complementation and countable unions. It specifies exactly which sets are measurable, providing the domain on which measures and probability are rigorously defined. It is the structural foundation of measure theory, integration, and the theory of stochastic processes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sigma-algebra", "labels": [ "Sigma-Algebra", "Sigma Algebra" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "sign-language-recognition", "title": "Sign Language Recognition", "domain": "ai", "domain_name": "Ai", "definition": "Sign language recognition is the computer-vision and sequence-modelling task of translating the manual and non-manual gestures of a signed language into text or speech. It must model hand shape, motion trajectory, facial expression, and grammatical structure that differs fundamentally from spoken languages. It is an accessibility-focused application that builds on robust hand tracking and temporal recognition.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sign-language-recognition", "labels": [ "Sign Language Recognition" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "signal-amplifier", "title": "Signal Amplifier", "domain": "robotics", "domain_name": "Robotics", "definition": "A signal amplifier is an electronic circuit that increases the power, voltage, or current of a weak signal while ideally preserving its waveform. In sensing applications it boosts low-amplitude transducer outputs above the noise floor so they can be digitised, with gain, bandwidth, noise figure, and linearity as key parameters. It is an essential front-end component of sensors and biosignal acquisition systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:signal-amplifier", "labels": [ "Signal Amplifier" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "signal-conditioning", "title": "signal conditioning", "domain": "robotics", "domain_name": "Robotics", "definition": "Signal conditioning is the ensemble of analogue and digital processing stages applied to raw electrical outputs from physical sensors \u2014 including amplification, filtering, analogue-to-digital conversion, isolation, linearisation, and calibration \u2014 to produce clean, scaled, noise-reduced representations in engineering units suitable for downstream control, estimation, and machine-learning pipelines. It occupies the critical interface between the physical world and digital computation, and its fidelity directly determines the accuracy of perception, control, and data-acquisition systems built upon it. Standard stages encompass instrumentation amplifiers, anti-aliasing filters, temperature compensation, galvanic isolation, sample-rate conversion, and offset or gain correction applied in hardware, firmware, or software. In robotics, industrial automation, medical devices, and IoT edge nodes, signal conditioning is a prerequisite for reliable sensor fusion, closed-loop control, and anomaly detection.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:signal-conditioning", "labels": [ "Signal Conditioning" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "signal-processing-unit", "title": "Signal Processing Unit", "domain": "robotics", "domain_name": "Robotics", "definition": "A signal processing unit is the hardware or firmware block that filters, transforms, and extracts features from raw sensor or transducer signals, often using DSP cores, FFTs, and digital filtering. In haptic and biosensing devices it converts captured analogue or digital streams into actionable control signals or feature vectors in real time. It is a core processing component of interactive sensing interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:signal-processing-unit", "labels": [ "Signal Processing Unit" ], "is_subclass_of": [ "Hardware" ], "wikilinks": [] }, { "id": "signal-processing", "title": "Signal Processing", "domain": "ai", "domain_name": "Ai", "definition": "Signal processing is the mathematical and engineering discipline concerned with the representation, analysis, transformation, synthesis, filtering, and compression of signals \u2014 time-varying or spatially varying quantities such as audio, video, sensor telemetry, radio-frequency waveforms, biomedical readings, and seismic data. It encompasses both continuous (analogue) and discrete (digital) domains, applying techniques from Fourier analysis, linear algebra, probability theory, and optimisation to extract information, remove noise, encode data, and control systems. Digital signal processing (DSP) executes these operations on sampled data using algorithms implemented in hardware or software, making it foundational to telecommunications, audio engineering, image processing, radar, and AI feature extraction pipelines. Modern signal processing increasingly fuses classical deterministic methods with statistical and machine-learning approaches, enabling adaptive filters, compressed sensing, and deep neural network-based feature representations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:signal-processing", "labels": [ "Signal Processing", "SignalProcessing" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "owl:Thing" ] }, { "id": "signal-protocol", "title": "Signal Protocol", "domain": "security", "domain_name": "Security", "definition": "Signal Protocol is an open-source cryptographic messaging protocol that provides end-to-end encryption for instant messaging applications, combining the Double Ratchet Algorithm with the X3DH (Extended Triple Diffie-Hellman) key agreement protocol. It achieves forward secrecy and break-in recovery (future secrecy) by continuously rotating encryption keys after each message exchange.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:signal-protocol", "labels": [ "Signal Protocol" ], "is_subclass_of": [ "Secure Messaging" ], "wikilinks": [] }, { "id": "signal-to-noise-ratio", "title": "Signal-to-noise Ratio", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:signal-to-noise-ratio", "labels": [ "Signal-to-noise Ratio" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "signaling-server", "title": "Signaling Server", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A signaling server is an intermediary service used in WebRTC and peer-to-peer systems to exchange session control messages between peers before a direct connection is established. It transmits session descriptions (SDP offers and answers) and ICE candidates so that peers can negotiate codec capabilities, network addresses, and connection parameters. Once the direct peer connection is set up, the signaling server is no longer involved in data transfer.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:signaling-server", "labels": [ "Signaling Server" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "signature-aggregation", "title": "Signature Aggregation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Signature Aggregation is a cryptographic technique that combines multiple digital signatures over distinct or identical messages into a single compact signature that can be verified with one operation. It reduces on-chain storage and verification cost in blockchain systems where many parties sign, such as validator committees. Schemes such as BLS and Schnorr support aggregation, improving scalability and bandwidth efficiency for distributed consensus.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:signature-aggregation", "labels": [ "Signature Aggregation" ], "is_subclass_of": [ "Digital Signature" ], "wikilinks": [] }, { "id": "signature-algorithm", "title": "Signature Algorithm", "domain": "security", "domain_name": "Security", "definition": "A signature algorithm is a cryptographic scheme comprising key generation, signing, and verification procedures that lets a holder of a private key produce a value provably tied to a message, which anyone with the public key can verify. It provides authenticity, integrity, and non-repudiation, with families including RSA, ECDSA, EdDSA, and Schnorr. It is the cryptographic primitive underpinning digital signatures.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:signature-algorithm", "labels": [ "Signature Algorithm" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "signature-scheme", "title": "Signature Scheme", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Signature Scheme is a cryptographic primitive that provides a triple of algorithms\u2014key generation, signing, and verification\u2014enabling a party holding a private key to produce an unforgeable authentication tag over arbitrary messages that any holder of the corresponding public key can verify. In blockchain systems, signature schemes authenticate transactions, authorise state transitions, and underpin identity and ownership semantics across the distributed ledger.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:signature-scheme", "labels": [ "Signature Scheme" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "signed-distance-function", "title": "Signed Distance Function", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A signed distance function (SDF) is a scalar field that returns, for any point in space, the distance to the nearest surface of a shape, with the sign indicating whether the point is inside (negative) or outside (positive). The surface itself is the zero level set where the function equals zero. SDFs provide a compact implicit representation of geometry that supports efficient ray marching, smooth shape blending, and analytic normals, and they underpin procedural rendering, collision queries, and learned 3D reconstruction in neural networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:signed-distance-function", "labels": [ "Signed Distance Function", "Signed Distance Field" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "sigstore", "title": "Sigstore", "domain": "security", "domain_name": "Security", "definition": "Sigstore is an open-source project and set of services for signing, verifying, and proving the provenance of software artefacts using short-lived keys and a public transparency log. It removes the burden of long-term key management by issuing ephemeral signing certificates bound to OpenID Connect identities, recording signatures in an append-only log (Rekor) for auditability. Sigstore underpins software supply-chain security through tools such as Cosign for container and artefact signing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:sigstore", "labels": [ "Sigstore" ], "is_subclass_of": [ "Supply Chain Security" ], "wikilinks": [] }, { "id": "sim-to-real-transfer", "title": "sim-to-real transfer", "domain": "robotics", "domain_name": "Robotics", "definition": "Sim-to-real transfer is the set of techniques and methodologies for training robotic policies, perception models, and control algorithms inside physics simulators and then deploying the resulting models on physical hardware with acceptable performance degradation. The fundamental barrier is the reality gap: discrepancies in physics fidelity, contact dynamics, sensor noise characteristics, visual appearance, and actuator latency cause policies optimised entirely in simulation to fail on real systems. Mitigation strategies include domain randomisation, domain adaptation, system identification, adaptive dynamics models, and photorealistic rendering to progressively close this gap. The field sits at the intersection of reinforcement learning, robot learning, and transfer learning, and underpins practical large-scale autonomous system deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sim-to-real-transfer", "labels": [ "Sim-to-Real Transfer", "Sim to Real Transfer", "Sim-to-Real Transfer Workflow", "Sim2Real Transfer" ], "is_subclass_of": [ "Transfer Learning", "Robot Learning" ], "wikilinks": [] }, { "id": "sim-clr", "title": "SimCLR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A self-supervised learning framework that learns visual representations through contrastive learning with data augmentation. A linear classifier on SimCLR representations achieves top-1 accuracy, matching supervised ResNet-50 performance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:sim-clr", "labels": [ "SimCLR" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "simplified-payment-verification", "title": "Simplified Payment Verification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Simplified Payment Verification (SPV) is a technique that lets a lightweight client confirm that a transaction is included in a blockchain without downloading the entire chain, by holding only block headers and requesting a Merkle proof linking the transaction to a header's Merkle root. Described in the original Bitcoin design, it trades the full validation guarantees of a complete node for drastically reduced storage and bandwidth, relying on the proof-of-work in headers and the honest-majority assumption. It is what makes mobile and embedded cryptocurrency wallets practical.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:simplified-payment-verification", "labels": [ "Simplified Payment Verification" ], "is_subclass_of": [ "Blockchain", "Protocol and Consensus" ], "wikilinks": [] }, { "id": "simulated-annealing", "title": "Simulated Annealing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A stochastic metaheuristic for global optimisation, inspired by the annealing of metals, that explores a solution space via local moves while accepting worsening solutions with a probability governed by a gradually decreasing temperature parameter, allowing early escape from local optima and increasingly greedy refinement as the temperature cools; widely applied to combinatorial problems such as travelling salesman routing, chip placement, scheduling and logistics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:simulated-annealing", "labels": [ "Simulated Annealing" ], "is_subclass_of": [ "Heuristic Methods" ], "wikilinks": [ "Heuristic Methods", "Local Search", "Combinatorial Optimisation" ] }, { "id": "simulation-engine", "title": "Simulation Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A simulation engine is the computational core responsible for advancing synthetic world state over discrete or continuous time steps, resolving inter-object interactions, enforcing physical or logical constraints, and exposing deterministic replay and instrumentation interfaces. It abstracts heterogeneous hardware (CPU thread pools, GPU compute shaders, distributed clusters) behind a unified loop that separates physics integration, collision detection, agent behaviour evaluation, and sensor data generation into composable subsystems. Specialised variants include rigid-body and soft-body physics engines (PhysX, Bullet, Havok), behaviour-simulation engines, robotics simulators (Isaac Sim, Gazebo), and synthetic-data factories used to train and evaluate machine-learning models at scale.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:simulation-engine", "labels": [ "Simulation Engine" ], "is_subclass_of": [ "Simulation Software" ], "wikilinks": [] }, { "id": "simulation-environment", "title": "Simulation Environment", "domain": "robotics", "domain_name": "Robotics", "definition": "A simulation environment is a software-defined system that models physical or virtual worlds with sufficient fidelity to support training, testing, or validation of agents, algorithms, or hardware without exposure to real-world risk or cost. It integrates a physics engine, sensor models, actuator dynamics, and observation/action interfaces, and is used across robotics, autonomous vehicle development, reinforcement learning research, and military mission planning. Fidelity-reality gaps \u2014 differences between simulated and real-world dynamics \u2014 are the primary challenge in transferring learned behaviours from simulation to deployment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:simulation-environment", "labels": [ "Simulation Environment", "SimulationEnvironment" ], "is_subclass_of": [ "Simulation" ], "wikilinks": [] }, { "id": "simulation-layer", "title": "Simulation Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Simulation Layer is the cross-cutting stratum that models system or environment behaviour to test and predict outcomes without acting on the real world. It sits above compute and model strata it uses and supports research, evaluation, and planning. It contains simulators, environment models, and the scenarios run within them.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:simulation-layer", "labels": [ "Simulation Layer" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Compute Layer", "Model Layer", "Research Layer", "Evaluation Layer", "Monte Carlo Method", "Digital Twin", "owl:Thing" ] }, { "id": "simulation-software", "title": "Simulation Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Application software that constructs, parameterises, and executes computational models of physical, biological, social, or engineered systems, enabling controlled experimentation and behaviour observation across time steps without manipulating real-world systems. Simulation software encompasses physics engines, agent-based modelling frameworks, discrete-event simulators, and real-time digital twin environments, serving domains ranging from aerospace engineering and molecular biology to urban planning and immersive training. Unlike general-purpose scientific computing, simulation software provides domain-specific abstraction layers, visualisation pipelines, and scenario management tools that allow non-specialists to configure and run experiments at scale. The field intersects spatial computing, machine learning, and high-performance computing as simulations grow to planetary scale and real-time fidelity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:simulation-software", "labels": [ "Simulation Software" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "simulation", "title": "Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Simulation is the computational reproduction of real-world systems, physical phenomena, or abstract processes through mathematical models that evolve over time, enabling experimentation, training, and prediction without risk to personnel or infrastructure. Simulations span a fidelity spectrum from simplified discrete-event models to high-fidelity continuous-physics environments powered by physics engines, rendering pipelines, and stochastic solvers. In spatial computing and metaverse contexts, real-time simulation underpins immersive training, digital twin synchronisation, and AI agent incubation. Simulation outputs are validated against empirical data and uncertainty quantified to ensure transferability of findings to the real world.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:simulation", "labels": [ "Simulation", "Construction Simulation", "Simulation-Driven Design" ], "is_subclass_of": [ "Digital Twin" ], "wikilinks": [] }, { "id": "simultaneous-localisation-and-mapping", "title": "simultaneous localisation and mapping", "domain": "robotics", "domain_name": "Robotics", "definition": "Simultaneous Localisation and Mapping (SLAM) is a computational technique by which a mobile robot or autonomous agent concurrently estimates its own pose and constructs a consistent map of a previously unknown environment from sequential sensor observations, resolving the fundamental circular dependency between localisation (which requires a map) and mapping (which requires a known pose). SLAM algorithms process data from sensors such as LiDAR, stereo cameras, RGB-D cameras, and inertial measurement units using probabilistic and optimisation-based frameworks \u2014 including extended Kalman filters, particle filters, and pose-graph optimisation \u2014 to maintain joint estimates of agent state and environmental structure. The problem is formulated as Bayesian inference over a high-dimensional joint distribution of robot trajectory and landmark positions, typically approximated through factor graphs. SLAM is foundational to autonomous vehicles, mobile robotics, augmented reality, and any system that must navigate without prior maps or GPS.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:simultaneous-localisation-and-mapping", "labels": [ "Simultaneous Localisation and Mapping", "Localisation and Mapping", "Simultaneous Localization and Mapping", "Visual Simultaneous Localisation and Mapping" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "singapore", "title": "Singapore", "domain": "finance", "domain_name": "Finance", "definition": "A sovereign island city-state in Southeast Asia and a major financial and technology hub. It is known for its role in global trade, finance, and as a centre for fintech and digital asset regulation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:singapore", "labels": [ "Singapore" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "China", "owl:Thing" ] }, { "id": "single-event-effect", "title": "Single Event Effect", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:single-event-effect", "labels": [ "Single Event Effect" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "single-event-upset", "title": "Single Event Upset", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:single-event-upset", "labels": [ "Single Event Upset" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "single-point-of-failure", "title": "Single Point Of Failure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A single point of failure is a component, service, or dependency within a system whose failure would cause the entire system to stop functioning, because no redundant alternative exists to take over its role. Identifying and eliminating single points of failure is a central goal of high-availability and fault-tolerant design. Mitigation strategies include redundancy, replication, clustering, and load balancing so that no individual element is indispensable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:single-point-of-failure", "labels": [ "Single Point Of Failure", "Single Point of Failure" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [] }, { "id": "single-sign-on", "title": "single sign-on", "domain": "security", "domain_name": "Security", "definition": "Single Sign-On (SSO) is an authentication architecture in which a user authenticates once to a central Identity Provider (IdP) and receives a cryptographically signed assertion or token that grants access to multiple independent service providers without re-entering credentials. Dominant protocol implementations include SAML 2.0 (XML-based assertions), OpenID Connect layered over OAuth 2.0 (JSON Web Token\u2013based), and Kerberos (ticket-granting ticket mechanism in Windows/Active Directory environments). SSO reduces credential exposure and attack surface by centralising authentication at a hardened IdP, simplifies user lifecycle management across enterprise systems, and is the cornerstone of modern Identity and Access Management platforms encompassing provisioning, deprovisioning, and audit.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:single-sign-on", "labels": [ "Single Sign-On", "Single Sign On", "Single Sign-On (SSO)" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "single-use-seals", "title": "Single Use Seals", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A single-use seal is a cryptographic primitive, proposed by Peter Todd in 2016, that is a uniquely identifiable object which can be closed over ('sealed to') a message exactly once, producing a publicly verifiable proof that the seal was closed over that specific message and no other. The canonical implementation uses a Bitcoin UTXO as the seal: because the consensus rules permit each UTXO to be spent only once, the transaction that spends it can commit to exactly one message, giving a tamper-evident, non-equivocable 'open once' guarantee. Single-use seals, combined with proof of publication, form the foundation of client-side validation and underpin Bitcoin smart-contract systems such as RGB and primitives such as Block Trails.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:single-use-seals", "labels": [ "Single Use Seals", "Single Use Seal", "Single-Use Seal", "Single-Use Seals" ], "is_subclass_of": [ "Cryptographic Commitment" ], "wikilinks": [] }, { "id": "single-agent-system", "title": "Single-Agent System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An architectural pattern in which one autonomous agent, equipped with its own model, memory, and tool access, carries an entire task from goal to completion within a single reasoning loop \u2014 with no delegation, inter-agent messaging, or coordination overhead; simpler to build, debug, and evaluate than multi-agent designs, and often the stronger baseline when a capable model with good tools can hold the whole problem in context.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:single-agent-system", "labels": [ "Single-Agent System" ], "is_subclass_of": [ "Architecture" ], "wikilinks": [ "Architecture", "Agent", "Multi-Agent System" ] }, { "id": "single-turn-inference", "title": "Single-Turn Inference", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The stateless request-response mode of using a language model: a prompt goes in, one completion comes out, and the interaction ends \u2014 with no intermediate tool execution, environmental feedback, self-correction, or persistent state; it is the cheapest and most predictable inference pattern, suited to classification, extraction, translation, and summarisation, and serves as the baseline against which iterative agentic workflows are defined and evaluated.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:single-turn-inference", "labels": [ "Single-Turn Inference" ], "is_subclass_of": [ "Inference" ], "wikilinks": [ "Inference", "Agentic Workflow", "Large Language Models" ] }, { "id": "singular-value-decomposition", "title": "Singular Value Decomposition", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Singular value decomposition (SVD) is a matrix factorisation that expresses any real or complex matrix as the product of two orthogonal (or unitary) matrices and a diagonal matrix of non-negative singular values. It generalises eigenvalue decomposition to arbitrary, non-square matrices and reveals the rank, range, and dominant directions of variation in data. SVD is foundational across machine learning and numerical linear algebra, underpinning dimensionality reduction, low-rank approximation, recommender systems, and the principal component analysis used to compress and denoise high-dimensional data.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:singular-value-decomposition", "labels": [ "Singular Value Decomposition" ], "is_subclass_of": [ "Dimensionality Reduction" ], "wikilinks": [] }, { "id": "singularity-analysis", "title": "Singularity Analysis", "domain": "robotics", "domain_name": "Robotics", "definition": "Singularity analysis is the study of robot configurations at which the manipulator Jacobian loses rank, causing a loss or gain of instantaneous degrees of freedom. At singularities the robot cannot move in certain Cartesian directions, joint velocities may diverge, and force or motion control degrades. Identifying and avoiding singularities is essential for safe, well-conditioned trajectory planning and control of robotic arms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:singularity-analysis", "labels": [ "Singularity Analysis" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [] }, { "id": "singularity", "title": "Singularity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Technological Singularity is a hypothesised future point at which artificial intelligence surpasses human cognitive capacity in all economically and strategically relevant domains, triggering a phase transition in civilisational development so rapid and so structurally discontinuous that extr...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:singularity", "labels": [ "Singularity", "Technological Singularity" ], "is_subclass_of": [ "AI Research Area", "Existential Risk", "Artificial General Intelligence", "Futures Studies", "Philosophy of Mind", "Transhumanism" ], "wikilinks": [ "AGI Timelines", "AI Impacts Surveys", "AI Policy", "AI Winter", "Bayesian Forecasting", "Cambridge CSER", "Compute Scaling", "Existential Risk", "Exponential Growth Models", "Future of Humanity Institute", "Futures Studies", "FuturesStudiesDomain", "Intelligence Amplification", "Intelligence Explosion", "Kurzweil Law of Accelerating Returns", "Longtermism", "Machine Intelligence Research Institute", "Metaculus", "Mind Uploading", "Narrow AI" ] }, { "id": "sink-mechanism", "title": "Sink Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Sink Mechanism is an economic design pattern that permanently removes virtual currency, tokens, or items from circulation in order to counterbalance sources of supply and maintain economic equilibrium, preventing hyperinflation in metaverse economies, play-to-earn platforms, and DeFi protocols. Effective sinks are integrated into desirable activities \u2014 such as cosmetic purchases, crafting costs, or transaction fees \u2014 so that value removal feels voluntary and rewarding to participants.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:sink-mechanism", "labels": [ "Sink Mechanism" ], "is_subclass_of": [ "DeFi and Economics", "Virtual Economy", "Blockchain" ], "wikilinks": [ "Economic Balance", "Game Economy", "Blockchain", "Tokenomics", "Virtual Currency", "Virtual Economy" ] }, { "id": "site-reliability-engineering", "title": "Site Reliability Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Site Reliability Engineering (SRE) is a discipline that applies software engineering principles and practices to operations work, treating infrastructure management and service reliability as software problems to be solved through automation, measurement, and iterative improvement. Originating at Google in the early 2000s, SRE defines explicit reliability targets via Service Level Objectives, manages risk through error budgets, and uses observability tooling to maintain confidence in production systems. SRE practitioners reduce operational toil through systematic automation, implement incident management processes, and balance the competing demands of feature velocity and system stability. The discipline formalises the role of the operations engineer as a software engineer who designs for reliability, scalability, and maintainability from the outset.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:site-reliability-engineering", "labels": [ "Site Reliability Engineering" ], "is_subclass_of": [ "Reliability Engineering" ], "wikilinks": [ "Observability", "Distributed Systems", "Reliability Engineering" ] }, { "id": "situational-awareness", "title": "Situational Awareness", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Situational awareness is an agent's perception of relevant elements in its environment, comprehension of their meaning and projection of their near-future state, supporting timely decisions.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:situational-awareness", "labels": [ "Situational Awareness" ], "is_subclass_of": [ "Robotics", "Perception and Sensing", "Robotics Domain" ], "wikilinks": [ "Computer Vision", "Autonomous Agent", "Multi-Agent System", "Digital Twin", "Robotics Domain" ] }, { "id": "six-degrees-of-freedom", "title": "Six Degrees Of Freedom", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Six degrees of freedom (6DoF) describes the full set of ways a rigid body can move in three-dimensional space: translation along three perpendicular axes and rotation about each of them. In spatial computing it characterises tracking systems that capture both an object's position and its orientation. Supporting 6DoF is essential for convincing virtual and augmented reality, robotics and motion-tracked interaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:six-degrees-of-freedom", "labels": [ "Six Degrees Of Freedom", "Six Degrees of Freedom" ], "is_subclass_of": [ "Tracking System" ], "wikilinks": [] }, { "id": "skeletal-animation", "title": "Skeletal Animation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Skeletal animation is a character animation technique in which a hierarchical rig of joints defines the articulated structure of a character or creature, and mesh vertices are deformed by the weighted influence of surrounding joints through a process known as skinning. Artists author motion by keyframing or procedurally driving joint transforms over time; the underlying mesh deforms accordingly in real time on the GPU via linear blend skinning or dual-quaternion skinning. The approach is the dominant method for animating characters in games, virtual reality, film visual effects, and virtual avatar systems due to its runtime efficiency, compact data representation, and amenability to motion capture retargeting and machine-learning-driven synthesis.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:skeletal-animation", "labels": [ "Skeletal Animation", "Skeleton Animation" ], "is_subclass_of": [ "Character Animation" ], "wikilinks": [] }, { "id": "skeletal-mesh", "title": "Skeletal Mesh", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A skeletal mesh is a 3D character or object model bound to an underlying bone hierarchy, or skeleton, so that mesh vertices deform smoothly as the skeleton is posed or animated. It is the standard representation for animatable characters in avatar systems and real-time engines, and it underlies procedural animation techniques that manipulate bone transforms algorithmically at runtime. Building one requires prior UV unwrapping and rigging work on the base mesh.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:skeletal-mesh", "labels": [ "Skeletal Mesh" ], "is_subclass_of": [ "Character Rigging" ], "wikilinks": [] }, { "id": "skeletal-rig", "title": "Skeletal Rig", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A skeletal rig is the hierarchy of bones and control structures bound to a 3D model that allows it to be posed and animated. Each bone influences nearby vertices through skinning weights, so that rotating or translating a bone deforms the surrounding mesh in a controlled way. Rigs typically combine forward and inverse kinematics with control handles, enabling animators to drive complex character motion from a compact set of controls.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:skeletal-rig", "labels": [ "Skeletal Rig" ], "is_subclass_of": [ "3D Modelling" ], "wikilinks": [] }, { "id": "skinning", "title": "Skinning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Skinning is the process of binding a deformable surface mesh to an underlying skeleton so that the mesh deforms naturally as the skeleton is animated. Each vertex is assigned weights that determine how strongly it follows each influencing bone, and these weights drive the deformation during playback. Skinning is essential to character animation, allowing a single rigged model to be posed and animated across many motions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:skinning", "labels": [ "Skinning" ], "is_subclass_of": [ "Character Rigging" ], "wikilinks": [] }, { "id": "skos-vocabulary", "title": "Skos Vocabulary", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A W3C recommendation providing an RDF-based data model for representing thesauri, taxonomies, classification schemes, and other structured controlled vocabularies, enabling publication and linking of knowledge organization systems on the semantic web through standardized concepts and relationships.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:skos-vocabulary", "labels": [ "Skos Vocabulary", "SKOS", "SKOS Concept Scheme" ], "is_subclass_of": [ "Standards and Interoperability", "Knowledge Organization System" ], "wikilinks": [ "Linked Vocabulary Publishing", "Knowledge Organization System", "metaverse" ] }, { "id": "slack", "title": "Slack", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Slack is a cloud-based team messaging and collaboration platform that organises workplace communication into persistent, searchable channels grouped by topic, project, or team, supplemented by direct messages and threaded conversations. Originally developed as an internal tool for the gaming company Tiny Speck in 2013 before pivoting to a standalone product, Slack is distinguished by its extensible integration ecosystem (over 2,500 app integrations), Workflow Builder for no-code automation, and its role as the de facto communication layer for software development teams. Acquired by Salesforce in 2021 for $27.7 billion, it competes primarily with Microsoft Teams.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:slack", "labels": [ "Slack" ], "is_subclass_of": [ "Collaboration Platform" ], "wikilinks": [] }, { "id": "slashing-condition", "title": "Slashing Condition", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A slashing condition is a rule in a proof-of-stake protocol that defines provable validator misbehaviour, such as double-signing or equivocation, and triggers the forfeiture of part of the offender's staked collateral. It makes attacks economically costly by penalising actions that threaten consensus safety, aligning validator incentives with honest participation. It is a core economic-security primitive of staking-based consensus.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:slashing-condition", "labels": [ "Slashing Condition" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "slashing-conditions", "title": "Slashing Conditions", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Slashing conditions are the set of cryptographically provable validator violations that a proof-of-stake blockchain protocol enforces by destroying or redistributing staked tokens. Encoded directly in protocol logic, they cover equivocation, surround votes, and prolonged unavailability, and their parameters set the strength of the chain's economic security. They are an implemented enforcement mechanism in modern blockchain protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:slashing-conditions", "labels": [ "Slashing Conditions" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "slashing", "title": "Slashing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Slashing is an automated penalty mechanism in proof-of-stake and delegated proof-of-stake blockchain networks that permanently destroys or confiscates a portion of a validator's bonded stake when that validator commits a provably attributable protocol violation, such as double-signing conflicting blocks or surround voting. The mechanism converts the cost of Byzantine behaviour into a concrete financial loss, aligning validator incentives with honest participation and securing the network's economic finality. Slashing conditions and penalty magnitudes are encoded in the consensus rules of each network and are adjudicated deterministically by all full nodes.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:slashing", "labels": [ "Slashing", "Economic Slashing", "Slashing Mechanism" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [ "Proof of Stake", "Validator", "Consensus Mechanism" ] }, { "id": "sliding-mode-control", "title": "Sliding Mode Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Sliding mode control is a nonlinear, robust control method that drives a system's state onto a designed sliding surface and constrains it there using high-frequency switching of the control input. Once on the surface, the closed-loop dynamics become insensitive to matched disturbances and parameter uncertainty, giving strong robustness. It is a control-theory technique widely applied to robotic actuators and power electronics, though it can induce chattering.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sliding-mode-control", "labels": [ "Sliding Mode Control" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "slippage", "title": "Slippage", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Slippage is the difference between the expected price of a trade and the price at which it actually executes, arising from price movement and limited liquidity between order submission and settlement. On automated-market-maker decentralised exchanges slippage is a direct function of trade size relative to pool depth, and traders set a slippage tolerance to bound acceptable execution price. Excessive slippage can be exploited through front-running and other maximal-extractable-value strategies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:slippage", "labels": [ "Slippage" ], "is_subclass_of": [ "Decentralized Exchange" ], "wikilinks": [] }, { "id": "slot-filling", "title": "Slot Filling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Slot filling is a natural language processing task in which a system extracts and populates predefined semantic fields (slots) from user utterances within a task-oriented dialogue context, enabling the system to gather the structured information required to fulfil a user request. It operates alongside intent classification to transform free-form text into actionable structured representations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:slot-filling", "labels": [ "Slot Filling" ], "is_subclass_of": [ "Information Extraction" ], "wikilinks": [] }, { "id": "small-language-models", "title": "Small Language Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Language models with significantly fewer parameters than frontier models, designed for efficiency, speed, and deployment on resource-constrained devices.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:small-language-models", "labels": [ "Small Language Models" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "small-modular-reactors", "title": "Small Modular Reactors", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Small modular reactors (SMRs) are nuclear fission reactors with an electrical output typically under 300 MWe whose components are factory-fabricated as standardised modules and assembled on site. Their smaller scale, passive safety features, and modular construction aim to lower capital cost and deployment time relative to gigawatt-scale plants. They are an emerging low-carbon power-infrastructure option, including for energy-intensive data centres.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:small-modular-reactors", "labels": [ "Small Modular Reactors" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "small-satellite", "title": "Small Satellite", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A small satellite is a spacecraft placed in a lower mass class than conventional large missions. The term is useful but not universal: agencies and market studies use different upper limits. NASA's small-spacecraft technology survey adopts 180 kilograms as a working boundary and subdivides the range into mini-, micro-, nano-, pico- and femtosatellites.[^1] That boundary belongs to NASA's scheme and should not be presented as an international legal definition.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:small-satellite", "labels": [ "Small Satellite" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [ "CubeSats" ] }, { "id": "small-solar-system-body", "title": "Small Solar System Body", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:small-solar-system-body", "labels": [ "Small Solar System Body" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "smart-building", "title": "Smart Building", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Smart Building is a facility that integrates networked sensors, actuators, control systems, and analytical software to automatically optimise energy consumption, occupant comfort, security, and operational efficiency in real time. Building systems \u2014 HVAC, lighting, access control, fire detection, lifts, and power distribution \u2014 are connected through a building management system (BMS) and increasingly exposed through open protocols such as BACnet, Modbus, MQTT, and OPC-UA to enable data-driven control strategies. Machine learning models analyse sensor streams to predict occupancy, detect anomalies, and schedule maintenance proactively. Smart buildings are fundamental nodes in [[Smart City]] infrastructure, integrating with district energy grids, electric vehicle charging, and urban mobility platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:smart-building", "labels": [ "Smart Building", "Smart Buildings" ], "is_subclass_of": [ "Building Automation" ], "wikilinks": [] }, { "id": "smart-cities", "title": "Smart Cities", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Smart Cities are urban environments that integrate digital technologies, data analytics, Internet of Things infrastructure, and connectivity to optimise the delivery of public services, improve quality of life, increase operational efficiency, and advance sustainability goals. They instrumentalise physical infrastructure \u2014 transport networks, utilities, buildings, and public spaces \u2014 with sensors and actuators connected through communication networks, and apply data-driven decision-making to city operations, planning, and citizen engagement.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:smart-cities", "labels": [ "Smart Cities" ], "is_subclass_of": [ "Digital Twin of Society (DToS)" ], "wikilinks": [] }, { "id": "smart-city-infrastructure", "title": "Smart City Infrastructure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Smart city infrastructure is the networked physical and digital substrate that enables a city to sense, communicate and respond to conditions in real time, integrating sensors, connectivity, edge and cloud computing, and data platforms across municipal systems. It underpins applications such as intelligent transport, smart grids, environmental monitoring and public-safety services by fusing data from distributed devices into analytics and control loops. The infrastructure couples Internet-of-Things endpoints with high-bandwidth networks and analytical platforms to improve efficiency, resilience and sustainability of urban services.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:smart-city-infrastructure", "labels": [ "Smart City Infrastructure" ], "is_subclass_of": [ "Smart City" ], "wikilinks": [] }, { "id": "smart-city", "title": "Smart City", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A smart city uses networked sensors, data platforms and analytics to manage urban systems such as transport, energy and public services, often paired with spatial digital twins and governed by open data standards and citizen-centric policies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:smart-city", "labels": [ "Smart City", "Smart City Ecosystem", "Smart City Platform" ], "is_subclass_of": [ "Computing and Cloud", "Spatial Computing Domain" ], "wikilinks": [ "Internet of Things", "Digital Twin", "Edge Computing", "Situational Awareness", "Spatial Computing Domain" ] }, { "id": "smart-contract-audit", "title": "Smart Contract Audit", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart contract audit is a structured security review of on-chain contract code that seeks to identify vulnerabilities, logic errors and economic flaws before deployment to an immutable ledger. Auditors combine manual code review with static analysis, automated scanners and, where warranted, formal verification to assess correctness against the intended specification. Because deployed contracts often custody substantial value and cannot easily be patched, auditing is a critical control in the blockchain and decentralised finance lifecycle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:smart-contract-audit", "labels": [ "Smart Contract Audit" ], "is_subclass_of": [ "Security Audit" ], "wikilinks": [] }, { "id": "smart-contract-coordination", "title": "Smart Contract Coordination", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Smart Contract Coordination is the use of blockchain smart contracts\u2014self-executing programs stored on distributed ledgers\u2014to automate coordination, task allocation, payment distribution, and milestone verification in distributed teams, enabling trustless collaboration through cryptographically enforced agreements that execute deterministically without centralised intermediaries. This approach removes the need for escrow agents and manual approvals by encoding collaboration rules as immutable on-chain logic triggered by verifiable real-world events. It is foundational to decentralised autonomous organisations and cryptocurrency-based remuneration workflows.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:smart-contract-coordination", "labels": [ "Smart Contract Coordination", "TELE-251-smart-contract-coordination" ], "is_subclass_of": [ "Protocol and Infrastructure", "Smart Contract" ], "wikilinks": [ "TELE-002-telecollaboration", "TELE-250-blockchain-collaboration", "TELE-252-dao-governance-telecollaboration", "TELE-253-cryptocurrency-remuneration", "Smart Contract", "SmartContracts" ] }, { "id": "smart-contract-deployment", "title": "Smart Contract Deployment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Smart contract deployment is the process of compiling contract source code to bytecode and publishing it to a blockchain, after which the contract acquires a fixed on-chain address and becomes callable by other accounts and contracts. It typically involves development frameworks that handle compilation, testing, gas estimation, and network submission. Deployment is irreversible on most chains, so verification and testing precede it as standard practice.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:smart-contract-deployment", "labels": [ "Smart Contract Deployment" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "smart-contract-enforcement", "title": "Smart Contract Enforcement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Smart contract enforcement is the property by which the terms encoded in a self-executing on-chain program are carried out automatically and irreversibly once predefined conditions are met, without requiring trust in a counterparty or intermediary. Enforcement derives from the deterministic execution of code on a distributed ledger and the immutability of recorded state. It reduces reliance on external legal recourse while raising questions of how code-based outcomes interact with conventional contract law.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:smart-contract-enforcement", "labels": [ "Smart Contract Enforcement" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "smart-contract-execution", "title": "smart contract execution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Smart contract execution is the deterministic, on-chain processing of immutable bytecode deployed at a specific blockchain address, triggered by an inbound transaction and evaluated identically by every validating node in the network. Execution occurs within a sandboxed virtual machine\u2014such as the Ethereum Virtual Machine (EVM) or Solana Virtual Machine (SVM)\u2014which enforces gas or fee metering to bound computation costs, maintains state isolation, and guarantees identical outputs across all nodes given the same inputs and world state. The execution model encompasses transaction ingestion, opcode interpretation, state transition application, event emission, and atomicity enforcement, with failed executions reverting all state changes. Smart contract execution underpins trustless multi-party agreements, decentralised finance protocols, non-fungible token operations, cross-chain messaging, and decentralised autonomous organisations without reliance on a central intermediary.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:smart-contract-execution", "labels": [ "Smart Contract Execution" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "smart-contract-governance", "title": "Smart Contract Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Smart contract governance is the set of mechanisms by which the rules, upgrades, and parameters of deployed on-chain contracts are proposed, decided, and applied, typically through token voting, multisig control, or upgrade proxies. It addresses how immutable code can nonetheless evolve safely and who holds authority to change protocol behaviour. Effective governance balances decentralisation against the need for timely security fixes and legal accountability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:smart-contract-governance", "labels": [ "Smart Contract Governance" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "smart-contract-layer", "title": "Smart Contract Layer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Smart Contract Layer is the stratum that holds self-executing programmes whose logic runs deterministically against agreed state. In the canonical stack it sits above the Middleware Layer and below the Application Layer, turning a shared ledger into a programmable platform. It contains contract code, virtual machine execution semantics, and the state these contracts read and write.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:smart-contract-layer", "labels": [ "Smart Contract Layer" ], "is_subclass_of": [ "Protocol and Consensus", "owl:Thing" ], "wikilinks": [ "Middleware Layer", "Consensus Layer", "Application Layer", "Ethereum Virtual Machine", "Decentralised Finance", "owl:Thing" ] }, { "id": "smart-contract-platform", "title": "Smart Contract Platform", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Smart Contract Platform is a blockchain-based infrastructure layer that provides a deterministic execution environment for self-executing programmable agreements, enabling decentralised application (dApp) development through virtual machine runtimes, developer toolchains, consensus-enforced state transitions, and economic incentive mechanisms such as gas metering. These platforms extend base-layer blockchains with Turing-complete or domain-specific scripting capabilities, allowing arbitrary business logic to be encoded and trustlessly enforced on a shared public ledger. Prominent examples include Ethereum and its EVM-compatible derivatives, Solana with its Sealevel parallel runtime, Cardano with its Plutus/eUTXO model, and Polkadot with its ink! WebAssembly runtime. Each platform makes distinct design trade-offs across the performance-security-decentralisation trilemma and imposes its own programming model, fee structure, and upgrade governance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:smart-contract-platform", "labels": [ "Smart Contract Platform" ], "is_subclass_of": [ "Blockchain", "Smart Contract" ], "wikilinks": [] }, { "id": "smart-contract-security", "title": "Smart Contract Security", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Smart contract security is the discipline of designing, reviewing, and verifying on-chain programs so they behave correctly and resist exploitation despite handling irreversible value transfers. Because deployed contracts are typically immutable and publicly visible, vulnerabilities such as reentrancy, access-control flaws, and arithmetic errors can lead to permanent loss of funds. The field combines secure coding patterns, automated analysis, formal verification, audits, and economic safeguards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:smart-contract-security", "labels": [ "Smart Contract Security" ], "is_subclass_of": [ "Security Audit" ], "wikilinks": [] }, { "id": "smart-contract", "title": "Smart Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Smart Contract is a self-executing digital program deployed on a blockchain that encodes contractual terms, business logic, and state-transition rules directly in code, automatically enforcing obligations when predetermined on-chain conditions are satisfied without requiring trusted interme...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:smart-contract", "labels": [ "Smart Contract", "Blockchain Smart Contract", "EVM Smart Contract", "Gateway Smart Contract", "Immutable Smart Contract", "Rental Smart Contract", "SmartContract", "Swap Smart Contract", "Token Smart Contract", "smart-contract" ], "is_subclass_of": [ "Network Component", "Blockchain", "Digital Agreement", "Executable Code", "Distributed Systems", "Protocol" ], "wikilinks": [ "ABI", "Atzei Bartoletti Cimoli 2017 SoK Attacks on Ethereum", "Bhargavan et al 2016 Formal Verification of Smart Contracts", "Blackshear et al 2019 Move Language", "Buterin 2013 Ethereum Whitepaper", "Cairo Language", "Centralised Database", "Certora Prover", "Certora Prover 2022", "Chainlink", "Compiler", "Constructor", "Contract Code", "Cryptographic Hash Function", "Cryptographic Proof", "CryptographyDomain", "Custodial Service", "Daian et al 2020 Flash Boys 2.0", "DeFi", "De Filippi Wright 2018 Blockchain and the Law" ] }, { "id": "smart-contracts", "title": "Smart Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Self-executing programs stored on a blockchain that automatically enforce and execute the terms of an agreement when predetermined conditions are met, eliminating the need for intermediaries and enabling trustless, transparent, and immutable transaction automation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:smart-contracts", "labels": [ "Smart Contracts", "BC-0013-smart-contracts", "BC-0142-smart-contract", "OpenZeppelin Contracts", "Smart Contracts on Bitcoin", "SmartContracts", "smart-contracts" ], "is_subclass_of": [ "Network Component", "Distributed Computing", "Blockchain" ], "wikilinks": [ "Cryptographic Hash Function", "Decentralized Application", "Automated Market Maker", "Blockchain", "Decentralized Finance (DeFi)", "Digital Signature", "Distributed Computing", "Ethereum", "Tokenization" ] }, { "id": "smart-grid", "title": "Smart Grid", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A smart grid is a modernised electricity distribution and transmission network that integrates digital communication, real-time sensing, automated control, and distributed intelligence to improve efficiency, reliability, resilience, and sustainability relative to the traditional centralised grid. It enables bidirectional power and information flows, accommodating distributed generation from renewable sources, battery energy storage systems, demand-side flexibility, and vehicle-to-grid interactions. Advanced metering infrastructure, distribution automation, phasor measurement units, and AI-driven optimisation allow operators to balance supply and demand dynamically across thousands of grid-edge resources. Smart grids are a foundational component of low-carbon energy infrastructure and are governed by a set of interoperability standards spanning communications, cybersecurity, and market protocols.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:smart-grid", "labels": [ "Smart Grid", "Smart Grid Integration", "Smart Grid Operations" ], "is_subclass_of": [ "Cyber Physical Systems" ], "wikilinks": [] }, { "id": "smart-home-automation", "title": "Smart Home Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Smart Home Automation is the networked control and coordination of domestic devices \u2014 lighting, heating, security, appliances, and entertainment \u2014 so that they can be monitored and operated remotely or triggered automatically by rules, schedules, and sensor input. It builds on Internet of Things connectivity, local hubs, and standard protocols to integrate heterogeneous devices into a single controllable environment. Voice assistants, mobile apps, and rule engines provide the user interface, while local processing increasingly handles latency-sensitive and privacy-sensitive logic. It enables energy management, accessibility, and convenience in residential settings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:smart-home-automation", "labels": [ "Smart Home Automation" ], "is_subclass_of": [ "Internet of Things" ], "wikilinks": [] }, { "id": "smart-home", "title": "Smart Home", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A smart home is a residence equipped with networked devices, sensors, and actuators that can be monitored and controlled remotely or automated to respond to conditions and occupant preferences. Smart homes integrate lighting, climate control, security, appliances, and entertainment through a common connectivity fabric and often a central hub or voice assistant. They build on Internet of Things technologies, wireless mesh networking, and interoperability standards to deliver convenience, energy efficiency, and accessibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:smart-home", "labels": [ "Smart Home" ], "is_subclass_of": [ "Internet Of Things" ], "wikilinks": [] }, { "id": "smart-manufacturing", "title": "Smart Manufacturing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Smart manufacturing is an integrated approach to industrial production that embeds advanced sensing, connectivity, data analytics, and artificial intelligence throughout the production system to achieve real-time visibility, adaptive control, and continuous optimisation of manufacturing processes. It combines cyber-physical systems, the Industrial Internet of Things, digital twins, machine learning, and cloud and edge computing to create intelligent factories capable of self-monitoring, predictive maintenance, and autonomous quality control. Smart manufacturing extends beyond factory automation to encompass supply chain integration, mass customisation, and sustainability optimisation, representing the operational realisation of the Industry 4.0 paradigm.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:smart-manufacturing", "labels": [ "Smart Manufacturing" ], "is_subclass_of": [ "Manufacturing Automation" ], "wikilinks": [] }, { "id": "smart-metering", "title": "Smart Metering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Smart Metering is the deployment of digital utility meters that record consumption of electricity, gas, water or heat at fine time resolution and communicate readings automatically to the utility and consumer. It replaces manual periodic readings with two-way communication, enabling near real-time monitoring, remote configuration and dynamic tariffs. Smart meters are a foundational component of advanced metering infrastructure and the broader smart grid.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:smart-metering", "labels": [ "Smart Metering" ], "is_subclass_of": [ "Energy Management" ], "wikilinks": [] }, { "id": "smart-royalties-ledger", "title": "Smart Royalties Ledger", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An automated tracking and distribution system that records creator royalty obligations, calculates payment amounts, and executes compensation transfers for digital content and NFT sales in virtual economy environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:smart-royalties-ledger", "labels": [ "Smart Royalties Ledger" ], "is_subclass_of": [ "Computing and Cloud", "Economic Layer" ], "wikilinks": [ "Audit Record", "Automated Creator Compensation", "Creator Economy Infrastructure", "Cross-Platform Attribution", "Distribution Queue", "ETSI GS MEC 003", "Identity System", "Oracle Service", "Payment Gateway", "Payment Tracking Engine", "Price Feed", "Royalty Calculator", "Settlement System", "Transaction Processor", "Treasury System", "Blockchain Network", "MiddlewareLayer", "Multi-Party Royalties", "Smart Royalty Contract", "Transparent Revenue Sharing" ] }, { "id": "smart-royalty-contract", "title": "Smart Royalty Contract", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A programmable blockchain contract that automatically enforces royalty payment terms, calculates compensation amounts, and triggers distributions to creators and rights holders upon qualifying transactions in NFT and digital asset ecosystems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:smart-royalty-contract", "labels": [ "Smart Royalty Contract", "Royalty Smart Contract" ], "is_subclass_of": [ "Governance and Safety", "Royalty Mechanism" ], "wikilinks": [ "Automated Payment Enforcement", "EIP-2981", "Enforcement Module", "Event Emitter", "Metadata Storage", "Multi-Recipient Distribution", "NFT Standard Implementation", "Oracle Service", "Payment Splitter", "Perpetual Creator Royalties", "Royalty Logic", "Royalty Registry", "Royalty Verification", "Wallet Infrastructure", "Blockchain Network", "Gas Fee Market", "MiddlewareLayer", "Payment Token", "Smart Contract Platform", "Token Standard" ] }, { "id": "snapshot-block", "title": "Snapshot Block", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A snapshot block is a specific block height at which token balances are recorded to determine voting weight in a governance proposal, freezing eligibility so that votes reflect holdings at one fixed moment rather than fluctuating balances. Using a past block prevents vote-buying or borrowing tokens after a proposal opens. It is a core primitive in off-chain and on-chain DAO voting systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:snapshot-block", "labels": [ "Snapshot Block", "Snapshot Block Mechanism" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "snapshot-governance", "title": "Snapshot Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Snapshot Governance is an off-chain voting mechanism used in decentralised autonomous organisations whereby token holders cast gasless ballots whose weight is determined by a verified snapshot of holdings at a specific block height. The system enables low-friction community governance without requiring on-chain transaction costs for each vote. Results are typically ratified by a multisig or on-chain executor that enforces the outcome. It represents a pragmatic compromise between full on-chain governance and centralised decision-making.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:snapshot-governance", "labels": [ "Snapshot Governance" ], "is_subclass_of": [ "Blockchain", "Governance and Regulation" ], "wikilinks": [] }, { "id": "snapshot-hub", "title": "Snapshot Hub", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Snapshot Hub is the backend service of the Snapshot off-chain governance platform that stores proposals, collects signed votes, and computes results using token balances read at a chosen block height. Votes are signed messages that incur no gas, while the hub indexes and tallies them according to a configurable voting strategy. It lets DAOs run gasless, verifiable governance while final execution remains optional and off-chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:snapshot-hub", "labels": [ "Snapshot Hub" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "snapshot-off-chain-voting", "title": "Snapshot Off-Chain Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A gasless governance platform that records votes off-chain by having participants sign messages, using token balances captured at a chosen block as voting weight. It avoids transaction fees while preserving a verifiable tally tied to on-chain holdings.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:snapshot-off-chain-voting", "labels": [ "Snapshot Off-Chain Voting", "Off-Chain Voting", "Snapshot Off-Chain Signed Voting" ], "is_subclass_of": [ "Decentralized Governance" ], "wikilinks": [ "Digital Signature", "Token", "DAOGovernance", "Governance", "Quadratic Voting", "Decentralized Governance", "https://snapshot.org/" ] }, { "id": "snapshot-voting", "title": "Snapshot Voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Snapshot Voting is the dominant off-chain gasless governance infrastructure operated by Snapshot Labs that enables DecentralizedAutonomousOrganization|decentralised autonomous organisations and DeFi protocols to conduct binding-or-advisory governance polls without spending GasFees|gas...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:snapshot-voting", "labels": [ "Snapshot Voting" ], "is_subclass_of": [ "Governance and Regulation", "Token and Asset", "DAO Governance", "Off-Chain Voting", "Cryptographic Signalling", "Gasless Governance", "Decentralised Voting", "Web3 Infrastructure" ], "wikilinks": [ "a16z Crypto 2023 State of Crypto Governance", "Aave", "Aave Governance", "Aave Governance AIP Archive", "ApeCoin DAO", "Aragon Governance", "Arbitrum DAO", "Arbitrum DAO Governance Documentation", "Barbereau et al 2022 DeFi Unregulated Governance", "Benet 2014 IPFS Specification", "Boardroom", "Boardroom DAO Governance Report 2024", "BrightID", "Buterin 2021 Moving Beyond Coin Voting Governance", "Colony Governance", "Compound Governor Bravo", "Cross-Chain Voting Aggregation", "CryptographicLayer", "Cryptographic Signalling", "DeFi" ] }, { "id": "snapshot", "title": "Snapshot", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Snapshot is an off-chain governance platform that lets token holders and decentralised autonomous organisations vote on proposals without paying on-chain transaction fees. Votes are signed cryptographically with a wallet and weighted according to token holdings recorded at a chosen block height, then aggregated and stored on the InterPlanetary File System. Because voting is gasless and non-binding at the protocol level, results are typically executed separately by multisignature wallets or on-chain modules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:snapshot", "labels": [ "Snapshot", "Snapshot Space", "Snapshot Strategy Specification" ], "is_subclass_of": [ "Decentralised Finance", "Decentralised Finance Domain" ], "wikilinks": [ "IPFS", "Ethereum", "Decentralised Autonomous Organisation", "Gnosis Safe", "Governance Domain", "Decentralised Finance Domain" ] }, { "id": "soc-2", "title": "Soc 2", "domain": "security", "domain_name": "Security", "definition": "SOC 2 (System and Organisation Controls 2) is an auditing standard developed by the American Institute of Certified Public Accountants (AICPA) that evaluates the controls of service organisations relevant to security, availability, processing integrity, confidentiality, and privacy. A SOC 2 report, issued by an independent CPA, is widely used by cloud service providers and SaaS companies to demonstrate trustworthiness to enterprise customers. Type I reports assess design at a point in time; Type II reports assess operating effectiveness over a period.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:soc-2", "labels": [ "Soc 2", "SOC 2" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "social-choice-theory", "title": "Social Choice Theory", "domain": "governance", "domain_name": "Governance", "definition": "Social choice theory is the formal study of how individual preferences can be aggregated into a collective decision, examining the properties, fairness, and impossibility constraints of voting rules and welfare functions. Foundational results such as Arrow's impossibility theorem show that no rank-aggregation rule can simultaneously satisfy a small set of seemingly reasonable fairness axioms. It provides the theoretical underpinning for voting mechanism design in both political and digital governance systems.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:social-choice-theory", "labels": [ "Social Choice Theory" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "social-consensus", "title": "Social Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Social consensus is the informal agreement reached among a blockchain's community of users, developers and node operators about the legitimate state or direction of a protocol, as distinct from the algorithmic consensus mechanism that orders transactions on-chain. It becomes decisive in situations that protocol rules alone cannot resolve, such as contested forks or disputed protocol governance changes, where the chain that the community recognises as legitimate prevails regardless of technical continuity. Social consensus is therefore the ultimate backstop of a decentralised network's legitimacy.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:social-consensus", "labels": [ "Social Consensus" ], "is_subclass_of": [ "Consensus" ], "wikilinks": [] }, { "id": "social-contract", "title": "Social Contract", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A philosophical and political theory positing that individuals consent, either explicitly or tacitly, to surrender certain freedoms and submit to authority in exchange for protection of their remaining rights and maintenance of social order.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:social-contract", "labels": [ "Social Contract" ], "is_subclass_of": [ "Legal and Regulatory", "Artificial Intelligence" ], "wikilinks": [ "Civil Society", "Constitutional Law", "Political Philosophy", "Artificial Intelligence", "Democratic Governance" ] }, { "id": "social-engineering", "title": "Social Engineering", "domain": "security", "domain_name": "Security", "definition": "Social engineering is the manipulation of people into divulging confidential information or performing actions that compromise security, exploiting human psychology rather than technical vulnerabilities. It includes techniques such as phishing, pretexting, baiting and impersonation that bypass technical controls by targeting trust, urgency and authority. It is one of the most effective and prevalent attack vectors in cybersecurity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:social-engineering", "labels": [ "Social Engineering" ], "is_subclass_of": [ "Attack Vector" ], "wikilinks": [] }, { "id": "social-impact-assessment-sia", "title": "Social Impact Assessment (SIA)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic eof potential social consequences of metaverse deployment on communities, stakeholder groups, and societal well-being.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:social-impact-assessment-sia", "labels": [ "Social Impact Assessment (SIA)" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Community Consultation", "Community Engagement", "Community Protection", "Compliance Management", "Data Collection", "Ethics Framework", "Impact Indicators", "Impact Metrics", "ISO 26000", "Policy Development", "Responsible Deployment", "Social Responsibility Policy", "Stakeholder Alignment", "Stakeholder Analysis", "Stakeholder Mapping", "UN SDG Toolkit", "Governance Framework", "MiddlewareLayer", "Risk Assessment", "TrustAndGovernanceDomain" ] }, { "id": "social-impact", "title": "Social Impact", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Social Impact refers to the full spectrum of effects \u2014 positive and negative, intended and unintended \u2014 that AI systems and digital technologies exert on individuals, communities, and social structures, encompassing changes to employment patterns, educational access, cultural practices, power distributions, social cohesion, and fundamental rights. Assessing social impact requires both quantitative metrics and qualitative analysis across affected stakeholder groups.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:social-impact", "labels": [ "Social Impact" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Ethan Mollick", "heiphetz2010training; @aldrich2005learning", "Nostr", "torok2017cascading", "Artificial Intelligence", "Cyber Security and Military", "Death of the Internet", "Deepfakes and fraudulent content", "Education and AI", "Equity", "Generative AI", "Humans, Avatars , Character", "Large Language Models", "legacy media", "Meta Platforms", "Metaverse and Telecollaboration", "MetaverseDomain", "Money", "Nostr protocol", "OpenAI" ] }, { "id": "social-interaction", "title": "Social Interaction", "domain": "ai", "domain_name": "Ai", "definition": "Social Interaction refers to the dynamic process by which two or more agents \u2014 human or artificial \u2014 mutually influence one another's behaviour, cognition, and emotional states through communicative acts, physical co-presence, or mediated channels. It is the fundamental unit of social life, encompassing verbal conversation, nonverbal signals, turn-taking, joint attention, empathic responsiveness, and negotiation of shared meaning. In the context of digital and AI systems, social interaction extends to human-computer interfaces, social robotics, virtual environments, and AI-mediated communication platforms, where the design of interaction modalities profoundly shapes social outcomes and wellbeing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:social-interaction", "labels": [ "Social Interaction", "Social Coordination", "Social Interaction Features", "Virtual Social Interaction" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "social-layer", "title": "Social Layer", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The Social Layer is the cross-cutting stratum that represents the human relationships, norms, and communities that surround and use a system. It sits above the institutional structures that formalise it and informs governance and application design. It contains community norms, reputation, communication channels, and the informal conventions that shape behaviour.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:social-layer", "labels": [ "Social Layer" ], "is_subclass_of": [ "Communication Technology", "owl:Thing" ], "wikilinks": [ "Institutional Layer", "Governance Layer", "Application Layer", "Social Capital", "Network Effect", "owl:Thing" ] }, { "id": "social-media-platform-infrastructure", "title": "Social Media Platform Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Digital platforms and applications that enable users to create, share, and interact with user-generated content within networked communities. Social media platforms combine content distribution infrastructure, algorithmic recommendation, identity and profile management, and real-time communication at scale. They are significant vectors for AI application (content moderation, recommendation, advertising optimisation) and raise governance challenges around bias, privacy, and misinformation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:social-media-platform-infrastructure", "labels": [ "Social Media Platform Infrastructure", "Social Media Algorithms", "Social Media Content", "Social Media Short-form", "Traditional Social Media Platform", "social media" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "social-navigation", "title": "Social Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Social Navigation is the discipline within robotics concerned with planning and executing robot motion through human-occupied spaces while respecting social norms, personal space conventions (proxemics), cultural context, and human comfort. It extends classical motion planning by incorporating models of human behaviour, pedestrian flow prediction, and non-verbal communication cues such as gaze and gesture, with the goal of producing movement that is perceived as natural and non-threatening.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:social-navigation", "labels": [ "Social Navigation" ], "is_subclass_of": [ "Human-Robot Interaction", "Robotics" ], "wikilinks": [ "HRI", "Mobile Robotics", "Robotics", "RoboticsDomain" ] }, { "id": "social-network-analysis", "title": "Social Network Analysis", "domain": "data", "domain_name": "Data", "definition": "Social network analysis (SNA) is a methodology for studying social structures through graph-theoretic and statistical techniques applied to networks of actors (nodes) and their relationships (edges). It measures structural properties such as centrality, clustering, path length, and community structure to identify influential actors, information bottlenecks, and emergent communities within social systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:social-network-analysis", "labels": [ "Social Network Analysis", "Social Graph Analysis" ], "is_subclass_of": [ "Network Analysis" ], "wikilinks": [] }, { "id": "social-network-graph", "title": "Social Network Graph", "domain": "data", "domain_name": "Data", "definition": "A graph-theoretic data structure in which nodes represent social actors (individuals, organisations, or automated agents) and edges encode directed or undirected social relations such as friendship, following, trust, or co-authorship. Social network graphs are analysed using network-science metrics including degree centrality, clustering coefficient, betweenness, and PageRank to reveal community structure, information diffusion pathways, and influential actors. They underpin decentralised identity federation protocols (ActivityPub, Nostr), recommendation engines, and adversarial analysis tasks such as Sybil detection. The formalism extends naturally to heterogeneous property graphs and hypergraphs when multi-typed relations or group memberships must be represented.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:social-network-graph", "labels": [ "Social Network Graph", "Open Social Graph", "Social Graph" ], "is_subclass_of": [ "Graph Data Model" ], "wikilinks": [ "owl:Thing" ] }, { "id": "social-platform", "title": "Social Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Social Platform is a networked software environment that mediates human connection, identity expression, content sharing, and community formation. It combines identity management, real-time communication, moderation tooling, and reputation mechanisms to sustain persistent social graphs, supporting both synchronous interaction (voice, video) and asynchronous content exchange across devices.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:social-platform", "labels": [ "Social Platform" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "social-presence-theory", "title": "Social Presence Theory", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"A theoretical framework explaining how communication media vary in their capacity to convey social cues (facial expressions, vocal intonation, body language, interpersonal warmth), thereby influencing the degree to which communicators perceive each other as psychologically present, real, and eng...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:social-presence-theory", "labels": [ "Social Presence Theory", "TELE-003-social-presence-theory" ], "is_subclass_of": [ "Telepresence", "CommunicationTheory" ], "wikilinks": [ "NonverbalCommunication", "TELE-002-telecollaboration", "TELE-004-media-richness-theory", "TELE-006-presence", "TELE-020-virtual-reality-telepresence", "TELE-100-ai-avatars", "TELE-110-spatial-audio-processing", "TELE-115-gaze-tracking", "CommunicationTheory", "TELE-001-telepresence" ] }, { "id": "social-presence", "title": "Social Presence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Subjective experience of feeling connected to and aware of other people in a mediated communication environment, fostering social interactions and relationships. Social presence is heightened by richer media, avatar fidelity, spatial audio, and shared immersive spaces, and is a key quality metric for telecollaboration and virtual world platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:social-presence", "labels": [ "Social Presence", "Digital Social Presence", "Social Presence System", "SocialPresence", "Virtual Social Presence" ], "is_subclass_of": [ "Telepresence", "Presence" ], "wikilinks": [ "Presence Research", "Presence", "Telecollaboration", "TelecollaborationDomain" ] }, { "id": "social-recovery", "title": "Social Recovery", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Social recovery is a wallet security mechanism that lets a user regain control of an account by relying on a set of trusted guardians rather than a single seed phrase. If a signing key is lost, a quorum of designated guardians can authorise a key rotation to a new owner address without ever holding the funds themselves. It mitigates catastrophic key loss while preserving self-custody, and is a flagship use case for smart-contract wallets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:social-recovery", "labels": [ "Social Recovery", "Social Recovery Wallet" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "social-robotics", "title": "Social Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "A field of robotics focused on designing robots capable of interacting and communicating with humans and other autonomous agents in a socially acceptable manner, following social rules, norms, and expectations of human interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:social-robotics", "labels": [ "Social Robotics", "Elder Care Robotics" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [ "Affective Computing", "AI", "Assistive Robotics", "Communication Capabilities", "Education", "Emotional Expression", "Healthcare", "Human", "Human-Robot Collaboration", "RB-1010-telepresence", "RB-1012-trust-in-automation", "Service Industry", "Social Agent", "Social Awareness", "Social Behavior", "Social Intelligence", "Social Sciences", "Computer Vision", "Human-Robot Interaction", "Natural Language Processing" ] }, { "id": "social-system", "title": "Social System", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A social system, in the context of virtual worlds, is the combination of features and rules that enable users to find, communicate with, and form relationships with one another, including presence, friend graphs, groups, voice and text chat, and reputation. It provides the connective fabric that turns a rendered space into an inhabited community. The design of these systems shapes safety, belonging, and the network effects that sustain a persistent virtual environment.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:social-system", "labels": [ "Social System" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "social-token-economy", "title": "Social Token Economy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Economic model where communities issue tokens representing reputation, participation value, or creator-fan relationships, enabling decentralized governance and value distribution.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:social-token-economy", "labels": [ "Social Token Economy" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Community Platform", "Community Token", "Creator Monetization", "Creator Token", "Fan Engagement", "Governance Rights", "Token Economy", "Token Economy Framework 2024", "Value Distribution", "Blockchain Infrastructure", "Community Governance", "Decentralized Exchange", "Digital Wallet", "MiddlewareLayer", "Reputation System", "Smart Contract Platform", "Token Standard", "VirtualEconomyDomain" ] }, { "id": "social-vr", "title": "Social VR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Social VR encompasses immersive virtual reality experiences specifically designed for real-time social interaction and collaboration among multiple users within shared persistent virtual spaces, leveraging avatar-based presence and synchronised communication.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:social-vr", "labels": [ "Social VR", "SocialVR" ], "is_subclass_of": [ "Virtual Reality" ], "wikilinks": [ "AltspaceVR", "AvatarInteraction", "BlockchainLedger", "CollaborativeSpace", "CommunityDAO", "DecentralisedIdentity", "DecentralizedID", "DecentralizedIdentity", "dt:authenticatedBy", "dt:enhancedBy", "dt:governedBy", "dt:monetizedVia", "dt:recordedOn", "enablesInteraction", "GestureRecognition", "hostsSession", "MetaHorizonWorlds", "NFT", "providesSpace", "supportsActivity" ] }, { "id": "societal-and-environmental-wellbeing", "title": "Societal and Environmental Wellbeing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Societal and Environmental Wellbeing is a trustworthiness dimension ensuring AI systems consider broader impacts on communities, environments, democratic processes, and human flourishing beyond immediate functional objectives. It promotes sustainable development, social cohesion, and alignment with the UN Sustainable Development Goals across the AI lifecycle.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:societal-and-environmental-wellbeing", "labels": [ "Societal and Environmental Wellbeing" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Ethics" ], "wikilinks": [ "Green AI", "IEA Data Center Report", "UN SDGs", "AIEthicsDomain", "ConceptualLayer", "EU AI Act" ] }, { "id": "sociotechnical-analysis", "title": "Sociotechnical Analysis", "domain": "governance", "domain_name": "Governance", "definition": "Sociotechnical analysis is a methodology that evaluates a technology by examining the interaction between its technical components and the human, organisational, and societal contexts in which it operates, rather than the artefact in isolation. Applied to AI, it traces how models, data, deployment settings, and affected communities jointly produce outcomes and harms. It is central to anticipating systemic risks that purely technical evaluation overlooks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sociotechnical-analysis", "labels": [ "Sociotechnical Analysis" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "sociotechnical-harm", "title": "Sociotechnical Harm", "domain": "governance", "domain_name": "Governance", "definition": "Sociotechnical harm is an adverse outcome that emerges from the interaction between a technical system and its social context rather than from a technical fault alone, such as representational harm, allocative discrimination, or erosion of public discourse. These harms are often diffuse, cumulative, and unevenly distributed across affected groups, making them hard to detect with conventional accuracy metrics. Identifying them requires attention to who is affected, how, and through what social mechanisms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:sociotechnical-harm", "labels": [ "Sociotechnical Harm" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "sociotechnical-risk", "title": "Sociotechnical Risk", "domain": "ai", "domain_name": "Ai", "definition": "Sociotechnical risk is the potential for harm arising from the interaction between a technical system and the social context in which it is built and used. In AI it covers harms that emerge from people, institutions and technology together rather than from code alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:sociotechnical-risk", "labels": [ "Sociotechnical Risk" ], "is_subclass_of": [ "AI Risk" ], "wikilinks": [ "Existential AI Risk", "AI Safety", "AI Risk" ] }, { "id": "sociotechnical-system", "title": "Sociotechnical System", "domain": "governance", "domain_name": "Governance", "definition": "A sociotechnical system is an analytical and design framework that treats technological artefacts and social structures as mutually constitutive elements of a unified system, recognising that technical components and human actors, organisations, and cultural practices co-evolve in ways that cannot be understood in isolation. Originating in organisational psychology and systems theory, it emphasises that optimising the technical subsystem alone is insufficient; effective system design requires joint optimisation of technical and social elements.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:sociotechnical-system", "labels": [ "Sociotechnical System" ], "is_subclass_of": [ "Complex Systems" ], "wikilinks": [] }, { "id": "soft-body-dynamics", "title": "Soft Body Dynamics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Soft Body Dynamics is the branch of physics simulation concerned with deformable objects \u2014 including cloth, flesh, elastic materials, vegetation, and fluids \u2014 that change shape in response to forces, collisions, and internal stresses. Unlike rigid body simulation, soft body methods must track per-vertex or per-element deformation states, typically using mass-spring networks, finite element methods (FEM), or position-based dynamics (PBD). It is essential for visual fidelity in real-time virtual environments, character animation, and embodied AI simulations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:soft-body-dynamics", "labels": [ "Soft Body Dynamics", "Soft Body Physics", "Soft Body Simulation" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "Game Physics", "Metaverse", "MetaverseDomain", "Physics Engine" ] }, { "id": "soft-fork", "title": "Soft Fork", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Soft Fork is a backward-compatible change to a blockchain's consensus rules in which the set of valid blocks is tightened so that newly produced blocks remain acceptable to non-upgraded nodes. Because old nodes still recognise the stricter blocks as valid, the network does not split provided a majority of hash power or stake enforces the new rules. Soft forks are commonly used to deploy protocol upgrades such as new script types without requiring every participant to update.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:soft-fork", "labels": [ "Soft Fork" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "soft-prompt-embedding", "title": "Soft Prompt Embedding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Soft Prompt Embedding is a parameter-efficient fine-tuning technique in which a small set of continuous, learnable token vectors (soft prompts) are prepended to the model's input embedding sequence and optimised via gradient descent, conditioning a frozen large language model's behaviour without modifying its weights. Unlike discrete (hard) prompts composed of natural-language tokens, soft prompt embeddings exist solely in the continuous embedding space and have no direct human-interpretable form. This approach enables task-specific adaptation of large models at a fraction of the computational and storage cost of full fine-tuning.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:soft-prompt-embedding", "labels": [ "Soft Prompt Embedding", "Soft Prompt Embeddings" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "soft-robotics", "title": "Soft Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robotics discipline employing compliant, flexible materials enabling safe human interaction and adaptation to unstructured environments, with applications across surgical robotics, food handling automation, wearable assistive devices, and collaborative manufacturing\u2014advancing through bio-inspired...", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:soft-robotics", "labels": [ "Soft Robotics", "SoftRobotics" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "3DPrinting", "AdaptiveGrasping", "BioInspiredDesign", "CompliantActuator", "controlledBy", "dt:controlledVia", "dt:designedWith", "dt:manufacturedBy", "dt:modeledBy", "dt:simulatedIn", "FiniteElementAnalysis", "GenerativeDesign", "hasActuator", "performsGrasping", "PneumaticActuation", "SafeInteraction", "usesMaterial", "MachineLearning", "PhysicsEngine", "RoboticsDomain" ] }, { "id": "soft-shadows", "title": "Soft Shadows", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Soft shadows are shadows with a graduated penumbra rather than a single hard edge, produced when a light source has physical area rather than being an idealised point. Rendering them requires sampling multiple points across the light's surface, or approximating the effect with techniques such as percentage-closer soft shadows, variance shadow maps, or area-light ray tracing. They are a significant contributor to perceived realism in rendered scenes because hard-edged shadows are rare under natural lighting conditions. Real-time soft shadow techniques trade accuracy for performance, while offline and ray-traced renderers can compute them with physically accurate light transport as part of global illumination.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:soft-shadows", "labels": [ "Soft Shadows" ], "is_subclass_of": [ "Shadow Mapping" ], "wikilinks": [] }, { "id": "softmax-function", "title": "Softmax Function", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The softmax function is a normalising transformation that maps a vector of real-valued scores (logits) into a probability distribution, where each output lies in the open interval (0, 1) and the outputs sum to one. It exponentiates each input and divides by the sum of all exponentials, amplifying larger scores while preserving rank order. Softmax is ubiquitous in machine learning as the final layer of multi-class classifiers and as the normalisation step inside attention mechanisms, and it pairs naturally with the cross-entropy loss whose gradient simplifies to the difference between predicted and target distributions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:softmax-function", "labels": [ "Softmax Function", "Softmax" ], "is_subclass_of": [ "Activation Function" ], "wikilinks": [] }, { "id": "software-architecture", "title": "Software Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Software Architecture for AI systems defines high-level structural patterns, component interactions, and design principles for building scalable, maintainable, and robust artificial intelligence applications. It encompasses microservices decomposition, event-driven designs, lambda and kappa architectures, feature stores, model registries, and observability pipelines, balancing modularity, reproducibility, and operational excellence.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:software-architecture", "labels": [ "Software Architecture", "Software Architecture Pattern" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Microservices", "MLOps", "System Design", "Distributed Systems" ] }, { "id": "software-as-a-service", "title": "Software As A Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software as a Service (SaaS) is a cloud computing delivery model in which fully functional applications are hosted and maintained by a provider and delivered to users over the internet via a web browser or thin client, with the provider managing all underlying infrastructure, platform, and application layers. SaaS eliminates the need for local installation and enables subscription-based access with automatic updates, multi-tenancy, and built-in scaling. Prominent examples include Salesforce CRM, Microsoft 365, Google Workspace, Slack, and Zoom.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:software-as-a-service", "labels": [ "Software As A Service", "Software as a Service", "Software-as-a-Service" ], "is_subclass_of": [ "Cloud Computing" ], "wikilinks": [] }, { "id": "software-bill-of-materials", "title": "Software Bill of Materials", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A software bill of materials is a formal, machine-readable inventory of the components, libraries and dependencies that make up a software artefact, together with their versions, suppliers and relationships. It provides transparency into what a piece of software actually contains, enabling vulnerability tracking, licence auditing and provenance verification across the supply chain. Standard formats such as SPDX and CycloneDX let an SBOM be generated, exchanged and consumed automatically by tooling.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:software-bill-of-materials", "labels": [ "Software Bill of Materials" ], "is_subclass_of": [ "Supply Chain Security" ], "wikilinks": [] }, { "id": "software-delivery-lifecycle", "title": "Software Delivery Lifecycle", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The software delivery lifecycle is the end-to-end sequence of stages a software change passes through, from planning and development through build, test, release, deployment and operation, typically supported by automated pipelines. It extends the traditional software development lifecycle by placing equal emphasis on the operational stages \u2014 continuous integration, continuous delivery, deployment and monitoring \u2014 that determine how reliably and quickly changes reach production. Organising these stages into a repeatable, largely automated lifecycle is a core practice of DevOps and underpins metrics such as deployment frequency and lead time for changes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:software-delivery-lifecycle", "labels": [ "Software Delivery Lifecycle" ], "is_subclass_of": [ "DevOps" ], "wikilinks": [] }, { "id": "software-development-automation", "title": "Software Development Automation", "domain": "ai", "domain_name": "Ai", "definition": "Software development automation is the use of tooling and AI agents to perform engineering tasks such as code generation, test writing, refactoring, dependency updates, and pull-request creation with minimal human intervention. Modern systems combine large language models with execution environments and version-control integrations to act on a codebase rather than merely suggest text. It aims to compress development cycles while keeping humans in the loop for review and intent.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:software-development-automation", "labels": [ "Software Development Automation", "Software Engineering Automation" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "software-development-kit", "title": "Software Development Kit", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A packaged collection of tools for building software against a specific platform, service, or hardware target, typically bundling libraries, application programming interfaces, compilers or build tooling, debuggers, emulators, documentation, and sample code, so that developers can create, test, and ship applications without assembling the platform-specific toolchain themselves.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:software-development-kit", "labels": [ "Software Development Kit" ], "is_subclass_of": [ "Application Software" ], "wikilinks": [ "Application Software", "Software Library", "Runtime Environment" ] }, { "id": "software-development-process", "title": "Software Development Process", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The structured set of methodologies, workflows, and best practices governing how software \u2014 including AI and machine learning systems \u2014 is conceived, built, tested, deployed, and maintained. For AI systems this encompasses data-centric workflows, experiment tracking, model validation protocols, MLOps pipelines, and cross-functional collaboration between data scientists, engineers, and domain experts across the full model lifecycle.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:software-development-process", "labels": [ "Software Development Process" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [ "Agile Development", "DevOps", "MLOps", "Version Control" ] }, { "id": "software-development", "title": "Software Development", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software development is the disciplined engineering practice of conceiving, specifying, designing, implementing, testing, deploying, and maintaining software systems through the coordinated application of programming languages, tooling, architectural patterns, and collaborative process frameworks. It encompasses the full software lifecycle \u2014 from requirements elicitation and system design through iterative coding, automated testing, continuous integration and delivery, and production operations \u2014 governed by methodologies such as Agile, Scrum, DevOps, and lean software development. The discipline has evolved from monolithic waterfall processes into fast-cycle, AI-augmented workflows in which developer tooling, cloud-native infrastructure, and open-source ecosystems are inseparable components of production capability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:software-development", "labels": [ "Software Development", "Interactive Software Development" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "software-engineering-agents", "title": "Software Engineering Agents", "domain": "ai", "domain_name": "Ai", "definition": "Software engineering agents are autonomous or semi-autonomous AI systems built on large language models that perform software development tasks \u2014 reading and editing codebases, running tests, debugging, and opening pull requests \u2014 by planning multi-step actions and invoking developer tools. They operate over real repositories using file navigation, shell execution, and version-control integration, and are evaluated on benchmarks such as SWE-bench that measure the rate at which they resolve genuine GitHub issues. They represent the application of agentic reasoning to the specific domain of writing and maintaining code.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:software-engineering-agents", "labels": [ "Software Engineering Agents", "Autonomous Software Engineering" ], "is_subclass_of": [ "LLM Agents" ], "wikilinks": [] }, { "id": "software-engineering-automation", "title": "Software Engineering Automation", "domain": "ai", "domain_name": "Ai", "definition": "Software engineering automation is the application of AI agents and tooling to perform software development tasks -- code generation, testing, refactoring and deployment -- with reduced human intervention. It builds on agent harnesses and autonomous task execution to let coding agents plan, execute and verify multi-step engineering work directly in a terminal or IDE. It underpins the current generation of terminal and IDE coding agents that operate over real codebases and toolchains.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:software-engineering-automation", "labels": [ "Software Engineering Automation" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "software-engineering", "title": "Software Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software Engineering is the disciplined application of systematic, quantifiable, and theoretically grounded approaches to the specification, design, development, testing, deployment, and maintenance of software systems. It draws on computer science, project management, and systems thinking to produce reliable, maintainable, scalable, and secure software at scale. The field encompasses a lifecycle spanning requirements elicitation, architectural design, implementation, verification and validation, and operational observability. Modern practice integrates agile and lean methodologies, continuous delivery pipelines, domain-driven design, and platform engineering to sustain high-velocity, high-quality development across distributed teams.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:software-engineering", "labels": [ "Software Engineering", "Software Engineering (Infrastructure)", "SoftwareEngineering" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [] }, { "id": "software-infrastructure", "title": "Software Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software Infrastructure comprises the foundational software layers \u2014 including operating systems, middleware, APIs, frameworks, and runtime environments \u2014 that underpin applications and services without being directly user-facing. It provides the shared services, communication channels, and execution contexts on which higher-level metaverse, AI, and distributed-systems applications are built, analogous to the role that physical infrastructure plays for built environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:software-infrastructure", "labels": [ "Software Infrastructure" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "software-library", "title": "Software Library", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Software Library is a collection of pre-compiled, reusable code modules that expose stable APIs, encapsulating common functionality such as rendering, physics simulation, networking, or cryptography. Libraries accelerate development by abstracting complexity, reducing defect rates, and enabling interoperability between components within a software stack.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:software-library", "labels": [ "Software Library" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "software-licence", "title": "Software Licence", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Software Licence is a legal instrument that governs the conditions under which software may be used, copied, modified, and distributed by parties other than the copyright holder. Licences range from permissive open-source agreements such as MIT and Apache 2.0 to copyleft licences such as GPL, and to proprietary end-user licence agreements. The licence type determines the legal obligations of users and downstream distributors with respect to attribution, source disclosure, and commercial use.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:software-licence", "labels": [ "Software Licence" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "software-moats", "title": "Software Moats", "domain": "economics", "domain_name": "Economics", "definition": "The durable competitive advantages, such as network effects, data accumulation, and switching costs, that protect software companies from competition and obsolescence.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:software-moats", "labels": [ "Software Moats" ], "is_subclass_of": [ "Software As A Service" ], "wikilinks": [] }, { "id": "software-platform", "title": "Software Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An integrated technology foundation providing common services, APIs, and infrastructure for developing, deploying, and running applications, including cloud-native platforms, container orchestration systems, and internal developer platforms that abstract underlying complexity while enabling scala...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:software-platform", "labels": [ "Software Platform" ], "is_subclass_of": [ "Platform and Environment", "Technology Infrastructure" ], "wikilinks": [ "Application Development", "metaverse", "Technology Infrastructure" ] }, { "id": "software-supply-chain", "title": "Software Supply Chain", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The software supply chain is the full set of components, processes, tools and actors involved in producing and delivering software, encompassing source code, third-party and open-source dependencies, build systems, package registries and deployment pipelines. Because modern applications assemble large amounts of external code, the integrity of every link matters for security and reliability. Securing it relies on practices such as software bills of materials, provenance attestation and dependency management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:software-supply-chain", "labels": [ "Software Supply Chain" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "software-system", "title": "Software System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An organised assembly of interacting software components \u2014 programs, libraries, services, configuration, and data \u2014 that together deliver a coherent set of functions on computing hardware for defined users and purposes. A software system is more than its source code: it encompasses runtime behaviour, interfaces and contracts with other systems, deployment environment, and the evolutionary pressures of maintenance, versioning, and backward compatibility over its operational lifetime.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:software-system", "labels": [ "Software System" ], "is_subclass_of": [ "System" ], "wikilinks": [ "System", "Algorithm", "Software Architecture", "Distributed System" ] }, { "id": "software-testing", "title": "Software Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software Testing is the systematic discipline of evaluating software systems and components by executing them under controlled conditions, observing emergent behaviour, and verifying conformance to specified requirements, design intent, and quality attributes. It encompasses a hierarchy of test granularity \u2014 from unit and integration tests that validate individual components to system, end-to-end, and acceptance tests that assess the complete product \u2014 as well as specialised dimensions such as performance, security, accessibility, regression, and exploratory testing. Testing is an integral quality assurance mechanism embedded throughout the software development lifecycle, supporting early defect detection, risk mitigation, and continuous delivery. Alongside static analysis and formal verification, it constitutes the primary empirical evidence base for software correctness and reliability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:software-testing", "labels": [ "Software Testing" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "software-defined-networking", "title": "Software-Defined Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Software-defined networking (SDN) is a network architecture that decouples the control plane, which decides how traffic is routed, from the data plane, which forwards packets, centralising control logic in a programmable software controller. By exposing the network through open interfaces, SDN allows traffic flows and policies to be configured dynamically and programmatically rather than device-by-device. This abstraction enables automation, virtualisation and centralised orchestration of network behaviour.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:software-defined-networking", "labels": [ "Software-Defined Networking", "Software Defined Networking" ], "is_subclass_of": [ "Network Architecture" ], "wikilinks": [] }, { "id": "soil-moisture-retrieval", "title": "Soil Moisture Retrieval", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:soil-moisture-retrieval", "labels": [ "Soil Moisture Retrieval" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solana", "title": "Solana", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Solana is a high-performance layer-1 blockchain designed by Anatoly Yakovenko and co-founded with Greg Fitzgerald, launched as mainnet-beta in 2020 by Solana Labs and the Solana Foundation. Its defining innovation, Proof of History, embeds a cryptographic clock \u2014 a verifiable delay function based on sequential SHA-256 hashing \u2014 into the ledger itself, enabling validators to agree on transaction ordering without peer-to-peer timestamp negotiation. This is complemented by the Sealevel parallel transaction execution runtime, the Gulf Stream mempool-less transaction forwarding protocol, Turbine block propagation, and the Tower BFT consensus algorithm, together enabling sustained throughputs of tens of thousands of transactions per second with sub-second finality at sub-cent fees. Solana underpins major DeFi ecosystems, NFT infrastructure, payments applications, and, more recently, AI-adjacent on-chain compute experiments.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:solana", "labels": [ "Solana" ], "is_subclass_of": [ "Smart Contract Platform", "Blockchain Network" ], "wikilinks": [] }, { "id": "solar-active-region", "title": "Solar Active Region", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-active-region", "labels": [ "Solar Active Region" ], "is_subclass_of": [ "Stellar Active Region" ], "wikilinks": [] }, { "id": "solar-cycle", "title": "Solar Cycle", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-cycle", "labels": [ "Solar Cycle" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-energetic-particle", "title": "Solar Energetic Particle", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-energetic-particle", "labels": [ "Solar Energetic Particle" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-flare", "title": "Solar Flare", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-flare", "labels": [ "Solar Flare" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-irradiance", "title": "Solar Irradiance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-irradiance", "labels": [ "Solar Irradiance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-magnetic-field", "title": "Solar Magnetic Field", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-magnetic-field", "labels": [ "Solar Magnetic Field" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-pv-installation-engineering-plan", "title": "Solar PV Installation Engineering Plan", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Solar plans are engineered configurations and optimisation analyses for solar photovoltaic (PV) installations, specifying panel tilt angles, array tiering, MPPT string layout, and economic projections for a given site's latitude and climate. This page documents a three-tier cascading solar panel system designed for Cumbria, England (54.5\u00b0N), where seasonal solar variation is extreme; the optimal configuration (43\u201345\u00b0 tilt, 20% horizontal overlap, 18 panels at 6.3 kW peak) achieves approximately 3,179 kWh/year \u2014 three times the output of a conventional single-layer array in the same 10 m\u00b2 footprint.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:solar-pv-installation-engineering-plan", "labels": [ "Solar PV Installation Engineering Plan", "solar plans" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "solar-physics", "title": "Solar Physics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-physics", "labels": [ "Solar Physics" ], "is_subclass_of": [ "Heliophysics" ], "wikilinks": [] }, { "id": "solar-radiation-pressure", "title": "Solar Radiation Pressure", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-radiation-pressure", "labels": [ "Solar Radiation Pressure" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-sail", "title": "Solar Sail", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-sail", "labels": [ "Solar Sail" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solar-wind", "title": "Solar Wind", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solar-wind", "labels": [ "Solar Wind" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solid-pod-storage", "title": "Solid Pod Storage", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Solid Pod Storage refers to the personal online data stores used in the Solid project, a web specification initiated by Tim Berners-Lee for decentralising data ownership. A Pod is a user-controlled storage space that holds data as linked-data resources, with access governed by the user rather than the applications that consume it. Applications authenticate against a Pod and read or write data only with the owner's permission, separating data from the services built on top of it.", "entityType": "Class", "qualityScore": 0.6, "maturity": "developing", "iri": "urn:ngm:class:solid-pod-storage", "labels": [ "Solid Pod Storage" ], "is_subclass_of": [ "Distributed Systems", "Distributed Systems Domain" ], "wikilinks": [ "Linked Data", "RDF", "WebID", "Data Sovereignty", "Decentralised Identity", "Semantic Web", "Self-Sovereign Identity", "Distributed Systems Domain" ] }, { "id": "solid-pod", "title": "Solid Pod", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Solid Pod (Personal Online Data store) is a W3C Linked Data Platform-compliant personal data vault that gives individuals and organisations sovereign control over their own data by hosting it at a user-chosen location and exposing it through a standards-based access control layer (WAC or ACP), ...", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:solid-pod", "labels": [ "Solid Pod", "Pod" ], "is_subclass_of": [ "Data Management", "Network and Communication", "Personal Data Store", "Linked Data Platform Resource" ], "wikilinks": [ "Access Control Policy", "ACP Access Control Policy", "Agent Personalisation", "AI Domain", "CSS Community Solid Server", "Data Sovereignty", "Decentralised Identity", "Decentralised Web Stack", "DPoP", "HTTP Protocol", "IdentityLayer", "Inrupt Pod Spaces", "Interoperable Personal Data", "LDP Container", "Linked Data Platform Resource", "Node Solid Server", "RDF Data Model", "RDF Resource", "RuVector Memory", "Solid Application Interoperability Specification" ] }, { "id": "solid-rocket-motor", "title": "Solid Rocket Motor", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:solid-rocket-motor", "labels": [ "Solid Rocket Motor" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "solid-oidc", "title": "Solid-OIDC", "domain": "data", "domain_name": "Data", "definition": "Solid-OIDC is the authentication specification for the Solid ecosystem, extending OpenID Connect with decentralised identity so a user authenticates with their own WebID and identity provider rather than a centralised platform. It issues DPoP-bound access tokens that resource servers (Solid Pods) verify, enabling a user to control which applications access their personal data store. It is the identity layer that makes Solid's separation of identity, data, and application possible.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:solid-oidc", "labels": [ "Solid-OIDC" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "solid", "title": "Solid", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Solid (Social Linked Data) is an open-source, W3C-specification-grounded decentralised web architecture conceived by Sir Tim Berners-Lee at MIT CSAIL (Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory) commercialised through the startup Inrupt and steer...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:solid", "labels": [ "Solid", "Solid Ecosystem", "Solid Protocol", "W3C Solid", "W3C Solid Protocol" ], "is_subclass_of": [ "Network and Communication", "Data Sovereignty", "Decentralized Web", "Linked Data", "Semantic Web Linked Data Standard", "Web Standards", "Privacy by Design", "Decentralised Protocol" ], "wikilinks": [ "Access Control Policy", "ActivityPub", "AT Protocol", "Blockchain Storage", "Centralised Data Silo", "Data Portability", "Data Sovereignty", "Decentralised Protocol", "DecentralisedWebDomain", "Decentralized Identity", "Decentralized Web", "DPoP Authentication", "DPoP Token", "Financial Data Portability", "FOAF", "GDPR", "GDPR Compliance", "Government Open Data", "Health Data Management", "HTTP" ] }, { "id": "solidity", "title": "Solidity", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Solidity is a statically-typed, curly-brace, contract-oriented programming language designed for writing smart contracts that execute on the Ethereum Virtual Machine (EVM) and EVM-compatible blockchain platforms. Released in 2014 by the Ethereum Foundation, it compiles to EVM bytecode and provides constructs including inheritance, interfaces, libraries, events, function modifiers, and custom error types. The language's semantics are shaped by on-chain execution constraints: every opcode costs gas, state is globally persistent, and deployed code is immutable absent explicit upgrade patterns. Solidity is the dominant language for decentralised finance, non-fungible tokens, and decentralised autonomous organisations, and its security vulnerability surface has driven a parallel industry of formal verification, auditing frameworks, and defensive programming libraries.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:solidity", "labels": [ "Solidity", "Solidity Contract", "Solidity Smart Contracts" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "solo-mining", "title": "Solo Mining", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Solo Mining is a blockchain participation strategy in which an individual miner independently operates hashing hardware and attempts to discover valid blocks without joining a mining pool, retaining the full block reward upon success but accepting high variance in earnings proportional to the miner's share of total network hash rate. It contrasts with pool mining by preserving full decentralisation of block production but is economically viable only when a miner controls a significant fraction of total hash power.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:solo-mining", "labels": [ "Solo Mining" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Consensus Protocol", "ConsensusProtocol" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "ConsensusDomain", "ConsensusProtocol", "ProtocolLayer" ] }, { "id": "soulbound-tokens", "title": "Soulbound Tokens", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Non-transferable tokens bound to a single account that represent credentials, affiliations or reputation rather than tradable value. Because they cannot be sold or moved, they encode persistent attributes of an identity.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:soulbound-tokens", "labels": [ "Soulbound Tokens", "Soulbound Token" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [ "Token Standard", "Sybil Resistance", "Verifiable Credentials", "Non-Fungible Token", "Identity", "Decentralized Identity", "https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4105763" ] }, { "id": "sound-money-principles", "title": "Sound Money Principles", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Sound money principles are the criteria by which a monetary asset is judged a reliable store of value and medium of exchange: scarcity, durability, divisibility, portability, fungibility, verifiability, and resistance to debasement by any issuer. Rooted in classical and Austrian monetary economics, they are frequently invoked to argue that a fixed-supply, credibly neutral asset preserves purchasing power over time. They form the economic case advanced for Bitcoin as a monetary good.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sound-money-principles", "labels": [ "Sound Money Principles" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "sound-money", "title": "Sound Money", "domain": "finance", "domain_name": "Finance", "definition": "Sound Money refers to a monetary medium characterised by a stable, predictable supply that cannot be arbitrarily expanded by a government or central authority, thereby preserving purchasing power over time and functioning as a reliable store of value. The concept is historically associated with commodity-backed currencies\u2014particularly the classical gold standard\u2014and in the contemporary context with Bitcoin's algorithmically fixed supply schedule. Sound money theorists, drawing heavily on the Austrian School, argue that monetary expansion through credit creation distorts price signals, misallocates capital, and produces inflationary cycles that systematically transfer wealth from savers to debtors and from the public to the financial sector. The notion stands in direct opposition to the discretionary monetary policy frameworks of modern central banking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sound-money", "labels": [ "Sound Money" ], "is_subclass_of": [ "Monetary System" ], "wikilinks": [] }, { "id": "source-routing", "title": "Source Routing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Source routing is a packet-forwarding technique in which the sender, rather than intermediate routers, specifies all or part of the path a packet must traverse, embedding the route as a list of hops within the packet itself. By moving path selection to the originator it enables explicit path control, multipath delivery, and privacy-preserving forwarding, at the cost of overhead and forwarding-table-bypass security considerations. It underpins onion-routing anonymity networks and payment-channel networks, where the sender constructs and conceals an end-to-end route across an overlay before any hop sees the full path.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:source-routing", "labels": [ "Source Routing" ], "is_subclass_of": [ "Routing" ], "wikilinks": [] }, { "id": "sovereign-ai", "title": "sovereign ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "National strategies for building indigenous AI capabilities\u2014foundational models trained on sovereign data using domestically controlled compute infrastructure\u2014to reduce dependence on foreign hyperscaler platforms. Encompasses government investment in national AI compute clusters, local language models, and data governance frameworks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:sovereign-ai", "labels": [ "Sovereign AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "sovereign-keyset", "title": "Sovereign Keyset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The cryptographic key material (BIP-340 Schnorr Keypair|BIP-340 Schnorr keypair) held securely by each VisionClaw Agentic Container|VisionClaw agent, used to prove identity via DID Nostr Identity|did:nostr DIDs, sign Verifiable Credential Surface|verifiable credentials, authentica...", "entityType": "Class", "qualityScore": 0.86, "maturity": "established", "iri": "urn:ngm:class:sovereign-keyset", "labels": [ "Sovereign Keyset" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [ "ADR-013", "AgenticSystemsDomain", "BIP-340", "BIP-340 Schnorr Keypair", "BIP-340 Schnorr Keypair", "BIP-340 Schnorr Keypair", "Blockchain Keypair", "Credential Issuance", "CryptographyDomain", "Decentralised Trust", "Hardware Security Module (optional)", "IdentityDomain", "Key Rotation Policy", "Key Storage", "Message Signing", "Nostr Relay", "Nostr Relay", "Peer Authentication", "Schnorr Signature Scheme", "secp256k1" ] }, { "id": "sovereign-mesh", "title": "Sovereign Mesh", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A peer-to-peer agent network where each node is identified by a DID Nostr Identity|did:nostr: DIDs and communicates via Nostr Relay|Nostr relays, enabling decentralised agent discovery, gossip, and work distribution without a centralised broker or coordinator, whilst respecting ea...", "entityType": "Class", "qualityScore": 0.87, "maturity": "established", "iri": "urn:ngm:class:sovereign-mesh", "labels": [ "Sovereign Mesh" ], "is_subclass_of": [ "Network Component", "Peer-to-Peer Network" ], "wikilinks": [ "ADR-009", "Agent Beacon", "Agent Beacon", "Agent Bead", "AgenticSystemsDomain", "BIP-340 Schnorr Keypair", "CoordinationLayer", "Decentralised Agent Coordination", "Decentralised Autonomous Organisation", "Decentralised Coordination", "Federation", "Gossip Consensus", "Message Signing", "Network Connectivity", "NIP-01 Nostr Protocol", "Nostr Protocol (NIP-01)", "Nostr Relay", "Nostr Relay", "P2PNetworkDomain", "PRD-006" ] }, { "id": "space-biology", "title": "Space Biology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-biology", "labels": [ "Space Biology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-data-centers", "title": "Space Data Centers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Orbital infrastructure designed to host high-density AI compute clusters, leveraging space-based energy and thermal management to scale beyond terrestrial constraints.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:space-data-centers", "labels": [ "Space Data Centers" ], "is_subclass_of": [ "Computational Infrastructure" ], "wikilinks": [] }, { "id": "space-environment", "title": "Space Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-environment", "labels": [ "Space Environment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-life-sciences-dataset", "title": "Space Life Sciences Dataset", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-life-sciences-dataset", "labels": [ "Space Life Sciences Dataset" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "space-life-sciences", "title": "Space Life Sciences", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-life-sciences", "labels": [ "Space Life Sciences" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-medicine", "title": "Space Medicine", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-medicine", "labels": [ "Space Medicine" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-physiology", "title": "Space Physiology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-physiology", "labels": [ "Space Physiology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-plasma", "title": "Space Plasma", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-plasma", "labels": [ "Space Plasma" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-probe", "title": "Space Probe", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-probe", "labels": [ "Space Probe" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "space-radiation-environment", "title": "Space Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-radiation-environment", "labels": [ "Space Radiation Environment" ], "is_subclass_of": [ "Space Environment" ], "wikilinks": [] }, { "id": "space-radiation", "title": "Space Radiation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-radiation", "labels": [ "Space Radiation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-reference-systems", "title": "Space Reference Systems", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-reference-systems", "labels": [ "Space Reference Systems" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-research-mission", "title": "Space Research Mission", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-research-mission", "labels": [ "Space Research Mission" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-robotics", "title": "Space Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Space robotics is the engineering discipline concerned with the design, development, and operation of robotic systems that function in the space environment, including planetary rovers, orbital manipulators, free-flying servicer spacecraft, and landers. These systems must contend with extreme thermal cycling, vacuum, radiation, communications latency, and the absence of gravity, requiring high degrees of autonomy, fault tolerance, and dexterous manipulation capability. Space robotics intersects closely with teleoperation research, autonomous navigation, computer vision, and the emerging field of on-orbit servicing and assembly.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:space-robotics", "labels": [ "Space Robotics" ], "is_subclass_of": [ "Autonomous Robot" ], "wikilinks": [] }, { "id": "space-science-and-systems", "title": "Space Science and Systems", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-science-and-systems", "labels": [ "Space Science and Systems" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-science", "title": "Space Science", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-science", "labels": [ "Space Science" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-segment", "title": "Space Segment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-segment", "labels": [ "Space Segment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-situational-awareness", "title": "Space Situational Awareness", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "Space situational awareness is the ability to identify, understand and assess conditions and hazards affecting space activity. In institutional use it can cover space surveillance and tracking, near-Earth objects and space weather.[^1] The related term **space domain awareness** often adds intent, behaviour and national-security interpretation. Authors should state which scope they mean rather than treating the terms as exact synonyms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-situational-awareness", "labels": [ "Space Situational Awareness" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-station", "title": "Space Station", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-station", "labels": [ "Space Station" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "space-systems-engineering", "title": "Space Systems Engineering", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-systems-engineering", "labels": [ "Space Systems Engineering" ], "is_subclass_of": [ "Systems Engineering" ], "wikilinks": [] }, { "id": "space-traffic-management", "title": "Space Traffic Management", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-traffic-management", "labels": [ "Space Traffic Management" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-weather-forecasting", "title": "Space Weather Forecasting", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-weather-forecasting", "labels": [ "Space Weather Forecasting" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-weather", "title": "Space Weather", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-weather", "labels": [ "Space Weather" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-to-ground-link", "title": "Space-to-ground Link", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-to-ground-link", "labels": [ "Space-to-ground Link" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "space-to-space-link", "title": "Space-to-space Link", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:space-to-space-link", "labels": [ "Space-to-space Link" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceborne-lidar", "title": "Spaceborne Lidar", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceborne-lidar", "labels": [ "Spaceborne Lidar" ], "is_subclass_of": [ "Earth Observation Instrument" ], "wikilinks": [] }, { "id": "spacecraft-attitude", "title": "Spacecraft Attitude", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-attitude", "labels": [ "Spacecraft Attitude" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-autonomy", "title": "Spacecraft Autonomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-autonomy", "labels": [ "Spacecraft Autonomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-battery-depth-of-discharge", "title": "Spacecraft Battery Depth of Discharge", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-battery-depth-of-discharge", "labels": [ "Spacecraft Battery Depth of Discharge" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-battery-thermal-runaway", "title": "Spacecraft Battery Thermal Runaway", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-battery-thermal-runaway", "labels": [ "Spacecraft Battery Thermal Runaway" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-battery", "title": "Spacecraft Battery", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-battery", "labels": [ "Spacecraft Battery" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-bus", "title": "Spacecraft Bus", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-bus", "labels": [ "Spacecraft Bus" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-communications", "title": "Spacecraft Communications", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-communications", "labels": [ "Spacecraft Communications" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-data-bus", "title": "Spacecraft Data Bus", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-data-bus", "labels": [ "Spacecraft Data Bus" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-detumbling", "title": "Spacecraft Detumbling", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-detumbling", "labels": [ "Spacecraft Detumbling" ], "is_subclass_of": [ "Attitude Control" ], "wikilinks": [] }, { "id": "spacecraft-docking", "title": "Spacecraft Docking", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-docking", "labels": [ "Spacecraft Docking" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-electrical-power-system", "title": "Spacecraft Electrical Power System", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-electrical-power-system", "labels": [ "Spacecraft Electrical Power System" ], "is_subclass_of": [ "Spacecraft Subsystem" ], "wikilinks": [] }, { "id": "spacecraft-environmental-testing", "title": "Spacecraft Environmental Testing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-environmental-testing", "labels": [ "Spacecraft Environmental Testing" ], "is_subclass_of": [ "Testing" ], "wikilinks": [] }, { "id": "spacecraft-guidance-navigation-and-control", "title": "Spacecraft Guidance Navigation and Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-guidance-navigation-and-control", "labels": [ "Spacecraft Guidance Navigation and Control" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-guidance", "title": "Spacecraft Guidance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-guidance", "labels": [ "Spacecraft Guidance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-habitat", "title": "Spacecraft Habitat", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-habitat", "labels": [ "Spacecraft Habitat" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-heater", "title": "Spacecraft Heater", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-heater", "labels": [ "Spacecraft Heater" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-housekeeping-data", "title": "Spacecraft Housekeeping Data", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-housekeeping-data", "labels": [ "Spacecraft Housekeeping Data" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "spacecraft-integration", "title": "Spacecraft Integration", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-integration", "labels": [ "Spacecraft Integration" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-magnetometer", "title": "Spacecraft Magnetometer", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-magnetometer", "labels": [ "Spacecraft Magnetometer" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "spacecraft-mechanism", "title": "Spacecraft Mechanism", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-mechanism", "labels": [ "Spacecraft Mechanism" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-module", "title": "Spacecraft Module", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-module", "labels": [ "Spacecraft Module" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-navigation-filter", "title": "Spacecraft Navigation Filter", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-navigation-filter", "labels": [ "Spacecraft Navigation Filter" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-navigation", "title": "Spacecraft Navigation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-navigation", "labels": [ "Spacecraft Navigation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-payload", "title": "Spacecraft Payload", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A spacecraft payload is the equipment that performs the mission's intended scientific, commercial, operational or technology-demonstration function. The spacecraft bus supports it with structure, power, thermal control, pointing, communications and data handling.[^1] The boundary depends on mission purpose: a star tracker is usually a bus attitude sensor, but can be flown as an experimental payload when testing the tracker itself.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-payload", "labels": [ "Spacecraft Payload" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-phase-change-thermal-storage", "title": "Spacecraft Phase Change Thermal Storage", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-phase-change-thermal-storage", "labels": [ "Spacecraft Phase Change Thermal Storage" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-pointing", "title": "Spacecraft Pointing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-pointing", "labels": [ "Spacecraft Pointing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-power-budget", "title": "Spacecraft Power Budget", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-power-budget", "labels": [ "Spacecraft Power Budget" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-power-and-thermal-control", "title": "Spacecraft Power and Thermal Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-power-and-thermal-control", "labels": [ "Spacecraft Power and Thermal Control" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-propulsion", "title": "Spacecraft Propulsion", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "Spacecraft propulsion produces controlled forces or impulses after launch. It changes a spacecraft's orbit or trajectory, maintains its assigned orbit, supports attitude control, avoids collisions, performs rendezvous and provides an end-of-life disposal manoeuvre. The propulsion architecture is chosen with the mission because thrust, total impulse, manoeuvre duration, power, mass, volume, safety and lifetime cannot be optimised independently.[^1]", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-propulsion", "labels": [ "Spacecraft Propulsion" ], "is_subclass_of": [], "wikilinks": [ "Electric Propulsion" ] }, { "id": "spacecraft-proximity-operations", "title": "Spacecraft Proximity Operations", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-proximity-operations", "labels": [ "Spacecraft Proximity Operations" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-radiator", "title": "Spacecraft Radiator", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-radiator", "labels": [ "Spacecraft Radiator" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-ranging", "title": "Spacecraft Ranging", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-ranging", "labels": [ "Spacecraft Ranging" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-redundancy", "title": "Spacecraft Redundancy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-redundancy", "labels": [ "Spacecraft Redundancy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-reliability", "title": "Spacecraft Reliability", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-reliability", "labels": [ "Spacecraft Reliability" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-rendezvous", "title": "Spacecraft Rendezvous", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-rendezvous", "labels": [ "Spacecraft Rendezvous" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-solar-array", "title": "Spacecraft Solar Array", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-solar-array", "labels": [ "Spacecraft Solar Array" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-solar-cell", "title": "Spacecraft Solar Cell", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-solar-cell", "labels": [ "Spacecraft Solar Cell" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-structure", "title": "Spacecraft Structure", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-structure", "labels": [ "Spacecraft Structure" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-subsystem", "title": "Spacecraft Subsystem", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-subsystem", "labels": [ "Spacecraft Subsystem" ], "is_subclass_of": [ "System" ], "wikilinks": [] }, { "id": "spacecraft-telemetry", "title": "Spacecraft Telemetry", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-telemetry", "labels": [ "Spacecraft Telemetry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-thermal-coating", "title": "Spacecraft Thermal Coating", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-thermal-coating", "labels": [ "Spacecraft Thermal Coating" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-thermal-control", "title": "Spacecraft Thermal Control", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-thermal-control", "labels": [ "Spacecraft Thermal Control" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-thermal-model", "title": "Spacecraft Thermal Model", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-thermal-model", "labels": [ "Spacecraft Thermal Model" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-transponder", "title": "Spacecraft Transponder", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-transponder", "labels": [ "Spacecraft Transponder" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-waste-management", "title": "Spacecraft Waste Management", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-waste-management", "labels": [ "Spacecraft Waste Management" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft-water-recovery", "title": "Spacecraft Water Recovery", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft-water-recovery", "labels": [ "Spacecraft Water Recovery" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spacecraft", "title": "Spacecraft", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spacecraft", "labels": [ "Spacecraft" ], "is_subclass_of": [ "System" ], "wikilinks": [] }, { "id": "spaced-repetition", "title": "Spaced Repetition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Spaced repetition is a learning technique that schedules reviews of material at increasing intervals timed to occur just before predicted forgetting, exploiting the psychological spacing effect to strengthen long-term retention efficiently. Algorithmic implementations track each item's review history and recall outcomes to compute the optimal next-review date, concentrating study effort on material at risk of being forgotten. It is a core mechanism in flashcard software and the memory-management components of adaptive learning systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:spaced-repetition", "labels": [ "Spaced Repetition" ], "is_subclass_of": [ "Adaptive Learning" ], "wikilinks": [] }, { "id": "spacefaring-data-centers", "title": "Spacefaring Data Centers", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Proposed orbital or space-based computational facilities designed to host high-density AI training and inference workloads.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spacefaring-data-centers", "labels": [ "Spacefaring Data Centers" ], "is_subclass_of": [ "AI Training Infrastructure" ], "wikilinks": [] }, { "id": "spaceflight-analogue", "title": "Spaceflight Analogue", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-analogue", "labels": [ "Spaceflight Analogue" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-biospecimen", "title": "Spaceflight Biospecimen", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-biospecimen", "labels": [ "Spaceflight Biospecimen" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-bone-loss", "title": "Spaceflight Bone Loss", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-bone-loss", "labels": [ "Spaceflight Bone Loss" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-cardiovascular-adaptation", "title": "Spaceflight Cardiovascular Adaptation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-cardiovascular-adaptation", "labels": [ "Spaceflight Cardiovascular Adaptation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-circadian-rhythm", "title": "Spaceflight Circadian Rhythm", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-circadian-rhythm", "labels": [ "Spaceflight Circadian Rhythm" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-experiment", "title": "Spaceflight Experiment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-experiment", "labels": [ "Spaceflight Experiment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-fluid-shift", "title": "Spaceflight Fluid Shift", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-fluid-shift", "labels": [ "Spaceflight Fluid Shift" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-immune-response", "title": "Spaceflight Immune Response", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-immune-response", "labels": [ "Spaceflight Immune Response" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-isolation", "title": "Spaceflight Isolation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-isolation", "labels": [ "Spaceflight Isolation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-microbiology", "title": "Spaceflight Microbiology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-microbiology", "labels": [ "Spaceflight Microbiology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-muscle-atrophy", "title": "Spaceflight Muscle Atrophy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-muscle-atrophy", "labels": [ "Spaceflight Muscle Atrophy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-omics", "title": "Spaceflight Omics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-omics", "labels": [ "Spaceflight Omics" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-plant-biology", "title": "Spaceflight Plant Biology", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-plant-biology", "labels": [ "Spaceflight Plant Biology" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-programme", "title": "Spaceflight Programme", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-programme", "labels": [ "Spaceflight Programme" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight-associated-neuro-ocular-syndrome", "title": "Spaceflight-associated Neuro-ocular Syndrome", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight-associated-neuro-ocular-syndrome", "labels": [ "Spaceflight-associated Neuro-ocular Syndrome" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spaceflight", "title": "Spaceflight", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spaceflight", "labels": [ "Spaceflight" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spam-prevention", "title": "Spam Prevention", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Spam prevention is the set of mechanisms that deter the flooding of a shared system with low-value or malicious activity. In blockchain networks it most commonly takes the form of transaction fees or proof-of-work requirements that impose a real cost on each submission, making large-scale spamming economically or computationally infeasible. Spam prevention protects scarce resources such as block space, mempool capacity and validator compute from denial-of-service style abuse. It is a key design constraint that interacts with throughput, fee markets and decentralisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:spam-prevention", "labels": [ "Spam Prevention" ], "is_subclass_of": [ "Blockchain", "Sybil Resistance" ], "wikilinks": [] }, { "id": "sparql-query-engine", "title": "Sparql Query Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A processor implementing the SPARQL Protocol and RDF Query Language (W3C standard) for retrieving, manipulating, and joining data stored in Resource Description Framework graphs, enabling semantic web queries through pattern matching of subject-predicate-object triples with support for filtering,...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:sparql-query-engine", "labels": [ "Sparql Query Engine" ], "is_subclass_of": [ "Standards and Interoperability", "Query Processor" ], "wikilinks": [ "Semantic Data Retrieval", "metaverse", "Query Processor" ] }, { "id": "sparse-autoencoders", "title": "Sparse Autoencoders", "domain": "ai", "domain_name": "Ai", "definition": "Sparse autoencoders are neural networks trained to reconstruct their input through a wide hidden layer subject to a sparsity penalty, so that only a small number of latent units activate for any given input. In mechanistic interpretability they are applied to the activations of large language models to decompose dense, polysemantic representations into more monosemantic, human-interpretable features. They have become a leading tool for understanding what concepts a model internally represents.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:sparse-autoencoders", "labels": [ "Sparse Autoencoders", "Sparse Autoencoder" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "sparse-matrix", "title": "Sparse Matrix", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A sparse matrix is a matrix in which the overwhelming majority of entries are zero, allowing specialised storage formats and algorithms that operate only on the non-zero elements. By storing and computing with structure rather than dense arrays, sparse representations dramatically reduce memory and arithmetic for large-scale linear algebra. They are foundational to optimisation problems such as bundle adjustment, graph algorithms and finite-element simulation where connectivity is local.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sparse-matrix", "labels": [ "Sparse Matrix" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "sparse-mixture-of-experts", "title": "Sparse Mixture of Experts", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An MoE variant with a trainable gating network that selects only a sparse combination of experts for each example, dramatically increasing model capacity whilst maintaining computational efficiency. Sparsely-gated MoE enables models with up to 137 billion parameters with manageable inference costs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:sparse-mixture-of-experts", "labels": [ "Sparse Mixture of Experts" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "sparse-representation", "title": "Sparse Representation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A sparse representation expresses a signal or data point as a linear combination of a small number of basis elements drawn from a larger, often overcomplete dictionary, with most coefficients zero. Sparsity yields compact, interpretable and noise-robust encodings, and underpins compressed sensing, dictionary learning and feature selection. It contrasts with dense distributed representations such as neural embeddings, and is recovered using L1 regularisation or greedy pursuit algorithms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:sparse-representation", "labels": [ "Sparse Representation" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "spatial-ai-assistant-agent", "title": "Spatial AI Assistant Agent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Spatial AI Associate is an AI-powered assistant role or software agent embedded within spatial computing and extended reality environments, responsible for contextual scene understanding, user intent interpretation, and proactive spatial guidance. It combines computer vision, SLAM-derived spatial maps, and large language model reasoning to deliver situated, location-aware assistance within physical or mixed-reality spaces.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:spatial-ai-assistant-agent", "labels": [ "Spatial AI Assistant Agent", "Spatial AI Associate" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "spatial-ai", "title": "Spatial AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Spatial AI is the application of machine learning and computer vision to understand, represent, and reason about three-dimensional physical space. It encompasses scene understanding, semantic mapping, depth estimation, and object detection, enabling robots, AR/VR systems, and autonomous agents to interpret their environment and plan actions within it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:spatial-ai", "labels": [ "Spatial AI" ], "is_subclass_of": [ "AI Research Area", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "spatial-anchor", "title": "Spatial Anchor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Coordinate reference that binds a virtual object's pose to a stable location in physical space, enabling persistent AR placement and physical-virtual registration.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:spatial-anchor", "labels": [ "Spatial Anchor", "Spatial Anchor Services" ], "is_subclass_of": [ "Platform and Environment", "Ar Scene Graph" ], "wikilinks": [ "Coordinate Transformation", "GPS Reference", "IEEE P2048-3", "Persistence Layer", "Pose Data", "Tracking Features", "Tracking System", "Visual Odometry", "World Coordinate Frame", "AR Scene Graph", "Coordinate System", "InteractionDomain", "NetworkLayer", "Persistent AR Placement", "Physical-Virtual Registration", "Shared AR Experiences", "SLAM", "Spatial Computing System", "Visual Marker" ] }, { "id": "spatial-anchoring", "title": "Spatial Anchoring", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial anchoring is the process of binding virtual or digital content to a precise, persistent location in physical space so that the content maintains a consistent position and orientation relative to the real world across multiple sessions, devices, and users. It relies on environmental mapping techniques such as Simultaneous Localisation and Mapping (SLAM), feature descriptor extraction, and cloud-synchronised anchor databases to re-localise digital objects reliably when the same physical environment is revisited. Anchors encode both geometric and semantic information about a surface or landmark, enabling persistent mixed-reality experiences, shared multi-user overlays, and location-aware services. The technology underpins applications ranging from indoor navigation and industrial training overlays to collaborative augmented reality workspaces and location-based gaming.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:spatial-anchoring", "labels": [ "Spatial Anchoring" ], "is_subclass_of": [ "Augmented Reality", "Environmental Mapping" ], "wikilinks": [ "Location Based AR", "metaverse", "Spatial Computing" ] }, { "id": "spatial-anchors", "title": "Spatial Anchors", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial anchors are persistent, georeferenced coordinate frames that bind virtual content to specific physical locations, enabling augmented and mixed reality experiences to survive device handoffs, multi-user sessions, and temporal gaps between visits. They are constructed by fusing visual feature maps (point clouds or learned descriptors), inertial measurements, and optionally GPS or Ultra-Wideband signals to produce a stable six-degrees-of-freedom pose estimate within a real-world coordinate system. Platform-level implementations such as ARKit, ARCore, Azure Spatial Anchors, and OpenXR's XR_MSFT_spatial_anchor extension expose this capability through APIs that serialise anchor state for later relocalisation. The technology underpins cross-device shared AR, indoor navigation, and location-aware digital twins by ensuring that virtual overlays remain semantically and geometrically coupled to the objects and surfaces they annotate.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:spatial-anchors", "labels": [ "Spatial Anchors", "Azure Spatial Anchors" ], "is_subclass_of": [ "AR Technology", "Platform and Environment" ], "wikilinks": [ "AR Technology", "metaverse", "Persistent AR Placement" ] }, { "id": "spatial-annotation", "title": "Spatial Annotation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Spatial annotation is the placement of digital notes, labels, or markers anchored to specific positions in a three-dimensional physical or virtual space, so that the content persists relative to real-world geometry as the viewer moves. It relies on spatial mapping and pose tracking to keep annotations registered to surfaces or objects. It is a foundational interaction for augmented reality collaboration, maintenance guidance, and shared spatial computing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:spatial-annotation", "labels": [ "Spatial Annotation" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "spatial-audio-processor", "title": "Spatial Audio Processor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Hardware or software component that renders three-dimensional sound using Head-Related Transfer Functions (HRTFs) and binaural synthesis, processing audio signals to simulate directional sound sources at specific positions in virtual space with real-time head tracking integration for immersive au...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-audio-processor", "labels": [ "Spatial Audio Processor" ], "is_subclass_of": [ "Display and Rendering", "Audio Processing System" ], "wikilinks": [ "3D Sound Rendering", "Audio Processing System", "metaverse" ] }, { "id": "spatial-audio-scene-description", "title": "Spatial Audio Scene Description", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A data model for encoding sound sources, listener positions, acoustic environments, and spatial audio metadata in three-dimensional space to enable immersive and realistic audio experiences in virtual environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:spatial-audio-scene-description", "labels": [ "Spatial Audio Scene Description" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "3D Coordinate System", "3D Scene Graph", "Acoustic Properties", "Acoustic Realism", "Ambisonics Playback", "Ambisonics Representation", "Audio Codec", "Audio Object", "Audio Streaming Protocol", "Binaural Rendering", "Dynamic Audio Mixing", "Head Tracking System", "ISO/IEC 23090-23 (MPEG-I Audio)", "Listener Position", "Object-Based Audio", "Room Acoustics Model", "SIGGRAPH Audio WG", "SMPTE ST 2128", "Sound Source Position", "Compute Layer" ] }, { "id": "spatial-audio-system", "title": "Spatial Audio System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An integrated audio technology framework creating three-dimensional soundscapes for VR, AR, and metaverse applications by simulating sound direction, distance, and environmental acoustics, enabling realistic audio experiences that respond dynamically to user movement and head orientation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-audio-system", "labels": [ "Spatial Audio System" ], "is_subclass_of": [ "Display and Rendering", "Immersive Audio Technology" ], "wikilinks": [ "Immersive Sound Experience", "Immersive Audio Technology", "metaverse" ] }, { "id": "spatial-audio", "title": "Spatial Audio", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Three-dimensional sound technology that uses head-related transfer functions (HRTF) and object-based mixing to position audio sources in 3D space around the listener, creating immersive auditory experiences that replicate natural sound perception.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:spatial-audio", "labels": [ "Spatial Audio", "Spatial Audio Experience", "Spatial Audio Technology", "SpatialAudio" ], "is_subclass_of": [ "Audio Technology" ], "wikilinks": [ "Audio Technology", "Immersive Experiences", "metaverse" ] }, { "id": "spatial-calibration", "title": "Spatial Calibration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of aligning virtual coordinate frames with physical world references in AR/VR systems, including camera calibration for video see-through displays and optical calibration for see-through displays, ensuring accurate registration of digital content with the real environment.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-calibration", "labels": [ "Spatial Calibration" ], "is_subclass_of": [ "Display and Rendering", "AR Registration" ], "wikilinks": [ "Accurate AR Alignment", "AR Registration", "metaverse" ] }, { "id": "spatial-computing", "title": "Spatial Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial computing is a computing paradigm in which digital information, computation, and interaction are anchored to and organised within three-dimensional physical or virtual space, enabling humans and machines to engage with data as if it inhabits the real world. It integrates continuous spatial sensing \u2014 through depth cameras, LiDAR, inertial measurement units, and computer vision \u2014 with real-time 3D rendering and multi-modal input (gaze, gesture, voice, touch) to create systems that perceive, reason about, and respond to the user's physical environment. The paradigm subsumes augmented reality, mixed reality, virtual reality, and autonomous spatial agents, unifying them under a common architectural requirement: the continuous, low-latency alignment of digital representations with physical geometry.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:spatial-computing", "labels": [ "Spatial Computing", "Spatial Computing Applications", "Spatial Computing Platform", "Spatial Computing Rendering", "Spatial Computing Standard", "Spatial Computing Streaming" ], "is_subclass_of": [ "Thing", "Human Computer Interaction" ], "wikilinks": [] }, { "id": "spatial-coordinates", "title": "Spatial Coordinates", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A unified reference frame system for positioning objects in three-dimensional virtual or mixed reality environments, including geographically-anchored poses (GeoPose) tied to Earth coordinates and local coordinate systems for scene-relative object placement with support for coordinate transformat...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-coordinates", "labels": [ "Spatial Coordinates" ], "is_subclass_of": [ "Standards and Interoperability", "Coordinate System" ], "wikilinks": [ "Object Positioning", "Coordinate System", "metaverse" ] }, { "id": "spatial-coverage", "title": "Spatial Coverage", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spatial-coverage", "labels": [ "Spatial Coverage" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spatial-data-structure", "title": "Spatial Data Structure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Spatial Data Structure is a data organisation scheme that partitions, indexes, or hierarchically organises geometric, geographic, or positional information to enable efficient queries over proximity, containment, intersection, and visibility. Classical families include space-partitioning trees (k-d trees, octrees, BSP trees, quadtrees), bounding-volume hierarchies (BVH), and spatial hash grids, each offering distinct trade-offs between construction cost, query throughput, memory footprint, and support for dynamic updates. These structures underpin performance-critical systems including real-time 3-D rendering, physics simulation, geographic information systems (GIS), robotics path planning, and spatial databases. The choice of structure is governed by data dimensionality, scene dynamism, query distribution, and hardware parallelism constraints.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:spatial-data-structure", "labels": [ "Spatial Data Structure", "Spatial Data Structures" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "spatial-database", "title": "Spatial Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A database system optimized for storing, indexing, and querying georeferenced data and geometric objects, supporting spatial data types (points, lines, polygons) and spatial operations (intersection, containment, proximity) per OGC Simple Features specification for GIS and metaverse applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-database", "labels": [ "Spatial Database" ], "is_subclass_of": [ "Data Management", "Database System" ], "wikilinks": [ "Geospatial Data Management", "Database System", "metaverse" ] }, { "id": "spatial-embodiment-harm-taxonomy", "title": "Spatial Embodiment Harm Taxonomy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse and Spatial Risks encompass the range of harms and hazards that arise specifically from spatially embodied, real-time virtual interactions \u2014 including harassment and abuse that mimics physical-world dynamics, miscommunication amplified by latency and avatar fidelity limitations, digital addiction, physical health impacts from prolonged immersive use, privacy violations through biometric data collection, and regulatory gaps as legislation lags behind technology deployment. These risks require multi-layered mitigation combining technical safeguards, governance frameworks, and international regulatory coordination.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:spatial-embodiment-harm-taxonomy", "labels": [ "Spatial Embodiment Harm Taxonomy", "Metaverse and Spatial Risks" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "rosenberg2022regulation" ] }, { "id": "spatial-index", "title": "Spatial Index", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Data structure optimized for efficient storage, retrieval, and querying of 3D spatial objects within virtual worlds using hierarchical geometric partitioning.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:spatial-index", "labels": [ "Spatial Index", "Spatial Indexing" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "Bounding Box", "Computational Geometry", "Distance Metric", "EWG/MSF taxonomy", "Geometric Primitives", "Grid-based Index", "I) Physical Layer", "IV) Data Layer", "Level of Detail Selection", "Octree", "Quadtree", "R-tree Structure", "View Frustum Culling", "Bounding Volume Hierarchy", "Collision Detection", "Coordinate System", "Data Structure", "Fast Spatial Queries", "InfrastructureDomain", "Nearest Neighbor Search" ] }, { "id": "spatial-interaction", "title": "Spatial Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The design and implementation of user input mods within extended reality environments that enable natural manipulation of virtual objects through gestures, eye tracking, voice commands, and physical movement in three-dimensional space.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-interaction", "labels": [ "Spatial Interaction" ], "is_subclass_of": [ "Interaction Technology", "Human Computer Interaction" ], "wikilinks": [ "Immersive User Experience", "Human Computer Interaction", "metaverse" ] }, { "id": "spatial-mapping-technology", "title": "Spatial Mapping Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial Mapping Technology refers to hardware and software systems that capture, process, and maintain geometric representations of physical environments in real time, enabling AR and MR devices to understand and interact with their surroundings. Core techniques include depth sensing, structured light, time-of-flight, and SLAM algorithms that fuse sensor data into persistent, updatable mesh models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:spatial-mapping-technology", "labels": [ "Spatial Mapping Technology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "spatial-mapping", "title": "Spatial Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of constructing and continuously updating three-dimensional representations of physical environments using sensor data\u2014including depth cameras, LiDAR, and IMU\u2014enabling AR/VR systems to understand surroundings for occlusion, collision, content placement, and navigation. Core algorithms include SLAM variants that fuse visual features with inertial measurements to produce dense mesh or voxel representations in real time.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:spatial-mapping", "labels": [ "Spatial Mapping", "3D Spatial Mapping" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "spatial-measurement", "title": "Spatial Measurement", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technology for determining distances, depths, and dimensional properties in physical and virtual environments using sensors such as Time-of-Flight cameras, structured light sensors, and LiDAR, enabling accurate 3D surface mapping, obstacle detection, and gesture tracking in AR/VR applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-measurement", "labels": [ "Spatial Measurement" ], "is_subclass_of": [ "Interaction Technology", "Depth Sensing" ], "wikilinks": [ "3D Environment Mapping", "Depth Sensing", "metaverse" ] }, { "id": "spatial-mesh", "title": "Spatial Mesh", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A three-dimensional geometric representation of real-world environments created through spatial mapping, where surfaces are reconstructed as interconnected polygonal meshes to enable accurate placement and occlusion of virtual objects in augmented reality applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-mesh", "labels": [ "Spatial Mesh" ], "is_subclass_of": [ "Content and Assets", "3D Reconstruction" ], "wikilinks": [ "AR Occlusion", "3D Reconstruction", "metaverse" ] }, { "id": "spatial-metadata", "title": "Spatial Metadata", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Descriptive information about geospatial and 3D content documenting location, coordinate system, projection, quality, lineage, and distribution attributes, following standards such as ISO 19115 and FGDC CSDGM to enable discovery, eand interoperability of spatial data resources.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-metadata", "labels": [ "Spatial Metadata" ], "is_subclass_of": [ "Standards and Interoperability", "Geospatial Information" ], "wikilinks": [ "Spatial Data Discovery", "Geospatial Information", "metaverse" ] }, { "id": "spatial-partitioning", "title": "Spatial Partitioning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial partitioning is the technique of subdividing a space into non-overlapping or hierarchically nested regions so that objects can be organised by location and queried efficiently. By grouping nearby objects and pruning regions that cannot contain a query result, it reduces the cost of operations such as collision detection, ray casting, nearest-neighbour search and visibility culling from quadratic toward logarithmic or linear scaling. Common structures include grids, quadtrees, octrees, k-d trees, binary space partitioning trees and bounding-volume hierarchies, each trading construction cost against query performance for particular workloads.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:spatial-partitioning", "labels": [ "Spatial Partitioning" ], "is_subclass_of": [ "Spatial Data Structure" ], "wikilinks": [] }, { "id": "spatial-presence", "title": "Spatial Presence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The psychological state of feeling physically located within a virtual environment, experiencing the sense of \"being there\" despite technological mediation, influenced by place illusion (perceived location), plausibility illusion (believable events), and temporal presence (immediacy of experience).", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-presence", "labels": [ "Spatial Presence" ], "is_subclass_of": [ "Platform and Environment", "VR Psychology" ], "wikilinks": [ "Immersive Experience", "metaverse", "VR Psychology" ] }, { "id": "spatial-queries", "title": "Spatial Queries", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Database queries that select geographic features based on location or spatial relationships such as intersection, containment, proximity, and adjacency, implemented using R-tree spatial indexing and filter-refine strategies per OGC Simple Features and SQL/MM Spatial standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-queries", "labels": [ "Spatial Queries" ], "is_subclass_of": [ "Content and Assets", "Database Query" ], "wikilinks": [ "Location Based Search", "Database Query", "metaverse" ] }, { "id": "spatial-reference-system", "title": "Spatial Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spatial-reference-system", "labels": [ "Spatial Reference System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spatial-resolution", "title": "Spatial Resolution", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spatial-resolution", "labels": [ "Spatial Resolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spatial-tracking-system", "title": "Spatial Tracking System", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technology for determining the six-degrees-of-freedom (6DOF) position and orientation of devices, controllers, or body parts in three-dimensional space, using visual-inertial SLAM, stereo cameras, and IMU sensors for precise real-time tracking in VR, AR, and mixed reality applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-tracking-system", "labels": [ "Spatial Tracking System" ], "is_subclass_of": [ "Interaction Technology", "Motion Tracking" ], "wikilinks": [ "Precise Spatial Positioning", "metaverse", "Motion Tracking" ] }, { "id": "spatial-tracking-technology", "title": "Spatial Tracking Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Hardware and software systems that determine the position and orientation of physical objects, users, or devices in three-dimensional space in real time. Spatial tracking underpins head-mounted display tracking, hand and eye tracking, and room-scale boundary definition in virtual and mixed reality systems, enabling accurate registration between physical movement and virtual scene response.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:spatial-tracking-technology", "labels": [ "Spatial Tracking Technology" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "spatial-tracking", "title": "Spatial Tracking", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technology that determines the position and orientation of objects in three-dimensional space using six degrees of freedom (6DoF), enabling precise tracking of headsets, controllers, and body movements through inside-out or outside-in sensor configurations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-tracking", "labels": [ "Spatial Tracking" ], "is_subclass_of": [ "Interaction Technology", "Motion Tracking" ], "wikilinks": [ "Immersive VR", "metaverse", "Motion Tracking" ] }, { "id": "spatial-user-interfaces", "title": "Spatial User Interfaces", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Three-dimensional interaction systems enabling users to engage with digital content through spatially-aware mods including hand gestures, gaze tracking, and voice commands, designed for natural movement within AR, VR, and mixed reality environments with consideration for user comfort zones and 6D...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:spatial-user-interfaces", "labels": [ "Spatial User Interfaces", "Spatial User Interface" ], "is_subclass_of": [ "Interaction Technology", "3D User Interface" ], "wikilinks": [ "Immersive Interaction", "3D User Interface", "metaverse" ] }, { "id": "spatial-web", "title": "Spatial Web", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Spatial Web is a proposed convergence of spatial computing, decentralised web technologies and the semantic web into a single addressable layer where digital information is anchored to physical locations, objects and people. It enables persistent spatial context, machine-readable semantics and decentralised identity so that humans and software agents can perceive, reason about and act on a blended physical-virtual environment. It is positioned as a successor architecture to the document-centric World Wide Web.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:spatial-web", "labels": [ "Spatial Web" ], "is_subclass_of": [ "Spatial Computing", "Semantic Spatial Web Layer" ], "wikilinks": [] }, { "id": "spatiotemporal-asset-catalogue", "title": "Spatiotemporal Asset Catalogue", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spatiotemporal-asset-catalogue", "labels": [ "Spatiotemporal Asset Catalogue" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spatiotemporal-interpolation", "title": "Spatiotemporal Interpolation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spatiotemporal-interpolation", "labels": [ "Spatiotemporal Interpolation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "speaker-diarisation", "title": "speaker diarisation", "domain": "ai", "domain_name": "Ai", "definition": "Speaker diarisation is the automated process of partitioning a continuous multi-speaker audio stream into temporally contiguous, speaker-homogeneous segments and assigning a speaker label to each segment, answering the question 'who spoke when?' without necessarily mapping labels to real-world identities. A canonical pipeline comprises voice activity detection, acoustic feature extraction, speaker embedding (via models such as x-vectors or ECAPA-TDNN), agglomerative or spectral clustering to group embeddings into speaker clusters, optional overlap detection, and a resegmentation or ViterBI refinement step. End-to-end neural approaches such as EEND (End-to-End Neural Diarisation) unify segmentation and assignment into a single sequence-labelling model capable of handling overlapping speech. Speaker diarisation is a foundational component in meeting transcription systems, clinical consultation recording, broadcast media indexing, and any dialogue-analytic pipeline that requires utterance attribution.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:speaker-diarisation", "labels": [ "Speaker Diarisation", "Diarisation Models", "Speaker Attribution", "Speaker Identity Model" ], "is_subclass_of": [ "Speech Processing" ], "wikilinks": [] }, { "id": "speaker-embedding", "title": "Speaker Embedding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Speaker Embedding is a fixed-dimensional vector representation of a speaker's vocal identity, extracted from variable-length speech segments by a neural network trained to encode speaker-discriminative acoustic features while remaining invariant to spoken content, channel conditions, and background noise. Models such as d-vectors (deep speaker embeddings), x-vectors (TDNN-based), and ECAPA-TDNN produce embeddings that cluster in a metric space where same-speaker utterances lie close together and different-speaker utterances are well-separated. Speaker embeddings enable downstream tasks including speaker verification, speaker identification, speaker diarisation, and personalised speech synthesis without storing raw audio. They are trained using discriminative objectives such as softmax classification over training speakers, generalised end-to-end loss, or angular prototypical loss.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:speaker-embedding", "labels": [ "Speaker Embedding" ], "is_subclass_of": [ "Embedding" ], "wikilinks": [] }, { "id": "speaker-recognition", "title": "Speaker Recognition", "domain": "ai", "domain_name": "Ai", "definition": "The automated identification or verification of a person's identity based on characteristics of their voice derived from acoustic speech signals. Speaker recognition encompasses two sub-tasks: speaker verification (confirming a claimed identity) and speaker identification (determining who among a set of known speakers produced a given utterance). Systems extract speaker-discriminative features such as MFCCs, i-vectors, or x-vectors from audio, then compare these against enrolled speaker models using distance metrics or neural classifiers. It is distinct from speech recognition, which transcribes words rather than identifies speakers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:speaker-recognition", "labels": [ "Speaker Recognition" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "specific-impulse", "title": "Specific Impulse", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:specific-impulse", "labels": [ "Specific Impulse" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "specification-gaming", "title": "Specification Gaming", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A failure mode of optimising systems in which an agent satisfies the literal specification of an objective while defeating its intended purpose \u2014 exploiting loopholes, simulator bugs, or proxy metrics to score highly without doing the task the designer actually wanted, as when a boat-racing agent loops through reward targets instead of finishing the race; it is the general phenomenon of which reward hacking in reinforcement learning is the canonical instance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:specification-gaming", "labels": [ "Specification Gaming" ], "is_subclass_of": [ "AI Risk" ], "wikilinks": [ "AI Risk", "Reward Hacking", "AI Safety Research", "Instruction Following" ] }, { "id": "specification", "title": "Specification", "domain": "standards", "domain_name": "Standards", "definition": "A precise, documented statement of the requirements, behaviour, interfaces, or characteristics that a system, product, or process must satisfy, written so that independent parties can implement against it and verify conformance to it. Specifications range from engineering requirements documents and CAD tolerances to protocol and file-format definitions published by standards bodies, and they are the normative core of every technical standard: the testable text that separates what conforms from what does not.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:specification", "labels": [ "Specification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "Technical Standard", "Open Standards", "Interoperability", "Standardisation" ] }, { "id": "specificity", "title": "Specificity", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Specificity (also called the True Negative Rate or TNR) is the proportion of actual negative instances that a binary classifier or diagnostic test correctly identifies as negative, computed as TN / (TN + FP). It quantifies a model's ability to avoid false positives, complementing sensitivity (recall) in characterising the full operating behaviour of a classifier. Together with sensitivity, specificity defines the two axes of the Receiver Operating Characteristic (ROC) curve, enabling principled threshold selection across tasks ranging from medical diagnostics to spam filtering and anomaly detection.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:specificity", "labels": [ "Specificity" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Confusion Matrix", "Precision", "Recall" ] }, { "id": "speckle-filtering", "title": "Speckle Filtering", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:speckle-filtering", "labels": [ "Speckle Filtering" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-analysis", "title": "Spectral Analysis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-analysis", "labels": [ "Spectral Analysis", "FrequencyDomainAnalysis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-band", "title": "Spectral Band", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-band", "labels": [ "Spectral Band" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-classification", "title": "Spectral Classification", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-classification", "labels": [ "Spectral Classification" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-clustering", "title": "Spectral Clustering", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A clustering technique that partitions data using the eigenvectors of a graph Laplacian built from a pairwise similarity matrix. By embedding points into the subspace spanned by the Laplacian's smallest eigenvectors and running a simple algorithm such as k-means in that spectral space, it relaxes NP-hard graph-cut objectives (ratio cut, normalised cut) into tractable eigenproblems, allowing it to recover non-convex, connectivity-defined clusters that centroid-based methods cannot separate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:spectral-clustering", "labels": [ "Spectral Clustering" ], "is_subclass_of": [ "Clustering" ], "wikilinks": [ "Clustering", "Community Detection", "Graph Theory", "Dimensionality Reduction", "Speaker Diarisation" ] }, { "id": "spectral-filtering", "title": "Spectral Filtering", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-filtering", "labels": [ "Spectral Filtering" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-radiance", "title": "Spectral Radiance", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-radiance", "labels": [ "Spectral Radiance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-resolution", "title": "Spectral Resolution", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-resolution", "labels": [ "Spectral Resolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-response-function", "title": "Spectral Response Function", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-response-function", "labels": [ "Spectral Response Function" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-synthesis", "title": "Spectral Synthesis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-synthesis", "labels": [ "Spectral Synthesis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectral-unmixing", "title": "Spectral Unmixing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectral-unmixing", "labels": [ "Spectral Unmixing" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spectrum-allocation", "title": "Spectrum Allocation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Spectrum allocation is the regulatory and technical process by which electromagnetic frequency bands are assigned to specific services, operators, or technologies to enable wireless communication without harmful interference. Governments and international bodies define how the radio spectrum is divided, licensed, and managed across uses ranging from mobile broadband and satellite communication to broadcasting and scientific research. Efficient allocation balances competing commercial, public safety, and scientific demands while adapting to evolving technologies such as 5G and millimetre-wave systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:spectrum-allocation", "labels": [ "Spectrum Allocation" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "spectrum-management", "title": "Spectrum Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Spectrum management is the regulatory and technical process of allocating, assigning, and coordinating use of the radio-frequency spectrum to minimise interference and maximise societal and economic value. It encompasses licensing, frequency planning, enforcement, and the design of sharing arrangements between competing wireless services. Spectrum management is administered by national regulators and harmonised internationally to enable cross-border wireless communication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:spectrum-management", "labels": [ "Spectrum Management" ], "is_subclass_of": [ "Infrastructure", "Radio Frequency Spectrum" ], "wikilinks": [] }, { "id": "spectrum", "title": "Spectrum", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spectrum", "labels": [ "Spectrum" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "speculative-decoding", "title": "speculative decoding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Speculative decoding is an inference acceleration technique for autoregressive language models in which a smaller, faster draft model proposes multiple candidate tokens that a larger target model verifies in a single parallel forward pass. Accepted tokens are committed to the output sequence; rejected tokens trigger a corrected sample from the residual distribution, ensuring the final output is statistically identical to sampling from the target model alone. Because modern GPU accelerators are memory-bandwidth-bound during autoregressive generation, batching verification of several candidate tokens substantially increases arithmetic utilisation and can raise effective throughput two to four times without altering the model's output distribution. The technique is now integrated into mainstream inference frameworks and sits at the intersection of model efficiency, hardware utilisation, and production LLM serving.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:speculative-decoding", "labels": [ "Speculative Decoding" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "speculative-memetic-token-propagation", "title": "Speculative Memetic Token Propagation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The phenomenon in which AI agents and autonomous social actors propagate memetic content\u2014ideas, tokens, political positions\u2014tied to speculative financial instruments (memecoins, programmatically-aligned tokens) that derive value from attention and community adoption rather than underlying utility. This creates feedback loops between AI terminal outputs, social media virality, and tokenised capital allocation with tangible real-world effects.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:speculative-memetic-token-propagation", "labels": [ "Speculative Memetic Token Propagation", "Financialised Agentic Memetics" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [ "Financial Nihilism" ] }, { "id": "speech-act-theory", "title": "Speech Act Theory", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Speech Act Theory is a framework from the philosophy of language holding that utterances perform actions rather than merely describing states of affairs. In distributed and multi-agent systems it underpins agent communication languages by classifying messages as performatives such as inform, request, commit, and declare, each carrying defined intent and felicity conditions. It distinguishes the locutionary content, illocutionary force, and perlocutionary effect of a communicative act, giving coordinating agents a shared semantics for interaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:speech-act-theory", "labels": [ "Speech Act Theory" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "speech-corpus", "title": "Speech Corpus", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A speech corpus is a curated collection of recorded audio, usually paired with transcriptions, speaker metadata or other annotations, used to train and evaluate speech and voice systems. Applications such as speech recognition and voice cloning both require a sufficiently large and diverse speech corpus to learn accurate acoustic and linguistic models. Corpus quality factors, including recording conditions, speaker diversity and transcription accuracy, directly bound the performance achievable by models trained on it.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:speech-corpus", "labels": [ "Speech Corpus" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "speech-processing", "title": "Speech Processing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The field concerned with the analysis, recognition, synthesis and transformation of human speech signals by computational systems.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:speech-processing", "labels": [ "Speech Processing", "Speech Technology" ], "is_subclass_of": [ "Audio Processing" ], "wikilinks": [ "Audio Processing", "Speech Recognition", "Text-to-Speech", "Natural Language Processing" ] }, { "id": "speech-recognition", "title": "Speech Recognition", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Speech Recognition (also called Automatic Speech Recognition, ASR) is the computational task of converting spoken acoustic signals into a textual representation, enabling machines to interpret and act on human voice input. Modern systems use end-to-end deep learning architectures \u2014 principally transformer-based encoder-decoder models such as Wav2Vec 2.0 and Whisper \u2014 trained on thousands of hours of labelled audio to achieve near-human word-error rates across diverse speakers, languages, and acoustic environments. Key technical challenges include robustness to background noise, speaker variability, dialectal and accent diversity, and code-switching; deployment challenges include latency constraints for real-time streaming, on-device inference under power and memory limits, and domain adaptation for specialised vocabularies such as clinical or legal terminology.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:speech-recognition", "labels": [ "Speech Recognition", "Audio-Visual Speech Recognition" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "owl:Thing" ] }, { "id": "speech-and-voice", "title": "Speech and Voice", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Speech and Voice AI is the computational subdomain of artificial intelligence encompassing mods for producing, recognising, transforming, cloning, and reasoning over human speech audio signals, implemented through neural architectures ranging from WaveNet autoregressive dilated convolutional voco...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:speech-and-voice", "labels": [ "Speech and Voice", "AI Speech, Voice AI, TTS, ASR, Speech Synthesis, Speech Recognition, Voice Technology, Conversational Voice AI, Text-to-Speech, Automatic Speech Recognition" ], "is_subclass_of": [ "AI Research Area", "Multimodal AI", "Natural Language Processing", "Audio Processing", "Machine Learning Discipline", "Conversational AI" ], "wikilinks": [ "AlgorithmLayer", "AudioAIDomain", "Audio Data", "Audio Processing", "Automatic Speech Recognition", "Customer Service Automation", "Education Technology", "Emotional AI", "Gaming AI", "GPU Inference", "Healthcare AI", "IEEE Audio Standards", "IETF WebRTC Standards", "Information Theory", "Language Models", "Linguistics", "Low-Latency Computing", "Media Production", "NaturalLanguageProcessingDomain", "Neural Networks" ] }, { "id": "speech-synthesis", "title": "SpeechSynthesis", "domain": "ai", "domain_name": "Ai", "definition": "Speech synthesis, or text-to-speech (TTS), is the artificial generation of human speech from text or linguistic representations using computational models that map phonetic, prosodic, and acoustic features to waveform output. Modern neural TTS systems use end-to-end deep learning pipelines\u2014typically a text front-end, an acoustic model (Tacotron, FastSpeech, or diffusion-based), and a neural vocoder (WaveNet, HiFi-GAN, Vocos)\u2014to produce speech that is perceptually natural, expressive, and stylistically controllable. Speech synthesis underpins screen readers, virtual assistants, navigation systems, voice cloning tools, and interactive conversational agents, and its quality is now at or approaching human parity on standard benchmarks for many languages.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:speech-synthesis", "labels": [ "SpeechSynthesis", "Speech Synthesis", "Speech synthesis", "Vocal Synthesis" ], "is_subclass_of": [ "Text-to-Speech" ], "wikilinks": [] }, { "id": "spherical-harmonic-analysis", "title": "Spherical Harmonic Analysis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spherical-harmonic-analysis", "labels": [ "Spherical Harmonic Analysis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spherical-harmonics", "title": "Spherical Harmonics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spherical harmonics are a complete set of orthogonal basis functions defined on the surface of a sphere, used to represent functions of direction compactly as a weighted sum of coefficients. Analogous to a Fourier series on the sphere, they allow smooth angular functions \u2014 such as incoming light or a directional colour \u2014 to be approximated with a small number of low-order coefficients. In computer graphics they underpin precomputed radiance transfer, irradiance environment lighting, and, more recently, view-dependent colour in Gaussian splatting and neural rendering.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:spherical-harmonics", "labels": [ "Spherical Harmonics" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "spin-stabilisation", "title": "Spin Stabilisation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spin-stabilisation", "labels": [ "Spin Stabilisation" ], "is_subclass_of": [ "Attitude Control" ], "wikilinks": [] }, { "id": "spiral-galaxy", "title": "Spiral Galaxy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spiral-galaxy", "labels": [ "Spiral Galaxy" ], "is_subclass_of": [ "Galaxy" ], "wikilinks": [] }, { "id": "spl-token", "title": "Spl Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "SPL Token is the token standard of the Solana Program Library, defining how fungible and non-fungible tokens are created and managed on the Solana blockchain through a shared on-chain program. Rather than deploying a separate contract per token as in ERC-20, all SPL tokens share a single canonical program and store balances in token accounts associated with each owner and mint. This account-model design enables high-throughput, low-cost token operations native to Solana's runtime.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:spl-token", "labels": [ "Spl Token", "SPL Token" ], "is_subclass_of": [ "Fungible Token" ], "wikilinks": [] }, { "id": "splicing", "title": "Splicing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Splicing is a Lightning Network operation that resizes an existing payment channel by adding or removing on-chain funds without closing and reopening it, preserving the channel's state and routing history. A splice-in increases capacity by committing additional bitcoin, while a splice-out withdraws funds to an on-chain address, both executed through a single funding transaction. This keeps the channel continuously available and reduces the on-chain cost and downtime of channel management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:splicing", "labels": [ "Splicing" ], "is_subclass_of": [ "Payment Channel" ], "wikilinks": [] }, { "id": "spline-interpolation", "title": "Spline Interpolation", "domain": "robotics", "domain_name": "Robotics", "definition": "Spline interpolation is a numerical method for constructing a smooth curve that passes through a given set of waypoints by fitting piecewise polynomial segments, most commonly cubic splines, that join with continuous position, velocity, and acceleration at each segment boundary. In robotics it is used to convert a sparse sequence of target waypoints into a continuous, dynamically feasible trajectory that a controller can track without abrupt changes in motion. Spline interpolation is a standard building block of trajectory generation and trajectory planning pipelines.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:spline-interpolation", "labels": [ "Spline Interpolation" ], "is_subclass_of": [ "Interpolation" ], "wikilinks": [] }, { "id": "sports-analytics", "title": "Sports Analytics", "domain": "ai", "domain_name": "Ai", "definition": "Sports analytics is the application of data capture, statistical modelling, and machine learning to athletic performance, tactics, and injury prevention, drawing on tracking data, biomechanics, and event logs. Computer-vision pipelines increasingly supply the underlying movement data by extracting player positions and body pose from video. It informs coaching decisions, talent evaluation, broadcast insight, and load management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sports-analytics", "labels": [ "Sports Analytics" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "spot-beam", "title": "Spot Beam", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:spot-beam", "labels": [ "Spot Beam" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "spot-bitcoin-etf", "title": "Spot Bitcoin ETF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A spot Bitcoin ETF is an exchange-traded fund that holds actual bitcoin and tracks its price directly, allowing investors to gain exposure through ordinary brokerage accounts without holding the asset or managing private keys. Approved by the US SEC in January 2024 after years of rejection, spot Bitcoin ETFs differ from earlier futures-based products by holding the underlying coin in regulated custody. They have channelled substantial institutional and retail capital into bitcoin and are widely regarded as a milestone in the asset's mainstream financial integration.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:spot-bitcoin-etf", "labels": [ "Spot Bitcoin ETF" ], "is_subclass_of": [ "Exchange-Traded Fund" ], "wikilinks": [] }, { "id": "spot-ethereum-etf", "title": "Spot Ethereum ETF", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A spot Ethereum ETF is an exchange-traded fund that holds actual ether and tracks its market price, allowing investors to gain regulated exposure to ETH through ordinary brokerage accounts without self-custody. Unlike futures-based products, it backs each share with the underlying asset held by a custodian. Its approval by securities regulators marked a significant step in integrating a major crypto asset into traditional capital markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:spot-ethereum-etf", "labels": [ "Spot Ethereum ETF", "Spot Ether ETF" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "spot-trading", "title": "Spot Trading", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Spot trading is the buying and selling of an asset for immediate delivery and settlement at the current market price, as opposed to settlement at a future date. In cryptocurrency and traditional markets it involves placing orders against an order book or liquidity pool, with ownership of the underlying asset transferring promptly, and it contrasts with derivatives and margin trading where exposure is taken without immediate full ownership.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:spot-trading", "labels": [ "Spot Trading" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "sprint-planning", "title": "Sprint Planning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sprint planning is the Scrum event in which a development team selects items from the product backlog and defines a goal and plan for the upcoming time-boxed iteration. The team negotiates scope against capacity, decomposes chosen items into actionable work, and commits to a realistic sprint backlog. It establishes shared understanding of what will be delivered and how, anchoring the iteration's focus.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sprint-planning", "labels": [ "Sprint Planning" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "squid-router", "title": "Squid Router", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Squid Router is a cross-chain liquidity and routing protocol built on the Axelar network that allows token swaps and transfers between different blockchains in a single transaction.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:squid-router", "labels": [ "Squid Router", "Router" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [ "Axelar", "Interoperability", "Cross-Chain Bridge" ] }, { "id": "ssl-termination", "title": "Ssl Termination", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "SSL termination is the practice of decrypting inbound TLS-encrypted traffic at a dedicated network endpoint, typically a reverse proxy or load balancer, before forwarding the now-plaintext requests to backend servers. By centralising the cryptographic handshake and certificate management at the edge, it offloads CPU-intensive encryption work from application servers and simplifies certificate lifecycle administration. The terminating node holds the private keys and performs the handshake, then routes traffic over the internal network. It is a foundational pattern in modern web infrastructure, often paired with re-encryption to backends for end-to-end security.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:ssl-termination", "labels": [ "Ssl Termination", "SSL Termination" ], "is_subclass_of": [ "Reverse Proxy" ], "wikilinks": [] }, { "id": "ssl", "title": "Ssl", "domain": "security", "domain_name": "Security", "definition": "SSL (Secure Sockets Layer) is a deprecated cryptographic protocol for establishing encrypted, authenticated connections between networked applications, and its name remains in common use as a colloquial label for its successor, TLS. SSL introduced the handshake model in which peers negotiate cipher suites, authenticate via certificates and derive symmetric session keys to protect subsequent traffic. All SSL versions are now considered insecure and have been superseded by Transport Layer Security. The term persists in product naming, certificates and developer vocabulary even though modern deployments use TLS.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:ssl", "labels": [ "Ssl", "SSL" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "stability-ai", "title": "Stability AI", "domain": "ai", "domain_name": "Ai", "definition": "Stability AI is a British AI company founded in 2019 (publicly prominent from 2022) that develops and releases open-weight generative models across image, audio, language, and video modalities, most notably the Stable Diffusion family of latent diffusion models for text-to-image synthesis. The company's open-weight release strategy, in which model weights are made freely downloadable rather than accessed only via hosted APIs, catalysed a large ecosystem of fine-tuned variants, community tools, and downstream commercial products. Stability AI also supports research into multimodal generation, efficient inference, and safety mechanisms for open foundation models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stability-ai", "labels": [ "Stability AI" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Stable Diffusion", "Text-to-Image", "Image Generation", "Generative AI" ] }, { "id": "stability-analysis", "title": "Stability Analysis", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Stability analysis is the systematic mathematical investigation of whether a dynamical system \u2014 physical, computational, financial, or ecological \u2014 will remain bounded, return to equilibrium, or diverge when subjected to perturbations from an operating point. Classical techniques include Lyapunov stability theory, eigenvalue analysis of linearised systems, Bode and Nyquist frequency-domain methods, and Floquet theory for periodic systems. In AI, stability analysis extends to training dynamics, gradient flow, and the behaviour of neural networks under input distribution shifts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:stability-analysis", "labels": [ "Stability Analysis", "StabilityAnalysis" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "stability", "title": "Stability", "domain": "robotics", "domain_name": "Robotics", "definition": "Stability is the property of a dynamical or control system whereby its state remains bounded and returns toward an equilibrium following a disturbance, rather than diverging. Formalised through notions such as Lyapunov stability and bounded-input bounded-output stability, it is a primary design objective for any feedback controller. Ensuring stability is prerequisite to performance, since an unstable system cannot be made to track references reliably.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stability", "labels": [ "Stability" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "stable-coins", "title": "Stable Coins", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stable Coins (stablecoins) are blockchain-native digital tokens engineered to maintain stable value relative to an external reference asset \u2014 most commonly the US dollar at 1:1 parity \u2014 through one of four primary stabilisation mechanisms: (i) fiat-collateralised custody, where off-chain reserves...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:stable-coins", "labels": [ "Stable Coins", "Stablecoins" ], "is_subclass_of": [ "Token and Asset", "Digital Asset", "Blockchain Network", "Payments Infrastructure", "Tokenised Asset", "Financial Instrument" ], "wikilinks": [ "Algorithmic Stabilisation", "Attestation Report", "Avalanche", "Base Network", "Basel III Crypto Exposure Rules", "BlackRock", "BlackRock", "Central Bank Policy", "Commercial Bank Money", "Corporate Treasury Management", "Cross-Border Remittances", "Cross-Border Remittances", "Custodian", "DeFi", "DeFi", "Decentralised Exchange", "Dollar Peg Mechanism", "Electronic Money", "Emerging Market Finance", "Eurodollar" ] }, { "id": "stable-diffusion-image-model", "title": "Stable Diffusion Image Model", "domain": "ai", "domain_name": "Ai", "definition": "Stable Diffusion is an open-source latent diffusion model developed by Stability AI in collaboration with CompVis and Runway, released in 2022, capable of generating high-quality images from text prompts. It operates in a compressed latent space rather than pixel space, dramatically reducing computational requirements compared to earlier diffusion models. The model supports text-to-image, image-to-image, and inpainting tasks, and has become the foundation for a large ecosystem of fine-tuned variants and extension tools.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:stable-diffusion-image-model", "labels": [ "Stable Diffusion Image Model", "Stable Diffusion", "Stable Diffusion Inpaint", "Stable Diffusion XL" ], "is_subclass_of": [ "Diffusion Model", "Text-to-Image" ], "wikilinks": [] }, { "id": "stable-diffusion", "title": "Stable Diffusion", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An open-weights family of latent diffusion models for text-to-image generation, first released by Stability AI, CompVis, and Runway in August 2022, which performs iterative denoising in a compressed VAE latent space conditioned on CLIP text embeddings, making photorealistic and stylised image synthesis feasible on consumer GPUs and seeding a vast open ecosystem of fine-tunes, LoRAs, and tooling.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:stable-diffusion", "labels": [ "Stable Diffusion" ], "is_subclass_of": [ "Diffusion Model" ], "wikilinks": [ "Diffusion Model", "Latent Diffusion", "Text-to-Image Generation", "Automatic1111" ] }, { "id": "stable-swap-invariant", "title": "Stable Swap Invariant", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The stable swap invariant is an automated market maker pricing formula, introduced by Curve Finance, that blends the constant-sum and constant-product curves to provide very low slippage for trades between assets expected to hold near-equal value, such as stablecoins or pegged tokens. Near the balanced point it behaves like a constant-sum market for tight pricing, while curving toward constant-product behaviour as reserves diverge to preserve liquidity. It is the core mechanism enabling efficient stablecoin exchange on-chain.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stable-swap-invariant", "labels": [ "Stable Swap Invariant", "StableSwap Invariant" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "stablecoin-regulation", "title": "Stablecoin Regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stablecoin regulation is the corpus of statutory, administrative, and supervisory rules that national and supranational authorities apply to Cryptocurrency|cryptocurrency tokens designed to maintain stable value relative to fiat currencies, commodities, or baskets of assets, spanning five con...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:stablecoin-regulation", "labels": [ "Stablecoin Regulation", "FCA Stablecoin Rules" ], "is_subclass_of": [ "Governance and Regulation", "Financial Regulation", "Payment Systems Regulation", "Electronic Money Regulation", "Crypto Regulation", "Consumer Protection", "Monetary Law" ], "wikilinks": [ "Attestation Regime", "Bank Deposit", "Banking Law", "Capital Adequacy", "Circle", "Circle", "ComplianceLayer", "Cross-Border Payment", "Crypto Regulation", "DeFi", "DeFi Integration", "Disclosure Standard", "EBA", "Electronic Money Regulation", "FCA", "FCA Regulatory Regime", "Financial Regulation", "FinancialRegulationDomain", "Financial Stability", "Financial Stability Board" ] }, { "id": "stablecoin-token", "title": "Stablecoin Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Stablecoin Token is a blockchain-native fungible token engineered to maintain a stable value, typically pegged to a fiat currency, commodity, or basket of assets, through one of three principal mechanisms: fiat-collateralised reserves held by a custodian (e.g. USDC, USDT), crypto-collateralised over-collateralisation enforced by smart contracts (e.g. DAI), or algorithmic supply adjustment that mints and burns tokens to defend the peg without direct collateral. Stablecoins serve as the primary medium of exchange, unit of account, and store of value within decentralised finance ecosystems, enabling lending, borrowing, and trading without exposure to cryptocurrency price volatility.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:stablecoin-token", "labels": [ "Stablecoin Token" ], "is_subclass_of": [ "Token and Asset", "Fungible Token" ], "wikilinks": [ "Blockchain", "Fungible Token" ] }, { "id": "stablecoin", "title": "Stablecoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A cryptocurrency whose value is algorithmically or institutionally pegged to a reserve asset to maintain price stability, enabling reliable medium of exchange and store of value in virtual economies.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:stablecoin", "labels": [ "Stablecoin", "Bridged Stablecoin", "Stablecoin Ecosystem", "Stablecoin Issuance" ], "is_subclass_of": [ "Crypto Token" ], "wikilinks": [ "Collateral Reserves", "Cross-Border Transactions", "IMF CBDC Notes", "ISO 24165", "Peg Mechanism", "Price Oracle", "Price Stability", "Reserve Asset", "Stabilization Algorithm", "Blockchain", "Crypto Token", "Digital Asset", "MiddlewareLayer", "Smart Contract", "TrustAndGovernanceDomain", "Value Transfer", "Virtual Commerce", "Virtual Currency", "VirtualEconomyDomain" ] }, { "id": "stablecoins-on-bitcoin", "title": "Stablecoins on Bitcoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stablecoins on Bitcoin are price-stable tokens issued and transferred on Bitcoin-based protocols such as Taproot Assets, Liquid, or RGB, rather than on smart-contract chains like Ethereum. They aim to bring dollar-denominated value transfer to the Bitcoin ecosystem, leveraging Bitcoin's settlement security and, where combined with the Lightning Network, near-instant low-cost payments. They extend Bitcoin's utility from a single monetary asset toward a broader settlement layer for fiat-pegged value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:stablecoins-on-bitcoin", "labels": [ "Stablecoins on Bitcoin" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "stack-machine", "title": "Stack Machine", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A stack machine is an execution model in which operands and intermediate results are held on a last-in-first-out stack rather than in named registers, with instructions implicitly consuming their inputs from the top of the stack and pushing their outputs back onto it. The model yields compact, position-independent bytecode and a simple deterministic evaluator, which is why it underpins many scripting languages and blockchain virtual machines such as Bitcoin Script and the Ethereum Virtual Machine.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stack-machine", "labels": [ "Stack Machine" ], "is_subclass_of": [ "Execution Model" ], "wikilinks": [] }, { "id": "stacker-news", "title": "Stacker News", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stacker News is an online community platform where users post and curate content and reward contributions with bitcoin payments over the Lightning Network.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:stacker-news", "labels": [ "Stacker News" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [ "Lightning Network", "Bitcoin", "community" ] }, { "id": "stacking", "title": "Stacking", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Stacking (stacked generalisation) is a hierarchical ensemble method that trains multiple diverse base models then combines their predictions using a meta-model, which learns the optimal weighting of base model outputs. Unlike bagging and boosting, stacking uses cross-validated out-of-fold predictions to train the meta-model, reducing information leakage and typically achieving superior predictive performance over any single base learner.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:stacking", "labels": [ "Stacking" ], "is_subclass_of": [ "AI Technique", "Ensemble Methods" ], "wikilinks": [ "Ensemble Methods" ] }, { "id": "stacks", "title": "Stacks", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Bitcoin Layer that enables smart contracts and applications whose state is anchored to the Bitcoin blockchain through its proof-of-transfer consensus.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:stacks", "labels": [ "Stacks" ], "is_subclass_of": [ "Bitcoin Proof-of-Work Protocol" ], "wikilinks": [ "Consensus Protocol", "Bitcoin", "Smart Contract", "Bitcoin Protocol" ] }, { "id": "stakeholder-analysis", "title": "Stakeholder Analysis", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder analysis is a systematic methodology for identifying, mapping, and evaluating the individuals, groups, and organisations that have an interest in\u2014or are affected by\u2014a system, project, or policy, assessing their interests, power, influence, and potential impact on outcomes. It is a foundational practice in requirements engineering, impact assessment, governance design, and change management, enabling decision-makers to anticipate conflicts, prioritise engagement, and design participation mechanisms that are inclusive and legitimate. In AI and technology contexts, stakeholder analysis extends to affected communities who may not self-identify or organise, requiring proactive identification of indirect and diffuse impacts.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:stakeholder-analysis", "labels": [ "Stakeholder Analysis" ], "is_subclass_of": [ "Stakeholder" ], "wikilinks": [] }, { "id": "stakeholder-consultation", "title": "Stakeholder Consultation", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder consultation is the structured process of seeking, gathering, and incorporating the views of parties affected by or interested in a decision, policy, or project. It enhances legitimacy, surfaces risks and local knowledge, and builds trust by giving affected groups a genuine voice before decisions are finalised. Consultation ranges from informing and gathering feedback through to collaborative and participatory decision-making.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stakeholder-consultation", "labels": [ "Stakeholder Consultation" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "stakeholder-coordination", "title": "Stakeholder Coordination", "domain": "governance", "domain_name": "Governance", "definition": "The process of aligning the interests, incentives, and actions of diverse participants \u2014 token holders, developers, validators, users, and institutional actors \u2014 so that a shared protocol, organisation, or ecosystem can reach collective decisions and execute them coherently. It combines formal mechanisms such as voting, delegation, and incentive design with informal channels such as forums, working groups, and signalling, and it is the central problem that governance systems exist to solve.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:stakeholder-coordination", "labels": [ "Stakeholder Coordination" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance", "Blockchain Governance", "Protocol Governance", "DAO" ] }, { "id": "stakeholder-engagement-in-ai", "title": "Stakeholder Engagement in AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Stakeholder Engagement in AI is a participatory process that systematically identifies, involves, and incorporates perspectives from individuals, groups, and communities affected by or having legitimate interests in AI systems, ensuring inclusive design, accountable deployment, and responsive governance. Engagement spans a spectrum from information provision and consultation through to co-design and empowerment mechanisms, and is required by frameworks including the EU AI Act and the OECD AI Principles.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:stakeholder-engagement-in-ai", "labels": [ "Stakeholder Engagement in AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "ISO 26000", "OECD AI Principles", "AIEthicsDomain", "ConceptualLayer", "EU AI Act" ] }, { "id": "stakeholder-engagement", "title": "Stakeholder Engagement", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder Engagement is the structured process through which organisations identify, communicate with, and incorporate the perspectives of individuals and groups who have interests in or are affected by a project, system, or policy. Effective engagement moves along a spectrum from one-way information provision through consultation to active co-design and ongoing collaborative governance. In the context of emerging technologies such as AI, robust stakeholder engagement is recognised as essential for identifying ethical risks, ensuring societal legitimacy, and building the trust necessary for adoption. Regulatory frameworks and voluntary standards increasingly mandate evidence of meaningful stakeholder engagement as a condition of approval or certification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stakeholder-engagement", "labels": [ "Stakeholder Engagement", "Multi-Stakeholder Engagement", "Stakeholder engagement throughout" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "stakeholder-mapping", "title": "Stakeholder Mapping", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder mapping is a visual technique for identifying the individuals, groups, and organisations that affect or are affected by a project, and plotting them against dimensions such as influence and interest to prioritise engagement effort. It produces a structured artefact, typically a grid or network diagram, that guides communication planning and risk management. Stakeholder mapping is commonly used as the initial, visual step within a broader stakeholder analysis process.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:stakeholder-mapping", "labels": [ "Stakeholder Mapping" ], "is_subclass_of": [ "Stakeholder Analysis" ], "wikilinks": [] }, { "id": "stakeholder-participation", "title": "Stakeholder Participation", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder participation is the structured involvement of affected and interested parties in the deliberation, design, and decision-making processes that govern a system, organisation, or protocol. It encompasses mechanisms for consultation, representation, and shared authority that aim to surface diverse interests and confer legitimacy on collective decisions. In governance contexts it is a precondition for accountable, inclusive, and durable outcomes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stakeholder-participation", "labels": [ "Stakeholder Participation" ], "is_subclass_of": [ "Stakeholder Engagement in AI" ], "wikilinks": [] }, { "id": "stakeholder-trust", "title": "Stakeholder Trust", "domain": "governance", "domain_name": "Governance", "definition": "Stakeholder trust is the confident reliance that individuals, groups, or organisations \u2014 including users, operators, regulators, and affected communities \u2014 place in a system, institution, or technology to behave reliably, safely, and in accordance with stated values. In the context of AI and automated systems, stakeholder trust encompasses both calculative assessments of technical competence and affective judgements about integrity and ethical alignment, and is recognised as a prerequisite for adoption, legitimate deployment, and effective governance. It must be earned through demonstrated performance, transparency, and accountability rather than assumed.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:stakeholder-trust", "labels": [ "Stakeholder Trust" ], "is_subclass_of": [ "Trust" ], "wikilinks": [] }, { "id": "stakeholder", "title": "Stakeholder", "domain": "governance", "domain_name": "Governance", "definition": "Any individual, group, organisation, or entity that has an interest in, is affected by, influences, or holds rights regarding an artificial intelligence system throughout its lifecycle, including those who develop, deploy, operate, use, regulate, are impacted by, or hold accountability for AI systems, as well as broader society and communities whose interests may be affected by AI system design, deployment, or outcomes, encompassing both direct participants in AI value chains and indirect parties with legitimate concerns about AI systems' societal implications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:stakeholder", "labels": [ "Stakeholder", "Stakeholder Buy-In", "Stakeholder Notification" ], "is_subclass_of": [ "Legal and Regulatory", "Regulatory Framework" ], "wikilinks": [ "Civil Society", "Participation", "Accountability", "AI Governance", "AI Impact Assessment", "AI Operator", "AI Provider", "AI User", "Fairness", "MetaverseDomain", "Transparency" ] }, { "id": "staking-reward", "title": "Staking Reward", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A staking reward is the compensation paid to participants who lock cryptocurrency to help secure a proof-of-stake network. Rewards are typically funded by protocol issuance and transaction fees and are distributed in proportion to the amount staked and the validator's correct, online participation in consensus. They create the economic incentive that aligns validator behaviour with network security and liveness.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:staking-reward", "labels": [ "Staking Reward" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "staking", "title": "Staking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The act of locking cryptocurrency as a stake to participate in a proof-of-stake network's validation process, earning rewards and risking penalties for misbehaviour. The staked amount aligns a validator's incentives with the network's correct operation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:staking", "labels": [ "Staking", "Cryptoeconomic Staking", "Solo Staking", "Staking Bond", "Staking Mechanism", "Staking Service", "Token Staking" ], "is_subclass_of": [ "Proof of Stake" ], "wikilinks": [ "Validator", "Proof of Stake", "Network Security", "Consensus Mechanism", "Yield Farming", "https://ethereum.org/en/staking/" ] }, { "id": "standard-contractual-clauses", "title": "Standard Contractual Clauses", "domain": "governance", "domain_name": "Governance", "definition": "Standard Contractual Clauses (SCCs) are pre-approved contractual templates issued by the European Commission that enable the lawful transfer of personal data from the European Economic Area to third countries lacking an adequacy decision, by binding data exporters and importers to GDPR-equivalent data protection obligations. The 2021 SCCs replaced older versions and introduced a modular structure covering controller-to-controller, controller-to-processor, processor-to-processor, and processor-to-controller transfers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:standard-contractual-clauses", "labels": [ "Standard Contractual Clauses", "GDPR Standard Contractual Clauses", "Standard Contract Clauses" ], "is_subclass_of": [ "Cross-Border Data Transfer Rule" ], "wikilinks": [] }, { "id": "standard-format-support", "title": "Standard Format Support", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Implementation of standardized data formats enabling interoperability across metaverse platforms, including 3D asset formats like glTF (ISO/IEC 12113:2022), Universal Scene Description and 3D Tiles for seamless exchange of geometry, materials, animations, and scene descriptions between diverse sy...", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:standard-format-support", "labels": [ "Standard Format Support" ], "is_subclass_of": [ "Standards and Interoperability", "Data Interoperability" ], "wikilinks": [ "Cross Platform Asset Exchange", "Data Interoperability", "metaverse" ] }, { "id": "standard-prompting", "title": "Standard Prompting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Standard prompting is the baseline method of querying a large language model by providing an instruction or question, optionally with input-output examples, and expecting an answer without an elicited reasoning process. It establishes the reference behaviour against which more elaborate strategies such as chain-of-thought, self-consistency and tool-augmented prompting are compared. Standard prompting subsumes zero-shot and few-shot formulations that map directly from prompt to answer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:standard-prompting", "labels": [ "Standard Prompting" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "standardisation", "title": "Standardisation", "domain": "standards", "domain_name": "Standards", "definition": "Standardisation is the process by which industry participants, standards bodies and regulators develop and agree formal specifications, protocols and best practices for a technology or process, so that independently built implementations remain interoperable. It typically proceeds through a working group or industry consortium drafting a specification, a review and consensus period, and formal ratification by a recognised standards organisation such as IEEE, ISO or JEDEC. Standardisation reduces fragmentation, lowers integration cost and is a prerequisite for broad ecosystem adoption of a new technology.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:standardisation", "labels": [ "Standardisation", "Standardization" ], "is_subclass_of": [ "Standards" ], "wikilinks": [] }, { "id": "standardization-bodies", "title": "Standardization Bodies", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Organizations developing technical standards for metaverse technologies, including IEEE (Metaverse Standards Committee), W3C (Immersive Web Working Group), ISO/IEC (3D formats), ITU (telecommunications), and the Metaverse Standards Forum coordinating member organizations for interoperability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:standardization-bodies", "labels": [ "Standardization Bodies" ], "is_subclass_of": [ "Standards and Interoperability", "Standards Organization" ], "wikilinks": [ "Technology Interoperability", "metaverse", "Standards Organization" ] }, { "id": "standardized-asset-classification", "title": "Standardized Asset Classification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Systematic taxonomies and frameworks for categorizing digital assets including cryptocurrencies, NFTs, and metaverse tokens based on characteristics such as issuance mod, value mechanism, rights conferred, fungibility, and redemption properties.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:standardized-asset-classification", "labels": [ "Standardized Asset Classification" ], "is_subclass_of": [ "AI Governance and Ethics", "Asset Management" ], "wikilinks": [ "Asset Management", "Blockchain", "metaverse", "Regulatory Compliance" ] }, { "id": "standardized-formats", "title": "Standardized Formats", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Standardized formats are openly specified, consistently structured file and data encodings, such as glTF for 3D assets, USD for scene description, or VRM for avatars, that allow content to be created in one tool and faithfully consumed in another. By fixing geometry, material, animation, and metadata conventions, they decouple content from any single application or platform. They are the precondition for portability, archival longevity, and interoperability across the metaverse content pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:standardized-formats", "labels": [ "Standardized Formats", "Common Data Formats" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [] }, { "id": "standards-based-taxonomy", "title": "Standards Based Taxonomy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A standards-based taxonomy is a hierarchical classification scheme whose categories, labels, and relationships are drawn from published technical standards rather than ad hoc convention. Within spatial computing it organises concepts using vocabularies from bodies such as the W3C, ISO, and the Khronos Group, supporting consistent categorisation and semantic interoperability across platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:standards-based-taxonomy", "labels": [ "Standards Based Taxonomy", "StandardsBasedTaxonomy" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "standards-body", "title": "Standards Body", "domain": "governance", "domain_name": "Governance", "definition": "A standards body is a formal organisation that develops, publishes, and maintains technical specifications, protocols, and reference implementations through structured, consensus-based processes involving multiple stakeholders. Such bodies define interoperability requirements that enable products and services from different vendors to work together reliably, and they may operate at national, regional, or international scope. Governance structures typically include member organisations, technical working groups, public review periods, and intellectual-property licensing frameworks (e.g. royalty-free or FRAND). Well-known examples include ISO, IEC, IEEE, W3C, IETF, NIST, Khronos Group, and the Open Geospatial Consortium.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:standards-body", "labels": [ "Standards Body", "National Standards Body" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "standards-compliance", "title": "Standards Compliance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of ensuring that systems, processes, products, or services meet the requirements of established technical standards, industry specifications, regulatory frameworks, and governance policies applicable to metaverse and virtual world technologies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:standards-compliance", "labels": [ "Standards Compliance" ], "is_subclass_of": [ "AI Governance and Ethics", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "standards-conformance-testing", "title": "Standards Conformance Testing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Testing or verification activities that determine whether a process, product, or service complies with the requirements of a specification, technical standard, contract, or regulation. Conformance testing verifies that implementations faithfully meet specified requirements through structured test suites, automated harnesses, and formal certification regimes \u2014 and is distinct from functional testing in that it judges compliance against an external normative reference rather than internal design intent.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:standards-conformance-testing", "labels": [ "Standards Conformance Testing", "Conformance Test Suite", "Conformance Testing" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "standards-conformance", "title": "Standards Conformance", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The state or process of meeting the requirements specified in a technical standard, specification, or guideline, ensuring that a product, process, or system aligns with internal or industry-specific standards for interoperability, quality, and compatibility.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:standards-conformance", "labels": [ "Standards Conformance", "Conformance Certification", "Conformance Clause", "Conformance Criteria", "System Conformance" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "standards-documentation", "title": "Standards Documentation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Established guidelines, rules, and best practices that govern the creation of technical documentation, including structure, formatting conventions, terminology usage, delivery methods, and presentation to ensure consistency, clarity, and accuracy across all content. Standards documentation underpins interoperability, regulatory compliance, and knowledge transfer in technical domains.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:standards-documentation", "labels": [ "Standards Documentation" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "Blockchain", "metaverse", "Standards" ] }, { "id": "standards-organization", "title": "Standards Organization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A standards organisation is a body that develops, publishes, and maintains technical standards through consensus among industry, academic, and governmental stakeholders. Bodies such as the W3C, ISO, IEEE, and the Khronos Group produce the specifications that underpin interoperability across the web, extended reality, and artificial intelligence systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:standards-organization", "labels": [ "Standards Organization", "SPEC Organization", "Standards Organisation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "standards-validation", "title": "Standards Validation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Standards validation is the process of confirming that a development product satisfies its intended use and stakeholder requirements\u2014distinct from verification, which checks only internal specification conformance. It employs testing, formal review, and compliance checks against published standards such as IEEE 1012 and FDA Computer Software Assurance to ensure the correct artefact was built.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:standards-validation", "labels": [ "Standards Validation" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "standards", "title": "Standards", "domain": "standards", "domain_name": "Standards", "definition": "Established tools, protocols, formats, services, and specifications that support the creation, operation, continuous improvement, and common use of metaverse technologies, enabling interoperability between disparate systems, platforms, and applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:standards", "labels": [ "Standards", "International Standards Ecosystem", "Modeling Standards", "Professional Standards", "Standards Development", "Standards Ecosystem" ], "is_subclass_of": [ "Standards and Interoperability", "Metaverse" ], "wikilinks": [ "Interoperability", "metaverse", "Metaverse" ] }, { "id": "star-cluster", "title": "Star Cluster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:star-cluster", "labels": [ "Star Cluster" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "star-formation", "title": "Star Formation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:star-formation", "labels": [ "Star Formation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "star-tracker", "title": "Star Tracker", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "A star tracker is an optical [[Sensor|sensor]] that estimates a spacecraft's absolute three-axis orientation by imaging stars and matching their pattern to an on-board catalogue.[^1] It measures attitude relative to the celestial reference field. It does not, by itself, provide the spacecraft's orbital position.", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:star-tracker", "labels": [ "Star Tracker" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [ "sensor" ] }, { "id": "star", "title": "Star", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:star", "labels": [ "Star" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "starknet", "title": "Starknet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Layer 2 validity rollup for Ethereum that uses STARK proofs to verify off-chain execution of transactions written in the Cairo programming language, enabling high-throughput, low-cost transactions while inheriting Ethereum's settlement security.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:starknet", "labels": [ "Starknet", "StarkNet" ], "is_subclass_of": [ "Layer 2 Networks" ], "wikilinks": [ "Zero-Knowledge Proof", "Ethereum", "Scalability", "Rollup", "Layer 2 Networks" ] }, { "id": "start-configuration", "title": "Start Configuration", "domain": "robotics", "domain_name": "Robotics", "definition": "A start configuration is the initial pose of a robot or articulated system, expressed as a point in its configuration space, from which a motion or path planner must compute a feasible route to a goal configuration. It encodes the complete set of joint values or positional parameters that fully describe the system's state at the beginning of a planned motion. Together with the goal configuration, it bounds the planning query.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:start-configuration", "labels": [ "Start Configuration" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "startup-ecosystem", "title": "Startup Ecosystem", "domain": "governance", "domain_name": "Governance", "definition": "A startup ecosystem is the interconnected network of founders, investors, talent, universities, accelerators, service providers, and regulators within a region that collectively enables new high-growth companies to form and scale. Its health depends on capital availability, dense knowledge spillovers, and supportive policy, which together produce self-reinforcing network effects. Concentrations such as Silicon Valley illustrate how these dynamics compound advantage and shape the trajectory of emerging technologies including AI.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:startup-ecosystem", "labels": [ "Startup Ecosystem" ], "is_subclass_of": [ "Economics" ], "wikilinks": [] }, { "id": "state-change", "title": "State Change", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A State Change is the transition from one defined configuration or status to another within a system, component, or entity, representing how systems evolve through discrete or continuous modifications to their properties. State changes are fundamental to distributed systems, blockchain ledgers (where account balances and smart-contract storage are updated atomically), digital twins (where physical sensor readings update virtual representations), and agentic AI systems (where goal statuses transition through planning and execution). Formal modelling of state changes enables audit trails, causality analysis, and consistency guarantees.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:state-change", "labels": [ "State Change" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "state-channel", "title": "State Channel", "domain": "blockchain", "domain_name": "Blockchain", "definition": "State channels are Layer 2 scaling solutions enabling off-chain interactions between participants through signed state updates, requiring only on-chain transactions for channel opening, closing, and dispute resolution, thereby achieving instant finality and near-zero marginal transaction costs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:state-channel", "labels": [ "State Channel", "Channel Protocol", "Ethereum State Channels", "State Channels", "StateChannel" ], "is_subclass_of": [ "Layer 2 Scaling", "Blockchain" ], "wikilinks": [ "Connext", "Layer 2", "Off-Chain Scaling", "Optimistic Rollup", "Payment Channel", "Perun", "Raiden Network", "Rollup", "Blockchain", "BlockchainDomain", "Digital Signature", "Digital Twin", "Lightning Network", "Micropayments" ] }, { "id": "state-estimation", "title": "State Estimation", "domain": "robotics", "domain_name": "Robotics", "definition": "State Estimation is the computational discipline of inferring the latent, unobservable internal state of a dynamical system \u2014 such as position, velocity, orientation, and joint angles \u2014 from a sequence of noisy, incomplete sensor measurements, using probabilistic inference frameworks. Core algorithms include the Kalman Filter and its nonlinear extensions (EKF, UKF), particle filters, and factor-graph optimisation methods that maintain a belief distribution over state variables over time. The field underpins autonomous navigation, robotic manipulation, aerospace guidance systems, and any cyber-physical system that must act on estimated rather than directly observed quantities. Modern formulations unify Bayesian filtering, maximum-a-posteriori smoothing, and deep-learning-based perception to achieve robust estimation under sensor failure, model mismatch, and adversarial environments.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:state-estimation", "labels": [ "State Estimation", "Probabilistic State Estimation", "Recursive State Estimation", "Robotics State Estimation", "State Estimator" ], "is_subclass_of": [ "Probabilistic Inference" ], "wikilinks": [ "Robotics Systems" ] }, { "id": "state-machine-replication", "title": "State Machine Replication", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "State Machine Replication (SMR) is a fault-tolerance and consistency technique in which multiple server replicas each maintain an identical deterministic state machine by processing the same totally-ordered sequence of client commands, ensuring that all correct replicas converge to the same state after executing every command. The approach was formalised by Leslie Lamport and later by Fred Schneider, and it provides the theoretical foundation for consensus protocols such as Paxos, Raft, and Viewstamped Replication. SMR simultaneously achieves high availability and strong consistency in the presence of crash or Byzantine failures by decoupling the agreement problem (ordering) from the execution problem (state transition). It is the core architectural abstraction underlying permissioned blockchains, cloud database replication, and coordination services such as Apache ZooKeeper.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:state-machine-replication", "labels": [ "State Machine Replication", "Replicated State Machine", "State Replication Protocol" ], "is_subclass_of": [ "Distributed Computing" ], "wikilinks": [ "owl:Thing" ] }, { "id": "state-machine", "title": "State Machine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "StateMachine is a computational abstraction \u2014 rooted in automata theory and formal language theory \u2014 representing a system whose behaviour is determined by a finite (or structured infinite) set of discrete states, a defined alphabet of inputs or events, a transition function mapping (state \u00d7 inpu...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:state-machine", "labels": [ "State Machine", "Status State Machine" ], "is_subclass_of": [ "Computing and Cloud", "Automata Theory", "Computational Model", "Formal Method", "Concurrency Theory", "Distributed Computing" ], "wikilinks": [ "Accepting State", "Actor Model", "AlgorithmLayer", "Automata Theory", "Behaviour Tree", "Compiler", "Computational Model", "ComputerScienceDomain", "Concurrency Theory", "Dataflow Model", "Determinism", "DistributedSystemsDomain", "Event Model", "Finite Automaton", "Formal Language Theory", "Formal Method", "FormalMethodsDomain", "Game AI", "IEEE 802.X", "IETF RFC 9562" ] }, { "id": "state-management", "title": "State Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "State Management is the set of patterns, architectures, and tools used to represent, update, synchronise, and persist the mutable state of an application or distributed system in a predictable and auditable way. In single-page web applications, state management libraries such as Redux implement unidirectional data flow and immutable state trees. In distributed systems, state management encompasses consensus protocols, state machine replication, and event sourcing patterns that ensure all nodes maintain consistent views of shared state despite failures and network partitions. Blockchain ledgers represent a specialised form of state management with Byzantine fault-tolerant distributed consensus.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:state-management", "labels": [ "State Management", "UI State Management", "World State Management" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "state-observer", "title": "State Observer", "domain": "robotics", "domain_name": "Robotics", "definition": "A state observer is a dynamical system that estimates the internal state of a controlled plant from its measured inputs and outputs when the full state cannot be directly sensed. By running a model of the plant in parallel and correcting it with the measurement error, an observer reconstructs unmeasured variables for use in feedback control. Classical examples include the Luenberger observer for deterministic systems and the Kalman filter for stochastic systems, making observers essential to state-feedback control of partially observable processes.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:state-observer", "labels": [ "State Observer", "Observer", "StateObserver" ], "is_subclass_of": [ "State Estimation" ], "wikilinks": [] }, { "id": "state-proof", "title": "State Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A compact cryptographic attestation that a particular state \u2014 an account balance, storage slot, or block commitment \u2014 is part of a blockchain's canonical history, verifiable by anyone without replaying the chain or trusting an intermediary; ranging from Merkle inclusion proofs against a state root to Algorand-style aggregate-signature certificates, state proofs are the primitive that lets light clients and cross-chain bridges verify one chain's state from another vantage point trustlessly.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:state-proof", "labels": [ "State Proof" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [ "Cryptographic Proof", "Light Client", "Algorand", "Merkle Tree" ] }, { "id": "state-representation", "title": "State Representation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "State representation is the encoding scheme used to capture the relevant information about a system or environment at a given point, transforming raw observations into a form suitable for prediction, planning, or synchronisation. Choices range from hand-crafted low-dimensional feature vectors to learned latent embeddings produced by an encoder network, and the choice materially affects downstream sample efficiency and generalisation. It is a prerequisite for constructing a world model and for synchronising state across distributed or networked simulations.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:state-representation", "labels": [ "State Representation" ], "is_subclass_of": [ "State Space" ], "wikilinks": [] }, { "id": "state-space-control", "title": "State Space Control", "domain": "robotics", "domain_name": "Robotics", "definition": "State Space Control is a modern control theory framework that describes dynamic systems as a set of first-order differential equations over an internal state vector, enabling full-state feedback design via techniques such as pole placement, LQR, and model predictive control. Unlike classical frequency-domain methods, state-space representations directly support multi-input multi-output systems, observer design, and optimal control synthesis, making them foundational for robotic and aerospace applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:state-space-control", "labels": [ "State Space Control", "LQR Control" ], "is_subclass_of": [ "Actuation and Control", "Feedback Control" ], "wikilinks": [ "Feedback Control", "Robotics" ] }, { "id": "state-space-model", "title": "State Space Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A State Space Model (SSM) is a mathematical framework that represents a dynamical system through a hidden (latent) state vector whose evolution over discrete or continuous time is governed by linear or learnable recurrence equations, paired with an output equation mapping states to observations. Originally formalised in control theory and signal processing \u2014 with the Kalman filter as a canonical inference algorithm \u2014 SSMs have been re-parameterised as structured sequence layers in deep learning, enabling sub-quadratic scaling in sequence length as an alternative to self-attention. Modern deep SSM variants such as S4, Mamba, and RWKV exploit diagonal or low-rank structure in the state transition matrix to achieve hardware-efficient training and inference on long sequences.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:state-space-model", "labels": [ "State Space Model", "Mamba State-Space Model" ], "is_subclass_of": [ "AI Model Architecture", "Sequence Model" ], "wikilinks": [ "Probabilistic Model", "Language Model", "Transformer", "Neural Network" ] }, { "id": "state-space-models", "title": "State Space Models", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "State space models are sequence models that maintain a hidden state evolving over time according to linear dynamics, used as an alternative to attention for long sequences. Recent deep learning variants make the dynamics input-dependent to capture context.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:state-space-models", "labels": [ "State Space Models" ], "is_subclass_of": [ "State Space Model" ], "wikilinks": [ "Neural Network", "Generative Models", "Recurrent Neural Network", "Attention Mechanisms", "State Space Model" ] }, { "id": "state-space-representation", "title": "State Space Representation", "domain": "robotics", "domain_name": "Robotics", "definition": "State space representation is a mathematical model that describes a dynamical system through a set of state variables, capturing all information needed to determine the system's future behaviour given its inputs. It expresses the system as first-order differential or difference equations relating states, inputs and outputs, typically in matrix form. The formulation underpins modern control, estimation, planning and search by providing a compact, computable description of the system over time.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:state-space-representation", "labels": [ "State Space Representation", "State-Space Representation" ], "is_subclass_of": [ "Dynamical Systems Theory" ], "wikilinks": [] }, { "id": "state-space-search", "title": "State Space Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "State space search is a foundational problem-solving paradigm in artificial intelligence in which a problem is formalised as an initial state, a set of operators that transform states, and a goal test; solving the problem means finding a path through the implicit graph of reachable states from the initial state to a goal state. Uninformed strategies such as breadth-first and depth-first search enumerate states systematically, whilst informed strategies exploit heuristic estimates to focus effort, underpinning classical planning, game playing and combinatorial optimisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:state-space-search", "labels": [ "State Space Search" ], "is_subclass_of": [ "Search Algorithm" ], "wikilinks": [ "Search Algorithm", "Automated Planning", "Heuristic Search", "State Space" ] }, { "id": "state-space-sequence-models", "title": "State Space Sequence Models", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "State Space and Other Approaches encompasses structured state space models (SSMs) and related sequence modelling architectures, most notably the Mamba family, which apply selective state-space mechanisms to achieve linear-time sequence processing. These approaches offer a compelling alternative to Transformers by combining efficient recurrence with hardware-aware algorithms, enabling superior throughput on long sequences across language, vision, and genomics domains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:state-space-sequence-models", "labels": [ "State Space Sequence Models", "State Space and Other Approaches" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "state-space", "title": "State Space", "domain": "ai", "domain_name": "Ai", "definition": "A state space is the mathematical set of all possible configurations (states) of a dynamical system, together with the transition rules that govern how the system evolves from one state to another over time. In control theory it is represented by first-order differential or difference equations relating state variables, inputs, and outputs; in artificial intelligence it denotes the complete set of system configurations that a search or planning algorithm may explore.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:state-space", "labels": [ "State Space" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "state-synchronisation", "title": "State Synchronisation", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "State synchronisation is the process of ensuring that multiple distributed nodes or replicas of a system maintain consistent views of shared mutable state despite network partitions, concurrent updates, and node failures. It encompasses techniques including leader-based replication, consensus protocols, conflict-free replicated data types (CRDTs), and operational transformation to reconcile divergent states. State synchronisation defines the trade-off between consistency, availability, and partition tolerance as described by the CAP theorem, choosing different points on that spectrum depending on application requirements. It is fundamental to distributed databases, real-time collaboration tools, multiplayer games, and blockchain networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:state-synchronisation", "labels": [ "State Synchronisation", "Distributed State Synchronisation", "State Synchronization", "State Synchronization Engine" ], "is_subclass_of": [ "Distributed System Protocol" ], "wikilinks": [] }, { "id": "state-transition-function", "title": "State Transition Function", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A state transition function is the deterministic rule of a blockchain protocol that maps a current ledger state and a validated block of transactions to the next state, defining exactly how balances, contract storage, and other state evolve. Because every honest node applies the same function to the same inputs, all nodes converge on identical state, which is the basis of replicated consensus. It is the formal heart of a blockchain understood as a replicated state machine.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:state-transition-function", "labels": [ "State Transition Function", "State Transition", "State Transition Model", "Transition Function" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "stateless-architecture", "title": "Stateless Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Stateless architecture is a design approach in which each request carries all the information needed to process it, and the serving component retains no client session state between requests. By externalising state to caches, databases, or tokens, stateless services can be freely replicated, replaced, and load-balanced, which simplifies horizontal scaling and fault recovery at the cost of pushing state management to dedicated stores.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stateless-architecture", "labels": [ "Stateless Architecture" ], "is_subclass_of": [ "Cloud Native" ], "wikilinks": [] }, { "id": "stateless-protocol", "title": "Stateless Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A stateless protocol is a communication protocol in which each request from a client to a server is treated independently, carrying all the information needed to be understood without relying on stored context from previous requests. The server retains no session state between requests, which simplifies server design and improves scalability and resilience because any server instance can handle any request. HTTP is the canonical example; where continuity is needed, state is reintroduced at a higher layer through tokens, cookies or explicit session management.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stateless-protocol", "labels": [ "Stateless Protocol" ], "is_subclass_of": [ "Protocol" ], "wikilinks": [] }, { "id": "static-analysis", "title": "Static Analysis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Static analysis is the examination of software source code, byte code or binaries without executing the program, in order to detect defects, security vulnerabilities, style violations and correctness properties. Techniques range from simple pattern-based linting to formal abstract interpretation and data-flow analysis over the program's control structure. It is commonly integrated into editors and continuous-integration pipelines to provide early feedback before code runs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:static-analysis", "labels": [ "Static Analysis" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "station-keeping", "title": "Station Keeping", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:station-keeping", "labels": [ "Station Keeping" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "statistical-analysis", "title": "Statistical Analysis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Statistical analysis is the systematic application of statistical methods to collect, describe, model, and draw inferences from data, quantifying uncertainty and supporting evidence-based conclusions. It spans descriptive summarisation, exploratory examination, hypothesis testing, and predictive modelling, relying on probability theory to generalise from samples to populations. In artificial intelligence and data science it provides the inferential foundation for validating models, interpreting results, and reasoning under uncertainty.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:statistical-analysis", "labels": [ "Statistical Analysis" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "statistical-hypothesis-testing", "title": "Statistical Hypothesis Testing", "domain": "ai", "domain_name": "Ai", "definition": "Statistical hypothesis testing is a formal procedure for determining whether an observed difference between two outcomes, such as the performance of two models, is unlikely to have arisen by chance under a stated null hypothesis. It computes a test statistic and associated p-value from sample data and compares it against a significance threshold to accept or reject the null hypothesis. In machine learning it is used to compare model or benchmark scores across runs, guarding against overinterpreting differences that fall within noise. Common tests include the paired t-test, Wilcoxon signed-rank test, and bootstrap resampling for non-normal score distributions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:statistical-hypothesis-testing", "labels": [ "Statistical Hypothesis Testing" ], "is_subclass_of": [ "Model Evaluation" ], "wikilinks": [] }, { "id": "statistical-inference", "title": "Statistical Inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Statistical inference is the process of drawing conclusions about a population or data-generating process from a finite sample, while quantifying the uncertainty of those conclusions. It encompasses estimation of parameters, hypothesis testing and prediction, grounded in probability theory. Both frequentist and Bayesian frameworks provide formal machinery for reasoning under sampling variability, underpinning model evaluation in machine learning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:statistical-inference", "labels": [ "Statistical Inference" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [] }, { "id": "statistical-learning-theory", "title": "Statistical Learning Theory", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A framework that studies the conditions under which algorithms can generalise from finite training data to unseen data, providing theoretical bounds on prediction error via concepts such as VC dimension, PAC learnability, and Rademacher complexity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:statistical-learning-theory", "labels": [ "Statistical Learning Theory", "Statistical Learning" ], "is_subclass_of": [ "Machine Learning Discipline", "AI Research Area" ], "wikilinks": [ "Probability Theory", "Statistics", "Supervised Learning", "Deep Learning", "Overfitting", "Machine Learning" ] }, { "id": "statistical-mechanics", "title": "Statistical Mechanics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The branch of physics that derives the macroscopic behaviour of matter from the statistical properties of its microscopic constituents, explaining thermodynamic quantities such as temperature, pressure, and entropy as averages over ensembles of particle configurations. Built on probability theory and the Boltzmann distribution, it provides the mathematical machinery \u2014 partition functions, ensembles, phase transitions \u2014 that now underpins complexity science, information theory, and energy-based machine learning models.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:statistical-mechanics", "labels": [ "Statistical Mechanics" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [ "Probability Theory", "Thermodynamics", "Complex Systems" ] }, { "id": "statistical-model", "title": "Statistical Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A statistical model is a formal mathematical representation of a data-generating process, expressed as a set of probability distributions over observed and latent variables. It provides the foundation for Bayesian inference, where prior distributions are updated with observed evidence, and for behavioural modelling, where such models describe patterns in agent or system behaviour. Statistical models range from simple parametric forms to complex hierarchical and graphical structures.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:statistical-model", "labels": [ "Statistical Model" ], "is_subclass_of": [ "Probabilistic Model" ], "wikilinks": [] }, { "id": "statistical-modelling", "title": "Statistical Modelling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Statistical modelling is the practice of representing data-generating processes with mathematical structures built on probability theory, in order to describe relationships, quantify uncertainty, test hypotheses, and make inferences or predictions. It encompasses approaches such as regression, generalised linear models, time-series models, and Bayesian methods, emphasising interpretable parameters and explicit assumptions. It provides the formal foundation on which much of machine learning and data analysis is built.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:statistical-modelling", "labels": [ "Statistical Modelling" ], "is_subclass_of": [ "Data Analysis" ], "wikilinks": [] }, { "id": "statistical-process-control", "title": "Statistical Process Control", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Statistical Process Control (SPC) is a quality-control methodology that applies statistical methods to monitor and control a manufacturing or service process. By tracking process variation over time against statistically derived control limits, SPC distinguishes ordinary common-cause variation from assignable special-cause variation, enabling operators to intervene only when a process is genuinely out of control. It underpins continuous improvement and defect prevention in industrial production.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:statistical-process-control", "labels": [ "Statistical Process Control" ], "is_subclass_of": [ "Quality Control" ], "wikilinks": [] }, { "id": "statistical-testing", "title": "Statistical Testing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Statistical testing is the practice of using sample data to assess evidence for or against a hypothesis about a population or process. It formalises a null and alternative hypothesis, computes a test statistic and an associated significance level, and decides whether observed effects are likely to be genuine rather than due to chance. It underpins rigorous evaluation of models, experiments and measurements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:statistical-testing", "labels": [ "Statistical Testing", "Statistical Significance Testing" ], "is_subclass_of": [ "Statistical Analysis" ], "wikilinks": [] }, { "id": "statistics", "title": "Statistics", "domain": "data", "domain_name": "Data", "definition": "The discipline concerned with collecting, analysing, interpreting, and drawing conclusions from data under uncertainty, encompassing both frequentist and Bayesian frameworks for inference, estimation, and hypothesis testing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:statistics", "labels": [ "Statistics", "Comparative Statistics", "Mean and Variance Statistics", "National Statistics" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [ "Probability Theory", "Statistical Learning Theory", "Bayesian Inference", "Machine Learning" ] }, { "id": "stdio-transport", "title": "Stdio Transport", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Stdio transport is a communication mechanism in which a client and server exchange messages over the standard input and standard output streams of a locally spawned process. It is one of the primary transports defined by the Model Context Protocol, where the host launches the server as a subprocess and frames JSON-RPC messages through stdin/stdout. It suits local, single-machine integrations because it requires no network sockets and inherits the operating system's process isolation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:stdio-transport", "labels": [ "Stdio Transport", "stdio" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "stealth-address", "title": "Stealth Address", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A stealth address is a privacy-enhancing technique that lets a recipient publish a single static address while every payment is sent to a unique, unlinkable one-time address derived on-chain. The sender combines the recipient's public scan and spend keys with ephemeral randomness, using elliptic-curve Diffie-Hellman to compute a destination only the recipient can detect and spend. Stealth addresses break the public linkage between a recipient's identity and their incoming transactions without requiring interaction or a shared secret beforehand.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:stealth-address", "labels": [ "Stealth Address" ], "is_subclass_of": [ "Privacy Preserving Blockchain" ], "wikilinks": [] }, { "id": "steganography", "title": "Steganography", "domain": "security", "domain_name": "Security", "definition": "Steganography is the practice of concealing the existence of a message by embedding it within an innocuous carrier such as an image, audio file or text, so that only the intended recipient is aware that hidden information is present. Unlike encryption, which scrambles content but leaves it visibly protected, steganography aims to make the very presence of communication undetectable. The two techniques are complementary and are often combined for layered confidentiality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:steganography", "labels": [ "Steganography" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "stellar-active-region", "title": "Stellar Active Region", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-active-region", "labels": [ "Stellar Active Region" ], "is_subclass_of": [ "Stellar Realm" ], "wikilinks": [] }, { "id": "stellar-astrophysics", "title": "Stellar Astrophysics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-astrophysics", "labels": [ "Stellar Astrophysics" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "stellar-atmosphere", "title": "Stellar Atmosphere", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-atmosphere", "labels": [ "Stellar Atmosphere" ], "is_subclass_of": [ "Stellar Realm" ], "wikilinks": [] }, { "id": "stellar-core", "title": "Stellar Core", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-core", "labels": [ "Stellar Core" ], "is_subclass_of": [ "Stellar Interior" ], "wikilinks": [] }, { "id": "stellar-corona", "title": "Stellar Corona", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-corona", "labels": [ "Stellar Corona" ], "is_subclass_of": [ "Stellar Atmosphere" ], "wikilinks": [] }, { "id": "stellar-evolution", "title": "Stellar Evolution", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-evolution", "labels": [ "Stellar Evolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "stellar-interior", "title": "Stellar Interior", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-interior", "labels": [ "Stellar Interior" ], "is_subclass_of": [ "Stellar Realm" ], "wikilinks": [] }, { "id": "stellar-nucleosynthesis", "title": "Stellar Nucleosynthesis", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-nucleosynthesis", "labels": [ "Stellar Nucleosynthesis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "stellar-realm", "title": "Stellar Realm", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-realm", "labels": [ "Stellar Realm" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "stellar-remnant", "title": "Stellar Remnant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stellar-remnant", "labels": [ "Stellar Remnant" ], "is_subclass_of": [ "Astronomical Body" ], "wikilinks": [] }, { "id": "stellar", "title": "Stellar", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Stellar is an open blockchain network and protocol designed for fast, low-cost transfer and exchange of digital representations of value, including fiat-backed tokens and stablecoins.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:stellar", "labels": [ "Stellar", "Stellar Network" ], "is_subclass_of": [ "Blockchain", "Layer 1 Blockchain", "Blockchain Domain" ], "wikilinks": [ "Blockchain", "Cross-Border Payments", "Stablecoin", "Payment Network", "Blockchain Domain" ] }, { "id": "stepper-motor", "title": "Stepper Motor", "domain": "robotics", "domain_name": "Robotics", "definition": "A stepper motor is an electromechanical actuator that divides a full rotation into a fixed number of discrete angular steps, enabling precise open-loop position control without requiring feedback sensors. By energising coils in sequence, the rotor advances one step per pulse, making stepper motors essential for applications demanding repeatable positioning such as CNC machining, 3D printing, and robotic joints.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:stepper-motor", "labels": [ "Stepper Motor" ], "is_subclass_of": [ "Actuation and Control", "Electric Motor" ], "wikilinks": [ "Electric Motor", "Robotics" ] }, { "id": "stereo-camera", "title": "Stereo Camera", "domain": "robotics", "domain_name": "Robotics", "definition": "A dual-lens imaging system that captures two horizontally offset images of a scene to compute disparity maps and recover metric 3D depth information. Stereo cameras are widely used in robotic navigation, obstacle avoidance, and spatial mapping because they provide passive depth sensing without emitting light.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stereo-camera", "labels": [ "Stereo Camera" ], "is_subclass_of": [ "Perception and Sensing", "Camera" ], "wikilinks": [ "Camera", "Robotics" ] }, { "id": "stereo-rectification", "title": "Stereo Rectification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Stereo rectification is the process of transforming a pair of stereo camera images so that corresponding points lie on the same horizontal scan line, reducing the search for correspondences from a two-dimensional problem to a one-dimensional one along epipolar lines. It relies on the epipolar geometry of the camera pair, computed from calibration or estimated from matched features, to derive the reprojection homographies applied to each image. Rectified image pairs are the standard input to dense stereo matching algorithms used for depth estimation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:stereo-rectification", "labels": [ "Stereo Rectification" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "stereo-vision", "title": "Stereo Vision", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Stereo vision is a computational technique that recovers metric depth information from a scene by analysing the horizontal displacement (disparity) between corresponding points in two or more rectified images captured from laterally separated viewpoints, mimicking the binocular parallax used by human and animal visual systems. The resulting dense disparity maps are converted into 3D point clouds or depth maps for downstream perception tasks.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:stereo-vision", "labels": [ "Stereo Vision", "Middlebury Stereo", "RAFT-Stereo" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "stewart-platform", "title": "Stewart Platform", "domain": "robotics", "domain_name": "Robotics", "definition": "A Stewart Platform (also known as a Gough-Stewart platform) is a type of parallel manipulator consisting of six variable-length prismatic actuators (struts) connecting a fixed base plate to a moveable top plate via universal or spherical joints, enabling six degrees of freedom\u2014three translational and three rotational\u2014within a compact, high-stiffness mechanical structure. First described by V.E. Gough in 1954 for tyre testing and later analysed by D. Stewart in 1965 for flight simulation, the architecture is characterised by high load-bearing capacity, positional accuracy, and mechanical rigidity compared to serial manipulators.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:stewart-platform", "labels": [ "Stewart Platform" ], "is_subclass_of": [ "Robot Type", "Parallel Robot" ], "wikilinks": [ "Parallel Robot", "Robotics" ] }, { "id": "sticky-notes", "title": "Sticky Notes", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Sticky Notes are digital equivalents of physical adhesive notes used on virtual whiteboards or collaboration surfaces to capture brief ideas, action items, or annotations. Multiple participants can add, move, and colour-code notes simultaneously, making them a lightweight tool for brainstorming and retrospectives in remote teams. They are widely supported in tools such as Miro, FigJam, and Microsoft Whiteboard.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:sticky-notes", "labels": [ "Sticky Notes" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "stigmergy", "title": "Stigmergy", "domain": "robotics", "domain_name": "Robotics", "definition": "An indirect coordination mechanism in which agents interact through modifications to a shared environment rather than through direct peer-to-peer communication. Originating in the study of social insects, stigmergy underlies emergent collective behaviour in swarm robotics and multi-agent systems: individual agents leave environmental signals (analogous to pheromone trails) that guide subsequent agents, producing globally coordinated outcomes from purely local rules.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:stigmergy", "labels": [ "Stigmergy" ], "is_subclass_of": [ "Swarm Robotics", "Robotics" ], "wikilinks": [ "Robotics Research", "Swarm Intelligence", "Robotics", "RoboticsDomain" ] }, { "id": "stochastic-differential-equation", "title": "Stochastic Differential Equation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A stochastic differential equation (SDE) is a differential equation in which one or more terms incorporate a stochastic process, typically Brownian motion or white noise, making the solution itself a stochastic process. SDEs generalise ordinary differential equations by including a diffusion term driven by a Wiener process, and their solutions are interpreted via Ito or Stratonovich calculus. SDEs are foundational in financial mathematics, physics, biology, and machine learning \u2014 particularly in score-based generative modelling and diffusion models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stochastic-differential-equation", "labels": [ "Stochastic Differential Equation" ], "is_subclass_of": [ "Differential Equations" ], "wikilinks": [] }, { "id": "stochastic-gradient-descent", "title": "Stochastic Gradient Descent", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Stochastic Gradient Descent (SGD) is an iterative optimisation algorithm that updates model parameters by computing gradients from randomly sampled mini-batches rather than the full training dataset, trading gradient accuracy for computational efficiency and the ability to escape shallow local minima. SGD and its adaptive variants (Adam, RMSprop, AdaGrad) are the primary training algorithms for deep neural networks across vision, language, and reinforcement learning domains.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:stochastic-gradient-descent", "labels": [ "Stochastic Gradient Descent", "Mini-Batch Gradient" ], "is_subclass_of": [ "Gradient Descent" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "stochastic-optimisation", "title": "Stochastic Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The branch of mathematical optimisation concerned with minimising or maximising objectives that involve randomness \u2014 either because the objective and constraints are expectations over uncertain data, or because the algorithm itself deliberately injects randomness, as in stochastic gradient methods, simulated annealing, and evolutionary search. It provides the convergence theory and algorithmic machinery, from Robbins-Monro stochastic approximation to Adam, that makes training large-scale machine learning models on sampled mini-batches tractable.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:stochastic-optimisation", "labels": [ "Stochastic Optimisation" ], "is_subclass_of": [ "Mathematical Optimisation" ], "wikilinks": [ "Mathematical Optimisation", "Stochastic Gradient Descent", "Convex Optimisation", "Random Search", "Monte Carlo Methods" ] }, { "id": "stochastic-process", "title": "Stochastic Process", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A stochastic process is a mathematically rigorous framework for describing the probabilistic evolution of a system over time (or another index set), defined formally as a collection of random variables {X_t : t \u2208 T} all defined on a common probability space (\u03a9, \u2131, \u2119) and indexed by a parameter se...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:stochastic-process", "labels": [ "Stochastic Process" ], "is_subclass_of": [ "AI Technique", "Dynamical Systems", "Probability Theory", "Mathematical Statistics", "Time Series Analysis", "Dynamical System", "Measure Theory" ], "wikilinks": [ "AlgorithmLayer", "Annals of Probability", "Autoregressive Model", "Bayesian Inference", "Bayesian Optimisation", "Black & Scholes 1973 Option Pricing", "Chapman-Kolmogorov Equation", "Climate Modelling", "Dalalyan 2017 ULA Guarantees", "Detailed Balance", "Deterministic Dynamical System", "Differential Equations", "Doob 1953 Stochastic Processes", "Dynamical System", "Epidemiology Modelling", "Filtration", "Financial Derivatives Pricing", "FinancialMathematicsDomain", "Fokker-Planck Equation", "Frequentist Statistics" ] }, { "id": "stochastic-processes", "title": "Stochastic Processes", "domain": "data", "domain_name": "Data", "definition": "A stochastic process is a collection of random variables indexed by time or another parameter, used to model systems that evolve with randomness. Examples include random walks, Markov chains, Poisson processes and Brownian motion, each characterised by how its future behaviour depends on its present and past states. Stochastic processes are foundational in probability theory and are applied across physics, biology, queueing theory and quantitative finance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stochastic-processes", "labels": [ "Stochastic Processes" ], "is_subclass_of": [ "Probability Theory", "owl:Thing" ], "wikilinks": [ "Markov Chain", "Brownian Motion", "Probability Theory", "Measure Theory", "Quantitative Finance", "Reinforcement Learning", "Time Series Analysis", "owl:Thing" ] }, { "id": "storage-architecture", "title": "Storage Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Storage architecture is the structured design of how data is persisted, organised, and accessed across hardware and software layers, encompassing block, file, and object paradigms as well as tiering, replication, and consistency models. It defines the trade-offs between durability, latency, throughput, and cost for a given workload. Sound storage architecture underpins databases, data lakes, and distributed systems by matching access patterns to the right storage substrate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:storage-architecture", "labels": [ "Storage Architecture" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "storage-engine", "title": "Storage Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A storage engine is the component of a database system responsible for how data is physically laid out, written, indexed and retrieved on durable media. It implements the on-disk data structures, transaction and concurrency control, write-ahead logging and recovery that guarantee durability and consistency. Different engines, such as B-tree and log-structured merge-tree designs, optimise for different read, write and space trade-offs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:storage-engine", "labels": [ "Storage Engine" ], "is_subclass_of": [ "Data Persistence" ], "wikilinks": [] }, { "id": "storage-hardware", "title": "Storage Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Storage Hardware comprises the physical devices and subsystems responsible for persistently recording, retrieving, and managing digital data within computing infrastructure. This category covers technologies ranging from solid-state drives and NVMe arrays to distributed storage nodes and archival media. In AI and spatial computing contexts, storage hardware is a critical constraint on model training throughput, inference latency, and the scalability of data pipelines.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:storage-hardware", "labels": [ "Storage Hardware" ], "is_subclass_of": [ "Platform and Environment", "Hardware Component" ], "wikilinks": [ "Hardware Component" ] }, { "id": "storage-infrastructure", "title": "Storage Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Comprehensive ecosystem of physical and logical systems that provide durable, scalable, and performant data retention, retrieval, and replication across distributed computing environments.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:storage-infrastructure", "labels": [ "Storage Infrastructure", "StorageInfrastructure" ], "is_subclass_of": [ "Data Management", "Data Layer", "Infrastructure", "Cloud Computing", "Data Engineering", "Distributed Systems", "Information Systems" ], "wikilinks": [ "AES-256 Encryption", "AI Model Training", "Apache Hudi", "Apache Iceberg", "Backup and Recovery", "Big Data Analytics", "Blake3", "Block Storage", "Ceph", "Cloud Native", "Consensus Mechanisms", "Consistency Protocols", "Consistent Hashing", "Content Delivery Networks", "CRDTs", "CRUSH Algorithm", "Cryptographic Hash Functions", "CXL Consortium", "Data Engineering", "Data Lakes" ] }, { "id": "storage-layer", "title": "Storage Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hardware and software infrastructure responsible for persistent retention, retrieval, and management of data and digital assets across distributed systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:storage-layer", "labels": [ "Storage Layer", "StorageLayer" ], "is_subclass_of": [ "Data Management", "Infrastructure Layer" ], "wikilinks": [ "Block Storage", "Cache Layer", "Content Delivery", "Data Persistence", "Data Storage Infrastructure", "File System", "MSF Taxonomy 2025", "Network Connectivity", "Object Storage", "Replication Protocol", "Asset Management", "Data Management", "Database System", "InfrastructureDomain", "Infrastructure Layer", "Storage Hardware" ] }, { "id": "storage-systems", "title": "Storage Systems", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Storage Systems are the hardware and software architectures responsible for persisting, organising, retrieving, and protecting digital data across the full hierarchy from on-chip registers and DRAM through local SSDs and HDDs to distributed cloud object stores and decentralised peer-to-peer networks. The discipline encompasses storage media technology, file systems, block and object storage interfaces, data durability through redundancy (RAID, erasure coding), consistency and replication protocols for distributed deployments, and the performance-cost-durability tradeoffs that govern system design. Storage Systems are foundational infrastructure for every computing application, with particular complexity arising in distributed and decentralised configurations where network partitions, node failures, and latency variability must be handled.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:storage-systems", "labels": [ "Storage Systems", "Storage", "Storage Solutions", "Storage System" ], "is_subclass_of": [ "Storage Infrastructure" ], "wikilinks": [] }, { "id": "store-of-value", "title": "Store of Value", "domain": "finance", "domain_name": "Finance", "definition": "A store of value is any asset that retains purchasing power over time, enabling the owner to defer consumption and retrieve real wealth in the future with minimal loss to inflation, confiscation, or degradation. Classical monetary theory identifies store-of-value function alongside medium-of-exchange and unit-of-account as the three primary functions of money, though assets can serve one function without the others.", "entityType": "Class", "qualityScore": 0.85, "maturity": "mature", "iri": "urn:ngm:class:store-of-value", "labels": [ "Store of Value" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "store-and-forward-communication", "title": "Store-and-forward Communication", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:store-and-forward-communication", "labels": [ "Store-and-forward Communication" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "storj", "title": "Storj", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Storj is a decentralised cloud storage network in which files are encrypted, split into pieces, and distributed across independent node operators who are paid for storage and bandwidth.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:storj", "labels": [ "Storj" ], "is_subclass_of": [ "Decentralised Storage" ], "wikilinks": [ "Cryptography", "Decentralised Storage" ] }, { "id": "storytelling-structure", "title": "Storytelling Structure", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Storytelling Structure is a formal organisation of narrative elements \u2014 acts, arcs, beats, and character relationships \u2014 that governs the temporal and causal progression of events in interactive and immersive experiences. In spatial computing contexts it shapes how virtual environments present branching narrative paths, cinematic sequences, and player-driven story arcs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:storytelling-structure", "labels": [ "Storytelling Structure" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "storytelling", "title": "Storytelling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The craft and practice of structuring narrative content to communicate meaning, emotion, and experience to an audience. In spatial computing and metaverse contexts, storytelling encompasses interactive narrative design, world-building, and immersive experience creation that leverage real-time 3D environments to place participants inside the story.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:storytelling", "labels": [ "Storytelling", "Transmedia Storytelling" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "stranded-energy-monetisation", "title": "Stranded Energy Monetisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stranded energy monetisation is the practice of converting otherwise wasted or non-transmittable electricity\u2014including curtailed renewable generation, flared natural gas, and grid-overflow surplus\u2014into economic value through flexible, interruptible load processes such as Bitcoin mining or electrolyser hydrogen production. By co-locating computation or industrial loads at the energy source, it captures value that cannot reach load centres due to grid constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:stranded-energy-monetisation", "labels": [ "Stranded Energy Monetisation" ], "is_subclass_of": [ "Stranded Energy" ], "wikilinks": [] }, { "id": "stranded-energy", "title": "Stranded Energy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stranded energy is generated power that cannot be economically transmitted to demand centres because it is produced in remote locations, lacks grid connection, or is curtailed when supply exceeds local demand. Examples include flared associated gas at oil wells, over-built hydro and wind capacity, and geothermal sites far from population. Because it would otherwise be wasted, stranded energy can be monetised by co-locating energy-intensive computation such as Bitcoin mining at the source.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stranded-energy", "labels": [ "Stranded Energy" ], "is_subclass_of": [ "Sustainability" ], "wikilinks": [] }, { "id": "strategic-bitcoin-reserve", "title": "Strategic Bitcoin Reserve", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A strategic Bitcoin reserve is a holding of bitcoin maintained by a state, corporation, or institution as a long-term store of value and hedge against fiat debasement, analogous to gold or foreign-currency reserves. Proponents argue that bitcoin's fixed supply and censorship resistance make it a credible reserve asset, while critics cite volatility and custody risk. The concept gained prominence as governments and large treasuries began formal accumulation programmes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:strategic-bitcoin-reserve", "labels": [ "Strategic Bitcoin Reserve", "Bitcoin Treasury Reserve", "Sovereign Bitcoin Reserve" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "strategic-planning", "title": "Strategic Planning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Strategic Planning is an organisational process for defining long-term direction, priorities, and resource allocation to achieve goals within a given domain. In spatial computing and metaverse contexts, it encompasses technology roadmaps, governance design, and stakeholder alignment for platform development.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:strategic-planning", "labels": [ "Strategic Planning" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Blockchain", "owl:Thing" ] }, { "id": "stratopause", "title": "Stratopause", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stratopause", "labels": [ "Stratopause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "stratosphere", "title": "Stratosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:stratosphere", "labels": [ "Stratosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "stratum-protocol", "title": "Stratum Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Stratum Protocol is a lightweight, JSON-based line protocol used to coordinate work between cryptocurrency mining pools and individual miners. The pool server distributes block templates and difficulty targets to connected mining clients, which return valid share submissions, allowing aggregated hashing power to be measured and rewarded. Stratum reduces bandwidth and latency compared with earlier polling schemes and remains the dominant pool communication standard for proof-of-work coins.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stratum-protocol", "labels": [ "Stratum Protocol" ], "is_subclass_of": [ "Mining Pool" ], "wikilinks": [] }, { "id": "stratum-v2", "title": "Stratum V2", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Stratum V2 is the second-generation protocol for communication between Bitcoin miners and mining pools, redesigned to improve efficiency, security, and decentralisation over the original Stratum. Its headline feature is Job Declaration, which lets individual miners construct their own block templates and choose which transactions to include rather than blindly mining the pool operator's template, redistributing transaction-selection power. It also adds end-to-end encryption, binary framing for lower bandwidth, and reduced susceptibility to man-in-the-middle hashrate hijacking.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:stratum-v2", "labels": [ "Stratum V2", "Stratum V2 Specification" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "stream-cipher", "title": "Stream Cipher", "domain": "security", "domain_name": "Security", "definition": "A stream cipher is a symmetric encryption technique that combines plaintext with a pseudorandom keystream, one bit or byte at a time, to produce ciphertext, in contrast to block ciphers which operate on fixed-size chunks. The security of a stream cipher depends critically on high-quality random number generation for keystream and nonce material, since keystream reuse catastrophically breaks confidentiality. Stream ciphers are favoured where low latency or resource-constrained encryption of continuous data streams is required.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:stream-cipher", "labels": [ "Stream Cipher" ], "is_subclass_of": [ "Symmetric Encryption" ], "wikilinks": [] }, { "id": "stream-processing", "title": "Stream Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Stream processing is a data processing paradigm in which computations are performed continuously on unbounded sequences of records as they arrive, rather than on static stored datasets, enabling low-latency analytics, transformations, and reactions to events within milliseconds to seconds of their occurrence. It is characterised by windowing operations, stateful operators, time-based semantics (event time versus processing time), and exactly-once or at-least-once delivery guarantees.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stream-processing", "labels": [ "Stream Processing", "Data Stream Processing", "Event Stream Processing" ], "is_subclass_of": [ "Real-time Processing" ], "wikilinks": [] }, { "id": "streaming-payment", "title": "Streaming Payment", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Streaming Payment is a continuous, real-time value transfer mechanism enabled by programmable money protocols \u2014 particularly Layer 2 payment channel networks such as the Bitcoin Lightning Network \u2014 whereby funds flow incrementally over time in proportion to ongoing service consumption rather than in discrete lump-sum transactions. Streaming payments enable pay-per-second, pay-per-byte, or pay-per-computation billing models that align payment precisely with value delivered, eliminating invoicing cycles and reducing counterparty risk. They are foundational to micropayment-based business models for media, APIs, AI inference, bandwidth, and real-time data feeds.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:streaming-payment", "labels": [ "Streaming Payment" ], "is_subclass_of": [ "Blockchain", "DeFi and Economics" ], "wikilinks": [] }, { "id": "streaming-payments", "title": "Streaming Payments", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Streaming payments are a payment model in which value flows continuously from payer to payee in real time, accruing per second or per block rather than in discrete lump sums. On programmable ledgers and payment channels this is implemented by smart contracts or channel updates that let a recipient withdraw the proportion earned at any moment. The model suits salaries, subscriptions, rentals, and machine-to-machine billing where settlement should track elapsed time or usage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:streaming-payments", "labels": [ "Streaming Payments", "Streaming Payment" ], "is_subclass_of": [ "Programmable Money" ], "wikilinks": [] }, { "id": "stress-testing", "title": "Stress Testing", "domain": "finance", "domain_name": "Finance", "definition": "Stress testing is a risk-management technique that evaluates how an institution, portfolio or system would perform under severe but plausible adverse conditions. It applies hypothetical or historical shock scenarios to estimate losses, capital depletion and liquidity strain beyond normal expectations. Supervisory stress tests are a central tool of prudential regulation for assessing resilience and informing capital requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:stress-testing", "labels": [ "Stress Testing" ], "is_subclass_of": [ "Risk Management" ], "wikilinks": [] }, { "id": "strict-encoding", "title": "Strict Encoding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Strict encoding is a deterministic binary serialisation scheme used in the RGB protocol that guarantees a single canonical byte representation for any given data structure. By forbidding ambiguous orderings and optional layout choices, it ensures that independently computed commitments and hashes match exactly, which is essential for client-side validation. It pairs with strict types to give RGB its reproducible, consensus-free verification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:strict-encoding", "labels": [ "Strict Encoding" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "strict-types", "title": "Strict Types", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Strict types is a type system and schema language used by the RGB protocol to define data structures with a fully deterministic memory layout and a content-addressable type identity. Each type has a unique hash derived from its definition, so schemas can be referenced and verified without ambiguity. It provides the typed foundation that strict encoding serialises during client-side validation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:strict-types", "labels": [ "Strict Types", "Strict Types Specification" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "strike", "title": "Strike", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Strike is a financial technology company founded by Jack Mallers that uses the Bitcoin Lightning Network as a settlement rail to deliver instant, low-cost global payments and remittances denominated in fiat currencies. It abstracts the technical complexity of Bitcoin and Lightning from end users, converting value at the edges so that senders and recipients may transact in their preferred local currencies without needing to hold cryptocurrency. Strike competes with legacy remittance operators and correspondent banking channels by exploiting Lightning's near-zero fee structure and sub-second finality to offer dramatically lower-cost international money movement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:strike", "labels": [ "Strike" ], "is_subclass_of": [ "Cross-Border Payments" ], "wikilinks": [ "Lightning Network", "Bitcoin", "Cross-Border Payments" ] }, { "id": "stripe-atlas", "title": "Stripe Atlas", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A service provided by Stripe that enables entrepreneurs to incorporate and launch their business in the United States, handling legal, banking, and compliance requirements.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:stripe-atlas", "labels": [ "Stripe Atlas" ], "is_subclass_of": [ "Stripe" ], "wikilinks": [] }, { "id": "stripe", "title": "Stripe", "domain": "finance", "domain_name": "Finance", "definition": "Stripe is a financial technology company founded in 2010 that provides programmable payment infrastructure, APIs, and financial services software enabling businesses to accept online and in-person payments, manage subscriptions, issue cards, and access embedded financial tooling. It operates as a payment service provider and merchant acquirer, abstracting the complexity of card networks, banking rails, and compliance into developer-friendly interfaces. Stripe's platform spans payment processing, revenue management, fraud detection, tax computation, and banking-as-a-service, making it a foundational layer of the internet economy. Its products serve businesses ranging from independent developers to large enterprises across more than 135 countries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:stripe", "labels": [ "Stripe" ], "is_subclass_of": [ "Payment Network" ], "wikilinks": [ "Payment Network", "Financial Services", "Stablecoin" ] }, { "id": "strong-consistency", "title": "Strong Consistency", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Strong consistency is a property of a distributed data system in which every read returns the result of the most recently completed write, so that all clients observe a single, up-to-date view of the data. It typically corresponds to linearizability, where operations appear to take effect instantaneously at some point between their invocation and response. Achieving strong consistency requires coordination such as consensus and, under the CAP theorem, trades availability for consistency during network partitions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:strong-consistency", "labels": [ "Strong Consistency" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "structural-member", "title": "Structural Member", "domain": "robotics", "domain_name": "Robotics", "definition": "A Structural Member is a discrete load-bearing component of a robotic system's physical frame, such as a link, beam, bracket, or chassis element, that transmits forces and torques between joints or actuators. Structural members define the kinematic chain of a robot and their geometric and material properties directly constrain workspace, payload, and dynamic performance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:structural-member", "labels": [ "Structural Member", "StructuralMember" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "structure-from-motion", "title": "structure-from-motion", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Structure-from-Motion (SfM) is a photogrammetric pipeline that simultaneously recovers three-dimensional scene geometry and camera motion parameters from an unordered collection of overlapping two-dimensional images. Sparse feature correspondences\u2014detected via descriptors such as SIFT, ORB, or learned alternatives\u2014seed an incremental or global pose-estimation stage that triangulates a sparse point cloud. Bundle adjustment then jointly refines all camera extrinsics, intrinsics, and 3D point coordinates by minimising reprojection error across the full image set. SfM is a foundational technique in computer vision, photogrammetry, autonomous navigation, and spatial computing, serving as the upstream stage for multi-view stereo densification and neural scene representations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:structure-from-motion", "labels": [ "Structure-from-Motion", "Structure From Motion", "Structure from Motion" ], "is_subclass_of": [ "Photogrammetry" ], "wikilinks": [] }, { "id": "structured-data", "title": "Structured Data", "domain": "data", "domain_name": "Data", "definition": "Structured data is information organised according to a predefined schema or model so that its meaning and relationships are explicit and machine-readable. By conforming to fixed fields, types, and constraints, it can be reliably queried, validated, exchanged, and reasoned over, in contrast to unstructured text or media. On the web it commonly takes the form of annotations such as Schema.org markup that let machines interpret page content.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:structured-data", "labels": [ "Structured Data" ], "is_subclass_of": [ "Data Model" ], "wikilinks": [] }, { "id": "structured-light", "title": "Structured Light", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Structured light is a 3D surface measurement technique that projects one or more known patterns \u2014 typically binary stripe sequences, sinusoidal fringes, or dot grids \u2014 onto a scene using a projector or laser, then captures the deformed pattern with one or more calibrated cameras. Because the geometry of the projection and capture system is precisely known, the per-pixel deformation of the projected pattern encodes depth, allowing a complete 3D point cloud or depth map of the object surface to be recovered through triangulation. Structured light systems achieve sub-millimetre to micrometre-scale depth accuracy and are widely deployed in industrial metrology, 3D scanning, face recognition, robotic guidance, and augmented reality depth sensing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:structured-light", "labels": [ "Structured Light", "Structured Light Scanning" ], "is_subclass_of": [ "Depth Sensing" ], "wikilinks": [] }, { "id": "structured-output", "title": "Structured Output", "domain": "ai", "domain_name": "Ai", "definition": "Structured output is a technique for constraining a language model to emit responses that conform to a predefined schema such as JSON, a regular grammar, or a typed object. It is enforced through prompt instructions, constrained decoding, or function-calling interfaces so that downstream systems can parse results reliably. Structured output bridges free-form generation and deterministic software by guaranteeing machine-readable, validatable responses.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:structured-output", "labels": [ "Structured Output", "Structured Output Generation", "Structured Outputs Strict Mode" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "structurizr-dsl", "title": "Structurizr DSL", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Structurizr DSL is a text-based domain-specific language for defining software architecture models conforming to Simon Brown's C4 Model, enabling teams to describe system context, containers, components, and code-level elements as code that can be version-controlled and rendered into multiple diagram formats. It is the primary input format for the Structurizr toolchain, which produces interactive, filterable architecture diagrams from a single workspace definition.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:structurizr-dsl", "labels": [ "Structurizr DSL", "Structurizr DSL Specification" ], "is_subclass_of": [ "Diagrams as Code" ], "wikilinks": [] }, { "id": "style-transfer", "title": "Style Transfer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Style Transfer is the technique of applying the artistic style of one image (style image) to the content of another image (content image), creating a new image that combines content from one source with the aesthetic style of another. Neural style transfer employs convolutional neural networks to separate and recombine content and style representations via Gram matrix optimisation, enabling artistic rendering, photo enhancement, and creative visual effects.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:style-transfer", "labels": [ "Style Transfer", "Garment Transfer", "IPAdapter Style Transfer", "Neural Style Transfer" ], "is_subclass_of": [ "Image to Image Translation" ], "wikilinks": [ "presentation", "rework based on gpt", "Artificial Intelligence", "Convolutional Neural Network", "Cyber Security and Military", "Digital Society Surveillance", "Education and AI", "Image Generation", "Image to Image Translation", "MetaverseDomain", "Microsoft Copilot", "Politics, Law, Privacy", "style transfer" ] }, { "id": "sub-millisecond-latency", "title": "Sub-Millisecond Latency", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Sub-Millisecond Latency is a network performance characteristic in which end-to-end transaction confirmation or consensus completion occurs in under one millisecond, enabling near-real-time settlement on blockchain networks. It depends on highly optimised peer-to-peer propagation, deterministic finality mechanisms, and minimal block time, distinguishing high-performance chains from conventional systems with multi-second confirmation times.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:sub-millisecond-latency", "labels": [ "Sub-Millisecond Latency" ], "is_subclass_of": [ "Network Component", "Transaction Confirmation" ], "wikilinks": [ "Blockchain" ] }, { "id": "subagent", "title": "Subagent", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A subagent is a subordinate autonomous agent spawned by a parent or orchestrating agent to carry out a delegated subtask within its own isolated context window, returning only a distilled result to the parent. By running in a fresh context, a subagent keeps the noise of its intermediate exploration \u2014 long file reads, search output, failed attempts \u2014 out of the parent's limited context, while allowing many subtasks to proceed in parallel. Subagents may be specialised by role, tools, or model tier, and are the unit of work distribution in hierarchical multi-agent architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:subagent", "labels": [ "Subagent" ], "is_subclass_of": [ "AI Agent", "AIAgent" ], "wikilinks": [ "AIAgent", "MultiAgentSystem", "TaskDelegation", "AgentLoop" ] }, { "id": "suborbital-spaceflight-vehicle", "title": "Suborbital Spaceflight Vehicle", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:suborbital-spaceflight-vehicle", "labels": [ "Suborbital Spaceflight Vehicle" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "suborbital-spaceflight", "title": "Suborbital Spaceflight", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:suborbital-spaceflight", "labels": [ "Suborbital Spaceflight" ], "is_subclass_of": [ "Spaceflight" ], "wikilinks": [] }, { "id": "subscription-model", "title": "Subscription Model", "domain": "finance", "domain_name": "Finance", "definition": "A revenue model in which customers pay a recurring fee \u2014 monthly or annually \u2014 for continuing access to a product or service rather than purchasing it outright, converting one-off sales into predictable recurring revenue; the economic backbone of Software as a Service, media streaming, and the membership tiers of the creator economy, it shifts the commercial burden from closing sales to retaining customers, measured through metrics such as MRR, churn, and customer lifetime value.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:subscription-model", "labels": [ "Subscription Model" ], "is_subclass_of": [ "Economic Model" ], "wikilinks": [ "Economic Model", "Software As A Service", "Streaming Payment", "Creator Economy" ] }, { "id": "subsetted-data-state", "title": "Subsetted Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:subsetted-data-state", "labels": [ "Subsetted Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "subsumption-architecture", "title": "Subsumption Architecture", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Subsumption architecture is a layered control method for autonomous agents in which simple reactive behaviours are stacked, with higher layers able to suppress or override lower ones to produce competent behaviour without a central world model.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:subsumption-architecture", "labels": [ "Subsumption Architecture" ], "is_subclass_of": [ "Robotics", "Actuation and Control", "Robotics Domain" ], "wikilinks": [ "Autonomous Agent", "Swarm Intelligence", "Multi-Agent System", "Robotics", "Robotics Domain" ] }, { "id": "subsurface-ocean", "title": "Subsurface Ocean", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:subsurface-ocean", "labels": [ "Subsurface Ocean" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "subsurface-scattering", "title": "Subsurface Scattering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Subsurface scattering is a light-transport phenomenon in which light penetrates the surface of a translucent material, scatters within it, and exits at a different point, producing the soft, diffuse glow characteristic of skin, wax, marble, and milk. In computer graphics it is modelled as part of global illumination to render such materials realistically, since a purely surface-level reflectance model cannot capture the way light bleeds beneath and through them. Approaches range from physically based diffusion and path-traced volumetric scattering to fast screen-space and pre-integrated approximations used in real-time rendering. It is essential to photorealistic depiction of organic and semi-transparent matter.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:subsurface-scattering", "labels": [ "Subsurface Scattering" ], "is_subclass_of": [ "Global Illumination" ], "wikilinks": [] }, { "id": "subword-tokenisation", "title": "Subword Tokenisation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A tokenisation strategy that segments text into sub-word units (via algorithms such as Byte-Pair Encoding, WordPiece, or SentencePiece), balancing vocabulary size against the ability to represent rare, morphologically complex, and out-of-vocabulary words. Subword tokenisation is the de facto standard preprocessing step for large language models.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:subword-tokenisation", "labels": [ "Subword Tokenisation", "Subword Segmentation", "WordPiece Tokenisation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "suggested-reading-order", "title": "Suggested Reading Order", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Suggested Reading Order is a curated navigation guide that sequences the principal pages of the NarrativeGoldmine knowledge graph into a coherent pedagogical path. It maps the intended conceptual dependencies \u2014 from foundational web technologies through decentralisation, digital assets, spatial computing, and AI \u2014 enabling readers to build understanding progressively across the graph's interconnected topic areas.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:suggested-reading-order", "labels": [ "Suggested Reading Order" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "NFTs", "RGB", "Adoption of Convergent Technologies", "Bitcoin", "Bitcoin As Money", "Blockchain", "BTC Layer 3", "Convergence", "Decentralised Web", "Definitions and frameworks for Metaverse", "Digital Asset Risks", "Digital Objects", "Disruption", "Distributed Identity", "Dreamlab", "Ethereum", "Introduction to me", "Lightning and Similar L2", "Metaverse and Spatial Risks", "Metaverse and Telecollaboration" ] }, { "id": "sun-sensor", "title": "Sun Sensor", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sun-sensor", "labels": [ "Sun Sensor" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "sun-sensor-geometry", "title": "Sun-sensor Geometry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sun-sensor-geometry", "labels": [ "Sun-sensor Geometry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sun-synchronous-orbit", "title": "Sun-synchronous Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sun-synchronous-orbit", "labels": [ "Sun-synchronous Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "sunspot", "title": "Sunspot", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:sunspot", "labels": [ "Sunspot" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "super-resolution", "title": "Super Resolution", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Super-Resolution is the process of enhancing the resolution and quality of low-resolution images by predicting and synthesising high-frequency details using deep learning models.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:super-resolution", "labels": [ "Super Resolution", "Image Super-Resolution", "Super-Resolution Model", "Super-resolution" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "computer vision", "Computer Vision", "Convolutional Neural Network", "Image Generation", "MetaverseDomain" ] }, { "id": "superchain", "title": "Superchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A superchain is a network of independent Layer-2 blockchains that share a common technical standard, security model, and communication layer so they interoperate as a unified system rather than isolated silos. Pioneered by the OP Stack ecosystem around Optimism, superchains let many rollups inherit the same upgrade path, governance, and eventually a shared sequencer, enabling low-latency cross-chain messaging and a consistent developer experience. The model addresses blockchain fragmentation by treating horizontal scaling as a coordinated mesh of homogeneous chains that settle to a common Layer 1.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:superchain", "labels": [ "Superchain" ], "is_subclass_of": [ "Rollup" ], "wikilinks": [] }, { "id": "superfluid", "title": "Superfluid", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A phase of matter that flows without viscosity, exhibiting effects such as frictionless flow and the ability to climb container walls. It arises in certain quantum systems at very low temperatures.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:superfluid", "labels": [ "Superfluid" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "Quantum Computing", "Entropy", "owl:Thing" ] }, { "id": "supergiant-star", "title": "Supergiant Star", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:supergiant-star", "labels": [ "Supergiant Star" ], "is_subclass_of": [ "Star" ], "wikilinks": [] }, { "id": "superimpose", "title": "Superimpose", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:superimpose", "labels": [ "Superimpose" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "superintelligence", "title": "Superintelligence", "domain": "ai", "domain_name": "Ai", "definition": "Superintelligence refers to a hypothetical form of artificial intelligence that surpasses the cognitive performance of any human across virtually all domains of interest, including scientific reasoning, social manipulation, and strategic planning. Nick Bostrom distinguishes speed superintelligence (same algorithms, faster hardware), collective superintelligence (many coordinated AI agents), and quality superintelligence (genuinely superior algorithms), each presenting distinct safety and control challenges.", "entityType": "Class", "qualityScore": 0.85, "maturity": "emerging", "iri": "urn:ngm:class:superintelligence", "labels": [ "Superintelligence", "Artificial Superintelligence" ], "is_subclass_of": [ "Artificial Superintelligence Theory" ], "wikilinks": [] }, { "id": "supernova-remnant", "title": "Supernova Remnant", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:supernova-remnant", "labels": [ "Supernova Remnant" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "supernova", "title": "Supernova", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:supernova", "labels": [ "Supernova" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "supervised-fine-tuning", "title": "Supervised Fine Tuning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A fine-tuning approach that uses labelled training data to adapt a pre-trained model to specific tasks, optimising performance through supervised learning on input-output pairs. Supervised fine-tuning (SFT) represents the most direct path from general pre-training to task-specific capability, and serves as the foundational first stage in multi-stage alignment pipelines such as InstructGPT and Constitutional AI.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:supervised-fine-tuning", "labels": [ "Supervised Fine Tuning", "Supervised Fine-Tuning" ], "is_subclass_of": [ "Fine Tuning", "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "supervised-learning", "title": "Supervised Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Supervised Learning is the machine learning paradigm where models learn from labeled training data to predict outputs for new, unseen inputs. The learning algorithm finds patterns mapping input features to target labels, guided by a loss function measuring prediction errors. Key tasks include classification (discrete labels) and regression (continuous values), spanning linear models, decision trees, support vector machines, neural networks, and ensemble methods.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:supervised-learning", "labels": [ "Supervised Learning", "SupervisedLearning" ], "is_subclass_of": [ "Machine Learning", "AI Technique" ], "wikilinks": [ "Classification", "Regression", "Deep Learning", "Training Data" ] }, { "id": "supervisor-worker-pattern", "title": "Supervisor-Worker Pattern", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The supervisor-worker pattern is a multi-agent coordination architecture in which a single supervisor agent decomposes a goal, delegates the resulting subtasks to a set of worker agents, and integrates their returned results into a coherent whole, without the workers communicating directly with one another. The supervisor owns planning, routing, verification, and error handling; the workers own execution of their assigned subtask. Centralising control this way makes the system's behaviour easy to reason about and to recover, at the cost of the supervisor becoming a throughput bottleneck and single point of failure.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:supervisor-worker-pattern", "labels": [ "Supervisor-Worker Pattern" ], "is_subclass_of": [ "AI Agent Coordination", "AIAgentCoordination" ], "wikilinks": [ "AIAgentCoordination", "TaskDelegation", "Orchestration", "MultiAgentSystem", "Subagent" ] }, { "id": "supervisory-authority", "title": "Supervisory Authority", "domain": "governance", "domain_name": "Governance", "definition": "A supervisory authority is an independent public body empowered by law to monitor, enforce, and provide guidance on compliance with a regulatory regime within a defined jurisdiction. In data-protection law, for example, supervisory authorities oversee how organisations process personal data, investigate complaints, conduct audits, and impose corrective measures or fines. More broadly the term covers sectoral regulators that license, examine, and sanction the entities they govern, deriving their legitimacy from statute and exercising powers of investigation, enforcement, and rule-setting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supervisory-authority", "labels": [ "Supervisory Authority" ], "is_subclass_of": [ "Regulation" ], "wikilinks": [] }, { "id": "supply-cap", "title": "Supply Cap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Supply Cap is a hard-coded or governance-determined upper bound on the total number of tokens or coins that will ever be issued by a blockchain protocol, creating a form of programmatic scarcity that underpins deflationary monetary policy. Bitcoin's 21 million coin cap is the canonical example, encoded in the protocol's halvening schedule and providing a predictable issuance curve. Supply caps differentiate blockchain-native assets from fiat currencies and influence long-run security models, since block rewards approach zero as the cap is neared and transaction fees must compensate.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:supply-cap", "labels": [ "Supply Cap", "21 Million Supply Cap" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "supply-chain-automation", "title": "Supply Chain Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "Supply chain automation is the application of robotics, software and data-driven control to execute and coordinate logistics, warehousing, procurement and fulfilment tasks with minimal manual intervention. It spans physical automation such as robotic picking and conveyance and digital automation such as automated ordering, demand forecasting and exception handling. The goal is higher throughput, lower error rates and end-to-end visibility across the flow of goods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supply-chain-automation", "labels": [ "Supply Chain Automation" ], "is_subclass_of": [ "Automation" ], "wikilinks": [] }, { "id": "supply-chain-blockchain", "title": "Supply Chain Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Supply chain blockchain refers to the application of distributed ledger technology to track, trace, and verify the provenance of goods as they move through supply chain networks from raw materials to end consumers, providing an immutable shared record of transactions that enables real-time visibility, authentication of origin, quality verification, and automated compliance through smart contracts.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:supply-chain-blockchain", "labels": [ "Supply Chain Blockchain" ], "is_subclass_of": [ "Blockchain Application" ], "wikilinks": [ "Product Traceability", "Blockchain" ] }, { "id": "supply-chain-data", "title": "Supply Chain Data", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Supply chain data is the structured information captured as goods, materials, and assets move through production and distribution networks, including provenance, location, condition, custody, and transaction records. When anchored on a blockchain or distributed ledger, this data gains tamper-evidence and shared, verifiable lineage across organisational boundaries, enabling end-to-end traceability. Reliable supply chain data underpins recall management, compliance, anti-counterfeiting, and sustainability reporting across multi-party supply chains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:supply-chain-data", "labels": [ "Supply Chain Data" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [] }, { "id": "supply-chain-decarbonisation", "title": "Supply Chain Decarbonisation", "domain": "governance", "domain_name": "Governance", "definition": "Supply chain decarbonisation is the systematic reduction of greenhouse gas emissions arising across an organisation's upstream and downstream value chain, including the indirect Scope 3 emissions that typically dominate a company's total footprint. It combines supplier engagement, low-carbon procurement, logistics optimisation, materials substitution and product redesign with rigorous emissions accounting. Because most value-chain emissions lie outside a firm's direct operational control, decarbonisation depends on data sharing, contractual incentives and collaboration across many tiers of suppliers. It is a core operational lever within broader climate governance and net-zero strategy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:supply-chain-decarbonisation", "labels": [ "Supply Chain Decarbonisation" ], "is_subclass_of": [ "Decarbonisation Strategy" ], "wikilinks": [] }, { "id": "supply-chain-finance", "title": "Supply Chain Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Supply chain finance is a set of financing and risk-mitigation techniques that optimise working capital and liquidity across the parties in a supply chain. It typically allows suppliers to receive early payment on approved invoices while buyers extend their payment terms, using the buyer's stronger credit standing to lower financing costs. Blockchain and distributed-ledger implementations add shared visibility, automated settlement and tamper-evident provenance to these flows.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supply-chain-finance", "labels": [ "Supply Chain Finance" ], "is_subclass_of": [ "Trade Finance" ], "wikilinks": [] }, { "id": "supply-chain-management", "title": "Supply Chain Management", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply Chain Management (SCM) is the end-to-end coordination of the flow of goods, services, information, and finances from raw-material suppliers through manufacturers, distributors, and retailers to end consumers, with the goal of maximising value and minimising waste. It encompasses strategic planning, demand forecasting, procurement, inventory control, logistics, and supplier relationship management as an integrated discipline. Modern SCM increasingly employs digital technologies \u2014 including IoT sensors, AI-driven demand planning, blockchain-based provenance tracking, and digital twins \u2014 to achieve visibility, resilience, and sustainability across multi-tier networks. The field draws on operations research, systems theory, and data science to optimise trade-offs between cost, service levels, lead times, and risk.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:supply-chain-management", "labels": [ "Supply Chain Management", "BC-0044-supply-chain-management" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [ "Blockchain Technology", "Blockchain Entity" ] }, { "id": "supply-chain-optimisation", "title": "Supply Chain Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Supply chain optimisation is the application of analytical and machine-learning techniques to plan and operate the flow of goods, information, and funds across sourcing, production, inventory, and distribution so as to minimise cost and risk while meeting service targets. It combines demand forecasting, inventory and network design, and routing decisions, often formulated as mathematical programmes or learned policies. AI-driven approaches increasingly use predictive analytics and digital twins to make these decisions adaptive to real-time conditions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supply-chain-optimisation", "labels": [ "Supply Chain Optimisation" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "supply-chain-provenance", "title": "Supply Chain Provenance", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply chain provenance is the verified documentary record of an item's origin, material composition, transformation steps, custody changes, and transportation history from raw material extraction through to the end consumer, enabling authenticity verification, ethical sourcing attestation, and regulatory compliance. It extends data provenance principles to physical goods, typically combining IoT sensor data, third-party audit records, and cryptographic anchors (hashes or blockchain transactions) to create an immutable chain of evidence that resists falsification and supports granular attribution.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:supply-chain-provenance", "labels": [ "Supply Chain Provenance" ], "is_subclass_of": [ "Provenance Tracking" ], "wikilinks": [] }, { "id": "supply-chain-resilience", "title": "Supply Chain Resilience", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply chain resilience is the capacity of a supply network to anticipate, absorb, adapt to, and recover from disruptions while maintaining the continuity of supply at acceptable cost and service levels. It combines visibility into the network, redundancy and flexibility in sourcing and capacity, and the ability to reconfigure flows when shocks occur. Resilience strategies balance efficiency against the buffers and diversification needed to withstand events such as supplier failure, natural disaster, or demand volatility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supply-chain-resilience", "labels": [ "Supply Chain Resilience" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [] }, { "id": "supply-chain-risk-management", "title": "Supply Chain Risk Management", "domain": "governance", "domain_name": "Governance", "definition": "Supply chain risk management is the systematic identification, assessment, mitigation, and monitoring of risks that arise from an organisation's network of suppliers, vendors, and logistics dependencies. It addresses disruption, quality, financial, geopolitical, cyber, and integrity risks across multiple tiers of suppliers, including the software supply chain. Practitioners apply frameworks such as the NIST risk management approach, conduct vendor due diligence, and use artefacts like the software bill of materials to gain visibility. The goal is resilient, continuous operation in the face of upstream uncertainty and threats.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:supply-chain-risk-management", "labels": [ "Supply Chain Risk Management" ], "is_subclass_of": [ "Risk Management Framework" ], "wikilinks": [] }, { "id": "supply-chain-security", "title": "Supply Chain Security", "domain": "security", "domain_name": "Security", "definition": "Supply chain security is the practice of protecting the integrity, provenance, and trustworthiness of the components, dependencies, and processes that compose a product, with particular emphasis on software supply chains. It addresses threats such as compromised dependencies, malicious build tooling, and tampered artefacts through measures like signed releases, software bills of materials, and reproducible builds. It has become a critical discipline as systems increasingly assemble third-party code and hardware.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:supply-chain-security", "labels": [ "Supply Chain Security", "Software Supply Chain Security" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "supply-chain-traceability", "title": "Supply Chain Traceability", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply chain traceability employs blockchain and smart contracts to create immutable, transparent records of a product's journey from origin through manufacturing, distribution, and final delivery, enabling verification of authenticity and regulatory compliance at each stage. Implementations track provenance, environmental conditions, and certifications, providing cryptographic proof of authenticity and preventing counterfeiting across food safety, pharmaceuticals, luxury goods, and circular-economy initiatives.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:supply-chain-traceability", "labels": [ "Supply Chain Traceability", "BC-0441-supply-chain-traceability", "BC-0446-supply-chain-traceability", "Content Supply Chain Integrity", "Supply Chain Attestation" ], "is_subclass_of": [ "Blockchain Application" ], "wikilinks": [ "BC-0142-smart-contract", "BC-0245-internet-of-things", "BC-0426-hyperledger-fabric", "BC-0447-anti-counterfeiting", "BC-0448-cold-chain-monitoring", "BC-0449-circular-economy", "BlockchainTechnology", "Counterfeiting", "BlockchainDomain", "CircularEconomy", "ColdChainMonitoring", "HyperledgerFabric", "SmartContract" ] }, { "id": "supply-chain-tracking", "title": "Supply Chain Tracking", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Supply chain tracking is the end-to-end recording of a product's movement, custody, and condition as it passes through producers, logistics, and retailers. Blockchain-based approaches anchor these events to an immutable, shared ledger so that participants can verify provenance and detect tampering without trusting a single intermediary. It enables traceability for food safety, anti-counterfeiting, and regulatory compliance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:supply-chain-tracking", "labels": [ "Supply Chain Tracking" ], "is_subclass_of": [ "Blockchain", "Supply Chain Traceability" ], "wikilinks": [] }, { "id": "supply-chain-transparency", "title": "Supply Chain Transparency", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply chain transparency is the degree to which information about the origins, provenance, custody chain, working conditions, and environmental and social impacts of goods and services is visible, accessible, and verifiable to stakeholders across the entire value network \u2014 including manufacturers, tier-n suppliers, logistics operators, retailers, regulators, and end consumers. It encompasses voluntary disclosure practices, mandatory reporting frameworks (such as the UK Modern Slavery Act and the EU Corporate Sustainability Due Diligence Directive), and technical systems \u2014 including blockchain-based provenance tracking, IoT sensor data chains, and Digital Product Passports \u2014 that create auditable, tamper-resistant records of supply chain events. Transparency is increasingly mandated through due diligence regulations and ESG market expectations, requiring organisations to verify and disclose conditions several tiers deep into their supplier networks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:supply-chain-transparency", "labels": [ "Supply Chain Transparency", "Value Chain Transparency" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "supply-chain-visibility", "title": "Supply Chain Visibility", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Supply chain visibility is the capability to track, monitor, and share accurate, real-time information about the location, status, and condition of goods, components, and materials as they move through a supply network from raw material sourcing through production, distribution, and delivery to end customers. It requires integration of data from disparate logistics systems, IoT sensors, carrier APIs, and ERP platforms, and provides the situational awareness needed for proactive disruption management, regulatory compliance, and demand-driven inventory optimisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:supply-chain-visibility", "labels": [ "Supply Chain Visibility" ], "is_subclass_of": [ "Supply Chain Management" ], "wikilinks": [] }, { "id": "supply-chain", "title": "supply chain", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A supply chain is the end-to-end network of organisations, people, activities, information flows, and resources involved in transforming raw materials into finished products and delivering them to end consumers. It spans procurement, manufacturing, quality assurance, logistics, customs compliance, and retail fulfilment across multiple tiers of suppliers and distributors. Modern supply chains are subject to geopolitical risk, demand variability, regulatory traceability mandates, and sustainability-reporting requirements under frameworks such as CSRD and the GHG Protocol. Digital technologies including blockchain-based provenance tracking, AI-driven demand forecasting, digital twins, and IoT-enabled visibility platforms are increasingly deployed to improve resilience, transparency, and efficiency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:supply-chain", "labels": [ "Supply Chain", "Resilient Supply Chains", "Retail Supply Chain", "Software Supply Chain Integrity", "Supply Chain Confidentiality", "Supply Chain Risk" ], "is_subclass_of": [ "Logistics Optimization" ], "wikilinks": [] }, { "id": "supply-and-demand", "title": "Supply and Demand", "domain": "finance", "domain_name": "Finance", "definition": "The foundational economic model describing how the price and traded quantity of a good emerge from the interaction of sellers' willingness to supply at each price and buyers' willingness to purchase, with markets tending towards the equilibrium where the two schedules intersect; shifts in either schedule \u2014 driven by costs, income, preferences, or expectations \u2014 move prices and quantities in predictable directions, making the model the basic engine of price formation in markets from commodities to currencies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:supply-and-demand", "labels": [ "Supply and Demand" ], "is_subclass_of": [ "Economics" ], "wikilinks": [ "Economics", "Microeconomics", "Price Discovery", "Exchange Rate" ] }, { "id": "support-vector-machine", "title": "Support Vector Machine", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A support vector machine (SVM) is a supervised learning model that finds the hyperplane separating classes with the maximum margin between the nearest training examples, called support vectors. Through the kernel trick it can construct non-linear decision boundaries by implicitly mapping inputs into higher-dimensional feature spaces. SVMs are grounded in statistical learning theory and are effective for classification and regression on small to medium, high-dimensional datasets.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:support-vector-machine", "labels": [ "Support Vector Machine" ], "is_subclass_of": [ "Supervised Learning" ], "wikilinks": [] }, { "id": "surface-marine-robot", "title": "Surface Marine Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Surface Marine Robot, also known as an Unmanned Surface Vehicle (USV), is an autonomous or remotely operated robotic platform that operates on the surface of bodies of water\u2014oceans, lakes, rivers, and coastal zones. It performs tasks including hydrographic survey, environmental monitoring, maritime patrol, search-and-rescue support, and data relay without requiring an onboard crew, thereby reducing operational cost and human risk in hazardous or remote marine environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:surface-marine-robot", "labels": [ "Surface Marine Robot" ], "is_subclass_of": [ "Marine Robot" ], "wikilinks": [ "Marine Robot", "Robotics" ] }, { "id": "surface-normal", "title": "Surface Normal", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A surface normal is a vector perpendicular to a surface at a given point, indicating the orientation of that surface in space. Normals are fundamental to lighting and shading calculations, because the angle between a normal and a light direction governs how much illumination a surface receives. In computer graphics and 3D reconstruction, surface normals drive realistic shading, normal mapping, and the recovery of fine geometric detail.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:surface-normal", "labels": [ "Surface Normal" ], "is_subclass_of": [ "Scene Geometry" ], "wikilinks": [] }, { "id": "surface-radiation-environment", "title": "Surface Radiation Environment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:surface-radiation-environment", "labels": [ "Surface Radiation Environment" ], "is_subclass_of": [ "Space Radiation Environment" ], "wikilinks": [] }, { "id": "surface-reflectance", "title": "Surface Reflectance", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:surface-reflectance", "labels": [ "Surface Reflectance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "surface-water", "title": "Surface Water", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:surface-water", "labels": [ "Surface Water" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "surgical-navigation", "title": "Surgical Navigation", "domain": "robotics", "domain_name": "Robotics", "definition": "Surgical Navigation is the intraoperative technology and methodology that provides surgeons with real-time spatial guidance by registering preoperative imaging data (CT, MRI) with the patient's anatomy and continuously tracking the position of surgical instruments relative to that registered model. It is analogous to GPS navigation but for the operating room, enabling submillimetre accuracy in procedures where anatomical landmarks are obscured, such as orthopaedic, neurosurgery, and spinal operations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:surgical-navigation", "labels": [ "Surgical Navigation", "Surgical Guidance" ], "is_subclass_of": [ "Navigation System" ], "wikilinks": [] }, { "id": "surgical-robot", "title": "Surgical Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Surgical Robot is a teleoperated or semi-autonomous robotic system designed to assist surgeons in performing minimally invasive procedures with enhanced precision, dexterity, and haptic feedback. Systems such as the da Vinci platform translate the surgeon's hand movements\u2014filtered for tremor\u2014into sub-millimetre instrument motions within the patient's body, enabling laparoscopic, urological, and cardiac procedures through small incisions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:surgical-robot", "labels": [ "Surgical Robot" ], "is_subclass_of": [ "Robot Type", "Service Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "surgical-robotics", "title": "surgical robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Surgical robotics is the application of robotic systems, real-time computer vision, and AI-driven control to assist surgeons or perform procedures with enhanced precision, dexterity, tremor cancellation, and minimally invasive access. Systems range from teleoperated master-slave platforms\u2014where a surgeon's hand movements are scaled, filtered, and transmitted to slave arms\u2014to semi-autonomous robots capable of tissue manipulation and suturing under human supervision. Key technical challenges include sub-millimetre motion tracking, intraoperative tissue deformation modelling, real-time haptic feedback, safe human-robot interaction, and regulatory compliance under medical device frameworks such as ISO 13485, IEC 62304, and EU MDR. The field sits at the intersection of mechatronics, control theory, computer vision, AI inference, and clinical medicine.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:surgical-robotics", "labels": [ "Surgical Robotics", "Surgical Robotics Vision", "SurgicalRobotics" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "surgical-simulation", "title": "Surgical Simulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Surgical simulation is the use of interactive virtual environments to model surgical procedures for training, planning and assessment. It combines real-time graphics, deformable tissue modelling, collision detection and haptic feedback to recreate the visual and tactile experience of operating. Surgical simulators allow clinicians to rehearse techniques and acquire skills without risk to patients.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:surgical-simulation", "labels": [ "Surgical Simulation" ], "is_subclass_of": [ "Medical Simulation" ], "wikilinks": [] }, { "id": "surrogate-model", "title": "Surrogate Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A surrogate model (or metamodel) is an inexpensive, data-driven approximation of an expensive-to-evaluate function, simulation or experiment, used to predict outcomes without running the full computation. It is fitted to a sample of evaluations and then queried cheaply to explore the design space, drive optimisation or quantify uncertainty. Surrogate models are central to Bayesian optimisation, where a Gaussian process guides where to evaluate next.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:surrogate-model", "labels": [ "Surrogate Model" ], "is_subclass_of": [ "Machine Learning", "Bayesian Optimisation" ], "wikilinks": [] }, { "id": "surveillance-capitalism", "title": "Surveillance Capitalism", "domain": "governance", "domain_name": "Governance", "definition": "Surveillance Capitalism is an economic logic, theorised by Shoshana Zuboff (2019), in which human experience is unilaterally claimed as a free raw material to be translated into behavioural data, processed by machine intelligence into prediction products, and sold in behavioural futures markets to business customers who seek to influence human behaviour. The model originated in digital advertising \u2014 Google's repurposing of surplus behavioural data from search to predict and influence click-through \u2014 and has since colonised social media, IoT devices, retail analytics, health apps, and smart-city infrastructure. Unlike industrial capitalism's exploitation of natural resources, surveillance capitalism exploits human behaviour and psyche as its primary resource, generating epistemic asymmetries between platform owners and the surveilled population and challenging the foundational conditions for autonomous selfhood, democratic deliberation, and market competition.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:surveillance-capitalism", "labels": [ "Surveillance Capitalism" ], "is_subclass_of": [ "Platform Economy" ], "wikilinks": [] }, { "id": "surveillance", "title": "Surveillance", "domain": "governance", "domain_name": "Governance", "definition": "Surveillance is the systematic monitoring, observation and collection of information about people, behaviour or activities, typically by state authorities, corporations or other organisations. It encompasses physical observation, electronic interception, video monitoring and the aggregation of digital data trails. Surveillance practices raise significant questions about proportionality, oversight and the balance between security objectives and individual privacy and civil liberties.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:surveillance", "labels": [ "Surveillance" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "suspicious-activity-report", "title": "Suspicious Activity Report", "domain": "finance", "domain_name": "Finance", "definition": "A Suspicious Activity Report (SAR) is a formal regulatory filing through which a financial institution notifies a national financial intelligence unit of transactions or behaviours that may indicate money laundering, terrorist financing, fraud, or other financial crime. SARs are mandated under anti-money-laundering frameworks and must be submitted within prescribed deadlines whenever staff or monitoring systems form a reasonable suspicion. The report documents the parties, accounts, transaction patterns, and the analyst's narrative rationale, while strict confidentiality (tipping-off prohibitions) prevents disclosure to the subject.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:suspicious-activity-report", "labels": [ "Suspicious Activity Report", "Suspicious Activity Reporting", "Suspicious Activity Reports" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "sustainability-framework", "title": "Sustainability Framework", "domain": "governance", "domain_name": "Governance", "definition": "A sustainability framework is a structured set of principles, metrics, and reporting standards used to measure and manage the environmental, social, and governance impact of an organisation or system. It defines what to measure, how to compute indicators such as carbon footprint, and how to disclose results so that performance can be compared and audited. Frameworks like GHG Protocol and emerging crypto-specific schemes provide the methodological backbone for sustainability claims.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sustainability-framework", "labels": [ "Sustainability Framework" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "sustainability-tool", "title": "Sustainability Tool", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Sustainability Tool is a software or analytical instrument used to measure, monitor, and optimise the environmental impact of digital systems, virtual environments, and technology infrastructure. In the metaverse context such tools address energy monitoring of rendering workloads, carbon footprint tracking for data-centre operations, and lifecycle assessment of hardware, enabling organisations to meet regulatory and voluntary environmental commitments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:sustainability-tool", "labels": [ "Sustainability Tool" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "sustainability", "title": "Sustainability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Sustainability is the capacity of systems \u2014 technological, social, economic, and environmental \u2014 to meet present needs without compromising the ability of future generations to meet their own. In digital and infrastructure contexts, it encompasses energy-efficient design, responsible resource consumption, equitable access, and long-term economic viability. Sustainability integrates environmental stewardship (carbon reduction, circular material flows), social equity (accessibility, labour fairness), and economic resilience (viable creator economies, non-exploitative business models). It provides the normative and operational framework within which infrastructure systems must be planned, governed, and evolved.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sustainability", "labels": [ "Sustainability" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "sustainability-reporting", "title": "sustainabilityreporting", "domain": "governance", "domain_name": "Governance", "definition": "Sustainability reporting is the structured, standardised practice through which organisations disclose their environmental, social, and governance (ESG) performance to investors, regulators, and other stakeholders. It encompasses the measurement, aggregation, and public presentation of data covering greenhouse gas emissions (Scope 1, 2, and 3), resource use, labour practices, board diversity, and supply chain impacts in conformance with frameworks such as GRI, ISSB IFRS S1/S2, TCFD, and the EU Corporate Sustainability Reporting Directive (CSRD). The discipline bridges quantitative carbon accounting, qualitative narrative disclosure, and third-party assurance to enable comparable, decision-useful ESG information. Emerging technologies including AI-assisted data collection, blockchain-based provenance, and real-time sensor integration are reshaping how organisations assemble and verify material sustainability data.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:sustainability-reporting", "labels": [ "SustainabilityReporting", "Sustainability Report", "Sustainability Reporting", "Sustainability Reporting Framework" ], "is_subclass_of": [ "Corporate Governance" ], "wikilinks": [] }, { "id": "sustainable-bitcoin-certificates", "title": "Sustainable Bitcoin Certificates", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Sustainable Bitcoin certificates are tradable attestations that a defined quantity of bitcoin mining was powered by clean or low-carbon energy, allowing holders to claim the environmental attributes of that hashrate. They function analogously to renewable energy certificates, decoupling the green-energy claim from the mined coins so it can be sold to ESG-conscious buyers. The mechanism aims to channel capital toward decarbonising proof-of-work mining.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:sustainable-bitcoin-certificates", "labels": [ "Sustainable Bitcoin Certificates" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "sustainable-bitcoin-protocol", "title": "Sustainable Bitcoin Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Sustainable Bitcoin Protocol (SBP) is a certification and tokenisation framework that attests the use of renewable or stranded energy in Bitcoin mining and issues on-chain Sustainable Bitcoin Certificates to verified miners. It enables green bitcoin premiums and provides institutional investors with auditable environmental provenance for Bitcoin holdings.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:sustainable-bitcoin-protocol", "labels": [ "Sustainable Bitcoin Protocol" ], "is_subclass_of": [ "Blockchain Sustainability" ], "wikilinks": [] }, { "id": "sustainable-consensus", "title": "Sustainable Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Sustainable consensus mechanisms are distributed agreement protocols designed for Blockchain Network|blockchain and distributed ledger systems that achieve Byzantine fault-tolerant finality whilst minimising energy consumption, carbon emissions, and physical resource expenditure\u2014contrasting s...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:sustainable-consensus", "labels": [ "Sustainable Consensus" ], "is_subclass_of": [ "Protocol and Consensus", "Consensus Mechanism", "Distributed Systems Protocol", "Blockchain Infrastructure", "Energy Efficient Computing", "Climate Technology" ], "wikilinks": [ "Algorand", "Algorand Pure Proof of Stake", "ASIC Mining", "Avalanche", "Avalanche Snow Consensus", "BIS Green Finance Standards", "Block Proposer Selection", "BLS Signature Aggregation", "CCRI Energy Benchmarks", "Centralised Database", "Chia Network", "Climate Technology", "ConsensusLayer", "Cosmos IBC Specification", "Crypto Climate Accord", "CryptoEconomicsDomain", "Cryptographic Hash Functions", "Cryptographic Sortition", "DeFi", "Decentralised Finance" ] }, { "id": "sustainable-development-goals", "title": "Sustainable Development Goals", "domain": "governance", "domain_name": "Governance", "definition": "The Sustainable Development Goals (SDGs) are a set of 17 interlinked global goals adopted by all United Nations member states in 2015 as part of the 2030 Agenda for Sustainable Development, spanning the elimination of poverty, zero hunger, quality education, gender equality, clean energy, decent work, reduced inequalities, climate action, and partnerships for implementation. Each goal is operationalised through targets and indicators that define measurable progress benchmarks, providing a shared framework for national governments, international organisations, civil society, and the private sector to align strategies and report progress. The SDGs extend and deepen the Millennium Development Goals, recognising that sustainable development requires integrated action across social, economic, and environmental dimensions simultaneously.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:sustainable-development-goals", "labels": [ "Sustainable Development Goals", "SDGs", "UN Sustainable Development Goals", "United Nations Sustainable Development Goals" ], "is_subclass_of": [ "Sustainable Development" ], "wikilinks": [] }, { "id": "sustainable-development", "title": "Sustainable Development", "domain": "governance", "domain_name": "Governance", "definition": "AI should contribute to the United Nations Sustainable Development Goals by addressing global challenges including climate change, resource depletion, biodiversity loss, and environmental degradation whilst ensuring development meets present needs without compromising future generations' ability to meet their own needs. Grounded in the Brundtland Commission (1987) definition and operationalised through the 17 SDGs adopted in 2015, it encompasses economic, social, and environmental pillars requiring coordinated policy, governance, and technological action.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:sustainable-development", "labels": [ "Sustainable Development" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "computational sustainability", "Green AI", "sustainable AI operations", "MetaverseDomain" ] }, { "id": "sustainable-finance", "title": "Sustainable Finance", "domain": "finance", "domain_name": "Finance", "definition": "Sustainable finance refers to the integration of environmental, social, and governance (ESG) criteria into financial decision-making, product design, and market regulation with the goal of channelling capital flows toward activities that support a sustainable economy. It encompasses green bonds, sustainability-linked loans, ESG investing, climate risk management, and regulatory taxonomies that define what qualifies as environmentally sustainable economic activity.", "entityType": "Class", "qualityScore": 0.82, "maturity": "established", "iri": "urn:ngm:class:sustainable-finance", "labels": [ "Sustainable Finance", "Sustainable Investing" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "sustainable-technology", "title": "Sustainable Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technology designed, deployed, and operated to minimise long-term environmental impact through energy efficiency, use of renewable energy sources, responsible lifecycle management, and reduction of waste and carbon emissions. In IT and spatial computing contexts this encompasses energy-efficient data centres, green cloud computing, virtualisation, sustainable software engineering practices, circular economy principles applied to hardware, and AI-driven optimisation of resource consumption. Sustainable technology sits at the intersection of governance, infrastructure design, and environmental accountability, and is increasingly mandated by corporate ESG commitments and governmental net-zero regulations.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:sustainable-technology", "labels": [ "Sustainable Technology" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "swarm-control", "title": "Swarm Control", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Swarm control coordinates many simple agents through local rules and interactions so that useful collective behaviour emerges without centralised command.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:swarm-control", "labels": [ "Swarm Control", "Swarm Coordination" ], "is_subclass_of": [ "Robotics", "Actuation and Control", "Robotics Domain" ], "wikilinks": [ "Multi-Agent System", "Swarm Intelligence", "Autonomous Agent", "Subsumption Architecture", "Robotics Domain" ] }, { "id": "swarm-intelligence", "title": "Swarm Intelligence", "domain": "robotics", "domain_name": "Robotics", "definition": "Swarm intelligence is the collective behaviour that emerges from many simple agents following local rules and interactions, producing coordinated global behaviour without central control.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:swarm-intelligence", "labels": [ "Swarm Intelligence" ], "is_subclass_of": [ "Navigation and Planning", "Autonomous Systems Domain" ], "wikilinks": [ "Swarm Robotics", "Multi-Robot Systems", "Decentralized Swarm Control", "Distributed Systems", "Autonomous Systems Domain" ] }, { "id": "swarm-robot", "title": "Swarm Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Swarm Robot is a member of a multi-agent robotic system in which large numbers of simple, decentralised agents coordinate through local interactions to achieve complex collective behaviours without centralised control. Drawing from biological models such as ant colonies and flocking birds, swarm robotics enables robustness through redundancy, scalability, and emergent task execution across domains including environmental monitoring, logistics, and disaster response.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:swarm-robot", "labels": [ "Swarm Robot" ], "is_subclass_of": [ "Robot Type", "Autonomous Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "swarm-robotics", "title": "Swarm Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Swarm robotics employs large numbers of simple, autonomous agents that exhibit sophisticated collective behaviours through local interactions and decentralised control without centralised coordination, inspired by biological swarms like ant colonies and bird flocks. Individual robots with limited sensing, computation, and actuation communicate locally with neighbours, creating emergent system-level intelligence enabling tasks like coordinated navigation, object transport, and environmental sensing that exceed individual capabilities.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:swarm-robotics", "labels": [ "Swarm Robotics", "SwarmRobotics" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "BioInspiredRobotics", "CollectiveIntelligence", "communicatesVia", "consistsOfRobots", "DecentralisedControl", "DecentralizedControl", "DecentralizedControl", "dt:coordinatedBy", "dt:governedBy", "dt:optimizedVia", "dt:simulatedIn", "dt:trackedOn", "EmergentBehavior", "EmergentBehavior", "EvolutionaryAlgorithm", "exhibitsBehavior", "MultiAgentSystem", "SearchAndRescue", "SelfOrganization", "usesCoordination" ] }, { "id": "swath", "title": "Swath", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:swath", "labels": [ "Swath" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "swi-glu", "title": "SwiGLU", "domain": "ai", "domain_name": "Ai", "definition": "SwiGLU is a gated activation function for neural networks that combines the Swish (SiLU) nonlinearity with a Gated Linear Unit, computing the element-wise product of a Swish-activated projection and a linear gate projection. It is widely used in the feed-forward sublayers of modern transformer models because it empirically improves quality over ReLU or GELU at comparable cost. Its gating mechanism gives the network a learnable, input-dependent pathway through each feed-forward block.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:swi-glu", "labels": [ "SwiGLU" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "sybil-attack", "title": "Sybil Attack", "domain": "security", "domain_name": "Security", "definition": "A Sybil Attack is a network security threat in which a single adversary creates a large number of pseudonymous identities to gain disproportionate influence over a peer-to-peer system. In blockchain and distributed ledger contexts it can subvert reputation systems, distort consensus voting, facilitate eclipse attacks, and undermine proof-of-stake weighting. Countermeasures include proof-of-work, stake-weighted voting, identity verification, and peer diversity requirements.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:sybil-attack", "labels": [ "Sybil Attack" ], "is_subclass_of": [ "Security and Identity", "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "sybil-resistance", "title": "sybil resistance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Sybil resistance is a security property of distributed and decentralised networks that constrains the ability of a single adversary to gain disproportionate influence by fabricating multiple pseudonymous or fake identities. The property is foundational to any permissionless system where voting power, reputation, resource allocation, or participation rights are tied to the concept of a unique participant. Mechanisms achieving sybil resistance range from resource-binding consensus protocols \u2014 proof-of-work and proof-of-stake \u2014 to cryptographic proof-of-personhood schemes and social trust-graph analysis, each balancing security guarantees against privacy requirements.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:sybil-resistance", "labels": [ "Sybil Resistance", "Sybil Resistance Metrics" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "symbol-grounding", "title": "Symbol Grounding", "domain": "ai", "domain_name": "Ai", "definition": "Symbol grounding is the problem and process of connecting abstract symbols manipulated by an intelligent system to their referents in the perceptual and physical world, so that the symbols carry intrinsic meaning rather than being defined only by other symbols. First articulated by Stevan Harnad, it asks how a system can avoid an infinite regress of definitions and acquire understanding tied to sensory experience. It is central to debates about whether language models truly understand the concepts they process.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:symbol-grounding", "labels": [ "Symbol Grounding" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "symbolic-ai", "title": "Symbolic AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Symbolic AI (Good Old-Fashioned AI, GOFAI) is an approach to artificial intelligence based on explicit symbolic representations of knowledge, logical inference rules, and search algorithms. It underpins expert systems, automated planning, knowledge graphs, and formal reasoning systems, and is currently experiencing a renaissance as a complement to neural approaches in neurosymbolic AI.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:symbolic-ai", "labels": [ "Symbolic AI", "Symbolic AI System" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "AAAI", "Academic AI Research", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "symbolic-object-library", "title": "Symbolic Object Library", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A collection or repository of standardized 3D digital objects, assets, and components designed for use across metaverse platforms, including models, textures, animations, and interactive elements that can be shared and reused in virtual environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:symbolic-object-library", "labels": [ "Symbolic Object Library" ], "is_subclass_of": [ "Content and Assets", "Digital Asset" ], "wikilinks": [ "Digital Asset", "metaverse" ] }, { "id": "symbolic-reasoning", "title": "Symbolic Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A family of AI techniques that represent knowledge as explicit symbols, rules, and logical relations and derive new conclusions through formal inference mechanisms such as resolution, forward/backward chaining, or constraint propagation. Symbolic reasoning underpins expert systems, knowledge graphs, and logic programming, and contrasts with sub-symbolic connectionist approaches while being combined with them in neuro-symbolic architectures.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:symbolic-reasoning", "labels": [ "Symbolic Reasoning" ], "is_subclass_of": [ "AI Technique", "Reasoning", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "symmetric-cryptography", "title": "Symmetric Cryptography", "domain": "security", "domain_name": "Security", "definition": "Symmetric cryptography is a class of cryptographic techniques in which the same secret key is used for both encryption and decryption. It includes block ciphers and stream ciphers and underpins fast bulk data protection and authenticated encryption. Its security depends on keeping the shared key secret and on secure key distribution, which is often handled by asymmetric methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:symmetric-cryptography", "labels": [ "Symmetric Cryptography" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "symmetric-encryption", "title": "Symmetric Encryption", "domain": "security", "domain_name": "Security", "definition": "A cryptographic method using a single shared secret key for both encryption and decryption, requiring secure key exchange between parties before communication. Provides high-throughput confidentiality for data at rest and in transit, making it the standard approach for bulk data encryption across security, infrastructure, and distributed-systems domains.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:symmetric-encryption", "labels": [ "Symmetric Encryption", "AES-128 Encryption", "Symmetric Cipher", "Symmetric Key Cryptography" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain" ], "wikilinks": [ "Confidential Transactions", "Hybrid Encryption", "Asymmetric Encryption", "Blockchain", "Cryptography", "Digital Signature", "Hash Function", "Key Derivation Function", "Random Number Generation" ] }, { "id": "symmetric-key", "title": "Symmetric Key", "domain": "security", "domain_name": "Security", "definition": "A symmetric key is a single shared secret used by both the sender and receiver to encrypt and decrypt data in symmetric-key cryptography. Because the same key performs both operations, it must be kept secret and distributed securely between the communicating parties. Symmetric-key algorithms such as AES are fast and efficient for bulk data encryption, which makes them the workhorse of confidentiality in practice, often combined with asymmetric techniques that solve the problem of securely exchanging the shared key.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:symmetric-key", "labels": [ "Symmetric Key" ], "is_subclass_of": [ "Cryptographic Key" ], "wikilinks": [] }, { "id": "synchronisation", "title": "Synchronisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Synchronisation is the coordination of multiple processes, devices, or data replicas so that they reach a consistent state or act in a precise temporal relationship. In computing it covers concurrency primitives like locks and barriers; in distributed and robotic systems it covers clock alignment and coordinated motion. Accurate synchronisation is essential wherever independent components must agree on order, timing, or shared state.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:synchronisation", "labels": [ "Synchronisation", "Data Synchronisation", "Delta Synchronisation", "Differential Synchronisation", "IoT Synchronisation", "Peer-to-peer Synchronisation", "Synchronisation Mechanism" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "synchronization-protocol", "title": "Synchronization Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A synchronization protocol is a set of rules by which distributed parties bring their state, clocks, or data into agreement despite operating independently and communicating over unreliable channels. It encompasses clock-synchronisation protocols (NTP, PTP) that align time across machines, and data-synchronisation protocols that reconcile divergent replicas using version vectors, operational transforms, or conflict-free replicated data types. Synchronization protocols are foundational to distributed databases, collaborative editing, mobile offline-first applications, and any system where multiple nodes must converge on a consistent view.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:synchronization-protocol", "labels": [ "Synchronization Protocol" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "synchronous-collaboration", "title": "Synchronous Collaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"Real-time interaction mode where distributed participants engage simultaneously through technology-mediated channels, enabling immediate feedback, spontaneous ideation, and social presence comparable to co-located teamwork.\"", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:synchronous-collaboration", "labels": [ "Synchronous Collaboration", "TC-0010-Synchronous-Collaboration" ], "is_subclass_of": [ "Communication Technology", "Telecollaboration", "TC-0001-telecollaboration-domain" ], "wikilinks": [ "TC-0001-telecollaboration-domain", "TC-0011-Video-Conferencing", "TC-0020-Asynchronous-Collaboration", "TC-0040-Communication-Protocols", "TC-0080-Team-Coordination", "TELE-028-horizon-workrooms", "TELE-100-ai-avatars", "TELE-105-real-time-language-translation", "TELE-150-webrtc", "TELE-153-5g-telepresence", "TELE-251-smart-contract-coordination", "TELE-252-dao-governance-telecollaboration", "TELE-301-virtual-office-spaces", "Social Presence" ] }, { "id": "synchronous-communication", "title": "Synchronous Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Synchronous communication is an interaction pattern in which participants exchange messages in real time, with sender and receiver engaged simultaneously and responses expected without significant delay. In collaboration it covers video conferencing, live chat and voice calls; in distributed systems it covers blocking request-response exchanges where a caller waits for a reply. It favours immediacy and tight coordination at the cost of requiring co-presence and tolerance to latency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:synchronous-communication", "labels": [ "Synchronous Communication" ], "is_subclass_of": [ "Real-Time Communication" ], "wikilinks": [] }, { "id": "synchronous-execution", "title": "Synchronous Execution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Synchronous Execution is a computational execution model in which operations are performed sequentially, with each call blocking the invoking thread until a result is returned before the next operation begins. This model provides deterministic, predictable control flow and simplifies error handling, making it well-suited for transactional operations, authentication flows, and ACID-compliant database interactions, though it constrains throughput and scalability under high-concurrency workloads.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:synchronous-execution", "labels": [ "Synchronous Execution" ], "is_subclass_of": [ "Protocol and Consensus", "Execution Model" ], "wikilinks": [ "API Request-Response", "Blocking Operation", "Database Operations", "Execution Model", "Thread Management", "Asynchronous Execution", "Distributed Systems", "Transaction Processing" ] }, { "id": "syntactic-interoperability", "title": "Syntactic Interoperability", "domain": "data", "domain_name": "Data", "definition": "Syntactic interoperability is the ability of two or more systems to exchange data using compatible formats, structures, and encoding rules so that the receiving system can correctly parse the transmitted information. It concerns agreement on data formats, message structure, and serialisation grammar, independent of the meaning of the data. Syntactic interoperability is a prerequisite layer beneath semantic interoperability in interoperability frameworks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:syntactic-interoperability", "labels": [ "Syntactic Interoperability" ], "is_subclass_of": [ "Interoperability" ], "wikilinks": [] }, { "id": "synthetic-aperture-radar", "title": "Synthetic Aperture Radar", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:synthetic-aperture-radar", "labels": [ "Synthetic Aperture Radar" ], "is_subclass_of": [ "Active Remote Sensing", "Microwave Remote Sensing" ], "wikilinks": [] }, { "id": "synthetic-asset", "title": "Synthetic Asset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Synthetic Asset is a tokenised financial instrument on a blockchain whose value tracks an underlying reference asset, such as a fiat currency, commodity, equity, or index, without requiring direct ownership or custody of that asset. Synthetic assets derive their price through collateralisation and price oracles rather than through a one-to-one backing of the underlying, distinguishing them from wrapped tokens. Protocols such as Synthetix mint synthetic exposures (synths) backed by over-collateralised pools, allowing on-chain trading of real-world price feeds. They enable permissionless access to traditional markets but carry oracle, liquidation, and collateral-risk dependencies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:synthetic-asset", "labels": [ "Synthetic Asset" ], "is_subclass_of": [ "Cryptocurrency Token" ], "wikilinks": [] }, { "id": "synthetic-biology", "title": "Synthetic Biology", "domain": "data", "domain_name": "Data", "definition": "Synthetic biology is a field that applies engineering principles to design and construct biological systems and organisms. It combines biology, genetics, and engineering.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:synthetic-biology", "labels": [ "Synthetic Biology" ], "is_subclass_of": [ "AI Research Area", "owl:Thing" ], "wikilinks": [ "FDA", "owl:Thing", "https://en.wikipedia.org/wiki/Synthetic_biology", "https://www.genome.gov/genetics-glossary/Synthetic-Biology" ] }, { "id": "synthetic-data-generation", "title": "Synthetic Data Generation", "domain": "ai", "domain_name": "Ai", "definition": "Synthetic Data Generation is the process of algorithmically producing artificial datasets that statistically mirror real-world distributions without exposing sensitive personal information. Techniques include generative adversarial networks, diffusion models, physics simulation, and rule-based sampling, enabling model training where real data is scarce, private, or costly to label.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:synthetic-data-generation", "labels": [ "Synthetic Data Generation", "Synthetic Dataset Generation", "Synthetic Training Data Generation" ], "is_subclass_of": [ "Data Engineering" ], "wikilinks": [ "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "synthetic-data-generator", "title": "Synthetic Data Generator", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AI-powered system that produces artificial datasets preserving statistical properties and structural characteristics of original data while protecting privacy and enabling testing scenarios.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:synthetic-data-generator", "labels": [ "Synthetic Data Generator" ], "is_subclass_of": [ "Content and Assets", "Data Anonymization Pipeline" ], "wikilinks": [ "AI Pipeline", "AILayer", "Data Augmentation", "Data Management Platform", "Data Simulator", "IEEE P2048-9", "ISO/IEC 5259", "OECD AI", "Privacy Metrics", "Privacy Validator", "Statistical Analyzer", "Statistical Models", "Testing Dataset Creation", "ComputationAndIntelligenceDomain", "Computer Vision", "DataLayer", "Differential Privacy", "Generative Adversarial Network", "Generative Model", "InfrastructureDomain" ] }, { "id": "synthetic-data", "title": "synthetic data", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Synthetic data is algorithmically generated content that preserves the statistical properties, distributional characteristics, and structural patterns of real-world datasets without containing actual personal or proprietary records. It is produced using techniques such as generative adversarial networks, variational autoencoders, diffusion models, rule-based simulators, and statistical resampling methods. Synthetic data serves to augment scarce or imbalanced training corpora, enable privacy-compliant data sharing under regulations such as GDPR, and stress-test machine-learning pipelines with rare, hazardous, or counterfactual edge-case scenarios. Quality is typically benchmarked via fidelity metrics such as Fr\u00e9chet Inception Distance, train-on-synthetic-test-on-real accuracy, and statistical divergence measures.", "entityType": "Class", "qualityScore": 0.73, "maturity": "established", "iri": "urn:ngm:class:synthetic-data", "labels": [ "Synthetic Data" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "synthetic-media", "title": "Synthetic Media", "domain": "ai", "domain_name": "Ai", "definition": "Synthetic Media refers to audio, video, image, and text content that is wholly generated or substantially manipulated by computational systems \u2014 primarily AI models \u2014 rather than captured directly from reality. It encompasses outputs from generative adversarial networks, diffusion models, large language models, neural text-to-speech systems, and neural rendering pipelines, including deepfakes, AI-generated images, cloned voices, and fully synthetic video sequences. The concept spans both creative and malicious applications, from cinematic visual effects and interactive media to disinformation campaigns and identity fraud, making detection, provenance tracking, and regulatory disclosure central concerns. Synthetic media sits at the convergence of generative AI capability, content authenticity infrastructure, and digital governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:synthetic-media", "labels": [ "Synthetic Media", "Synthetic Media Creation" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "synthetix", "title": "Synthetix", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A protocol on Ethereum that issues synthetic assets (synths) tracking the price of external references such as fiat currencies, commodities, and indices, backed by a pooled collateral of its native SNX token and other staked assets; trades execute against the shared debt pool rather than an order book.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:synthetix", "labels": [ "Synthetix" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Collateral Management", "Price Oracle", "Liquidity Provision", "Decentralised Exchange", "Smart Contract" ] }, { "id": "system-architecture", "title": "System Architecture", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "System Architecture defines the holistic structure integrating hardware, software, data infrastructure, and operational components to deliver capabilities at scale, encompassing distributed training infrastructure, inference servers, data pipelines, model registries, monitoring systems, and edge-cloud coordination.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:system-architecture", "labels": [ "System Architecture" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [ "Infrastructure as Code", "MLOps", "Cloud Computing", "Distributed Systems" ] }, { "id": "system-dynamics", "title": "System Dynamics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A modelling methodology, originated by Jay Forrester at MIT in the 1950s, that represents a complex system as aggregate stocks, flows, and feedback loops governed by coupled differential (or difference) equations, simulating how policy choices and delays produce counter-intuitive behaviour over time; it takes a top-down, population-level view, in deliberate contrast to agent-based modelling's bottom-up simulation of heterogeneous individuals.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:system-dynamics", "labels": [ "System Dynamics" ], "is_subclass_of": [ "Computational Modelling" ], "wikilinks": [ "Computational Modelling", "Agent-Based Modelling", "Feedback Loop", "Complex Systems" ] }, { "id": "system-identification", "title": "system identification", "domain": "robotics", "domain_name": "Robotics", "definition": "System identification is the discipline of constructing or refining a mathematical model of a dynamical system\u2014parametric or non-parametric\u2014from observed input-output experimental data, enabling accurate simulation and model-based control design. It spans classical methods such as prediction error minimisation, subspace identification, and autoregressive modelling for linear time-invariant systems, through to Gaussian process regression, neural ordinary differential equations, and physics-informed learning for nonlinear and hybrid systems. In robotics and mechatronics, system identification calibrates rigid-body dynamics parameters\u2014link inertia tensors, joint friction coefficients, and actuator gains\u2014required by whole-body controllers and model predictive controllers to generate physically consistent torque commands. The field occupies an intersection of statistical estimation theory, control engineering, and machine learning, and is foundational to closing the sim-to-real gap in data-driven robot learning.", "entityType": "Class", "qualityScore": 0.73, "maturity": "mature", "iri": "urn:ngm:class:system-identification", "labels": [ "System Identification" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "system-integration", "title": "System Integration", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "System Integration is the engineering discipline of connecting and orchestrating disparate software components, services, databases, and external platforms so that they behave as a unified, interoperable system. It spans interface design via APIs and messaging protocols, middleware orchestration, data transformation pipelines, identity federation, and governance of information flows across organisational and technical boundaries. Unlike simple point-to-point coupling, mature system integration imposes coherent contracts, observability, and error-recovery strategies so that independently developed or procured subsystems can exchange information reliably. The discipline underpins enterprise IT, cloud-native architectures, IoT deployments, and the composable software stacks that drive spatial computing and AI platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:system-integration", "labels": [ "System Integration" ], "is_subclass_of": [ "Software Architecture" ], "wikilinks": [] }, { "id": "system-interoperability", "title": "System Interoperability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "System interoperability is the ability of different information technology systems, applications, and devices to exchange data, interpret shared information, and use it in a mutually useful way without requiring special translation or middleware. It enables diverse systems to communicate and work together seamlessly through adherence to common standards, protocols, and data formats, facilitating efficient information flow across organisational and technical boundaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:system-interoperability", "labels": [ "System Interoperability" ], "is_subclass_of": [ "Network and Communication", "System Capability" ], "wikilinks": [ "Common Data Formats", "Core Technology", "Enterprise Connectivity", "Standard Protocols", "System Capability", "Data Sharing", "System Integration" ] }, { "id": "system-prompt", "title": "system prompt", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A system prompt is a structured instruction block injected into the context window of a large language model at the start of an inference session, establishing operational context, persona, tool descriptions, safety constraints, and behavioural guidelines before any user turn is processed. Unlike user messages, system prompts are authored by operators rather than end users and govern the model's allowed behaviours, response style, and tool-use policy throughout the session. In agentic and multi-agent architectures, system prompts serve as the primary mechanism for role specialisation, capability scoping, safety guardrail enforcement, and task decomposition. The security boundary between the system prompt and user-controlled input is a central concern in prompt injection defence.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:system-prompt", "labels": [ "System Prompt" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [] }, { "id": "system-software", "title": "System Software", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Low-level software that directly manages hardware resources and provides foundational services upon which application software operates. In XR and spatial computing contexts, system software encompasses device drivers, operating system kernels, firmware, and hardware abstraction layers that expose display, tracking, and input peripherals to higher-level runtimes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:system-software", "labels": [ "System Software" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "system-on-chip", "title": "System-on-Chip", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A system-on-chip (SoC) is an integrated circuit that combines most or all of the components of a computer system onto a single die, typically including one or more processor cores, memory, input and output interfaces and specialised accelerators. By co-locating these subsystems, an SoC reduces physical size, power consumption and inter-component latency compared with multi-chip designs. SoCs are foundational to mobile devices, embedded systems and edge computing, where space and energy efficiency are critical.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:system-on-chip", "labels": [ "System-on-Chip", "System on Chip" ], "is_subclass_of": [ "Hardware Component" ], "wikilinks": [] }, { "id": "system", "title": "System", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A bounded, purposeful arrangement of components\u2014hardware, software, data, and processes\u2014that interact to perform functions beyond those of any individual component. In AI contexts, a system integrates models, inference engines, data pipelines, and interfaces into deployable solutions that operate within broader sociotechnical environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:system", "labels": [ "System" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "systemic-risk-management", "title": "Systemic Risk Management", "domain": "finance", "domain_name": "Finance", "definition": "Systemic risk management is the set of practices, institutions and policy tools aimed at identifying, monitoring and mitigating risks that could destabilise an entire financial system rather than a single firm. It combines macroprudential regulation, stress testing, capital and liquidity buffers, and resolution regimes to limit contagion across interconnected institutions and markets. Coordinated by bodies such as the Financial Stability Board and central banks, it seeks to preserve financial stability and prevent crises from cascading through the economy.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:systemic-risk-management", "labels": [ "Systemic Risk Management" ], "is_subclass_of": [ "Financial Stability" ], "wikilinks": [] }, { "id": "systemic-risk", "title": "Systemic Risk", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Risk specific to high-impact capabilities of general-purpose AI models with significant impact on the Union market due to reach, or actual or foreseeable negative effects on public health, safety, fundamental rights, environment, democracy, or rule of law. Under EU AI Act Article 51, models exceeding 10^25 FLOPs training compute or matching the most advanced GPAI capabilities are presumed to carry systemic risk and face enhanced obligations including adversarial testing, incident reporting, and cybersecurity protection.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:systemic-risk", "labels": [ "Systemic Risk", "Systemic Risk Mitigation", "Systemic Risk Oversight" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "systems-design", "title": "Systems Design", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A design discipline focused on the structure and layout of websites and applications, as opposed to the creation of individual visual assets.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:systems-design", "labels": [ "Systems Design" ], "is_subclass_of": [ "Image Generation" ], "wikilinks": [] }, { "id": "systems-engineering", "title": "Systems Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systems Engineering is an interdisciplinary methodology for designing, integrating, and managing complex systems over their entire lifecycle, from concept definition through disposal. It applies structured processes \u2014 requirements analysis, functional decomposition, architecture definition, interface control, verification, and validation \u2014 to ensure that the emergent behaviour of integrated subsystems meets overall system objectives within defined cost, schedule, and performance constraints. Systems Engineering is distinguished from narrower disciplines by its holistic perspective: it treats the system as a whole rather than optimising individual components in isolation. Governing standards include ISO/IEC/IEEE 15288 (Systems and Software Life Cycle Processes) and the INCOSE Systems Engineering Handbook.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:systems-engineering", "labels": [ "Systems Engineering" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "systems-theory", "title": "Systems Theory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Systems theory is an interdisciplinary study of systems as sets of interrelated components, focusing on the relationships, feedback, and emergent behaviour that arise from interaction.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:systems-theory", "labels": [ "Systems Theory" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Cybernetics", "owl:Thing" ] }, { "id": "systolic-array", "title": "Systolic Array", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Systolic Array is a specialised parallel computing architecture composed of a homogeneous network of processing elements (PEs) that rhythmically compute and pass data through the array in a pipelined fashion, analogous to the rhythmic pumping of the heart. Each PE performs a fixed local computation and passes results to neighbours without centralised control or global memory access. This architecture is highly efficient for matrix multiplication and convolution operations, making it the dominant microarchitecture in modern AI accelerators such as Google's Tensor Processing Units.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:systolic-array", "labels": [ "Systolic Array" ], "is_subclass_of": [ "Infrastructure", "AI Hardware" ], "wikilinks": [] }, { "id": "tc-0080-team-coordination", "title": "TC-0080-Team-Coordination", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Team coordination is the management of dependencies between people, tasks and resources so that members of a group act in a consistent and timely way towards shared objectives.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tc-0080-team-coordination", "labels": [ "TC-0080-Team-Coordination", "Distributed Team Coordination", "Team Coordination" ], "is_subclass_of": [ "Collaboration", "Workspace Tools" ], "wikilinks": [ "Collaboration", "Remote Collaboration", "Project Management" ] }, { "id": "tcfd", "title": "TCFD", "domain": "governance", "domain_name": "Governance", "definition": "The Task Force on Climate-related Financial Disclosures (TCFD) is a disclosure framework established by the Financial Stability Board in 2015 under Mark Carney's initiative, designed to help companies and financial institutions consistently report material climate-related risks and opportunities in a format useful to investors, lenders, and insurers. The framework is structured around four thematic pillars\u2014Governance, Strategy, Risk Management, and Metrics and Targets\u2014and emphasises scenario analysis to disclose how different climate futures (aligned with 1.5\u00b0C, 2\u00b0C, or 4\u00b0C pathways) affect an organisation's financial position. TCFD recommendations have been incorporated into mandatory regulatory regimes in the UK, EU, New Zealand, and other jurisdictions, and serve as the conceptual foundation for the IFRS Sustainability Disclosure Standards (ISSB S2).", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tcfd", "labels": [ "TCFD", "TCFD Framework", "TCFD Recommendations", "TCFD Reporting" ], "is_subclass_of": [ "SustainabilityReporting" ], "wikilinks": [] }, { "id": "tcp-ip", "title": "TCP/IP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "TCP/IP is the layered suite of communication protocols that underpins the Internet, named after its two core members: the Transmission Control Protocol and the Internet Protocol. It organises networking into link, internet, transport, and application layers, providing addressing, routing, and reliable end-to-end byte streams over heterogeneous physical networks. Its packet-switched, best-effort design with end-to-end reliability above the network layer enabled global interoperability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:tcp-ip", "labels": [ "TCP/IP" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "tcp", "title": "TCP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Transmission Control Protocol (TCP) is a connection-oriented transport-layer protocol that provides reliable, ordered, and error-checked delivery of a byte stream between applications over an IP network. It establishes connections via a three-way handshake, segments data, acknowledges receipt, retransmits lost segments, and applies flow and congestion control to share network capacity fairly. TCP is one of the core protocols of the Internet protocol suite, underlying most application protocols that demand reliability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:tcp", "labels": [ "TCP" ], "is_subclass_of": [ "Network Transport" ], "wikilinks": [] }, { "id": "tee", "title": "TEE", "domain": "security", "domain_name": "Security", "definition": "A trusted execution environment, a secure area of a processor that isolates code and data so that they are protected from the rest of the system, including a compromised operating system. It provides confidentiality and integrity for sensitive computation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tee", "labels": [ "TEE", "Trusted Execution Environment" ], "is_subclass_of": [], "wikilinks": [ "Hardware", "Data Confidentiality", "Trusted Execution Environment" ] }, { "id": "tele-001-telepresence", "title": "TELE 001 telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "TELE 001 Telepresence is a foundational concept within the telecollaboration domain that defines the sense of physical presence transmitted across a distance through audio-visual, haptic, and immersive communication technologies. It serves as the baseline specification for how distributed participants experience co-presence in collaborative environments, underpinning higher-order telecollaboration and virtual meeting systems.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:tele-001-telepresence", "labels": [ "TELE 001 telepresence", "TELE-001-telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "owl:Thing" ] }, { "id": "tele-050-neuralrenderingtelepresence", "title": "TELE 050 neuralrenderingtelepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Neural Rendering Telepresence (TELE 050) is a class of telepresence systems that replace conventional rasterisation pipelines with neural rendering techniques \u2014 including neural radiance fields, Gaussian splatting, and differentiable rendering \u2014 to reconstruct and transmit photorealistic volumetric representations of remote participants in real time. These approaches substantially reduce capture hardware requirements while improving perceptual fidelity and supporting free-viewpoint synthesis.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:tele-050-neuralrenderingtelepresence", "labels": [ "TELE 050 neuralrenderingtelepresence" ], "is_subclass_of": [ "Telepresence", "TELE 001 telepresence" ], "wikilinks": [ "owl:Thing" ] }, { "id": "tele-005-common-ground-theory", "title": "TELE-005-common-ground-theory", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Common ground theory is an account of communication holding that participants build and rely on shared knowledge, beliefs and assumptions, which they update through interaction to coordinate meaning.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-005-common-ground-theory", "labels": [ "TELE-005-common-ground-theory" ], "is_subclass_of": [ "Communication Theory" ], "wikilinks": [ "Communication Theory", "Collaboration" ] }, { "id": "tele-025-microsoft-hololens", "title": "TELE-025-microsoft-hololens", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Microsoft HoloLens is a self-contained mixed reality head-mounted display that overlays interactive holographic content onto the wearer's view of the real world.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-025-microsoft-hololens", "labels": [ "TELE-025-microsoft-hololens", "HoloLens", "Microsoft HoloLens" ], "is_subclass_of": [ "Head-Mounted Display" ], "wikilinks": [ "Spatial Mesh", "Spatial Anchors", "Mixed Reality", "Head-Mounted Display" ] }, { "id": "tele-026-microsoft-mesh", "title": "TELE-026-microsoft-mesh", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Microsoft Mesh is a platform for shared mixed reality experiences that lets people in different physical locations join a common virtual space using avatars or holographic representations, integrating with Microsoft Teams to deliver immersive collaboration scenarios such as design review, training and distributed meetings.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tele-026-microsoft-mesh", "labels": [ "TELE-026-microsoft-mesh" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Avatar System", "Spatial Anchors", "Remote Collaboration", "Mixed Reality", "Telepresence" ] }, { "id": "tele-027-spatial-platform", "title": "TELE-027-spatial-platform", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Spatial is a collaboration platform that hosts virtual rooms where participants meet as avatars to share documents, 3D models and media for meetings, events and presentations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-027-spatial-platform", "labels": [ "TELE-027-spatial-platform" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Avatar System", "Remote Collaboration", "Metaverse", "Telepresence" ] }, { "id": "tele-050-neural-rendering-telepresence", "title": "TELE-050-neural-rendering-telepresence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Neural rendering for telepresence is the application of learned, image-synthesis models to reconstruct and display remote people and scenes with photorealistic appearance and free viewpoint control.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-050-neural-rendering-telepresence", "labels": [ "TELE-050-neural-rendering-telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Neural Rendering", "Novel View Synthesis", "Telepresence" ] }, { "id": "tele-053-volumetric-video-conferencing", "title": "TELE-053-volumetric-video-conferencing", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Volumetric video conferencing is real-time meeting communication in which participants are captured as three-dimensional representations and rendered so others can view them from arbitrary angles.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-053-volumetric-video-conferencing", "labels": [ "TELE-053-volumetric-video-conferencing", "Volumetric Video Conferencing" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Volumetric Capture", "Point Cloud", "Telepresence", "Video Conferencing" ] }, { "id": "tele-060-instant-ngp", "title": "TELE-060-instant-ngp", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Instant-NGP (Instant Neural Graphics Primitives) is a method that uses a multi-resolution hash encoding to train and evaluate neural graphics representations such as neural radiance fields far faster than earlier approaches.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-060-instant-ngp", "labels": [ "TELE-060-instant-ngp", "Instant NGP" ], "is_subclass_of": [ "Neural Rendering" ], "wikilinks": [ "Neural Radiance Fields", "Differentiable Rendering", "Novel View Synthesis", "NeRF", "Neural Rendering" ] }, { "id": "tele-102-codec-avatars", "title": "TELE-102-codec-avatars", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Codec Avatars are a research effort by Meta to produce photorealistic, real-time avatars of people that are learned from capture data and driven by sensors to reproduce expression and appearance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-102-codec-avatars", "labels": [ "TELE-102-codec-avatars" ], "is_subclass_of": [ "Avatar System" ], "wikilinks": [ "Neural Rendering", "Volumetric Capture", "Telepresence", "Avatar System" ] }, { "id": "tele-107-ai-meeting-assistants", "title": "TELE-107-ai-meeting-assistants", "domain": "metaverse", "domain_name": "Metaverse", "definition": "AI meeting assistants are software agents that join or process meetings to provide transcription, summarisation, action-item extraction and live support using speech recognition and language models.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-107-ai-meeting-assistants", "labels": [ "TELE-107-ai-meeting-assistants" ], "is_subclass_of": [ "Telecollaboration" ], "wikilinks": [ "Speech Recognition", "Large Language Models", "Meeting Transcription", "Conversational AI", "Telecollaboration" ] }, { "id": "tele-110-spatial-audio-processing", "title": "TELE-110-spatial-audio-processing", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Spatial audio processing is the set of signal-processing techniques that position sound sources in three-dimensional space for a listener, using cues such as direction, distance and room acoustics.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-110-spatial-audio-processing", "labels": [ "TELE-110-spatial-audio-processing" ], "is_subclass_of": [ "Spatial Audio" ], "wikilinks": [ "Binaural Rendering", "Telepresence", "Spatial Audio" ] }, { "id": "tele-151-real-time-protocols", "title": "TELE-151-real-time-protocols", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Real-time protocols are network communication standards designed to deliver time-sensitive media and data with low latency and predictable timing, accepting some loss in exchange for timeliness.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-151-real-time-protocols", "labels": [ "TELE-151-real-time-protocols", "Real-Time Protocols" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Network Protocol", "Low Latency", "WebRTC" ] }, { "id": "tele-153-5-g-telepresence", "title": "TELE-153-5g-telepresence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "5G telepresence is the delivery of real-time immersive communication over fifth-generation mobile networks, using their higher bandwidth and lower latency to support volumetric and high-resolution remote presence on mobile devices.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-153-5-g-telepresence", "labels": [ "TELE-153-5g-telepresence" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "5G Network", "Network Slicing", "Low Latency", "5G", "Telepresence" ] }, { "id": "tele-154-edge-computing-telepresence", "title": "TELE-154-edge-computing-telepresence", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Edge computing for telepresence places processing such as encoding, rendering and reconstruction close to users at the network edge to reduce latency and bandwidth for immersive remote communication.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-154-edge-computing-telepresence", "labels": [ "TELE-154-edge-computing-telepresence" ], "is_subclass_of": [ "Telepresence", "Edge Computing" ], "wikilinks": [ "Edge Computing", "Content Delivery Network", "Low Latency", "Cloud Computing", "Telepresence" ] }, { "id": "tele-157-predictive-tracking", "title": "TELE-157-predictive-tracking", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Predictive tracking is the estimation of a user's future head or body pose from recent motion so that rendering can be aligned to where the user will be, reducing perceived latency.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-157-predictive-tracking", "labels": [ "TELE-157-predictive-tracking" ], "is_subclass_of": [ "Telepresence" ], "wikilinks": [ "Kalman Filter", "Sensor Fusion", "Low Latency", "Inertial Measurement Unit", "Telepresence" ] }, { "id": "tele-252-dao-governance-telecollaboration", "title": "TELE-252-dao-governance-telecollaboration", "domain": "metaverse", "domain_name": "Metaverse", "definition": "DAO governance for telecollaboration is the use of decentralised autonomous organisation mechanisms, such as token-weighted or membership voting, to coordinate decisions and resource allocation within distributed collaborative communities.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tele-252-dao-governance-telecollaboration", "labels": [ "TELE-252-dao-governance-telecollaboration" ], "is_subclass_of": [ "Telecollaboration" ], "wikilinks": [ "DAO Governance", "DAO", "Remote Collaboration", "Collaboration", "Telecollaboration" ] }, { "id": "tele020virtualrealitytelepresence", "title": "TELE020virtualrealitytelepresence", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Virtual Reality Telepresence (TELE020) combines immersive head-mounted display technology with real-time communication infrastructure to create a compelling shared sense of presence among geographically distributed participants. Key technical components include stereoscopic rendering, head-related transfer function spatial audio, full-body motion tracking for avatar animation, and bandwidth-adaptive network transport; modern platforms extend to eye tracking, facial capture, haptic feedback, and mixed-reality blending of physical and virtual participants.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:tele020virtualrealitytelepresence", "labels": [ "TELE020virtualrealitytelepresence" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "term-index", "title": "TERM INDEX", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Term Index is a comprehensive reference catalogue mapping domain terminology, identifiers, and conceptual relationships within ontologies, enabling discovery and navigation of complex knowledge spaces across metaverse, blockchain, and robotics ecosystems. Well-structured indices provide hierarchical classification, cross-references, and semantic linking that facilitate knowledge retrieval, API design, and knowledge-graph integration, serving both user-facing navigation and machine-readable specifications supporting automated reasoning and SPARQL query optimisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:term-index", "labels": [ "TERM INDEX" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "KnowledgeGraphIntegration", "OWLOntology", "RoboticsOntology", "SPARQLQuery", "MetaverseDomain" ] }, { "id": "tf-idf", "title": "TF-IDF", "domain": "data", "domain_name": "Data", "definition": "Term frequency\u2013inverse document frequency, a classical term-weighting scheme that scores a term's importance to a document as the product of how often it occurs in that document and the logarithm of how rare it is across the collection, producing sparse vector representations that underpin lexical search ranking, document similarity, and feature extraction for text mining.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:tf-idf", "labels": [ "TF-IDF" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [ "Information Retrieval", "BM25", "Keyword Search", "Cosine Similarity" ] }, { "id": "thorchain", "title": "THORChain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decentralised cross-chain liquidity protocol that lets users swap native assets across different blockchains without wrapping them or relying on a single custodian. It uses bonded validators and liquidity pools denominated in its native asset.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:thorchain", "labels": [ "THORChain" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [ "Liquidity Pool", "Validator", "Atomic Swap", "Interoperability", "Decentralized Exchange", "Automated Market Maker", "Cross-Chain Bridge", "https://thorchain.org/" ] }, { "id": "tls-1-3", "title": "TLS 1.3", "domain": "security", "domain_name": "Security", "definition": "TLS 1.3 (Transport Layer Security version 1.3, standardised in RFC 8446, August 2018) is the current major version of the TLS protocol, redesigned to eliminate legacy cryptographic weaknesses, reduce handshake round-trips from two to one (zero for session resumption), mandate forward secrecy on every connection, and restrict the cipher suite to a small set of authenticated encryption algorithms. It replaces TLS 1.2 as the baseline secure transport for HTTPS, QUIC, and virtually all authenticated internet communications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:tls-1-3", "labels": [ "TLS 1.3" ], "is_subclass_of": [ "Transport Layer Security", "Encryption Protocol" ], "wikilinks": [] }, { "id": "tls-encryption", "title": "TLS Encryption", "domain": "security", "domain_name": "Security", "definition": "TLS (Transport Layer Security) encryption is the cryptographic protocol layer that provides confidentiality, integrity, and server authentication for communications over TCP/IP networks, most visibly as the foundation of HTTPS. It operates through a handshake that negotiates cipher suites, authenticates servers (and optionally clients) via X.509 certificates, establishes ephemeral session keys using asymmetric key exchange (ECDHE), and then encrypts all subsequent data using symmetric ciphers (AES-GCM). TLS 1.3 (RFC 8446, 2018) is the current standard, eliminating obsolete constructs and reducing handshake latency to one round trip.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tls-encryption", "labels": [ "TLS Encryption" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "tls", "title": "TLS", "domain": "security", "domain_name": "Security", "definition": "Transport Layer Security (TLS) is an IETF-standardised cryptographic protocol that provides authenticated, confidential, and integrity-protected communication channels over reliable transports such as TCP and QUIC. TLS 1.3 (RFC 8446, 2018) achieves a one-round-trip handshake using ephemeral Diffie-Hellman key exchange with mandatory forward secrecy, encrypting the server certificate within the handshake to prevent passive fingerprinting. It is the security foundation of HTTPS, gRPC, MQTT, SMTP-over-TLS, LDAPS, and virtually every application-layer protocol requiring channel security, with mutual TLS (mTLS) extending the model to bidirectional client and server authentication.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:tls", "labels": [ "TLS", "SSL/TLS", "TLS 1.2", "TLS Protocol", "TLS/SSL", "Transport Layer Security" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [] }, { "id": "tlv-encoding", "title": "TLV Encoding", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Type-Length-Value (TLV) encoding is a compact, extensible binary serialisation scheme in which each data field is represented by a type identifier, a length descriptor, and the value payload. Its self-describing structure allows parsers to skip unknown record types while preserving forward and backward compatibility, which is why protocols such as Lightning and Taproot Assets use it to embed optional metadata in transaction records.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tlv-encoding", "labels": [ "TLV Encoding" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "tnfd", "title": "TNFD", "domain": "governance", "domain_name": "Governance", "definition": "The Taskforce on Nature-related Financial Disclosures (TNFD) is a global framework that guides organisations in assessing, reporting, and acting on nature- and biodiversity-related risks, dependencies, and impacts. Modelled on the climate-focused TCFD, it provides recommended disclosures structured around governance, strategy, risk management, and metrics so that capital can be redirected toward nature-positive outcomes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tnfd", "labels": [ "TNFD", "TNFD Recommendations" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "toe-framework", "title": "TOE Framework", "domain": "governance", "domain_name": "Governance", "definition": "The Technology-Organization-Environment (TOE) framework is an organisational-level theory explaining how a firm's technological context, organisational characteristics, and external environment jointly shape the adoption and assimilation of technological innovations. It is widely applied in information-systems research to model the determinants of enterprise uptake of emerging technologies such as AI and cloud computing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:toe-framework", "labels": [ "TOE Framework" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "tpu", "title": "TPU", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Tensor Processing Unit \u2014 Google's custom ASIC optimised for the dense matrix multiplications that dominate neural network training and inference. TPUs use systolic arrays to achieve high throughput on 8-bit and 16-bit arithmetic at significantly lower energy per FLOP than general-purpose GPUs, and are available via Google Cloud as Cloud TPUs for large-scale model training.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tpu", "labels": [ "TPU", "Google TPU", "Google TPU v4" ], "is_subclass_of": [ "AI Infrastructure", "AI Hardware", "Artificial Intelligence" ], "wikilinks": [ "Google Cloud", "AI Hardware", "Artificial Intelligence", "ArtificialIntelligenceDomain" ] }, { "id": "tabular-iceberg", "title": "Tabular Iceberg", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tabular-iceberg", "labels": [ "Tabular Iceberg" ], "is_subclass_of": [ "Iceberg" ], "wikilinks": [] }, { "id": "tachocline", "title": "Tachocline", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tachocline", "labels": [ "Tachocline" ], "is_subclass_of": [ "Stellar Interior" ], "wikilinks": [] }, { "id": "tactile-sensing", "title": "Tactile Sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Tactile sensing is the ability of a robotic or prosthetic system to detect and interpret physical contact information \u2014 including contact force magnitude, direction, distribution, texture, slip, and temperature \u2014 through sensors embedded in or on the surface of an end effector or robotic skin. It is the mechanical analogue of the human sense of touch and provides information that visual sensing alone cannot supply, such as the internal stress distribution of a grasped object or the onset of slippage. Tactile sensing is fundamental to dexterous manipulation, safe human-robot interaction, and feedback-controlled assembly, and has become a defining capability differentiating next-generation robotic systems from conventional industrial manipulators.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:tactile-sensing", "labels": [ "Tactile Sensing" ], "is_subclass_of": [ "Perception System" ], "wikilinks": [] }, { "id": "tactile-sensor", "title": "Tactile Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A sensor embedded in a robot's end-effector or body surface that measures contact forces, pressure distributions, vibrations, and slip, enabling dexterous manipulation and safe physical interaction. Tactile sensing complements proprioception and vision to provide robots with fingertip-level awareness analogous to the human sense of touch.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tactile-sensor", "labels": [ "Tactile Sensor", "Optical Tactile Sensor" ], "is_subclass_of": [ "Perception and Sensing", "Exteroceptive Sensor" ], "wikilinks": [ "Exteroceptive Sensor", "Robotics" ] }, { "id": "tagged-hash", "title": "Tagged Hash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Tagged Hash is a domain-separated hash construction formalised in BIP-340 and used throughout Bitcoin's Taproot/Schnorr signature ecosystem, computed as SHA256(SHA256(tag) || SHA256(tag) || msg), where tag is a human-readable string identifying the protocol context. The double-hashing of the tag prefix creates a unique domain separator that prevents cross-protocol hash collisions \u2014 ensuring that a hash computed in one context (e.g. key tweaking) cannot be misinterpreted or replayed in another context (e.g. signature nonce generation). Tagged hashes improve security proofs and simplify protocol composition by making each domain's hash computationally distinct.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:tagged-hash", "labels": [ "Tagged Hash" ], "is_subclass_of": [ "Blockchain", "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "talent-concentration", "title": "Talent Concentration", "domain": "governance", "domain_name": "Governance", "definition": "Talent concentration is the clustering of scarce, highly skilled researchers and engineers within a small number of organisations, geographic hubs, or research groups. In frontier AI the pattern is pronounced: a few thousand specialists capable of training state-of-the-art models are concentrated in a handful of laboratories and metropolitan clusters, reinforced by compensation escalation, compute access, and network effects. The resulting asymmetry shapes competitive dynamics, national capability, and the diffusion of expertise across the wider economy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:talent-concentration", "labels": [ "Talent Concentration" ], "is_subclass_of": [ "AI Talent" ], "wikilinks": [ "AI Talent", "Competition in AI", "AI Talent War", "AI Investment" ] }, { "id": "tally", "title": "Tally", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A web platform and toolset for creating and managing on-chain decentralised autonomous organisations, providing interfaces for proposal creation, delegation and governance voting across widely used governance contract frameworks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tally", "labels": [ "Tally", "Tally Frontend", "Tally Governance", "Tally Governor Standard", "Tally On-chain Voting", "Tally Protocol" ], "is_subclass_of": [ "DAO Tooling" ], "wikilinks": [ "Decentralised Autonomous Organisation", "Governance Token", "Governance", "DAO", "DAO Tooling" ] }, { "id": "tamper-detection", "title": "Tamper Detection", "domain": "security", "domain_name": "Security", "definition": "Tamper detection is the set of technical mechanisms and protocols that identify whether a digital asset, data record, physical device, or communication has been unauthorisedly modified since its creation or last verified state. In the digital domain it employs cryptographic hash functions, digital signatures, Merkle proofs, and content-addressed storage to generate verifiable commitments that reveal any change to the protected content. Applied to media authenticity, hardware integrity, and data provenance, tamper detection is a foundational component of trust architectures for critical systems, supply chains, digital forensics, and content authenticity verification.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tamper-detection", "labels": [ "Tamper Detection", "Tamper-Evident Seal" ], "is_subclass_of": [ "Data Integrity" ], "wikilinks": [] }, { "id": "tamper-evidence", "title": "Tamper Evidence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Tamper evidence is the property of a system, record, or physical artefact whereby any unauthorised alteration leaves detectable traces that can be subsequently verified. It relies on mechanisms such as cryptographic hashing, digital signatures, and immutable logs so that the integrity of an item can be assessed after the fact. Tamper evidence does not necessarily prevent modification, but it guarantees that modification cannot occur undetected.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:tamper-evidence", "labels": [ "Tamper Evidence" ], "is_subclass_of": [ "Spatial Computing", "Immutability" ], "wikilinks": [] }, { "id": "tamper-resistance", "title": "Tamper Resistance", "domain": "security", "domain_name": "Security", "definition": "The property of a device or system that makes unauthorised physical access, modification, or extraction of its protected contents actively difficult, achieved through measures such as hardened enclosures, potted or shielded circuitry, mesh sensors that detect penetration, and logic that zeroises cryptographic keys when intrusion is sensed. Tamper resistance aims to prevent or frustrate an attack in progress, in contrast to tamper evidence, which merely ensures that interference leaves a detectable trace. It is a defining requirement for secure hardware including trusted platform modules, hardware security modules, smartcards, and payment terminals, and is graded by certification schemes such as FIPS 140-3 physical security levels and Common Criteria evaluations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:tamper-resistance", "labels": [ "Tamper Resistance" ], "is_subclass_of": [ "Hardware Security" ], "wikilinks": [ "Hardware Security", "Tamper Evidence", "Trusted Platform Module", "Hardware Security Module" ] }, { "id": "tamper-evident-storage", "title": "Tamper-Evident Storage", "domain": "security", "domain_name": "Security", "definition": "Tamper-evident storage is a data storage architecture designed so that any unauthorised modification of stored records leaves a detectable trace, typically achieved through cryptographic hash chaining, content addressing, or append-only write structures such as a Merkle DAG. It provides the storage-layer implementation of tamper evidence, ensuring that audit logs and other sensitive records can be verified for integrity after the fact rather than merely trusted. Tamper-evident storage is a foundational requirement for audit logs used in regulated or security-critical systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:tamper-evident-storage", "labels": [ "Tamper-Evident Storage" ], "is_subclass_of": [ "Tamper Evidence" ], "wikilinks": [] }, { "id": "taproot-assets", "title": "Taproot Assets", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taproot Assets (formerly Taro, renamed mid-2022) is a Bitcoin-native asset-issuance and transfer protocol developed by Lightning Labs, formally specified in BIPs 327-330, that enables arbitrary fungible tokens, non-fungible tokens, and stablecoins to be issued on the Bitcoin base layer using ...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:taproot-assets", "labels": [ "Taproot Assets" ], "is_subclass_of": [ "Protocol and Consensus", "Token and Asset", "Bitcoin Proof-of-Work Protocol", "Bitcoin Protocol", "Asset Issuance Protocol", "Client-Side Validation", "Lightning Network Extension", "Cryptographic Commitment Scheme" ], "wikilinks": [ "AI Agent Economies", "AI agents", "Amboss", "Asset Group Key", "Asset Issuance Protocol", "Asset Witness Proof", "Atomic Cross-Asset Swaps", "Atomic Swaps", "BIP-327", "BIP-328", "BIP-329", "BIP-330", "BIP-341", "BIP-342", "Bip39.io", "BitcoinBaseLayer", "Bitcoin-Native Stablecoins", "Bitcoin Network", "Bitcoin Protocol", "Bitfinex" ] }, { "id": "taproot", "title": "Taproot", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taproot is a soft-fork upgrade to the Bitcoin protocol, activated at block 709,632 in November 2021, comprising BIPs 340, 341, and 342. It introduces Schnorr signatures (BIP 340), Pay-to-Taproot (P2TR) outputs with Merkelised Abstract Syntax Tree (MAST) spending-condition commitments (BIP 341), and Tapscript \u2014 an updated Bitcoin Script dialect (BIP 342). Together these improvements enhance transaction privacy by making complex multi-condition spends indistinguishable from simple key-path spends, improve efficiency via Schnorr signature aggregation, and expand smart-contract expressiveness on Bitcoin's base layer.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:taproot", "labels": [ "Taproot", "Taproot Soft Fork", "Taproot Upgrade", "Taproot Witnesses" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "target-tracking", "title": "Target Tracking", "domain": "ai", "domain_name": "Ai", "definition": "Target tracking is a computational and signal-processing discipline concerned with estimating the state \u2014 typically position, velocity, and orientation \u2014 of one or more moving objects over time from sequences of noisy sensor observations. It encompasses algorithms such as Kalman filters, particle filters, and multi-hypothesis trackers, applied across radar, sonar, computer vision, and LiDAR modalities. Applications span aerospace surveillance, autonomous vehicle perception, robotic manipulation, sports analytics, and augmented reality. Modern deep-learning-based trackers jointly perform detection and tracking, achieving robust performance in complex, cluttered environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:target-tracking", "labels": [ "Target Tracking", "Multi-Target Tracking" ], "is_subclass_of": [ "Object Detection and Tracking" ], "wikilinks": [] }, { "id": "targeted-advertising", "title": "Targeted Advertising", "domain": "governance", "domain_name": "Governance", "definition": "Targeted advertising is the practice of delivering advertisements to specific individuals or audience segments selected on the basis of inferred characteristics, behaviour, or context. It relies on collecting and analysing personal data to predict which users are most likely to respond, and on automated systems that match adverts to audiences in real time. Because it depends on extensive data collection and profiling, targeted advertising sits at the centre of debates about privacy, consent, and the governance of digital platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:targeted-advertising", "labels": [ "Targeted Advertising" ], "is_subclass_of": [ "Advertising" ], "wikilinks": [] }, { "id": "task-adaptation", "title": "Task Adaptation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The process of specialising a pre-trained or general-purpose machine learning model to perform well on a specific downstream task by adjusting its parameters, architecture, or inference behaviour. Task adaptation encompasses techniques such as fine-tuning, instruction tuning, prompt engineering, and parameter-efficient methods (LoRA, adapters) that bridge the gap between a model's pre-training distribution and the requirements of a target application.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:task-adaptation", "labels": [ "Task Adaptation", "Rapid Task Adaptation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "task-allocation", "title": "Task Allocation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Task allocation is the process of assigning work items to agents in a multi-agent system so that collective objectives are met efficiently, taking into account agent capabilities, current load, task dependencies, deadlines, and communication costs. Approaches range from centralised optimisation and market-based auctions to fully distributed negotiation, and the choice of mechanism determines the system's scalability, robustness to agent failure, and responsiveness to changing conditions.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:task-allocation", "labels": [ "Task Allocation" ], "is_subclass_of": [ "Coordination Mechanisms" ], "wikilinks": [ "Coordination Mechanisms", "Multi-Agent Coordination", "Task Delegation", "Game Theory" ] }, { "id": "task-analysis", "title": "Task Analysis", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Task analysis is the systematic study of how users accomplish goals, decomposing activities into the sequence of actions, decisions, and cognitive operations required to complete them. It produces structured descriptions such as hierarchical task models and workflow maps that inform interface design, error prevention, and training. By making implicit work practices explicit, it grounds design decisions in observed user behaviour rather than assumption.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:task-analysis", "labels": [ "Task Analysis" ], "is_subclass_of": [ "Interaction Design" ], "wikilinks": [] }, { "id": "task-automation", "title": "Task Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "The automation of individual, often repetitive tasks so that they run without manual effort. It addresses discrete actions rather than coordinating entire business processes.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:task-automation", "labels": [ "Task Automation" ], "is_subclass_of": [ "Automation" ], "wikilinks": [ "Automation", "Business Process Automation", "Workflow Automation" ] }, { "id": "task-decomposition", "title": "Task Decomposition", "domain": "ai", "domain_name": "Ai", "definition": "Task decomposition is the process of breaking a complex goal into smaller, ordered sub-tasks that can be planned, delegated, and executed independently. In AI agent systems it lets a model or orchestrator turn an open-ended request into a tractable plan, often assigning sub-tasks to specialised agents or tools. Effective decomposition improves reliability, parallelism, and the ability to recover from partial failure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:task-decomposition", "labels": [ "Task Decomposition", "Problem Decomposition" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "task-delegation", "title": "Task Delegation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Task Delegation is the mechanism by which an agent assigns a sub-task to another agent or tool deemed more capable, available or specialised for that work. In multi-agent and agentic systems it underpins division of labour, allowing a coordinating agent to decompose a goal and route components to subordinate executors. Effective delegation requires shared task representation, capability awareness and result aggregation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:task-delegation", "labels": [ "Task Delegation" ], "is_subclass_of": [ "Multi-Agent System" ], "wikilinks": [] }, { "id": "task-execution", "title": "Task Execution", "domain": "robotics", "domain_name": "Robotics", "definition": "The process of carrying out a defined unit of work, including scheduling, resource allocation, and tracking of completion. In computing and robotics it covers running operations in response to plans or requests.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:task-execution", "labels": [ "Task Execution", "Robot Task Execution" ], "is_subclass_of": [ "Task Planning" ], "wikilinks": [ "Task Planning", "Automation", "Robotics" ] }, { "id": "task-planning", "title": "Task Planning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Task Planning is the computational sub-field of artificial intelligence and robotics concerned with automatically synthesising finite sequences of discrete actions \u2014 called plans \u2014 that transform a given initial world state into a state satisfying a specified goal condition, subject to action pre...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:task-planning", "labels": [ "Task Planning", "Robotic Task Planning" ], "is_subclass_of": [ "AI Technique", "Automated Planning", "Symbolic AI", "Artificial Intelligence", "Robotics", "Sequential Decision Making" ], "wikilinks": [ "Action Model", "Action Schema", "AlgorithmLayer", "AutomatedReasoningDomain", "Autonomous Vehicles", "Behaviour-Based Robotics", "Behaviour Tree Execution", "Behaviour Trees", "Chain-of-Thought Reasoning", "Constraint Satisfaction", "Diffusion Policy", "End-to-End Learning", "Feasibility Checker", "Formal Logic", "Game AI", "Goal-Oriented Action Planning", "Goal Specification", "Heuristic Function", "HTN Decomposition", "HTN Planning" ] }, { "id": "task-scheduling", "title": "Task Scheduling", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "The process of deciding which tasks run on which computational resources and when, subject to constraints such as priorities, deadlines, dependencies, and resource capacity. Task scheduling appears at every scale of computing: operating-system schedulers multiplex threads across CPU cores, real-time schedulers such as rate-monotonic and earliest-deadline-first guarantee that control loops meet hard deadlines, cluster orchestrators place jobs across machines, and workflow engines order dependent steps expressed as directed acyclic graphs. Because optimal scheduling is NP-hard in most general formulations, practical schedulers rely on priority-based heuristics, often implemented over priority queues, trading strict optimality for predictability, fairness, and low overhead.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:task-scheduling", "labels": [ "Task Scheduling" ], "is_subclass_of": [ "Resource Allocation" ], "wikilinks": [ "Resource Allocation", "Operating System", "Workflow Automation", "Priority Queue", "Real-Time Control" ] }, { "id": "task-space-control", "title": "Task Space Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Task space control, also called operational space control, regulates a robot's end-effector directly in Cartesian task coordinates rather than in joint space. Control laws are formulated in terms of end-effector position, orientation and force, with the manipulator Jacobian mapping task-space commands to joint actuation. This approach simplifies specification of interaction tasks such as following a path, applying a force or maintaining compliance against the environment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:task-space-control", "labels": [ "Task Space Control" ], "is_subclass_of": [ "Robot Control" ], "wikilinks": [] }, { "id": "task-specific-head", "title": "Task Specific Head", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Task-Specific Head is a shallow neural network module appended to a frozen or fine-tuned pre-trained model to adapt its representations for a particular downstream task. Architecturally it may be a single linear projection for classification, start/end span predictors for question answering, or a lightweight decoder for generation; it is randomly initialised and optimised during fine-tuning whilst the shared base model provides task-agnostic representations.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:task-specific-head", "labels": [ "Task Specific Head", "Mask Head" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Hardware and Edge", "MetaverseDomain", "software engineering" ] }, { "id": "task-specific-model", "title": "Task Specific Model", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Task Specific Model is a machine learning model trained or adapted to perform a single, narrowly defined task such as sentiment classification, named entity recognition or defect detection. Unlike general-purpose foundation models, it optimises parameters against the distribution of one objective, often yielding higher accuracy and lower inference cost for that task. Such models are typically produced by training from scratch on labelled data or by fine-tuning a larger base model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:task-specific-model", "labels": [ "Task Specific Model", "Task-Specific Model" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "task-and-motion-planning", "title": "Task and Motion Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Task and motion planning is an approach in robotics that combines high-level symbolic task planning with low-level geometric motion planning to produce fully executable robot plans. It jointly addresses what actions to perform and how to physically execute each movement, interleaving logical goal satisfaction with collision-free trajectory synthesis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:task-and-motion-planning", "labels": [ "Task and Motion Planning", "Task Planner" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [ "Planning", "Robot Control", "Pathfinding Algorithm", "Motion Planning", "https://en.wikipedia.org/wiki/Motion_planning", "https://arxiv.org/abs/2010.01083" ] }, { "id": "task-oriented-dialogue", "title": "Task-Oriented Dialogue", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Task-Oriented Dialogue is a conversational-AI paradigm in which a system engages in multi-turn interaction with a user to accomplish a specific, bounded goal, such as booking a flight or resetting a password, rather than conversing openly. It typically relies on intent classification to determine what the user wants and slot filling to extract the specific parameters needed to complete that intent. Its success is measured by task completion rate rather than by open-ended conversational quality.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:task-oriented-dialogue", "labels": [ "Task-Oriented Dialogue" ], "is_subclass_of": [ "Dialogue System" ], "wikilinks": [] }, { "id": "task-specific-dataset", "title": "Task-Specific Dataset", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A task-specific dataset is a curated collection of labelled or structured examples assembled to train or evaluate a machine learning model on a single, narrowly defined task, as distinct from the broad, general-purpose corpora used for pretraining. It typically supports fine-tuning techniques such as LoRA and DoRA, where a smaller, high-quality dataset adapts a pretrained model to a specific domain or behaviour. The quality and relevance of a task-specific dataset directly bound the ceiling of performance achievable through fine-tuning.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:task-specific-dataset", "labels": [ "Task-Specific Dataset" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "tax-treatment-crypto", "title": "Tax Treatment Crypto", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The application of domestic and international taxation principles, regulations, and compliance frameworks to transactions involving cryptocurrency and digital assets, encompassing capital gains taxation, income taxation, value-added tax treatment, and reporting obligations under frameworks such as HMRC guidance, IRS Notice 2014-21, and the OECD Crypto-Asset Reporting Framework.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:corporate-tax-compliance-framework-treatment-crypto", "labels": [ "Tax Treatment Crypto" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Compliance" ], "wikilinks": [ "26 USC Internal Revenue Code", "Council Directive 2006/112/EC VAT", "DeFi", "Decentralised Finance", "ECJ Skatteverket v Hedqvist C-264/14", "EU DAC8 Directive", "HMRC", "HMRC Cryptoassets Manual", "Infrastructure Investment and Jobs Act 2021", "Internal Revenue Service", "IRS Notice 2014-21", "IRS Revenue Ruling 2019-24", "OECD", "OECD Crypto-Asset Reporting Framework CARF", "TCGA 1992 UK Taxation of Chargeable Gains Act", "BlockchainDomain", "Cryptocurrency" ] }, { "id": "taxation", "title": "Taxation", "domain": "finance", "domain_name": "Finance", "definition": "The compulsory levying of financial charges by a state on individuals, businesses, and transactions to fund public expenditure and steer economic behaviour. Taxation spans income, corporation, consumption, capital gains, and property taxes, and its administration depends on financial reporting, valuation rules, and compliance infrastructure \u2014 questions made newly complex by digital assets, cross-border platforms, and decentralised finance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:taxation", "labels": [ "Taxation" ], "is_subclass_of": [ "Fiscal Policy" ], "wikilinks": [ "Fiscal Policy", "HM Treasury", "Financial Reporting", "Crypto Regulation", "Compliance" ] }, { "id": "taxonomic-framework", "title": "Taxonomic Framework", "domain": "robotics", "domain_name": "Robotics", "definition": "A Taxonomic Framework is a principled hierarchical classification system that organises concepts within a domain into superclass-subclass relationships, enabling consistent identification, comparison, and retrieval of entities. In the robotics ontology it structures robot types, actuators, sensors, and control strategies into a formal class hierarchy that supports OWL reasoning, SPARQL queries, and interoperability across ontologies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:corporate-tax-compliance-frameworkonomic-framework", "labels": [ "Taxonomic Framework", "TaxonomicFramework" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "owl:Thing" ] }, { "id": "taxonomy", "title": "Taxonomy", "domain": "data", "domain_name": "Data", "definition": "A taxonomy is a hierarchical classification scheme that organises concepts or entities into nested categories based on shared characteristics, typically expressing broader-than and narrower-than relationships. It provides a controlled vocabulary that supports consistent naming, navigation, and retrieval within a domain. Taxonomies are a simpler precursor to fuller ontologies, which add richer relations and formal axioms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:taxonomy", "labels": [ "Taxonomy", "Taxonomy Development" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "teacher-student-training", "title": "Teacher Student Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Teacher-Student Training is a machine learning paradigm in which a larger, higher-capacity teacher model supervises the training of a smaller student model, transferring knowledge through soft probability distributions (dark knowledge), intermediate feature representations, or attention maps rather than hard labels alone. The approach underpins knowledge distillation for model compression and is widely used to deploy efficient models on edge hardware without sacrificing task performance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:teacher-student-training", "labels": [ "Teacher Student Training", "Teacher Model" ], "is_subclass_of": [ "AI Technique", "Training Method" ], "wikilinks": [ "Computer Vision", "MetaverseDomain" ] }, { "id": "technical-architecture-framework", "title": "Technical Architecture Framework", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Technical Architecture Framework is a structured methodology or reference model that defines the components, interfaces, and integration patterns needed to design and evaluate complex technology systems \u2014 such as metaverse platforms, distributed applications, or spatial-computing stacks. Frameworks such as ETSI's metaverse domain model or IEEE P2874 provide vocabulary, layered decompositions, and conformance criteria that guide architectural decision-making and interoperability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technical-architecture-framework", "labels": [ "Technical Architecture Framework", "TechnicalArchitectureFramework" ], "is_subclass_of": [ "Platform and Environment", "Definitions and frameworks for Metaverse" ], "wikilinks": [ "owl:Thing" ] }, { "id": "technical-architecture", "title": "Technical Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical Architecture is the structured design of a system's components, their relationships, and the principles governing their evolution. It defines how hardware, software, data, and communication elements are organised to satisfy functional and non-functional requirements within a spatial computing or platform context.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:technical-architecture", "labels": [ "Technical Architecture" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "technical-committee", "title": "Technical Committee", "domain": "standards", "domain_name": "Standards", "definition": "A Technical Committee is a chartered working body within a standards organisation responsible for developing, reviewing and maintaining standards in a defined subject area. It convenes domain experts and stakeholders to draft specifications, resolve comments and reach consensus through structured procedures before publication. Technical committees are the engine of formal standardisation, balancing competing interests to produce interoperable, vendor-neutral specifications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:technical-committee", "labels": [ "Technical Committee" ], "is_subclass_of": [ "Standards Organization" ], "wikilinks": [] }, { "id": "technical-debt", "title": "Technical Debt", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Technical debt is the implied future cost incurred when a software team chooses an expedient solution over a better but slower approach, accruing rework that must eventually be paid down through refactoring. Like financial debt it carries interest: the longer suboptimal code persists, the more effort future changes require. Managing it involves making the debt visible, prioritising repayment, and balancing delivery speed against long-term maintainability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:technical-debt", "labels": [ "Technical Debt" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "technical-documentation", "title": "Technical Documentation", "domain": "data", "domain_name": "Data", "definition": "Technical documentation is the body of written material that explains how to use, operate, maintain, or build a product or system, spanning API references, user guides, architecture documents, runbooks, and specifications. Good technical documentation reduces onboarding time, encodes institutional knowledge, supports interoperability, and is increasingly treated as a versioned, tested artefact within the docs-as-code discipline. It is essential to the adoption and maintainability of software, hardware, and standards, and serves as a primary source for both human understanding and machine consumption.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:technical-documentation", "labels": [ "Technical Documentation", "Documentation Practices", "Technical Specification Document", "Technical Writing" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "technical-expertise", "title": "Technical Expertise", "domain": "governance", "domain_name": "Governance", "definition": "Technical expertise is the specialised knowledge, skill, and practical judgement required to design, build, or operate within a complex technical domain. It is a prerequisite and a gating factor for advanced work, and its scarcity can both enable progress and act as a barrier to misuse. The level and distribution of expertise materially affects how quickly capabilities such as AI development or biotechnology proliferate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:technical-expertise", "labels": [ "Technical Expertise" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "technical-history-extended-cv", "title": "Technical History (extended CV)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Technical History (extended CV) is the knowledge-graph representation of the author's professional research trajectory, spanning 15+ years leading immersive technology laboratories, securing multi-million-pound grant funding, and developing open-source metaverse and AI systems. It contextualises published research, institutional roles, and ongoing projects \u2014 including Knowhere, Flossverse, and FutureFleet \u2014 within the graph's broader topics of spatial computing, AI, and decentralised systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technical-history-extended-cv", "labels": [ "Technical History (extended CV)" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Agents", "Bitcoin", "Digital Objects", "flossverse", "Gemini", "Hyper personalisation", "Knowhere", "Location Based Experience", "Metaverse and Telecollaboration", "Mixed Reality", "National Industrial Centre for Virtual Environments", "Proprietary Large Language Models", "Stable Coins", "Trust and Safety", "Unreal Engine" ] }, { "id": "technical-robustness-and-safety", "title": "Technical Robustness and Safety", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Technical Robustness and Safety is a core AI trustworthiness dimension that requires AI systems to perform reliably under varied and adversarial conditions, implement fallback mechanisms for graceful degradation, and maintain operational safety throughout their lifecycle. It encompasses resilience to adversarial attacks, accurate uncertainty quantification, comprehensive risk assessment, and incident response protocols mandated by frameworks such as the EU AI Act Article 15 and NIST AI RMF.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:technical-robustness-and-safety", "labels": [ "Technical Robustness and Safety" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Trustworthiness Dimensions" ], "wikilinks": [ "EU AI Act Article 15", "EU HLEG AI", "NIST AI RMF", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "technical-standard", "title": "Technical Standard", "domain": "standards", "domain_name": "Standards", "definition": "A technical standard is a formal document that establishes uniform engineering or technical criteria, methods, processes, and practices to ensure that products, services, and systems are safe, reliable, interoperable, and consistently perform as intended. Standards are developed through consensus-based processes by authoritative bodies and provide the foundation for quality control, innovation, and global technology interoperability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:technical-standard", "labels": [ "Technical Standard", "Technical Specification" ], "is_subclass_of": [ "Legal and Regulatory", "Specification" ], "wikilinks": [ "Best Practices", "Core Technology", "Specification", "Technical Requirements", "Interoperability", "Quality Assurance", "Safety" ] }, { "id": "technical-standards", "title": "Technical Standards", "domain": "standards", "domain_name": "Standards", "definition": "Formal specifications, protocols, and guidelines that define how technologies, components, and systems should operate and interact, ensuring interoperability, quality, and compatibility across different platforms and implementations.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:technical-standards", "labels": [ "Technical Standards", "Binding Technical Standards" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "technique", "title": "Technique", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Technique is a specific, repeatable method or procedure applied within a domain to achieve a defined outcome, distinguishing it from broader algorithms or frameworks by its focus on concrete application steps. In artificial intelligence, techniques encompass approaches such as supervised learning, fine-tuning, and prompt engineering that are composed into pipelines and evaluated against benchmarks. Techniques may be realised by models or algorithms, and their effectiveness is assessed through empirical evaluation on datasets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:technique", "labels": [ "Technique", "Normalisation Technique" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "technological-leadership", "title": "Technological Leadership", "domain": "ai", "domain_name": "Ai", "definition": "Technological leadership is the position a company, nation, or research institution attains when it consistently sets the pace of innovation and capability in a given technology domain ahead of its rivals. It confers strategic, economic, and often military advantage by shaping standards, capturing talent, and directing the trajectory of adjacent industries. In artificial intelligence, it is contested through model capability benchmarks, compute access, talent acquisition, and patent output. Maintaining it requires sustained investment because leads erode quickly as rivals close capability gaps.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:technological-leadership", "labels": [ "Technological Leadership" ], "is_subclass_of": [ "Technology Race" ], "wikilinks": [] }, { "id": "technology-acceptance-model", "title": "Technology Acceptance Model", "domain": "governance", "domain_name": "Governance", "definition": "The Technology Acceptance Model (TAM) is an information-systems theory, introduced by Fred Davis, that predicts user adoption of a technology from two key beliefs: perceived usefulness and perceived ease of use. These beliefs shape attitudes and behavioural intention, which in turn drive actual usage. TAM is one of the most widely applied frameworks for explaining and forecasting the uptake of new technologies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:technology-acceptance-model", "labels": [ "Technology Acceptance Model" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "technology-adoption", "title": "Technology Adoption", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The process by which individuals, organizations, and societies integrate new technologies into their workflows, practices, and systems, encompassing awareness, etrial, implementation, and sustained use across AI, blockchain, metaverse, and robotics domains.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:technology-adoption", "labels": [ "Technology Adoption", "TechnologyAdoption" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [ "Change Management", "Competitive Advantage", "Ecosystem Maturity", "Evaluation", "Gartner Hype Cycle", "Implementation", "Innovation Diffusion", "ISO/IEC TR 24030", "McKinsey", "Rogers Diffusion of Innovations", "Stakeholder Buy-In", "Sustained Use", "Trial", "AI Governance", "AI-GroundedDomain", "Awareness", "BlockchainDomain", "ConceptualLayer", "Digital Transformation", "Infrastructure" ] }, { "id": "technology-diffusion", "title": "Technology Diffusion", "domain": "economics", "domain_name": "Economics", "definition": "The process and mechanisms by which new technologies are adopted, spread, and integrated into various sectors and populations over time.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:technology-diffusion", "labels": [ "Technology Diffusion" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "technology-disruption-dynamics", "title": "Technology Disruption Dynamics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Disruption describes the process by which emerging technologies, platforms, or economic models radically displace incumbent systems by offering superior capability, lower cost, or fundamentally new value propositions. In the context of AI, blockchain, and spatial computing, disruption manifests as wholesale replacement of legacy industries\u2014finance, knowledge work, supply chain, governance\u2014rather than incremental improvement.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-disruption-dynamics", "labels": [ "Technology Disruption Dynamics", "Disruption" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Blockchain", "Death of the Internet" ] }, { "id": "technology-ecosystem", "title": "Technology Ecosystem", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Technology Ecosystem is the interconnected network of platforms, developer tools, standards bodies, runtime environments, and community participants that collectively sustain a technology domain. In the metaverse context it encompasses rendering engines, spatial-computing SDKs, open standards (OpenXR, USD, WebXR), cloud infrastructure, hardware manufacturers, and application developers whose interdependencies determine adoption trajectories and competitive dynamics.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-ecosystem", "labels": [ "Technology Ecosystem" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "technology-infrastructure", "title": "Technology Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Technology Infrastructure encompasses the foundational hardware, networking, cloud services, and software platforms upon which higher-level applications and services are built. In the metaverse and spatial-computing domains it includes compute clusters, edge nodes, low-latency networking, content delivery networks, and the operating standards and APIs that enable interoperable, scalable, and resilient digital environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-infrastructure", "labels": [ "Technology Infrastructure", "Infrastructure Metrics" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "owl:Thing" ] }, { "id": "technology-power-structure-visualisation", "title": "Technology Power Structure Visualisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Calculating Empires is a large-scale research visualisation project by Kate Crawford and Vladan Joler that maps the co-evolution of technology and power structures from 1500 to the present. The work contextualises contemporary AI within five centuries of imperial infrastructure, surveillance, and control systems, foregrounding continuities between historical colonialism and digital power.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:technology-power-structure-visualisation", "labels": [ "Technology Power Structure Visualisation", "Calculating Empires" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Artificial Intelligence", "Cyber Security and Military", "Digital Society Surveillance", "Education and AI", "Politics, Law, Privacy" ] }, { "id": "technology-programme-delivery-planning", "title": "Technology Programme Delivery Planning", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Delivery Planning is the structured process of decomposing a technology programme into milestones, tasks, resource assignments, and timelines to ensure coherent execution. In the context of spatial computing and AI platform development, delivery planning coordinates parallel workstreams\u2014GenAI model development, platform infrastructure, digital asset creation, human tracking integration, and USD file format implementation\u2014typically represented as Gantt charts or sprint boards.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-programme-delivery-planning", "labels": [ "Technology Programme Delivery Planning", "Delivery Planning" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "AI Agent System", "Blockchain" ] }, { "id": "technology-race", "title": "Technology Race", "domain": "ai", "domain_name": "Ai", "definition": "A technology race is a competitive dynamic in which multiple actors, such as companies, nations, or research labs, accelerate investment and development in a given technology to secure strategic, economic, or military advantage ahead of rivals.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-race", "labels": [ "Technology Race" ], "is_subclass_of": [ "Competition in AI" ], "wikilinks": [] }, { "id": "technology-sector-landscape-analysis", "title": "Technology Sector Landscape Analysis", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Landscape, in the context of this knowledge graph, is a structured market-analysis overview of the current state of an AI or technology sector, mapping key players, technology stacks, investment flows, and emerging patterns. Landscape analyses such as those from a16z or Cowboy Ventures provide practitioners with orientation in rapidly shifting fields and inform infrastructure strategy.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:technology-sector-landscape-analysis", "labels": [ "Technology Sector Landscape Analysis", "Landscape" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Landscape" ] }, { "id": "technology-stack", "title": "Technology Stack", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The integrated set of software components, frameworks, libraries, runtime environments, and infrastructure services that together support the development and operation of a digital application or platform. In spatial computing contexts, a technology stack spans from hardware drivers and operating systems through XR runtimes, game engines, and application frameworks up to user-facing interfaces and cloud services.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:technology-stack", "labels": [ "Technology Stack", "Creative Technology Stack", "Technology Stack Integration", "TechnologyStack", "XR Technology Stack" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "technology-transfer", "title": "Technology Transfer", "domain": "ai", "domain_name": "Ai", "definition": "Technology transfer is the process of moving knowledge, skills, methods and inventions from the setting where they are created, such as a university or research laboratory, to organisations that can develop and commercialise them. It encompasses the legal, organisational and economic mechanisms, including licensing, spin-outs and collaborative research, by which research outputs become products, services and capabilities. Effective technology transfer is a central channel through which research and development drives economic and societal impact.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:technology-transfer", "labels": [ "Technology Transfer" ], "is_subclass_of": [ "Innovation" ], "wikilinks": [] }, { "id": "telecollaboration-and-telepresence", "title": "Telecollaboration and Telepresence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Telecollaboration and Telepresence are complementary technologies enabling geographically distributed participants to share a common virtual or augmented workspace with a sense of physical co-presence. They combine real-time audiovisual communication, spatial audio, avatar embodiment, and low-latency networking to support synchronous collaborative tasks across professional, educational, and social domains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:telecollaboration-and-telepresence", "labels": [ "Telecollaboration and Telepresence" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "1099 Generation", "2D Websites", "201-inch Virtual Display", "3D Asset Import", "3D Audio Rendering", "3D Models", "3D Object Manipulation", "3D Rendering", "4K", "5D Input", "5G", "5G mmWave", "5G Network Slicing", "5G Networks", "5G Remote Surgery", "5K", "6G", "6G Research", "6G Wireless", "70\u00b0 FOV" ] }, { "id": "telecollaboration", "title": "Telecollaboration", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The process of individuals or groups working toger towards shared goals across geographical distances through technology-mediated communication and coordination tools, integrating synchronous and asynchronous interaction modalities to achieve collaborative outcomes comparable to or exceeding co-...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:telecollaboration", "labels": [ "Telecollaboration", "TELE-002-telecollaboration", "Telecollaboration Stack" ], "is_subclass_of": [ "Communication Technology", "DistributedWork" ], "wikilinks": [ "3D Object Manipulation", "Communication Infrastructure", "CoordinationMechanisms", "Cross-Cultural Teamwork", "Distributed Knowledge Work", "Global Research Collaboration", "Real-Time Communication", "Real-Time Protocols", "Remote Team Productivity", "TELE-003-social-presence-theory", "TELE-005-common-ground-theory", "TELE-010-synchronous-collaboration", "TELE-011-asynchronous-collaboration", "TELE-026-microsoft-mesh", "TELE-027-spatial-platform", "TELE-028-horizon-workrooms", "TELE-100-ai-avatars", "TELE-105-real-time-language-translation", "TELE-107-ai-meeting-assistants", "TELE-150-webrtc" ] }, { "id": "telecommand", "title": "Telecommand", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:telecommand", "labels": [ "Telecommand" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "telecommunications-infrastructure", "title": "Telecommunications Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Telecommunications Infrastructure encompasses the physical and logical systems \u2014 including fibre-optic cable networks, cellular radio access networks, satellite constellations, submarine cables, internet exchange points, and backbone routers \u2014 that transport digital signals across local, regional, and global scales. It forms the foundational transport layer upon which internet protocols, mobile broadband, and converged communications services operate. As a critical national and commercial resource, it is subject to regulatory oversight, spectrum management, and increasingly stringent security requirements. Modern deployments integrate software-defined networking and network-function virtualisation to increase flexibility and reduce capital expenditure.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:telecommunications-infrastructure", "labels": [ "Telecommunications Infrastructure" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "telecommunications", "title": "Telecommunications", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Telecommunications is the sector and technical discipline concerned with the transmission of information \u2014 voice, data, and video \u2014 across distances through electronic, optical, or electromagnetic means, encompassing the physical infrastructure, protocols, standards, and regulatory frameworks that underpin global connectivity. It includes mobile and fixed-line networks, satellite systems, fibre optic backbones, and the software-defined and virtualised network functions that increasingly replace dedicated hardware.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:telecommunications", "labels": [ "Telecommunications" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "telemedicine", "title": "Telemedicine", "domain": "ai", "domain_name": "Ai", "definition": "Telemedicine is the delivery of clinical healthcare services \u2014 including diagnosis, consultation, monitoring, and treatment guidance \u2014 at a distance using telecommunications technology such as video conferencing, secure messaging, and remote sensing devices. It encompasses synchronous consultations (live video visits), asynchronous store-and-forward exchanges (images, test results), and remote patient monitoring (RPM) using wearable sensors and IoT devices. Telemedicine dramatically expands access to specialist care for patients in rural, underserved, or mobility-constrained situations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:telemedicine", "labels": [ "Telemedicine" ], "is_subclass_of": [ "Healthcare Technology" ], "wikilinks": [] }, { "id": "telemetry-and-analytics", "title": "Telemetry & Analytics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic collection and analysis of usage and performance data from metaverse applications and platforms to enable monitoring, optimization, and decision-making.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:telemetry-and-analytics", "labels": [ "Telemetry & Analytics" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Analytics Engine", "Capacity Planning", "Data Collection Pipeline", "Event Logging", "EWG/MSF Taxonomy", "Real-Time Data Streaming", "Statistical Analysis", "Usage Analytics", "User Behavior Analysis", "Data Layer", "Data Management", "Data Storage", "InfrastructureDomain", "Monitoring Dashboard", "Performance Metrics", "Performance Optimization", "Quality Assurance" ] }, { "id": "telemetry-tracking-and-command", "title": "Telemetry Tracking and Command", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:telemetry-tracking-and-command", "labels": [ "Telemetry Tracking and Command" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "telemetry", "title": "Telemetry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Telemetry is the automated collection, transmission and aggregation of measurements and signals from remote or distributed systems to a central point for monitoring and analysis. In software and infrastructure it commonly refers to the emission of metrics, logs, traces and events that describe a system's behaviour and health. Telemetry data is the raw substrate on which observability, alerting and performance analysis are built.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:telemetry", "labels": [ "Telemetry" ], "is_subclass_of": [ "Observability" ], "wikilinks": [] }, { "id": "teleoperated-robot", "title": "Teleoperated Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Teleoperated Robot is a robotic system remotely controlled by a human operator via a communication link, combining human-level judgement with robotic precision and reach in environments inaccessible or hazardous to people. Key technical dimensions include haptic feedback, anthropomorphic dexterity, immersive operator interfaces (including VR headsets), latency management, and AI-assisted control to reduce operator cognitive load; applications span surgery, nuclear decommissioning, disaster response, and construction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:teleoperated-robot", "labels": [ "Teleoperated Robot", "Teleoperated System" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "teleoperation-systems", "title": "Teleoperation Systems", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Teleoperation Systems are robotic control architectures enabling human operators to command and manipulate remote robots through master-slave or supervisory interfaces, transmitting operator intentions to robot actuators whilst providing sensory feedback (visual, auditory, haptic) from the robot to the operator, creating bidirectional human-machine coupling for remote physical interaction.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:teleoperation-systems", "labels": [ "Teleoperation Systems", "TELE-201-teleoperation-systems" ], "is_subclass_of": [ "Telepresence", "Robotic Telepresence" ], "wikilinks": [ "da Vinci", "TELE-200-robotic-telepresence", "TELE-203-haptic-feedback-telepresence", "Robotic Telepresence", "Robotics-Telepresence Bridge" ] }, { "id": "teleoperation", "title": "Teleoperation", "domain": "robotics", "domain_name": "Robotics", "definition": "Teleoperation is the remote control of robots and physical systems by human operators in real-time, enabling task execution in hazardous, distant, or inaccessible environments whilst maintaining human supervision and decision-making authority.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:teleoperation", "labels": [ "Teleoperation", "Bilateral Teleoperation", "Remote Teleoperation", "Robotic Teleoperation", "Teleoperation Interface" ], "is_subclass_of": [ "Human-Robot Interaction" ], "wikilinks": [ "5G", "AI", "BilateralControl", "controlsRobot", "Dashboard", "DeepSeaExploration", "dt:assistedBy", "dt:enhancedBy", "dt:monitoredOn", "dt:securedVia", "dt:uses", "enablesFeedback", "EncryptedChannel", "HapticInterface", "ImmersiveControl", "providesInterface", "RemoteControl", "RescueOperations", "SharedAutonomy", "SurgicalRobotics" ] }, { "id": "telepresence", "title": "Telepresence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Telepresence is a suite of technologies and interaction paradigms that create a compelling perceptual sense of physical presence in a remote or virtual location, enabling natural communication, social interaction, and physical manipulation across geographic distance. Systems span high-definition video conferencing with matched room geometries and spatial audio, robotic telepresence platforms that provide mobile embodied presence, and fully immersive extended-reality environments that deliver volumetric representations and haptic feedback. The defining characteristic is the degree of presence \u2014 the subjective feeling of 'being there' \u2014 which depends on sensory fidelity, low latency, spatial coherence, and natural interaction affordances. Convergent advances in spatial computing, network infrastructure, and AI-driven media processing are progressively closing the gap between telepresent and co-located experience.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:telepresence", "labels": [ "Telepresence", "5G Telepresence", "Immersive Telepresence", "Presence and Telepresence", "Telepresence (Distributed Collaboration)" ], "is_subclass_of": [ "Distributed Collaboration" ], "wikilinks": [] }, { "id": "telescope", "title": "Telescope", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:telescope", "labels": [ "Telescope" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "telethermometer", "title": "Telethermometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:telethermometer", "labels": [ "Telethermometer" ], "is_subclass_of": [ "Thermometer" ], "wikilinks": [] }, { "id": "temperature-sensor", "title": "Temperature Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A temperature sensor is a transducer that converts thermal energy into an electrical signal, enabling robotic and autonomous systems to monitor ambient, surface, or internal temperatures. Common types include thermocouples, RTDs, and infrared sensors, each suited to different accuracy, range, and response-time requirements in industrial, healthcare, and field robotics.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:temperature-sensor", "labels": [ "Temperature Sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "temporal-coverage", "title": "Temporal Coverage", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:temporal-coverage", "labels": [ "Temporal Coverage" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "temporal-difference-learning", "title": "Temporal Difference Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Temporal difference (TD) learning is a class of model-free reinforcement learning methods that estimate value functions by bootstrapping: each value estimate is updated towards a target composed of the immediate reward plus the discounted estimate of the successor state, rather than waiting for a full episode return. The TD error, the difference between the bootstrapped target and the current estimate, drives incremental updates and underpins algorithms such as TD(0), TD(lambda), SARSA, and Q-learning. By combining the sampling of Monte Carlo methods with the bootstrapping of dynamic programming, TD learning enables online, incremental learning from incomplete sequences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:temporal-difference-learning", "labels": [ "Temporal Difference Learning" ], "is_subclass_of": [ "Reinforcement Learning", "Reinforcement Learning Algorithm" ], "wikilinks": [] }, { "id": "temporal-entity", "title": "Temporal Entity", "domain": "data", "domain_name": "Data", "definition": "A temporal entity is an ontological category for anything that exists in or is defined by time, such as an instant, an interval, an event, or a process. In knowledge representation it serves as a shared superclass that anchors time-bearing concepts so they can be related by ordering, duration, and containment. The notion follows established time ontologies like the W3C OWL-Time vocabulary.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:temporal-entity", "labels": [ "Temporal Entity" ], "is_subclass_of": [ "Entity" ], "wikilinks": [] }, { "id": "temporal-logic", "title": "Temporal Logic", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Temporal logic is a formal system of logic that extends classical propositional or predicate logic with operators for reasoning about propositions whose truth changes over time. Operators such as 'eventually', 'always', 'next' and 'until' allow the specification of orderings and timing of events without explicit reference to clock values. It is the standard language for stating correctness properties of reactive and concurrent systems in formal verification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:temporal-logic", "labels": [ "Temporal Logic" ], "is_subclass_of": [ "Logic" ], "wikilinks": [] }, { "id": "temporal-motion-diffusion-adapter", "title": "Temporal Motion Diffusion Adapter", "domain": "ai", "domain_name": "Ai", "definition": "AnimateDiff is an open-source framework that inserts lightweight motion-module adapters into pre-trained text-to-image diffusion models to generate temporally consistent animated sequences without retraining the base image model. Developed by researchers at CUHK and ByteDance and released in 2023, it enables personalised diffusion model checkpoints to produce video clips by learning motion priors from video data in a plug-and-play manner. AnimateDiff integrates natively with the Stable Diffusion ecosystem, including ControlNet conditioning and LoRA fine-tuning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:temporal-motion-diffusion-adapter", "labels": [ "Temporal Motion Diffusion Adapter", "AnimateDiff Motion" ], "is_subclass_of": [ "Video Generation" ], "wikilinks": [] }, { "id": "temporal-reasoning", "title": "Temporal Reasoning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The branch of knowledge representation and reasoning concerned with representing time \u2014 instants, intervals, durations, and their qualitative and quantitative relations \u2014 and drawing inferences about the ordering, persistence, and change of events and states, underpinning planning, scheduling, narrative understanding, and any system that must reason about what held before, holds now, or will hold later.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:temporal-reasoning", "labels": [ "Temporal Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [ "Reasoning", "Temporal Logic", "Common Sense Reasoning", "Automated Planning" ] }, { "id": "temporal-resolution", "title": "Temporal Resolution", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:temporal-resolution", "labels": [ "Temporal Resolution" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "temporary-asset-access", "title": "Temporary Asset Access", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A mechanism enabling time-limited usage rights to digital assets in the metaverse without transferring ownership, implemented through smart contracts that separate user roles from owner roles with automatic expiration.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:temporary-asset-access", "labels": [ "Temporary Asset Access" ], "is_subclass_of": [ "Content and Assets", "Digital Asset Management" ], "wikilinks": [ "Digital Asset Management", "metaverse" ] }, { "id": "tendermint-consensus", "title": "Tendermint Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tendermint Consensus is a Byzantine Fault Tolerant consensus engine that combines Practical Byzantine Fault Tolerance (PBFT) with Proof-of-Stake validator sets to achieve immediate block finality and high transaction throughput. It underpins the Cosmos ecosystem and enables inter-blockchain communication by guaranteeing that committed blocks are never reverted, eliminating the probabilistic finality of Proof-of-Work chains.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:tendermint-consensus", "labels": [ "Tendermint Consensus", "Tendermint BFT" ], "is_subclass_of": [ "Protocol and Consensus", "Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Byzantine Fault Tolerance" ] }, { "id": "tendermint", "title": "tendermint", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tendermint is a Byzantine fault-tolerant (BFT) consensus engine that implements a partially synchronous, round-based, leader-elected protocol providing deterministic finality on each block. It separates the consensus layer from the application layer via the Application BlockChain Interface (ABCI), allowing any deterministic state machine to be driven by its consensus guarantees. Tendermint is the consensus core of the Cosmos SDK and underpins a large family of interoperable application-specific blockchains, offering safety guarantees so long as fewer than one-third of validator voting power is Byzantine. Its design derives from Practical Byzantine Fault Tolerance (PBFT) but introduces pragmatic modifications \u2014 rotating proposers, weighted voting, and a three-phase commit protocol \u2014 optimised for open, permissionless validator sets.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:tendermint", "labels": [ "Tendermint" ], "is_subclass_of": [ "Protocol and Consensus" ], "wikilinks": [] }, { "id": "tensiometer", "title": "Tensiometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tensiometer", "labels": [ "Tensiometer" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "tensor-computation", "title": "Tensor Computation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Tensor computation is the set of mathematical operations performed on multi-dimensional arrays (tensors) that form the core numerical substrate of modern machine learning. These operations include contraction, element-wise arithmetic, broadcasting, decomposition, and automatic differentiation, typically executed on specialised hardware such as GPUs or TPUs. Tensor computation frameworks provide graph-based or eager execution models that map operations to hardware acceleration primitives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:tensor-computation", "labels": [ "Tensor Computation" ], "is_subclass_of": [ "Tensor" ], "wikilinks": [] }, { "id": "tensor-core", "title": "Tensor Core", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Tensor Core is a specialised hardware execution unit within a GPU that performs small matrix multiply-accumulate operations in a single instruction, optimised for the dense linear algebra at the heart of deep learning. It operates on mixed-precision inputs, accumulating in higher precision while multiplying in reduced precision to maximise throughput. By accelerating matrix multiplication, Tensor Cores deliver large gains in training and inference performance over general-purpose GPU cores.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:tensor-core", "labels": [ "Tensor Core", "Tensor Cores" ], "is_subclass_of": [ "GPU Architecture" ], "wikilinks": [] }, { "id": "tensor-decomposition", "title": "Tensor Decomposition", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Tensor decomposition factorises a multi-dimensional array, or tensor, into a combination of simpler, typically lower-rank components, such as in CP or Tucker decomposition, that approximate the original tensor while using far fewer parameters. It is used to compress large weight tensors in neural networks, revealing latent structure and reducing memory and compute cost. It is a key technique underlying model quantisation and hardware-efficient inference pipelines.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:tensor-decomposition", "labels": [ "Tensor Decomposition" ], "is_subclass_of": [ "Tensor Computation" ], "wikilinks": [] }, { "id": "tensor-parallelism", "title": "Tensor Parallelism", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Tensor parallelism is a model-parallel training and inference technique that splits individual weight tensors, and the matrix multiplications that use them, across multiple accelerators so that each device holds and computes a shard of every layer. Partial results are combined with collective communication such as all-reduce. It lets models whose layers exceed a single device's memory run at scale, and is typically combined with data and pipeline parallelism.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tensor-parallelism", "labels": [ "Tensor Parallelism" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "tensor-processing-unit", "title": "Tensor Processing Unit", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Tensor Processing Unit (TPU) is Google's custom application-specific integrated circuit (ASIC) designed to accelerate machine learning workloads, particularly tensor operations in deep neural network training and inference. Built around a systolic array architecture for highly efficient matrix multiplication, TPUs prioritise throughput and power efficiency over general-purpose flexibility, and have been instrumental in training large-scale models such as BERT, PaLM, and Gemini.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:tensor-processing-unit", "labels": [ "Tensor Processing Unit" ], "is_subclass_of": [ "Inference Hardware" ], "wikilinks": [ "Inference Hardware", "Large Language Models" ] }, { "id": "tensor", "title": "Tensor", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A tensor is a multidimensional array of numerical values characterised by a rank, a shape and a data type, generalising scalars, vectors and matrices to arbitrary dimensions. In machine learning it is the fundamental data structure that holds inputs, parameters, activations and gradients as they flow through a model. Tensor operations such as contraction, broadcasting and elementwise functions are the computational primitives executed on accelerators during training and inference.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:tensor", "labels": [ "Tensor" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "tensor-flow-lite", "title": "TensorFlow Lite", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "TensorFlow Lite (now rebranded as LiteRT) is Google's open-source deep learning inference framework designed for on-device execution of machine learning models on mobile, embedded, and IoT edge hardware. It converts full TensorFlow or ONNX models into a FlatBuffers-based .tflite format and provides a lightweight interpreter that minimises memory footprint and latency without requiring network connectivity. The framework supports post-training optimisation techniques including quantisation, pruning, and weight clustering, and exposes hardware-acceleration delegates for GPUs, DSPs, NPUs, and the Coral Edge TPU. Its sister project TensorFlow Lite Micro (TFLite Micro) extends the same programming model to bare-metal microcontrollers with as little as 256 KB of RAM.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:tensor-flow-lite", "labels": [ "TensorFlow Lite" ], "is_subclass_of": [ "Machine Learning Framework" ], "wikilinks": [ "Machine Learning", "Machine Learning Framework" ] }, { "id": "tensor-flow", "title": "TensorFlow", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "An open-source machine learning framework developed by Google for building and deploying numerical computation and deep learning models across CPUs, GPUs and specialised accelerators, using a dataflow graph paradigm with automatic differentiation and a high-level Keras API.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:tensor-flow", "labels": [ "TensorFlow" ], "is_subclass_of": [ "Deep Learning Framework", "Machine Learning Discipline" ], "wikilinks": [ "Automatic Differentiation", "Hardware Acceleration", "Backpropagation", "PyTorch", "Machine Learning" ] }, { "id": "tensor-rt", "title": "TensorRT", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "TensorRT is NVIDIA's high-performance deep learning inference optimisation SDK that takes trained neural network models and compiles them into highly efficient inference engines tuned for specific NVIDIA GPU architectures. It performs a suite of graph-level and kernel-level optimisations including layer fusion, tensor fusion, kernel auto-tuning, precision calibration (FP32, FP16, INT8, FP8), and dynamic shape support to maximise throughput and minimise latency. Models from training frameworks are ingested primarily via the ONNX interchange format, then compiled offline into serialised engine files that are loaded at runtime. TensorRT underpins production AI inference across data-centre GPU clusters, autonomous vehicle compute stacks, and NVIDIA Jetson edge platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:tensor-rt", "labels": [ "TensorRT" ], "is_subclass_of": [ "Inference Engine" ], "wikilinks": [ "Hardware Acceleration", "Edge Inference", "Model Deployment", "Inference Engine" ] }, { "id": "term-definitions", "title": "Term Definitions", "domain": "data", "domain_name": "Data", "definition": "Term definitions are the authoritative, machine-readable descriptions that bind a label or identifier to its precise meaning within a controlled vocabulary, glossary, or ontology. Each definition fixes the intension of a term so that data producers and consumers interpret it consistently. They are the atomic content of glossaries and semantic registries and are essential for interoperability across systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:term-definitions", "labels": [ "Term Definitions" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "terminal-bench-2-0", "title": "Terminal Bench 2.0", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A standardized benchmark suite used to evaluate the agentic coding and terminal interaction capabilities of large language models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:terminal-bench-2-0", "labels": [ "Terminal Bench 2.0" ], "is_subclass_of": [ "GPT" ], "wikilinks": [] }, { "id": "terminal-coding-agents", "title": "Terminal Coding Agents", "domain": "ai", "domain_name": "Ai", "definition": "Terminal-native AI coding agents that operate through CLI interfaces with tool-call loops, providing autonomous software development capabilities via text-based interaction \u2014 includes opencode, Gemini CLI, Codex, crush, Open Interpreter, goose, and aider.", "entityType": "Class", "qualityScore": 0.91, "maturity": "established", "iri": "urn:ngm:class:terminal-coding-agents", "labels": [ "Terminal Coding Agents" ], "is_subclass_of": [ "Agent Harness", "Agentic AI", "Software Engineering Automation", "Autonomous Agent" ], "wikilinks": [ "Agent Harness", "IDE Coding Agents", "Harness Configuration Packs", "Agent Execution Sandboxes", "Agent Evaluation Benchmarks", "Multi-Agent Orchestration Frameworks", "Large Language Models", "Tool Use", "Function Calling", "Model Context Protocol", "Agent Loop", "Human-in-the-Loop", "Sandboxed Code Execution", "Agent Memory", "Context Window", "Software Engineering Automation", "Prompt Injection", "AI Safety", "Foundation Model", "Chain of Thought" ] }, { "id": "terminology-playbook", "title": "Terminology Playbook", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The canonical usage guide for knowledge-stack terminology across the DreamLab mesh: which artefact the words taxonomy, ontology, knowledge graph, reasoning, grounding and context graph each name, the one-line gloss each term carries at first use on a public surface, and the conflations the mesh's own copy must avoid. The source of truth is VisionFlow/docs/terminology.md; this page is its published mirror.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:terminology-playbook", "labels": [ "Terminology Playbook" ], "is_subclass_of": [ "Knowledge Graph Style Guide" ], "wikilinks": [ "Taxonomy", "Ontology", "Knowledge Graph", "Reasoning", "Context Graph", "Ontology Loom", "Knowledge Graph Style Guide" ] }, { "id": "terraform", "title": "Terraform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Terraform is an open-source infrastructure-as-code tool that lets engineers define cloud and on-premises resources in declarative configuration files and provision them through a consistent execution workflow. It maintains a state file that records the real-world mapping of declared resources, computes a plan describing the changes needed to reach the desired state, and applies those changes idempotently. Terraform supports many providers through a plugin model, enabling multi-cloud and hybrid provisioning from a single configuration language.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:terraform", "labels": [ "Terraform" ], "is_subclass_of": [ "Infrastructure as Code" ], "wikilinks": [] }, { "id": "terrain-aspect", "title": "Terrain Aspect", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrain-aspect", "labels": [ "Terrain Aspect" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "terrain-correction", "title": "Terrain Correction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrain-correction", "labels": [ "Terrain Correction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "terrain-slope", "title": "Terrain Slope", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrain-slope", "labels": [ "Terrain Slope" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "terrestrial-planet", "title": "Terrestrial Planet", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrestrial-planet", "labels": [ "Terrestrial Planet" ], "is_subclass_of": [ "Planet" ], "wikilinks": [] }, { "id": "terrestrial-reference-frame", "title": "Terrestrial Reference Frame", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrestrial-reference-frame", "labels": [ "Terrestrial Reference Frame" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "terrestrial-time", "title": "Terrestrial Time", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:terrestrial-time", "labels": [ "Terrestrial Time" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "tessellation-shader", "title": "Tessellation Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A tessellation shader is a programmable stage of the GPU graphics pipeline that subdivides coarse geometric patches into finer primitives at render time, controlling the level of detail of surfaces dynamically. It comprises a control phase that sets tessellation factors and an evaluation phase that positions the generated vertices. By adaptively refining meshes on the GPU, it produces smooth curved surfaces and displacement-mapped detail without inflating the source geometry.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:tessellation-shader", "labels": [ "Tessellation Shader" ], "is_subclass_of": [ "Shader Language" ], "wikilinks": [] }, { "id": "tessera", "title": "Tessera", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tessera is a private-transaction manager used with the Quorum enterprise Ethereum client to store and exchange encrypted payloads between authorised parties off the main chain. It enables confidential smart-contract state by keeping private data outside public blocks while anchoring only encrypted hashes on-chain. Tessera is a key component for enterprises that need permissioned, privacy-preserving distributed-ledger deployments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tessera", "labels": [ "Tessera", "Tessera Privacy Manager" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [] }, { "id": "test-automation", "title": "Test Automation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Test automation is the practice of using software tools to execute predefined tests against an application, compare actual outcomes with expected results and report defects without manual intervention. It accelerates feedback, improves repeatability and enables tests to run continuously as part of integration and delivery pipelines. Test automation spans unit, integration, regression, performance and end-to-end testing, and is central to maintaining quality in fast-moving software systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:test-automation", "labels": [ "Test Automation" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "test-data-management", "title": "Test Data Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Test data management is the discipline of provisioning, maintaining and controlling the datasets used in software testing, including generating synthetic data, masking or anonymising production data, and versioning fixtures so that tests are repeatable and do not leak sensitive information. It is a prerequisite for reliable automated testing processes, since flaky or stale test data is a common source of non-deterministic test failures. Mature test data management practices integrate with the software testing pipeline to refresh or reset datasets between test runs.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:test-data-management", "labels": [ "Test Data Management" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "test-dataset", "title": "Test Dataset", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A test dataset is a partition of data held out from model training and used exclusively to provide an unbiased estimate of a trained model's performance on unseen examples. Unlike the training and validation sets, it is touched only once final hyperparameters are fixed, preventing the leakage that would inflate reported accuracy. Its disjointness from training data is what makes it a credible measure of generalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:test-dataset", "labels": [ "Test Dataset" ], "is_subclass_of": [ "Machine Learning", "Dataset" ], "wikilinks": [] }, { "id": "test-driven-development", "title": "Test-Driven Development", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Test-driven development is a software practice in which developers write an automated test that defines a desired behaviour before writing the code to satisfy it, then refactor once the test passes. This red-green-refactor cycle drives design from the perspective of usage and ensures that every feature is covered by tests as it is built. The discipline tends to produce modular, well-specified code and a comprehensive regression suite.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:test-driven-development", "labels": [ "Test-Driven Development" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "testing-infrastructure", "title": "Testing Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The ensemble of tools, environments, frameworks, and automated pipelines used to validate software correctness, performance, and reliability across development and deployment stages. Testing infrastructure encompasses unit, integration, and end-to-end test runners, continuous integration services, containerised test environments, and monitoring dashboards that together ensure software meets quality standards before release.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:testing-infrastructure", "labels": [ "Testing Infrastructure" ], "is_subclass_of": [ "Software Engineering", "Quality Assurance" ], "wikilinks": [ "owl:Thing" ] }, { "id": "testing-process", "title": "Testing Process", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic execution of verification and validation operations to detect faults, verify functionality, and ensure quality standards in metaverse systems and applications.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:testing-process", "labels": [ "Testing Process" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Continuous Integration", "Defect Tracking System", "Development Workflow", "Integration Testing", "MSF Taxonomy 2025", "Performance Testing", "Quality Validation", "Security Testing", "Test Automation Framework", "Test Data Management", "Unit Testing", "User Acceptance Testing", "Compliance Verification", "Data Layer", "InfrastructureDomain", "Middleware Layer", "Performance Optimization", "Quality Assurance", "Risk Mitigation", "Telecollaboration" ] }, { "id": "testing", "title": "Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Testing is the software-engineering practice of verifying that a system behaves as specified and detecting defects by executing it against defined inputs and checking the outputs. It spans levels from unit and integration to system and acceptance testing, and styles from manual to fully automated. Systematic testing is a primary mechanism for establishing software reliability and guarding against regressions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:testing", "labels": [ "Testing", "Load Testing", "Testing Framework" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "testnet", "title": "Testnet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain testnet is a parallel blockchain network maintained by protocol developers and validators that mirrors the structure, consensus rules, and transaction semantics of a production mainnet whilst operating in a sandboxed environment with free or faucet-dispensed native tokens (typically 1 ...", "entityType": "Class", "qualityScore": 0.51, "maturity": "established", "iri": "urn:ngm:class:testnet", "labels": [ "Testnet" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain", "Blockchain Network", "Distributed System", "Testing Infrastructure", "Protocol Implementation", "Peer-to-Peer Network" ], "wikilinks": [ "Adler Quintyne-Collins 2019 Building Scalable Decentralized Payment Systems", "Al-Bassam Sonnino Buterin 2018 Fraud and Data Availability Proofs", "Arbitrum Sepolia", "Bahack 2013 Bitcoin PoW Theoretical Attacks", "Ben-Sasson et al 2014 zk-SNARKs Succinct", "Bitcoin Core", "Bitcoin Testnet3", "Block Explorer", "Blockchain-GroundedDomain", "Buchman Kwon Milosevic 2018 Tendermint BFT", "Bug Discovery", "Buterin 2014 Ethereum Yellow Paper", "Buterin & Griffith 2017 Casper FFG", "Buterin Hertzog Hsu 2019 Combining GHOST and Casper", "ConsensusLayer", "Consensus Liveness", "Consensus Research", "Consensus Rules", "Cross-Chain Integration", "Cross-Client Interoperability" ] }, { "id": "tether", "title": "Tether", "domain": "finance", "domain_name": "Finance", "definition": "Tether is the issuer of USDT, the largest fiat-collateralised stablecoin by circulating supply, designed to maintain a value pegged to the United States dollar. Tokens are issued on numerous blockchains and are intended to be redeemable one-to-one for dollars, backed by reserves held by the issuer. Tether is widely used for trading, settlement and as a dollar proxy on exchanges, and its reserve composition and transparency have been subjects of regulatory scrutiny.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tether", "labels": [ "Tether", "Tether Limited" ], "is_subclass_of": [ "Stablecoin", "Digital Asset Domain" ], "wikilinks": [ "Blockchain", "Reserve Backing", "Stablecoin", "Crypto Trading", "Decentralised Finance Domain", "Regulatory Domain", "Digital Asset Domain" ] }, { "id": "text-classification", "title": "Text Classification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Text classification is the supervised natural-language-processing task of assigning one or more predefined categorical labels to a span of text such as a document, sentence, or query. It maps variable-length text inputs to a fixed label space using learned representations and a decision function. Common formulations include binary, multi-class, and multi-label classification over topics, sentiment, intent, or content policy categories.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:text-classification", "labels": [ "Text Classification" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "text-embeddings", "title": "Text Embeddings", "domain": "ai", "domain_name": "Ai", "definition": "Text embeddings are dense numerical vector representations of words, sentences, or documents that place semantically similar text close together in a high-dimensional space. They are produced by neural models trained so that distance or cosine similarity in the vector space reflects meaning rather than surface form. Embeddings are the foundation of semantic search, clustering, and retrieval-augmented generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:text-embeddings", "labels": [ "Text Embeddings", "Text Embedding" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "text-encoder", "title": "text encoder", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A text encoder is a neural network component that maps raw text\u2014after tokenisation\u2014into dense, contextualised vector representations that capture semantic, syntactic, and relational information for use in downstream tasks. Transformer-based encoders such as BERT, RoBERTa, and ALBERT produce bidirectional contextual embeddings via masked language model pre-training, whilst CLIP's text tower produces contrastively aligned embeddings shared with a visual encoder. In generative image and video pipelines, the text encoder translates natural language prompts into conditioning vectors that guide the diffusion denoising process via cross-attention.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:text-encoder", "labels": [ "Text Encoder", "CLIP Text Encoder", "T5 Text Encoder" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [] }, { "id": "text-generation", "title": "Text Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Text Generation is the NLP task of producing coherent, contextually appropriate natural language text using neural language models, including applications such as story generation, article writing, code generation, and creative content production. Modern text generation employs transformer-based language models with autoregressive or sequence-to-sequence architectures, controllable generation techniques, and prompt engineering to produce human-quality text across diverse domains and styles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:text-generation", "labels": [ "Text Generation" ], "is_subclass_of": [ "Language Modeling" ], "wikilinks": [ "automation", "Large Language Model", "neural networks", "organisation", "Vercel", "Accessibility", "Agent Frameworks", "agents", "artificial intelligence", "ChatGPT", "Checkpoints", "ComfyUI", "Death of the Internet", "Deepfakes and fraudulent content", "Google", "GPT", "GPT Engineer", "Infrastructure", "Knowledge Graphing", "Language Modeling" ] }, { "id": "text-mining", "title": "Text Mining", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Text mining is the automated discovery of useful patterns, structure and knowledge from large collections of unstructured natural-language text. It combines natural-language processing, information retrieval and data-mining techniques to transform documents into structured representations amenable to analysis. Applications span extracting entities and relations, classifying and clustering documents, and surfacing trends across corpora too large to read manually.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:text-mining", "labels": [ "Text Mining" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "text-preprocessing", "title": "Text Preprocessing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Text preprocessing is the stage of a natural language processing pipeline that transforms raw text into a normalised, structured form suitable for tokenisation and modelling, encompassing steps such as lowercasing, punctuation and whitespace normalisation, removal of unwanted characters, and segmentation into sentences or tokens. It precedes and feeds directly into tokenisation, which converts the cleaned text into the discrete units a neural network consumes. Consistent text preprocessing reduces vocabulary sparsity and noise, materially affecting downstream model quality.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:text-preprocessing", "labels": [ "Text Preprocessing" ], "is_subclass_of": [ "Data Preprocessing" ], "wikilinks": [] }, { "id": "text-summarisation", "title": "Text Summarisation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Text summarisation is the NLP task of producing concise, coherent summaries that capture the essential information from longer documents or collections. Systems employ extractive methods (selecting key sentences) or abstractive methods (generating new text) using transformer architectures such as BART, PEGASUS, and T5, with applications spanning news aggregation, research synthesis, and document retrieval.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:text-summarisation", "labels": [ "Text Summarisation", "Abstractive Summarisation", "Summarisation", "Summarisation Models" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "chen2011leet", "cole2013call", "ComfyWorkFlows", "flux", "glas2013battlefields", "iser1993fictive", "serapis2008coming", "taylor2009play", "visionflow", "AI Video", "BART", "ComfyUI", "ControlNet and Similar Spatial Conditioning Systems", "Death of the Internet", "Education and AI", "Face Swap", "Flux.1", "KOHYA Dreambooth and similar", "Layoff tracker and threatened roles", "LoRA" ] }, { "id": "text-to-3-d", "title": "Text-to-3D", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Text-to-3D is a generative AI capability that synthesises three-dimensional geometry, texture, and material properties from natural-language descriptions, bridging the gap between linguistic intent and spatial representation. Dominant technical approaches include score-distillation sampling (SDS) that distils a 2D diffusion prior into a NeRF or 3D Gaussian Splatting field, multi-view diffusion models that jointly generate consistent images from multiple viewpoints before reconstructing a mesh, and image-conditioned 3D reconstruction pipelines. Text-to-3D has transformative applications in game asset creation, virtual production, digital twins, augmented-reality content authoring, and e-commerce visualisation, dramatically reducing the time and expertise required to populate 3D environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:text-to-3-d", "labels": [ "Text-to-3D" ], "is_subclass_of": [ "AI Application", "Generative AI" ], "wikilinks": [ "Artificial Intelligence", "Generative Ai" ] }, { "id": "text-to-image-benchmark", "title": "Text-to-Image Benchmark", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Standardized evaluation frameworks and leaderboards used to measure and compare the performance of generative image models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:text-to-image-benchmark", "labels": [ "Text-to-Image Benchmark" ], "is_subclass_of": [ "Text-to-Image" ], "wikilinks": [] }, { "id": "text-to-image-generation", "title": "Text-to-Image Generation", "domain": "ai", "domain_name": "Ai", "definition": "Text-to-image generation is a class of generative AI techniques that synthesise photorealistic or stylised images from natural-language textual descriptions, typically employing diffusion models, autoregressive transformers, or hybrid architectures trained on large paired datasets of images and captions. The synthesis process encodes a text prompt into a conditioning latent representation, then iteratively denoises random noise into structured visual output guided by that signal through cross-attention mechanisms. Leading systems such as DALL-E 3, Stable Diffusion XL, Midjourney, Imagen, and Flux exemplify the paradigm across proprietary and open-weight deployment modes. The field intersects creative tooling, computer vision, multimodal AI, and contested questions of copyright, consent, and synthetic media provenance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:text-to-image-generation", "labels": [ "Text-to-Image Generation", "Text to Image Generation", "Text-to-Image Synthesis" ], "is_subclass_of": [ "Image Generation" ], "wikilinks": [] }, { "id": "text-to-image", "title": "Text-to-Image", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A generative AI capability that synthesises visual imagery from natural language textual descriptions. Systems such as diffusion models iteratively denoise latent representations conditioned on text embeddings, enabling creation of photorealistic and artistic images from prompts without requiring explicit pixel-level instructions.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:text-to-image", "labels": [ "Text-to-Image", "Text To Image", "Text to Image" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [ "Artificial Intelligence", "Generative Ai" ] }, { "id": "text-to-speech", "title": "Text-to-Speech", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A speech-synthesis technology that converts written text into spoken audio output using neural vocoder models, enabling voice interfaces, accessibility tools, voice assistants, and real-time narration. Modern TTS systems leverage transformer-based architectures to produce natural, expressive speech with controllable prosody.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:text-to-speech", "labels": [ "Text-to-Speech", "Synthetic Speech", "Text-to-Speech Synthesis" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [ "Artificial Intelligence", "Natural Language Processing" ] }, { "id": "text-to-video-generation", "title": "Text-to-Video Generation", "domain": "ai", "domain_name": "Ai", "definition": "Text-to-video generation is a generative-AI task in which a model synthesises a coherent video clip directly from a natural-language description. It extends text-to-image diffusion and transformer methods with temporal modelling so that motion, object permanence, and scene consistency hold across frames. The field advanced rapidly with large latent-diffusion and spatiotemporal-transformer models capable of producing seconds of high-fidelity footage from a prompt.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:text-to-video-generation", "labels": [ "Text-to-Video Generation", "Text-to-Video" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "textual-inversion", "title": "Textual Inversion", "domain": "ai", "domain_name": "Ai", "definition": "Textual inversion is a fine-tuning technique for text-to-image diffusion models that learns a new embedding vector for a placeholder token from a handful of example images, capturing a specific subject or style without altering the model weights. The learned pseudo-word can then be composed into prompts like any ordinary token. It is lightweight and shareable because only a small embedding, not the full network, is trained.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:textual-inversion", "labels": [ "Textual Inversion" ], "is_subclass_of": [ "Parameter-Efficient Fine-Tuning" ], "wikilinks": [] }, { "id": "texture-atlas", "title": "Texture Atlas", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A texture atlas is a single large image that packs many smaller textures or sprites into one bitmap, accompanied by coordinate metadata mapping each sub-image to a region of the atlas. By consolidating textures, it lets a renderer draw many objects that share one bound texture, reducing state changes and draw calls and improving GPU efficiency in real-time graphics. UV coordinates of meshes are remapped to address sub-regions within the atlas. It is widely used in game engines, 2D sprite rendering, and font glyph caching to optimise throughput on the graphics pipeline.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:texture-atlas", "labels": [ "Texture Atlas" ], "is_subclass_of": [ "Real-Time Rendering" ], "wikilinks": [] }, { "id": "texture-compression", "title": "Texture Compression", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Texture compression is a set of techniques that store image textures in compact, GPU-decodable formats to reduce memory footprint and bandwidth during real-time rendering. Unlike general image compression, these formats support fast random access and fixed-rate block decoding directly in hardware, trading some image quality for large savings in storage and runtime cost. Common schemes include block-based formats such as BCn, ETC and ASTC.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:texture-compression", "labels": [ "Texture Compression" ], "is_subclass_of": [ "Spatial Computing", "Display and Rendering" ], "wikilinks": [] }, { "id": "texture-map", "title": "Texture Map", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Texture Map is a 2D image applied to the surface of a 3D model to add visual detail such as colour, roughness, normals, or emissive properties without increasing polygon count. Texture maps are indexed via UV coordinates that establish a correspondence between surface points and image pixels. They are a foundational component of real-time and offline rendering pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:texture-map", "labels": [ "Texture Map" ], "is_subclass_of": [ "Spatial Computing", "Uv Mapping" ], "wikilinks": [] }, { "id": "texture-mapping", "title": "Texture Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Texture mapping is the computer graphics technique of applying a 2D image (texture) to the surface of a 3D geometric model so as to simulate surface colour, roughness, reflectance, and fine structural detail without subdividing the underlying mesh. The process involves establishing a correspondence between 3D surface points and 2D texture coordinates (UV space), then sampling the texture through a pipeline that handles filtering, mipmapping, and perspective-correct interpolation. Modern pipelines extend the concept to multi-channel PBR texture sets (albedo, metalness, roughness, normal, ambient-occlusion, emissive) that together drive physically-based shading models, enabling photorealistic rendering in both real-time and offline contexts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:texture-mapping", "labels": [ "Texture Mapping" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [] }, { "id": "texture-sampling", "title": "Texture Sampling", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Texture sampling is the process of reading colour or data values from a texture image at arbitrary coordinates during rendering, typically involving interpolation between discrete texel values to produce a smooth result. Filtering strategies such as bilinear, trilinear, and anisotropic sampling trade computational cost against visual quality when textures are minified or viewed at oblique angles. Texture sampling is invoked heavily during post-processing passes and volume rendering, where many samples per pixel may be required.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:texture-sampling", "labels": [ "Texture Sampling" ], "is_subclass_of": [ "Texture Mapping" ], "wikilinks": [] }, { "id": "tezos", "title": "Tezos", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tezos is a self-amending, proof-of-stake Layer 1 blockchain platform launched in 2018 that enables on-chain governance through a formal, stakeholder-driven amendment process, allowing protocol upgrades to be proposed, debated, and applied without hard forks. It uses a Liquid Proof of Stake consensus mechanism in which token holders (XTZ holders) may either bake blocks themselves or delegate their stake to bakers, earning proportional rewards. The platform supports Turing-complete smart contracts written in Michelson and higher-level languages such as Ligo and SmartPy, targeting decentralised finance, NFT issuance, and digital asset custody. Its on-chain governance model is considered a pioneering example of decentralised protocol evolution and formal verification-friendly contract design.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tezos", "labels": [ "Tezos" ], "is_subclass_of": [ "Layer 1", "Layer 1 Blockchain" ], "wikilinks": [ "Liquid Proof of Stake", "Smart Contract", "Consensus Protocol", "Layer 1" ] }, { "id": "the-bitter-lesson", "title": "The Bitter Lesson", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A principle in artificial intelligence asserting that general methods leveraging computation and data, rather than human-designed domain knowledge, ultimately yield the most effective and scalable results.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:the-bitter-lesson", "labels": [ "The Bitter Lesson" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "the-sandbox", "title": "The Sandbox", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The Sandbox is a blockchain-based virtual world and game-creation platform where users build, own and monetise voxel-based experiences and assets recorded as non-fungible tokens on Ethereum, combining user-generated content authoring tools with a decentralised token economy centred on virtual land and in-world items.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:the-sandbox", "labels": [ "The Sandbox" ], "is_subclass_of": [ "Metaverse Platform" ], "wikilinks": [ "Ethereum", "Smart Contract", "Asset Tokenization", "Blockchain Gaming", "Digital Asset", "Metaverse Platform" ] }, { "id": "theorem-proving", "title": "Theorem Proving", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Theorem proving is the activity of establishing the truth of mathematical or logical statements by constructing rigorous, step-by-step deductions from axioms and inference rules. Automated theorem proving uses software to search for or verify such proofs, while interactive theorem proving combines machine checking with human guidance. It underpins formal verification of hardware, software and protocols, as well as the mechanisation of mathematics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:theorem-proving", "labels": [ "Theorem Proving" ], "is_subclass_of": [ "Automated Reasoning" ], "wikilinks": [] }, { "id": "theoretical-astrophysics", "title": "Theoretical Astrophysics", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:theoretical-astrophysics", "labels": [ "Theoretical Astrophysics" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "theoretical-computer-science", "title": "Theoretical Computer Science", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Theoretical computer science is the branch of computer science concerned with the mathematical foundations of computation, including computability, computational complexity, formal languages, and algorithm analysis.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:theoretical-computer-science", "labels": [ "Theoretical Computer Science" ], "is_subclass_of": [ "Computer Science" ], "wikilinks": [] }, { "id": "theory-of-mind", "title": "Theory of Mind", "domain": "ai", "domain_name": "Ai", "definition": "The cognitive capacity to attribute mental states \u2014 beliefs, desires, intentions, knowledge, and emotions \u2014 to oneself and to others, and to recognise that others' mental states may differ from one's own and from reality. Central to human social cognition and classically probed with false-belief tasks, theory of mind has become a benchmark capability for artificial agents, where machine analogues are pursued to support cooperation, communication, and safe interaction with people.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:theory-of-mind", "labels": [ "Theory of Mind" ], "is_subclass_of": [ "Cognitive Science" ], "wikilinks": [ "Cognitive Science", "Common Sense Reasoning", "Social Interaction", "Social Robotics" ] }, { "id": "therapeutic-vr", "title": "Therapeutic VR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The clinical application of virtual reality technology for treatment and rehabilitation purposes, including exposure therapy for phobias and PTSD, distraction-based pain management, cognitive behavioural therapy delivered in immersive digital environments, and motor rehabilitation. FDA regulatory authorisation and CMS billing codes established from 2021 onwards have formalised therapeutic VR as a reimbursable clinical modality.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:therapeutic-vr", "labels": [ "Therapeutic VR" ], "is_subclass_of": [ "Platform and Environment", "Virtual Reality Applications" ], "wikilinks": [ "metaverse", "Virtual Reality Applications" ] }, { "id": "thermal-balance", "title": "Thermal Balance", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermal-balance", "labels": [ "Thermal Balance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "thermal-infrared-remote-sensing", "title": "Thermal Infrared Remote Sensing", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermal-infrared-remote-sensing", "labels": [ "Thermal Infrared Remote Sensing" ], "is_subclass_of": [ "Remote Sensing" ], "wikilinks": [] }, { "id": "thermal-louver", "title": "Thermal Louver", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermal-louver", "labels": [ "Thermal Louver" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "thermal-management", "title": "Thermal Management", "domain": "robotics", "domain_name": "Robotics", "definition": "Thermal Management is the engineering discipline of controlling the temperature of electronic and electromechanical systems to keep components within safe operating limits. In robotics it addresses heat generated by motors, motor drivers and power electronics, using cooling systems, heat conduction paths and temperature sensing to dissipate or redistribute heat. Effective thermal management protects reliability, sustains performance under load and prevents thermal shutdown or damage.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:thermal-management", "labels": [ "Thermal Management", "ThermalManagement" ], "is_subclass_of": [ "Power Management" ], "wikilinks": [] }, { "id": "thermal-vacuum-testing", "title": "Thermal Vacuum Testing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermal-vacuum-testing", "labels": [ "Thermal Vacuum Testing" ], "is_subclass_of": [ "Spacecraft Environmental Testing" ], "wikilinks": [] }, { "id": "thermodynamics", "title": "Thermodynamics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Thermodynamics is the branch of physics concerned with heat, work, temperature and energy, and the laws governing their transformation and the direction of spontaneous processes. Its core principles - conservation of energy, the increase of entropy, and absolute-zero limits - underpin engineering of engines, cooling and energy systems. In computing infrastructure it constrains power consumption, heat dissipation and the efficiency of data-centre cooling.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:thermodynamics", "labels": [ "Thermodynamics" ], "is_subclass_of": [ "Infrastructure", "Computing and Cloud" ], "wikilinks": [] }, { "id": "thermometer", "title": "Thermometer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermometer", "labels": [ "Thermometer" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "thermosphere", "title": "Thermosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thermosphere", "labels": [ "Thermosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "thesaurus", "title": "Thesaurus", "domain": "data", "domain_name": "Data", "definition": "A thesaurus is a controlled vocabulary that organises preferred terms together with their synonyms and structured semantic relationships, principally broader, narrower and related-term links. By mapping non-preferred synonyms onto a single preferred descriptor, it enforces consistent indexing and improves retrieval by reconciling the many ways people express the same concept. Thesauri underpin information retrieval, library cataloguing and knowledge organisation, and are formalised by standards such as ISO 25964 and expressible in SKOS.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:thesaurus", "labels": [ "Thesaurus" ], "is_subclass_of": [ "Controlled Vocabulary" ], "wikilinks": [] }, { "id": "think-aloud-protocol", "title": "Think Aloud Protocol", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Think Aloud Protocol is a qualitative usability research method in which participants verbalise their thought processes continuously while performing tasks with a system or interface. Originating in cognitive psychology, it provides direct access to users' mental models, expectations, and confusion points. The method is widely used in human-computer interaction research and UX design to uncover usability issues that observation alone cannot reveal. Two variants exist: concurrent think-aloud (verbalising during the task) and retrospective think-aloud (recalling thoughts after completing it).", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:think-aloud-protocol", "labels": [ "Think Aloud Protocol", "Think-Aloud Protocol" ], "is_subclass_of": [ "Spatial Computing", "Human Computer Interface" ], "wikilinks": [] }, { "id": "third-party-auditing", "title": "Third Party Auditing", "domain": "governance", "domain_name": "Governance", "definition": "Third-party auditing is the independent examination of an organisation's systems, processes, controls, or claims by an external party that has no stake in the outcome, in order to provide credible assurance to stakeholders. By separating the auditor from the audited, it strengthens trust, accountability, and regulatory compliance beyond what self-assessment can offer. In technology and AI governance it covers security audits, conformity assessment, and verification of model or supply-chain claims.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:third-party-auditing", "labels": [ "Third Party Auditing", "Third-Party Auditing" ], "is_subclass_of": [ "Audit" ], "wikilinks": [] }, { "id": "third-party-auditor", "title": "Third Party Auditor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An independent security firm or organization that conducts comprehensive reviews of smart contracts, blockchain protocols, and digital systems to identify vulnerabilities, ensure code quality, and verify compliance before deployment.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:third-party-auditor", "labels": [ "Third Party Auditor", "Third-Party Auditor" ], "is_subclass_of": [ "Governance and Safety", "Security Services" ], "wikilinks": [ "metaverse", "Security Services" ] }, { "id": "third-party-certification", "title": "Third Party Certification", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Third-party certification is a conformity-assessment arrangement in which an independent accredited body, with no commercial interest in the outcome, evaluates a product, system or process against a defined standard and issues a certificate attesting to conformity. Its independence distinguishes it from first-party self-declaration and second-party customer assessment, lending the result greater credibility and market trust. In artificial intelligence it is increasingly proposed as a mechanism to assure that systems meet safety, fairness and governance requirements.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:third-party-certification", "labels": [ "Third Party Certification", "Third-Party Certification" ], "is_subclass_of": [ "Certification" ], "wikilinks": [] }, { "id": "third-party-verification", "title": "Third Party Verification", "domain": "finance", "domain_name": "Finance", "definition": "Third party verification is the independent assessment of a claim, asset, process, or record by an entity that has no stake in the outcome, providing impartial assurance that stated facts are accurate and conform to defined standards. By separating the verifier from the parties being assessed, it increases trust, reduces conflicts of interest, and supports accountability in markets, supply chains, and reporting. It is central to audit, certification, and attestation regimes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:third-party-verification", "labels": [ "Third Party Verification", "Third-Party Audit", "Third-Party Verification" ], "is_subclass_of": [ "VERIFICATION" ], "wikilinks": [] }, { "id": "third-party-assurance", "title": "Third-Party Assurance", "domain": "governance", "domain_name": "Governance", "definition": "Third-party assurance is the independent verification of an organisation's claims, disclosures, or controls by a party with no stake in the outcome, producing an assurance opinion that stakeholders can rely on more readily than self-reported data. It is commonly required for emissions reporting, carbon neutrality claims, and financial statements, where an accredited external auditor examines evidence against a recognised standard. It differs from internal audit in that independence from the reporting entity is the source of its credibility.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:third-party-assurance", "labels": [ "Third-Party Assurance" ], "is_subclass_of": [ "Audit" ], "wikilinks": [] }, { "id": "third-party-risk-management", "title": "Third-Party Risk Management", "domain": "governance", "domain_name": "Governance", "definition": "Third-party risk management (TPRM) is the discipline of identifying, assessing, and controlling the risks an organisation inherits from vendors, suppliers, and other external partners. It covers due diligence, contractual controls, ongoing monitoring, and offboarding across security, compliance, operational, and reputational dimensions. TPRM has become essential as organisations rely on extended ecosystems of cloud services and outsourced functions.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:third-party-risk-management", "labels": [ "Third-Party Risk Management" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "third-party-vulnerability", "title": "Third-Party Vulnerability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Third-Party Vulnerability is a security weakness originating in externally sourced software components, libraries, APIs, or services that are integrated into an AI system, creating risk vectors outside the direct control of the system developer or operator. Exploitation of these vulnerabilities can compromise model integrity, data confidentiality, or system availability.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:third-party-vulnerability", "labels": [ "Third-Party Vulnerability" ], "is_subclass_of": [ "AI Governance and Ethics", "Risk" ], "wikilinks": [ "Artificial Intelligence", "Risk" ] }, { "id": "thread-protocol", "title": "Thread Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Thread is an IPv6-based, low-power wireless mesh networking protocol designed for IoT devices in home and commercial environments, using IEEE 802.15.4 as its radio layer and providing self-healing, self-configuring mesh topology with native IP routing, secure device commissioning, and no single point of failure. Thread is managed by the Thread Group and forms the network and transport foundation for the Matter smart-home application protocol.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:thread-protocol", "labels": [ "Thread Protocol", "Thread" ], "is_subclass_of": [ "IEEE 802.15.4" ], "wikilinks": [] }, { "id": "threaded-messaging", "title": "Threaded Messaging", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Threaded Messaging is a structured asynchronous communication pattern in which replies are grouped beneath a parent message, forming discrete conversation threads that preserve context and reduce noise in shared channels. By isolating discussions, it enables parallel workstreams within a single collaboration platform while keeping notification volume manageable for participants. Platforms such as Slack and Discord popularised the pattern, which is now foundational to modern digital workplace communication.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:threaded-messaging", "labels": [ "Threaded Messaging" ], "is_subclass_of": [ "Communication Technology", "Asynchronous Collaboration" ], "wikilinks": [ "Collaboration Tools", "Discord", "Slack", "Asynchronous Collaboration", "TelecollaborationDomain" ] }, { "id": "threat-actor", "title": "Threat Actor", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Threat Actor is an individual, organised group, or nation-state entity that possesses the intent, capability, and opportunity to exploit vulnerabilities in digital systems. Actors are classified by motivation (financial, ideological, strategic, personal) and sophistication tier (opportunistic script-kiddies through state-sponsored APTs), with attribution performed via TTPs, infrastructure patterns, and targeting behaviour.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:threat-actor", "labels": [ "Threat Actor", "Threat Actor Profiling" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain" ], "wikilinks": [ "Attack Vector", "Blockchain", "Resilience", "Risk", "Security", "Vulnerability" ] }, { "id": "threat-detection", "title": "Threat Detection", "domain": "security", "domain_name": "Security", "definition": "Threat detection is the cybersecurity discipline concerned with identifying malicious or anomalous activity within a system or network in time to enable an effective defensive response, distinguishing genuine threats from benign anomalies across high-volume, noisy telemetry. It encompasses signature-based detection of known attack patterns, behavioural analytics for novel threats, and machine learning models that model normal system baselines and flag statistical deviations. Effective threat detection must balance sensitivity (catching real attacks) against specificity (avoiding alert fatigue from false positives).", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:threat-detection", "labels": [ "Threat Detection", "AI-Driven Threat Detection" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "threat-hunting", "title": "Threat Hunting", "domain": "security", "domain_name": "Security", "definition": "Threat hunting is a proactive cybersecurity practice in which analysts iteratively search through networks, endpoints, and data repositories for hidden adversarial activity that has evaded automated detection. Unlike reactive incident response, threat hunting is hypothesis-driven, combining threat intelligence, behavioural analytics, and expert intuition to identify indicators of compromise or attack techniques before they manifest as confirmed incidents. The discipline reduces dwell time and surfaces novel attacker tradecraft for which signatures do not yet exist.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:threat-hunting", "labels": [ "Threat Hunting" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "threat-intelligence-platform", "title": "Threat Intelligence Platform", "domain": "security", "domain_name": "Security", "definition": "A threat intelligence platform (TIP) is a security system that aggregates, normalises, and correlates indicators of compromise and adversary intelligence from multiple feeds into an actionable, queryable repository. It enriches and scores indicators, manages their lifecycle, and distributes them to detection and response tooling. A TIP turns raw threat data into context that defenders can use to anticipate, detect, and block attacks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:threat-intelligence-platform", "labels": [ "Threat Intelligence Platform" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "threat-intelligence", "title": "Threat Intelligence", "domain": "security", "domain_name": "Security", "definition": "Threat intelligence is the disciplined process of collecting, processing, analysing, and disseminating information about adversaries, their capabilities, intentions, and tactics in order to enable organisations to make informed defensive and strategic decisions. It transforms raw threat data \u2014 indicators of compromise, malware signatures, actor profiles, campaign timelines \u2014 into contextualised, actionable knowledge tailored to a specific audience. Intelligence is typically classified by time horizon and consumer: strategic intelligence informs executive risk decisions, operational intelligence guides incident response planning, and tactical intelligence feeds real-time detection and blocking systems. Sharing platforms such as STIX/TAXII and ISACs enable cross-organisational distribution of finished intelligence to raise collective defensive posture.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:threat-intelligence", "labels": [ "Threat Intelligence" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "Network Security", "Information Security", "Cybersecurity", "https://www.cisa.gov/topics/cyber-threats-and-advisories", "https://attack.mitre.org/" ] }, { "id": "threat-model", "title": "Threat Model", "domain": "security", "domain_name": "Security", "definition": "A threat model is a structured representation of the security properties, assumptions, and adversarial conditions relevant to a system, used to identify potential attack vectors, prioritise mitigations, and reason systematically about security guarantees. It defines who the adversary is (capabilities, motivations, access), what assets are worth protecting, and what attacks \u2014 such as those catalogued in STRIDE or MITRE ATT&CK \u2014 could compromise confidentiality, integrity, or availability. Threat modelling is applied during system design to surface architectural weaknesses before implementation, and updated continuously as the threat landscape evolves. It is a prerequisite for sound security architecture, cryptographic protocol design, and regulatory compliance.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:threat-model", "labels": [ "Threat Model" ], "is_subclass_of": [ "Security Framework" ], "wikilinks": [] }, { "id": "threat-modelling", "title": "Threat Modelling", "domain": "security", "domain_name": "Security", "definition": "Threat modelling is a structured security engineering process that identifies, enumerates, and prioritises potential threats to a system by reasoning systematically about adversaries, attack vectors, and mitigations before or during design. It produces an explicit model of what can go wrong, enabling security controls to be allocated proportionally to risk, and is applied across software, hardware, and AI systems throughout the development lifecycle.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:threat-modelling", "labels": [ "Threat Modelling", "Privacy Threat Modelling", "Threat Modeling" ], "is_subclass_of": [ "Risk Assessment" ], "wikilinks": [] }, { "id": "threat-surface-map", "title": "Threat Surface Map", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A comprehensive security visualization and inventory framework that identifies, catalogs, and models all potential attack vectors, vulnerability exposure points, and threat entry paths across network, application, data, identity, and infrastructure layers of a system.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:threat-surface-map", "labels": [ "Threat Surface Map" ], "is_subclass_of": [ "Governance and Safety", "Spatial Embodiment Harm Taxonomy" ], "wikilinks": [ "Asset Inventory", "Attack Vector Inventory", "Configuration Management Database", "Exposure Point Catalog", "Identity and Access Management", "Incident Response", "ISO 27001", "MITRE ATT&CK", "NIST Cybersecurity Framework", "OWASP", "Penetration Testing", "Penetration Testing Tools", "Risk Assessment Matrix", "Risk Management Framework", "Security Audit", "Security Control Mapping", "Security Information and Event Management", "Security Monitoring", "Security Posture Management", "Threat Intelligence Feed" ] }, { "id": "three-dimensional-graphics", "title": "Three Dimensional Graphics", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Three Dimensional Graphics encompasses the computational techniques for modelling, transforming, and rendering three-dimensional geometry onto two-dimensional display surfaces, encompassing rasterisation, ray tracing, shading, and scene graph management. It underpins game engines, virtual and augmented reality platforms, digital twins, and scientific visualisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:three-dimensional-graphics", "labels": [ "Three Dimensional Graphics" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "three-axis-stabilisation", "title": "Three-axis Stabilisation", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:three-axis-stabilisation", "labels": [ "Three-axis Stabilisation" ], "is_subclass_of": [ "Attitude Control" ], "wikilinks": [] }, { "id": "three-js", "title": "Three.js", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Three.js is an open-source JavaScript 3D graphics library that provides a high-level scene-graph abstraction over WebGL (and, increasingly, WebGPU), enabling interactive 3D experiences that run in any modern web browser without plugins. It supplies cameras, lights, materials, geometry primitives, model loaders, animation and post-processing systems, and first-class WebXR support, making it the most widely adopted foundation for browser-based 3D visualisation, product configurators and immersive web experiences.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:three-js", "labels": [ "Three.js" ], "is_subclass_of": [ "Graphics Library" ], "wikilinks": [ "Graphics Library", "WebGL", "Scene Graph", "Babylon Js", "WebXR" ] }, { "id": "threshold-cryptography", "title": "Threshold Cryptography", "domain": "security", "domain_name": "Security", "definition": "Threshold cryptography is a branch of cryptography in which a secret \u2014 such as a private key, decryption key, or signing capability \u2014 is distributed among a set of n parties such that any qualifying subset of at least t parties can jointly perform the cryptographic operation, while no coalition of fewer than t parties can do so alone. The t-of-n structure provides both redundancy and distributed access control: no single party holds complete key material, eliminating single points of compromise from theft, coercion, or insider malfeasance. The three primary primitives are threshold signatures, threshold decryption, and distributed key generation (DKG), with applications spanning cryptocurrency custody, distributed certificate authorities, multi-party computation protocols, and confidential smart contract execution. Modern protocols such as CGGMP21, GG20, and FROST have brought threshold operations to practical round counts suitable for production deployment.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:threshold-cryptography", "labels": [ "Threshold Cryptography" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "threshold-optimisation", "title": "Threshold Optimisation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Threshold optimisation is the process of selecting the decision boundary applied to a model's continuous scores so that discrete predictions best satisfy a chosen objective. By tuning where a probability or score is converted into a class label, practitioners trade off precision against recall, manage class imbalance, and satisfy fairness or cost constraints. It is a post-hoc technique that adjusts operating points without retraining the underlying model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:threshold-optimisation", "labels": [ "Threshold Optimisation" ], "is_subclass_of": [ "Model Evaluation", "Performance Metrics" ], "wikilinks": [] }, { "id": "threshold-selection", "title": "Threshold Selection", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Threshold selection is the process of choosing the decision boundary that converts a classifier's continuous scores or probabilities into discrete class labels. The chosen threshold trades off competing error types, moving along the trade-off captured by the ROC and precision-recall curves. Appropriate selection depends on the relative costs of false positives and false negatives and on class prevalence in the deployment setting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:threshold-selection", "labels": [ "Threshold Selection" ], "is_subclass_of": [ "Model Evaluation", "AI Technique" ], "wikilinks": [] }, { "id": "threshold-signature-scheme", "title": "Threshold Signature Scheme", "domain": "security", "domain_name": "Security", "definition": "A Threshold Signature Scheme (TSS) is a cryptographic protocol in which a private key is distributed among n parties such that any subset of at least t parties can jointly compute a valid digital signature without any single party ever possessing the complete private key, while fewer than t parties gain zero information about the key. TSS extends threshold secret sharing (Shamir's Secret Sharing) to the signing operation itself, producing a signature that is indistinguishable on-chain from a standard single-key signature, thereby enhancing privacy and reducing transaction fees compared to traditional on-chain multisig. TSS underpins distributed key management for exchanges, MPC wallets, and cross-chain bridge custody architectures.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:threshold-signature-scheme", "labels": [ "Threshold Signature Scheme", "FROST Threshold Signatures", "Threshold Signature", "Threshold Signatures" ], "is_subclass_of": [ "Signature Scheme" ], "wikilinks": [] }, { "id": "throughput-optimisation", "title": "Throughput Optimisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The systems discipline of maximising useful work completed per unit time \u2014 requests, tokens, or samples per second \u2014 on fixed hardware, typically by batching to raise arithmetic intensity, keeping accelerators saturated through scheduling and overlap, and managing memory so that capacity rather than stalls bounds concurrency, usually traded off explicitly against per-request latency.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:throughput-optimisation", "labels": [ "Throughput Optimisation" ], "is_subclass_of": [ "Inference Optimisation" ], "wikilinks": [ "Inference Optimisation", "Batch Inference", "Continuous Batching", "KV Cache" ] }, { "id": "throughput", "title": "Throughput", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Throughput is the rate at which a system completes useful work over a unit of time, such as requests served per second, tokens generated per second or bytes transferred per second. It measures sustained productive capacity rather than the time to complete a single operation. In machine learning serving it captures how many inferences or training samples a system can process under load. Maximising throughput typically involves batching, parallelism and resource utilisation, often trading off against per-request latency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:throughput", "labels": [ "Throughput" ], "is_subclass_of": [ "Performance Metrics", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "thrust", "title": "Thrust", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thrust", "labels": [ "Thrust" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "thruster", "title": "Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:thruster", "labels": [ "Thruster" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "tidal-heating", "title": "Tidal Heating", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tidal-heating", "labels": [ "Tidal Heating" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "tide-gauge", "title": "Tide Gauge", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tide-gauge", "labels": [ "Tide Gauge" ], "is_subclass_of": [ "Measurement Instrument" ], "wikilinks": [] }, { "id": "tim-reutermann-decentralised-governance-thinker", "title": "Tim Reutermann Decentralised Governance Thinker", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Tim Reutermann is a writer and technologist whose work synthesises cypherpunk philosophy, decentralised governance models, and AI-driven economic paradigm shifts. He advocates liquid democracy, universal basic income, and open-source systems as mechanisms for aligning economic incentives with social goods in the face of automation-driven disruption.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:tim-reutermann-decentralised-governance-thinker", "labels": [ "Tim Reutermann Decentralised Governance Thinker", "Tim Reutermann" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "time-dilution-of-precision", "title": "Time Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:time-dilution-of-precision", "labels": [ "Time Dilution of Precision" ], "is_subclass_of": [ "Dilution of Precision" ], "wikilinks": [] }, { "id": "time-lock", "title": "Time Lock", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A time lock is a cryptographic and protocol-level constraint that prevents the spending or execution of funds, transactions, or governance actions until a specified time or block height has been reached. Implemented on blockchains through absolute and relative locktime fields and dedicated script opcodes, time locks enforce delayed settlement, contestation windows, and staged execution without trusting a third party. They are foundational to payment channels, atomic swaps, vesting schedules, and the safety delays in DAO governance, where they give participants time to react before irreversible actions occur.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:time-lock", "labels": [ "Time Lock", "Time-Lock" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "time-of-flight", "title": "Time Of Flight", "domain": "robotics", "domain_name": "Robotics", "definition": "Time of Flight (ToF) is a sensing technique that determines the distance to an object by measuring the elapsed time between the emission of a signal \u2014 typically light, sound, or radio waves \u2014 and the detection of its reflection. In depth sensing, ToF cameras emit pulsed or modulated infrared light and record the per-pixel round-trip delay to construct a dense depth map. ToF sensors are widely deployed in robotics, autonomous vehicles, augmented reality, and gesture recognition due to their ability to produce real-time depth data without ambient light dependency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:time-of-flight", "labels": [ "Time Of Flight", "Time of Flight", "Time-of-Flight" ], "is_subclass_of": [ "Robotics", "Proximity Sensor" ], "wikilinks": [] }, { "id": "time-series-data", "title": "Time Series Data", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Time series data is a sequence of observations indexed in time order, typically recorded at regular or irregular intervals from sensors, systems or markets. Its temporal structure exposes trends, seasonality and autocorrelation that distinguish it from cross-sectional data and demand specialised storage, querying and analysis. Time series underpin monitoring, forecasting and anomaly detection across many domains.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:time-series-data", "labels": [ "Time Series Data", "Time-Series Data" ], "is_subclass_of": [ "Data" ], "wikilinks": [] }, { "id": "time-series-forecasting", "title": "Time Series Forecasting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Time orecasting (TSF) is the domain of statistical and machine learning mods that predict future values of ordered sequential observations indexed by time, encompassing classical statistical approaches (ARIMA \u2014 AutoRegressive Integrated Moving Average \u2014 modelling the s a linear combination of its...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:time-series-forecasting", "labels": [ "Time Series Forecasting", "Time-Series Forecasting" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline", "Statistical Modelling", "Predictive Analytics", "Sequential Data Processing", "Probabilistic Forecasting" ], "wikilinks": [ "AlgorithmLayer", "Anomaly Detection", "ARIMA", "Autocorrelation Analysis", "Automated Reporting", "Bayesian Inference", "Capacity Planning", "Causal Inference", "Classification", "Climate Modelling", "Covariates", "Cross-Validation Strategy", "CRPS Evaluation Standard", "Decision Support", "Demand Planning", "Econometrics", "Energy Forecasting", "Epidemiological Modelling", "Evaluation Metric", "Exponential Smoothing" ] }, { "id": "time-series", "title": "Time Series", "domain": "data", "domain_name": "Data", "definition": "A time series is a sequence of data points indexed in chronological order, typically sampled at consistent intervals, that captures how a measured quantity changes over time. Time series data underpins forecasting, trend analysis, and anomaly detection, and requires specialised handling for properties such as seasonality, trend, and autocorrelation that are absent from unordered datasets. Derived statistics such as moving averages are computed directly over a time series to smooth noise and reveal underlying patterns.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:time-series", "labels": [ "Time Series" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [] }, { "id": "time-synchronisation", "title": "time synchronisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Time synchronisation is the process of coordinating the clocks of networked devices to a shared reference time so that distributed computations, transactions, events, and audit logs share a consistent and ordered temporal frame. It is realised through layered protocols including the Network Time Protocol (NTP / RFC 5905), which delivers millisecond-level accuracy over wide-area networks via a stratum hierarchy anchored to atomic or GPS reference clocks, and the Precision Time Protocol (PTP / IEEE 1588-2019), which exploits hardware timestamping in network interface cards and PTP-aware switches to achieve sub-microsecond accuracy on local and carrier-grade networks. Accurate time synchronisation is foundational to distributed consensus algorithms, cryptographic certificate validation, financial transaction sequencing, telecommunications frequency synchronisation, and industrial real-time control; conversely, clock skew and drift are root causes of ordering anomalies, replay attacks, split-brain conditions, and regulatory non-compliance.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:time-synchronisation", "labels": [ "Time Synchronisation", "Time Synchronisation Protocol", "Timing Synchronisation" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "time-sensitive-networking", "title": "Time-Sensitive Networking", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Time-Sensitive Networking (TSN) is a set of IEEE 802.1 standards that add deterministic, bounded-latency delivery to standard Ethernet, enabling time-critical and best-effort traffic to share the same network. TSN provides precise time synchronisation, traffic scheduling, frame preemption, and reservation mechanisms so that control-loop and audio-video data arrive within guaranteed time windows. It is foundational to industrial automation, automotive in-vehicle networks, and professional media, replacing proprietary fieldbuses with converged standard Ethernet.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:time-sensitive-networking", "labels": [ "Time-Sensitive Networking" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "time-series-analysis", "title": "Time-Series Analysis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Time-series analysis is the statistical and computational study of data sequences indexed by time, encompassing methods for decomposition, modelling, forecasting, and anomaly detection of temporal patterns including trend, seasonality, and autocorrelation structures. Foundational techniques include ARIMA, exponential smoothing, spectral analysis, and state-space models, with deep learning approaches such as transformers and N-BEATS increasingly dominant in practice.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:time-series-analysis", "labels": [ "Time-Series Analysis", "Time Series Analysis" ], "is_subclass_of": [ "Data Analysis" ], "wikilinks": [] }, { "id": "time-series-database", "title": "Time-Series Database", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A time-series database (TSDB) is a database system optimised for storing, querying, and analysing data points indexed by time. It is designed for workloads dominated by high-volume, append-only writes of timestamped measurements and by queries that aggregate over time ranges. TSDBs employ time-aware partitioning, columnar layouts, and specialised compression to handle the scale and access patterns of metrics, events, and sensor readings efficiently.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:time-series-database", "labels": [ "Time-Series Database", "Time Series Database", "Time-Series Databases" ], "is_subclass_of": [ "Database Management System" ], "wikilinks": [] }, { "id": "time-domain-astronomy", "title": "Time-domain Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:time-domain-astronomy", "labels": [ "Time-domain Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "time-of-flight-sensor", "title": "Time-of-Flight Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A time-of-flight (ToF) sensor measures distance by emitting a light signal, typically infrared, and timing how long it takes to reflect back from a surface, yielding a per-pixel depth map. Because it directly measures travel time, it produces depth in real time without the baseline geometry needed by stereo cameras. ToF sensors are widely used for hand and gesture tracking, scene reconstruction, and mixed-reality depth sensing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:time-of-flight-sensor", "labels": [ "Time-of-Flight Sensor", "Time-of-Flight Camera", "Time-of-Flight Measurement", "Time-of-Flight Sensing" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "timelock-contract", "title": "Timelock Contract", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A smart contract or transaction condition that prevents funds or actions from being executed until a specified time or block height is reached. It is used to enforce delays and to coordinate conditional payments.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:timelock-contract", "labels": [ "Timelock Contract" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [ "Smart Contract", "Hash Time-Locked Contract", "Ethereum" ] }, { "id": "timelock-controller", "title": "Timelock Controller", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A timelock controller is a smart contract governance primitive that enforces a mandatory waiting period \u2014 the timelock delay \u2014 between the scheduling of an on-chain operation (such as a protocol upgrade, parameter change, or treasury disbursement) and its execution, giving token holders, security researchers, and affected parties an opportunity to review, object to, or exit before the change takes effect. Operations must be queued with their full parameters, remain in the queue for the configured delay, and then be explicitly executed; they may also be cancelled by authorised roles during the waiting period. Timelock controllers are a fundamental safety mechanism in DeFi and DAO governance, protecting against malicious or erroneous governance proposals.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:timelock-controller", "labels": [ "Timelock Controller", "TimelockController" ], "is_subclass_of": [ "On-chain Governance" ], "wikilinks": [] }, { "id": "timelock", "title": "Timelock", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A timelock is a cryptographic or smart-contract mechanism that prevents a transaction, function call, or asset transfer from executing until a specified block height or Unix timestamp has been reached. Timelocks enforce temporal constraints on blockchain operations, separating the proposal of an action from its execution to allow inspection, challenge, or cancellation during a mandatory delay window.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:timelock", "labels": [ "Timelock", "Timelock Delay", "Timelock Module" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "timer", "title": "Timer", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A timer is a software or hardware mechanism that measures elapsed time or schedules an action to occur after a delay or at a recurring interval. In collaborative and real-time applications it provides shared countdowns, time-boxing, and synchronised triggers that all participants observe consistently. Timers are a basic building block of event scheduling, animation, and time-bounded interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:timer", "labels": [ "Timer" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "timestamp-authority", "title": "Timestamp Authority", "domain": "security", "domain_name": "Security", "definition": "A Timestamp Authority (TSA) is a trusted third party that issues cryptographically signed timestamps attesting that a particular piece of data existed at or before a specified point in time, in accordance with the RFC 3161 Internet X.509 Public Key Infrastructure Time-Stamp Protocol. The TSA receives a hash of the document or data, signs it together with the current time using its private key, and returns a TimeStampToken that can be independently verified by any party holding the TSA's public key certificate. Timestamp tokens are widely used in digital signature workflows to prove long-term validity\u2014establishing that signatures were made before certificate revocation or expiry. Under eIDAS regulation in Europe, qualified TSAs form part of trust service infrastructure with legal standing equivalent to notarisation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:timestamp-authority", "labels": [ "Timestamp Authority" ], "is_subclass_of": [ "Certificate Authority" ], "wikilinks": [] }, { "id": "timestamp-service", "title": "Timestamp Service", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A timestamp service issues verifiable evidence that a particular piece of data existed at or before a specific point in time, without revealing the data's contents. It accepts the hash of a document, binds it to a trusted time reference, and returns a signed timestamp token that anyone can later verify. Timestamp services support non-repudiation, intellectual-property priority, regulatory record-keeping, and long-term signature validation, and may be anchored in a trusted timestamp authority or in a public blockchain for trust-minimised proof of existence.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:timestamp-service", "labels": [ "Timestamp Service" ], "is_subclass_of": [ "Timestamping Service" ], "wikilinks": [] }, { "id": "timestamp", "title": "Timestamp", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A recorded value indicating the time of block creation in a blockchain system, embedded in each block header to establish chronological ordering of the chain and enable time-based protocol rules. Timestamps support difficulty adjustment, consensus validity checks, and provide an immutable audit trail for transaction ordering and data provenance.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:timestamp", "labels": [ "Timestamp", "Electronic Timestamp", "Last Seen Timestamp", "Timestamp Ordering" ], "is_subclass_of": [ "Distributed Data Structure", "Blockchain Entity", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "timestamping-service", "title": "Timestamping Service", "domain": "security", "domain_name": "Security", "definition": "A timestamping service is a system that issues verifiable proof that a piece of data existed at or before a particular time, often using cryptographic methods.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:timestamping-service", "labels": [ "Timestamping Service" ], "is_subclass_of": [ "Cryptographic Protocol", "Trusted Timestamping" ], "wikilinks": [ "Hash Function", "Cryptography", "Content Provenance", "Timestamp", "Trusted Timestamping" ] }, { "id": "tiny-ml", "title": "TinyML", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine learning inference deployed on resource-constrained microcontrollers with kilobyte-scale RAM, milliwatt power budgets, and MHz-range processors, enabling always-on intelligent sensing in IoT devices, wearables, and embedded sensors without cloud connectivity. Requires aggressive model optimisation through INT8 quantisation, pruning, and knowledge distillation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tiny-ml", "labels": [ "TinyML" ], "is_subclass_of": [ "AI Technique", "Deep Learning" ], "wikilinks": [ "MLPerf Tiny", "TensorFlow Lite Micro", "TinyML Foundation", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "tls-handshake", "title": "Tls Handshake", "domain": "security", "domain_name": "Security", "definition": "The TLS handshake is the negotiation phase of the Transport Layer Security protocol in which a client and server agree on protocol version and cipher suite, authenticate via certificates, and establish shared session keys. It combines public-key cryptography for authentication and key agreement with symmetric cryptography for the subsequent record protocol. Modern versions complete in fewer round trips and provide forward secrecy through ephemeral key exchange.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:tls-handshake", "labels": [ "Tls Handshake", "TLS Handshake" ], "is_subclass_of": [ "Transport Layer Security" ], "wikilinks": [] }, { "id": "togaf", "title": "Togaf", "domain": "governance", "domain_name": "Governance", "definition": "TOGAF, The Open Group Architecture Framework, is a widely adopted methodology and framework for designing, planning, implementing and governing enterprise information architecture. Its core is the Architecture Development Method, an iterative cycle covering business, data, application and technology architecture. TOGAF provides a common vocabulary, reference models and governance structures that align IT strategy with business objectives.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:togaf", "labels": [ "Togaf", "TOGAF" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "token-bonding-curve", "title": "Token Bonding Curve", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Algorithmic pricing mechanism that defines token value as a mathematical function of circulating supply and reserve balance, providing continuous liquidity through automated market making.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:token-bonding-curve", "labels": [ "Token Bonding Curve", "Bonding Curve" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Automated Trading", "Continuous Liquidity", "DeFi Standards Alliance", "Liquidity Mechanism", "Mathematical Model", "Predictable Pricing", "Price Oracle", "Pricing Formula", "Reserve Pool", "Reserve Token", "Supply Function", "Token Economy", "Automated Market Maker", "Blockchain Infrastructure", "Decentralized Exchange", "Economic Parameters", "MiddlewareLayer", "Smart Contract", "VirtualEconomyDomain" ] }, { "id": "token-bridge", "title": "Token Bridge", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Token Bridge is a protocol that enables the transfer of token value or representation between two distinct blockchain networks that cannot natively communicate. It commonly locks or burns tokens on the source chain and mints or releases an equivalent representation on the destination chain, coordinated by validators, relayers or light-client proofs. Token bridges are a core mechanism for cross-chain liquidity but historically a significant locus of security risk.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:token-bridge", "labels": [ "Token Bridge" ], "is_subclass_of": [ "Blockchain Interoperability" ], "wikilinks": [] }, { "id": "token-cost", "title": "Token Cost", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Token cost is the economic and computational expense incurred when a large language model consumes and produces tokens, calculated as the sum of input (prompt) tokens and output (completion) tokens multiplied by their respective per-token prices, and it constitutes the dominant recurring cost driver of deployed generative-AI and agent-orchestration systems. Because providers meter usage in tokens rather than requests, token cost scales super-linearly with prompt length, conversation history, retrieved context, tool-call payloads, and multi-step agent fan-out, making it the primary budget variable an orchestrator must forecast, cap, and attribute. Managing token cost involves measurement (counting tokens per call and per session), attribution (allocating spend to tasks, agents, or tenants), and reduction levers such as prompt compression, context pruning, caching, model-tier routing, and truncation of chat history against the model context window.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:token-cost", "labels": [ "Token Cost" ], "is_subclass_of": [ "Performance Metrics" ], "wikilinks": [ "Cost-Efficient Inference", "Context Window", "Tokeniser", "Large Language Model", "Latency" ] }, { "id": "token-custody-service", "title": "Token Custody Service", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A secure infrastructure system for safeguarding digital tokens and cryptographic assets through multi-signature wallets, cold storage, and enterprise-grade custodial services in virtual economy environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:token-custody-service", "labels": [ "Token Custody Service" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Asset Recovery", "Audit Trail System", "Cold Storage System", "ETSI GS MEC 003", "Identity Verification System", "Institutional Trading", "Key Management Service", "Multi-Signature Wallet", "Secure Token Storage", "Security Module", "Access Control System", "Blockchain Network", "Compliance Framework", "Cryptographic Key Management", "Digital Asset Infrastructure", "MiddlewareLayer", "Regulatory Compliance", "VirtualEconomyDomain" ] }, { "id": "token-distribution", "title": "Token Distribution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token distribution is the process and schema by which a blockchain project allocates its native tokens across stakeholders such as founders, investors, the community, and a treasury. It specifies how many tokens each group receives, when they unlock through vesting, and through which mechanisms they are released, all of which shape decentralisation, incentive alignment, and market liquidity. A distribution design is a central lever of a project's tokenomics.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:token-distribution", "labels": [ "Token Distribution" ], "is_subclass_of": [ "Tokenomics" ], "wikilinks": [] }, { "id": "token-economics", "title": "Token Economics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenomics is the study and design of the economic systems governing cryptocurrency tokens, encompassing supply mechanics (inflation/deflation), distribution schedules, utility functions, governance rights, and incentive structures that determine token value, network security, and sustainable ecosystem growth.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:token-economics", "labels": [ "Token Economics", "TokenEconomics" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Economics" ], "wikilinks": [ "DeFi", "Decentralised Governance", "Network Incentives", "Staking", "Value Capture", "BlockchainDomain", "Blockchain Economics", "Cryptocurrency", "Governance Token", "MEV", "Smart Contract" ] }, { "id": "token-economy", "title": "Token Economy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A token economy is an economic system in which digital tokens \u2014 cryptographic representations of value, rights, or access \u2014 serve as the primary medium of exchange, incentive, and governance within a defined platform or protocol. Token economies leverage blockchain infrastructure to create programmable, permissionless economic systems where tokens encode ownership (utility tokens, security tokens, governance tokens, NFTs), align participant incentives through staking and reward mechanisms, and enable decentralised governance. The concept spans DeFi protocols, Web3 platforms, creator economies, and emerging AI agent economies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:token-economy", "labels": [ "Token Economy", "Tokenised Economy" ], "is_subclass_of": [ "Blockchain Economics" ], "wikilinks": [] }, { "id": "token-efficiency", "title": "Token Efficiency", "domain": "economics", "domain_name": "Economics", "definition": "The measure of how effectively an AI model utilizes input and output tokens to complete a task, where higher efficiency can reduce total costs even if per-token prices are higher.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:token-efficiency", "labels": [ "Token Efficiency" ], "is_subclass_of": [ "AI Economics" ], "wikilinks": [] }, { "id": "token-embedding", "title": "Token Embedding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A learnable lookup table that maps each discrete token in a vocabulary to a dense continuous vector, providing the initial semantic encoding fed into transformer model layers. Embeddings are trained end-to-end and combined with positional encodings; they may be tied with the output un-embedding matrix to reduce parameter count and improve training stability.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:token-embedding", "labels": [ "Token Embedding" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "token-emission", "title": "Token Emission", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token emission is the process by which new units of a protocol's native token are created and distributed over time, typically to reward liquidity providers, stakers, or validators for participating in the network. It is governed by an emission schedule that specifies the rate and decay of issuance, balancing the need to bootstrap participation against the dilution that excessive emission imposes on existing holders. In DeFi, protocols competing for liquidity, such as in curve wars and yield farming, use token emission as an incentive lever, directing rewards toward pools or strategies the protocol wants to attract capital into. Poorly calibrated emission schedules are a common cause of unsustainable yields that collapse once emission-driven demand outpaces genuine protocol usage.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:token-emission", "labels": [ "Token Emission" ], "is_subclass_of": [ "Tokenomics" ], "wikilinks": [] }, { "id": "token-engineering", "title": "Token Engineering", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The discipline of designing token systems and their incentive structures using methods from economics, control theory and systems engineering. It treats a token economy as a system to be specified, simulated and tested before deployment.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:token-engineering", "labels": [ "Token Engineering" ], "is_subclass_of": [ "Tokenomics" ], "wikilinks": [ "Mechanism Design", "Token Economics", "Token", "DeFi", "Tokenomics", "https://tokenengineeringcommunity.github.io/website/" ] }, { "id": "token-generation", "title": "Token Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Token generation is the autoregressive process by which a language model produces output tokens one at a time, sampling from a probability distribution over the vocabulary conditioned on the input context and all previously generated tokens. Each forward pass through the model produces logits over the vocabulary; a sampling strategy \u2014 greedy decoding, temperature sampling, top-k, or nucleus sampling \u2014 selects the next token, which is appended to the context for the subsequent step. Token generation is the primary inference workload of large language models and determines output quality, latency, and throughput.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:token-generation", "labels": [ "Token Generation" ], "is_subclass_of": [ "Artificial Intelligence", "AI Technique" ], "wikilinks": [] }, { "id": "token-governance", "title": "Token Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A decision-making mechanism for blockchain protocols and DAOs in which voting power is allocated in proportion to holdings of a governance token, so that proposals \u2014 parameter changes, treasury spending, protocol upgrades \u2014 are ratified by token-weighted ballots executed through smart contracts, trading the legitimacy problems of plutocratic weighting against the sybil resistance and skin-in-the-game that stake-based voting provides.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:token-governance", "labels": [ "Token Governance" ], "is_subclass_of": [ "Decentralised Governance" ], "wikilinks": [ "Decentralised Governance", "Governance Token", "DAO", "On-chain Governance" ] }, { "id": "token-issuance", "title": "Token Issuance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token issuance is the process by which a blockchain protocol or smart contract creates and distributes new digital tokens into circulation, establishing the initial and ongoing supply available to network participants. It encompasses the technical minting of token units according to predefined rules as well as the economic policies governing distribution, vesting schedules, and supply caps. Issuance mechanisms range from one-time generation events to continuous emission programmes tied to staking rewards or protocol activity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:token-issuance", "labels": [ "Token Issuance" ], "is_subclass_of": [ "Token Economics" ], "wikilinks": [] }, { "id": "token-maxing", "title": "Token Maxing", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The practice of maximizing the volume of tokens processed or generated by AI systems, often used as a proxy metric for user engagement or system utilization.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:token-maxing", "labels": [ "Token Maxing" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "token-metadata", "title": "Token Metadata", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token metadata is the structured descriptive data associated with a blockchain token, typically including attributes such as name, description, image references, traits, and provenance. For non-fungible tokens it is commonly stored as a JSON document referenced by a tokenURI and often hosted on decentralised storage such as IPFS to preserve immutability and addressability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:token-metadata", "labels": [ "Token Metadata", "Token Metadata Schema" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "token-processing-volume", "title": "Token Processing Volume", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The aggregate measure of linguistic units processed by large language models over a defined period, serving as a key indicator of AI infrastructure utilization and scale.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:token-processing-volume", "labels": [ "Token Processing Volume" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "token-router", "title": "Token Router", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A software intermediary that directs LLM inference requests to various API providers based on cost, latency, or performance criteria.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:token-router", "labels": [ "Token Router" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "token-sale", "title": "Token Sale", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Token Sale is a fundraising mechanism in which a blockchain project issues and sells digital tokens to investors or users in exchange for cryptocurrency or fiat currency, typically prior to or during the launch of a protocol or application. Token sales include initial coin offerings (ICOs), initial exchange offerings (IEOs), and initial DEX offerings (IDOs), each with differing levels of regulatory oversight and platform involvement. The proceeds are used to fund development, while purchasers receive tokens that may grant utility, governance rights, or speculative value within the ecosystem.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:token-sale", "labels": [ "Token Sale" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "token-standard", "title": "Token Standard", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Technical specification defining rules, interfaces, and functions that smart contract tokens must implement to ensure interoperability within a blockchain ecosystem, exemplified by ERC-20 (fungible tokens), ERC-721 (NFTs), and ERC-1155 (multi-token), establishing common APIs for transfers, balance queries, approvals, and metadata across wallets, exchanges, and decentralised applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:token-standard", "labels": [ "Token Standard", "CMTA Token Standard", "ERC Token Standard", "Interchain Token Standard", "Token Standard Protocol" ], "is_subclass_of": [ "Token and Asset", "Blockchain" ], "wikilinks": [ "DeFi", "Fungibility", "NFT", "Blockchain", "Decentralized Exchange", "Smart Contract" ] }, { "id": "token-swapping", "title": "Token Swapping", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token swapping is the on-chain exchange of one cryptocurrency or digital token for another without an intermediary custodian, executed atomically within a single transaction through smart contracts. The mechanism is implemented via automated market makers (AMMs), which price assets algorithmically using constant-function market-making formulas against liquidity pools, or via atomic swap protocols using hash time-locked contracts (HTLCs) that enable trustless cross-chain exchanges. Token swapping is the primary interaction primitive of decentralised exchanges and a building block for complex DeFi strategies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:token-swapping", "labels": [ "Token Swapping", "Token Swap" ], "is_subclass_of": [ "Decentralized Finance (DeFi)" ], "wikilinks": [] }, { "id": "token-transfer", "title": "Token Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A token transfer is the operation of moving ownership of a fungible or non-fungible blockchain token from one account to another, recorded as a state change on a distributed ledger. On smart-contract platforms it is typically realised by invoking a transfer function defined by a token standard such as ERC-20, which debits the sender's balance, credits the recipient, and emits an event. Token transfers are the elementary settlement primitive underlying payments, trading, and decentralised finance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:token-transfer", "labels": [ "Token Transfer" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "token-weighted-voting", "title": "token-weighted voting", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Token-weighted voting is an on-chain governance mechanism in which each participant's voting power is directly proportional to the quantity of governance tokens they hold, lock, or stake. It is the dominant decision-making primitive in decentralised autonomous organisations and DeFi protocols, enabling transparent, programmable, and censorship-resistant governance of parameter changes, treasury allocations, and protocol upgrades. The mechanism exhibits plutocratic tendencies in which large token holders disproportionately control outcomes, motivating research into alternative weighting schemes such as quadratic voting, conviction voting, and vote-escrow models. Implementations range from fully on-chain execution via Governor Bravo and OpenZeppelin Governor to off-chain gasless signalling via platforms such as Snapshot.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:token-weighted-voting", "labels": [ "Token-Weighted Voting", "Stake-weighted Voting", "Token Voting", "Token-Based Voting" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "token", "title": "Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital asset representation built on an existing blockchain platform that represents ownership, utility, or access rights, typically adhering to standardized protocols for transferability and interoperability, implemented as a cryptographically-secured unit that can be owned, transferred, and programmably controlled through smart contracts according to defined rules and token standards.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:token", "labels": [ "Token", "AXL Token", "Community Token", "Digital Token", "Game Token", "Secure Token", "Token Incentive", "Token Supply Mechanics", "Token Trading", "Token-Based Representation" ], "is_subclass_of": [ "Digital Asset", "Blockchain Entity", "Transferable Right" ], "wikilinks": [ "Aave", "AML", "Arbitrum", "Avalanche", "BEP-20", "BRC-20", "Bridge", "Chainlink", "Compound", "Cosmos", "Curve Finance", "DeFi", "ERC-1155", "ERC-20", "ERC-3643", "ERC-4626", "ERC-721", "ERC-7540", "Howey Test", "IPFS" ] }, { "id": "tokenised-real-world-asset-transfer", "title": "Tokenised Real World Asset Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenised real-world asset transfer is the movement of on-chain tokens representing claims on off-chain assets such as real estate, commodities, or securities between parties or across blockchain networks. It requires that the cryptographic transfer of the token remains legally and operationally bound to the underlying asset, often relying on cross-chain bridges and interoperability protocols to preserve provenance across ledgers.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tokenised-real-world-asset-transfer", "labels": [ "Tokenised Real World Asset Transfer" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "tokenised-real-world-assets", "title": "Tokenised Real World Assets", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenised real-world assets (RWAs) are blockchain tokens that represent ownership or economic rights in tangible or traditional financial assets such as property, bonds, equities, or commodities. By mapping off-chain value onto programmable tokens, they enable fractional ownership, continuous settlement, and integration with decentralised finance while still depending on legal frameworks and custodians for enforceability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tokenised-real-world-assets", "labels": [ "Tokenised Real World Assets", "Real World Assets", "Real-World Assets", "Tokenised Assets", "Tokenised Real-World Asset" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "tokenised-securities", "title": "Tokenised Securities", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenised securities are regulated financial instruments such as equities, bonds, or fund units issued and transferred as blockchain tokens that carry the same legal rights as their traditional counterparts. They embed compliance logic, including transfer restrictions and investor accreditation checks, directly into smart contracts so that securities laws are enforced programmatically at the point of transaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tokenised-securities", "labels": [ "Tokenised Securities", "Tokenised Securities Market", "Tokenised Securities Trading", "Tokenised Securities on Bitcoin", "Tokenized Securities" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [] }, { "id": "tokeniser", "title": "tokeniser", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A tokeniser is a preprocessing component that segments raw text into a sequence of discrete tokens \u2014 sub-words, words, or characters \u2014 and maps each token to an integer identifier in a fixed vocabulary, forming the numerical input representation consumed by language model architectures. Modern sub-word tokenisers such as Byte Pair Encoding (BPE), WordPiece, and SentencePiece balance vocabulary coverage with sequence length efficiency, enabling models to handle arbitrary Unicode text including rare words, multilingual content, and specialised domains without out-of-vocabulary failures. The tokeniser vocabulary and its associated embedding matrix are co-trained with the model and constitute a foundational design decision that governs sequence length, memory footprint, and cross-lingual fairness. Multimodal extensions of tokenisation discretise images, audio, and video frames into token sequences analogous to text tokens, enabling unified cross-modal architectures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:tokeniser", "labels": [ "Tokeniser" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "tokenization-system", "title": "Tokenization System", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A tokenization system is the end-to-end technical and operational infrastructure that converts rights, assets, or value into blockchain tokens and manages their lifecycle of issuance, transfer, custody, and redemption. It combines smart-contract templates, identity and compliance modules, and registry services to ensure that tokens reliably represent and preserve their underlying claims.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tokenization-system", "labels": [ "Tokenization System" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [] }, { "id": "tokenization", "title": "Tokenization", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Process of representing real-world or virtual assets as digital tokens on a blockchain through cryptographic mechanisms and smart contract protocols.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:tokenization", "labels": [ "Tokenization", "Tokenisation" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Asset Metadata", "Blockchain Economy", "ERC-1155", "ERC-721", "ETSI GR ARF 010", "Fractional Ownership", "ISO 24165", "OMA3 Media Working Group", "Ownership Transfer", "Reed Smith Legal Framework", "Reed Smith + OMA3", "Token Standard Protocol", "VirtualProcess", "Asset Trading", "Blockchain Network", "Blockchain Transaction", "Consensus Mechanism", "Cryptographic Keys", "Digital Asset Management", "Digital Signature" ] }, { "id": "tokenized-asset", "title": "Tokenized Asset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A tokenized asset is a representation of ownership or economic rights in a real-world or digital asset recorded as a blockchain token. The token acts as a programmable, transferable claim whose issuance, transfer, and settlement are governed by smart contracts, enabling fractional ownership, faster settlement, and broader access. Underlying assets range from securities, real estate, and commodities to art and intellectual property.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:tokenized-asset", "labels": [ "Tokenized Asset", "Tokenised Asset" ], "is_subclass_of": [ "Asset Tokenization" ], "wikilinks": [] }, { "id": "tokenizer", "title": "Tokenizer", "domain": "ai", "domain_name": "Ai", "definition": "A tokenizer is the component that converts raw text into the discrete units (tokens) a language model processes, and back again. Modern tokenizers use subword algorithms such as byte-pair encoding, WordPiece, or unigram language models to balance vocabulary size against sequence length, representing common words as single tokens and rare words as compositions of smaller pieces. The tokenizer defines the model's vocabulary and directly affects context-window usage, multilingual fairness, handling of code and numbers, and ultimately the cost and capability of the system.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tokenizer", "labels": [ "Tokenizer" ], "is_subclass_of": [ "Tokenization" ], "wikilinks": [] }, { "id": "tokenomics-governance", "title": "Tokenomics Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenomics governance is the discipline integrating token-economic design with decentralised decision-making mechanisms, defining how Governance Token holders in DeFi and DAO protocols gain, exercise, and lose voting power whilst simultaneously participating as economic stakeholders t...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:tokenomics-governance", "labels": [ "Tokenomics Governance" ], "is_subclass_of": [ "DeFi and Economics", "DAO", "Governance Token", "Mechanism Design", "Protocol Economics", "Decentralised Finance" ], "wikilinks": [ "Aave", "Balancer", "Bonding Curve", "Bonding Mechanism", "Bribe Market", "Buyback and Burn", "Centralised Exchange Token", "Compound", "Compound Governor Bravo", "Convex Finance", "Curve Finance", "Curve Wars", "DeFi", "Decentralised Autonomous Organisation", "Decentralised Finance", "DecentralisedFinanceDomain", "Delegation System", "EigenLayer", "EIP-1559", "Emission Direction Control" ] }, { "id": "tokenomics", "title": "Tokenomics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tokenomics is the study and design of the economic systems underpinning blockchain-based tokens, encompassing token supply mechanics (fixed supply, inflationary, or deflationary schedules), distribution models (initial coin offerings, airdrops, liquidity mining), utility functions (governance rights, access control, fee payment), and incentive alignment mechanisms. A well-designed tokenomic model balances value creation and capture across stakeholders \u2014 developers, users, liquidity providers, and governance participants \u2014 through mechanisms such as staking, burning, bonding curves, and vesting schedules. Tokenomics draws from monetary theory, game theory, and mechanism design to engineer sustainable digital economies where token holders are incentivised to act in ways that grow the ecosystem's long-term value.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tokenomics", "labels": [ "Tokenomics" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "owl:Thing" ] }, { "id": "tone-mapping", "title": "Tone Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An image processing operation that compresses the high dynamic range of luminance values captured or rendered in a scene into the limited range a display or print medium can reproduce, while preserving perceived contrast, detail, and colour appearance. Tone mapping operators range from simple global curves such as Reinhard and filmic ACES transforms to local, content-adaptive methods, and are a standard final stage in real-time rendering pipelines, HDR photography, and cinematic colour workflows.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:tone-mapping", "labels": [ "Tone Mapping" ], "is_subclass_of": [ "Image Processing" ], "wikilinks": [ "Image Processing", "Colour Grading", "Computational Photography", "Photorealism" ] }, { "id": "tool-call-loop", "title": "Tool Call Loop", "domain": "ai", "domain_name": "Ai", "definition": "The core execution cycle of an LLM agent, in which the model emits a structured tool invocation, the harness executes it against the real environment (shell, file system, API, browser), appends the result to the conversation, and re-invokes the model \u2014 repeating until the model judges the task complete and returns a final answer; this iterate-observe-act loop is what turns single-turn text generation into grounded, autonomous task execution.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:tool-call-loop", "labels": [ "Tool Call Loop" ], "is_subclass_of": [ "Agent Loop" ], "wikilinks": [ "Agent Loop", "Agentic Workflow", "Function Calling", "Tool Use" ] }, { "id": "tool-definition", "title": "Tool Definition", "domain": "ai", "domain_name": "Ai", "definition": "A Tool Definition is a structured specification that describes a capability that an AI language model or agent can invoke at runtime, including the tool's name, a natural language description of its purpose, and a formal schema defining its input parameters and expected output format. Tool definitions are the primary mechanism through which AI systems access external functionality, enabling them to call APIs, query databases, execute code, retrieve documents, or perform actions in the world beyond text generation. The format of tool definitions is standardised by providers such as Anthropic, OpenAI, and Google, typically using JSON Schema to describe parameters, and the quality of a tool definition's description and schema directly determines how reliably an AI model selects and invokes it.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tool-definition", "labels": [ "Tool Definition", "Tool Choice Parameter" ], "is_subclass_of": [ "Tool Schema" ], "wikilinks": [] }, { "id": "tool-registry", "title": "Tool Registry", "domain": "ai", "domain_name": "Ai", "definition": "A tool registry is a catalogue that stores the definitions, schemas, permissions, and metadata of the external tools and functions an AI agent can invoke. It enables agents to discover, validate, and route calls to available capabilities at runtime, providing a governed and extensible interface between language models and the systems they act upon.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tool-registry", "labels": [ "Tool Registry" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "tool-schema", "title": "tool schema", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A tool schema is a structured, machine-readable specification \u2014 typically expressed in JSON Schema format \u2014 that formally defines a function or external capability that an AI agent may invoke at inference time, encoding its name, natural-language description, parameter names and types, constraints, and required versus optional fields. Tool schemas are injected into an AI model's context window as part of the system prompt or capability manifest, enabling the model to reason about available operations and emit syntactically correct, structured function-call outputs that an orchestration layer can dispatch. They constitute the primary interface contract between large language models and the external systems, services, or data sources they interact with, determining both tool-selection accuracy and invocation correctness. The tool schema construct is foundational to agentic AI architectures and underpins modern function-calling APIs, the Model Context Protocol, and multi-agent orchestration frameworks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:tool-schema", "labels": [ "Tool Schema" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "tool-use", "title": "tool use", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Tool Use is the capability of large language models and AI agent systems to invoke external functions, APIs, databases, or services at inference time, extending the model's effective knowledge and action repertoire beyond pure text generation. The model receives a structured description of available tools \u2014 a tool schema encoding names, descriptions, and JSON Schema parameter specifications \u2014 reasons about which tool to call and with what arguments, executes that call via a surrounding orchestration layer, and incorporates the returned observation into its subsequent reasoning. This capability is foundational to agentic AI systems that must take real-world actions such as code execution, web search, database queries, file manipulation, or actuator control, bridging the boundary between language generation and executable computation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:tool-use", "labels": [ "Tool Use", "Anthropic Tool Use API", "LLM Tool Use", "Tool Invocation", "Tool Use Block", "Tool Use Capability", "Tool Use System" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "tool-augmented-llm", "title": "Tool-Augmented LLM", "domain": "ai", "domain_name": "Ai", "definition": "A tool-augmented large language model is an LLM that can invoke external tools, APIs, code execution, or retrieval systems during inference to overcome the limits of its parametric knowledge. By emitting structured calls and incorporating the returned results, it can perform calculations, access live data, and act on the world rather than relying solely on text generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:tool-augmented-llm", "labels": [ "Tool-Augmented LLM" ], "is_subclass_of": [ "Large Language Models" ], "wikilinks": [] }, { "id": "tool-augmented-reasoning", "title": "Tool-Augmented Reasoning", "domain": "ai", "domain_name": "Ai", "definition": "Tool-augmented reasoning is an approach in which a language model invokes external tools such as calculators, search or code execution to solve tasks beyond its parametric knowledge. The model interleaves reasoning steps with tool calls.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:tool-augmented-reasoning", "labels": [ "Tool-Augmented Reasoning" ], "is_subclass_of": [ "Reasoning" ], "wikilinks": [ "Tool Use", "Reasoning", "Agentic AI", "Hallucination" ] }, { "id": "tooling-layer", "title": "Tooling Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Tooling Layer is the cross-cutting stratum that provides the development, deployment, and observability instruments used to build and operate the rest of the system. It sits beside the production strata rather than within the runtime data path and supports operational and research work. It contains build systems, debuggers, monitors, and automation utilities.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:tooling-layer", "labels": [ "Tooling Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Runtime Layer", "Operational Layer", "Research Layer", "Continuous Integration", "Observability", "owl:Thing" ] }, { "id": "top-of-atmosphere-reflectance", "title": "Top-of-atmosphere Reflectance", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:top-of-atmosphere-reflectance", "labels": [ "Top-of-atmosphere Reflectance" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "topographic-correction", "title": "Topographic Correction", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:topographic-correction", "labels": [ "Topographic Correction" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "topological-map", "title": "Topological Map", "domain": "robotics", "domain_name": "Robotics", "definition": "Topological Map is a graph-structured spatial abstraction used in Mobile Robot Navigation and Autonomous Navigation, representing environments as discrete nodes (places, waypoints, landmarks) connected by edges (traversable paths, transitions) rather than storing explicit metric coord...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:topological-map", "labels": [ "Topological Map" ], "is_subclass_of": [ "Navigation and Planning", "Robot Type", "Simultaneous Localisation and Mapping", "Spatial Representation", "Mobile Robot Navigation", "Graph Database", "Cognitive Map" ], "wikilinks": [ "AlgorithmLayer", "Assistive Robotics", "Bag-of-Words Model", "Bag-of-Words Model", "Bayesian Estimation", "Bundle Adjustment", "Cognitive Map", "ComputerVisionDomain", "Convolutional Neural Networks", "Convolutional Neural Networks", "Dense 3D Reconstruction", "FAB-MAP", "Feature Descriptors", "Graph Neural Networks", "Graph Neural Networks", "Graph Optimiser", "Graph Representation", "Graph Theory", "IEEE RAS", "Lifelong Mapping" ] }, { "id": "topology", "title": "Topology", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Topology is the mathematical study of properties of spaces that are preserved under continuous deformation, such as connectivity, continuity and the presence of holes. It provides the formal language for reasoning about shape independent of exact distances or coordinates, underpinning fields from network analysis to topological data analysis. In machine learning it informs how the structure of data and networks can be characterised and exploited.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:topology", "labels": [ "Topology" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [] }, { "id": "tor", "title": "Tor", "domain": "security", "domain_name": "Security", "definition": "A free overlay network and software that provides anonymous communication by routing traffic through a series of volunteer-operated relays using layered encryption, enabling censorship resistance, traffic-analysis resistance, and onion service hosting.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tor", "labels": [ "Tor", "Tor Network" ], "is_subclass_of": [ "Onion Routing" ], "wikilinks": [ "Onion Routing", "Encryption", "Privacy", "Network Security" ] }, { "id": "tornado-cash", "title": "Tornado Cash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tornado Cash is a set of non-custodial smart contracts on Ethereum and compatible networks that obscure the on-chain link between a deposit and a withdrawal. Users deposit a fixed denomination of a token into a pool and later withdraw the same amount to a different address, using a zero-knowledge proof to demonstrate ownership of a valid deposit without revealing which one. It became widely known both as a privacy tool and as the subject of sanctions by the United States Office of Foreign Assets Control in 2022.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:tornado-cash", "labels": [ "Tornado Cash" ], "is_subclass_of": [ "Cryptographic Primitive", "Cryptographic Domain" ], "wikilinks": [ "Zero-Knowledge Proof", "Ethereum", "Transaction Privacy", "zk-SNARK", "Regulatory Domain", "Cryptographic Domain" ] }, { "id": "torque-control", "title": "Torque Control", "domain": "robotics", "domain_name": "Robotics", "definition": "A low-level actuation strategy that directly commands the output torque of joints or motors rather than position or velocity, enabling compliant, force-sensitive interaction between a robot and its environment. Torque control is essential for safe human-robot collaboration and dexterous manipulation tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:torque-control", "labels": [ "Torque Control", "Torque Control Loop" ], "is_subclass_of": [ "Actuation and Control", "Motion Control" ], "wikilinks": [ "Motion Control", "Robotics" ] }, { "id": "torque-sensor", "title": "Torque Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A torque sensor is a transducer that measures the rotational force, or torque, applied about an axis, commonly at a robot joint or drive shaft. By reporting the load a joint experiences, it enables force control, compliance, collision detection, and safe physical interaction between robots and humans. Torque sensors typically use strain gauges bonded to a deformable element whose deflection is proportional to the applied moment.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:torque-sensor", "labels": [ "Torque Sensor" ], "is_subclass_of": [ "Robot Sensor" ], "wikilinks": [] }, { "id": "torque", "title": "Torque", "domain": "robotics", "domain_name": "Robotics", "definition": "Torque is the rotational analogue of force, defined as the cross product of the moment arm and the applied force vector (\u03c4 = r \u00d7 F), measured in newton-metres. In robotics it governs joint actuation, grip force, and dynamic loading across mechanical transmissions, and is the primary physical quantity managed by torque-controlled servos and force-torque sensors during manipulation tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:torque", "labels": [ "Torque", "Torque Sensing" ], "is_subclass_of": [ "Actuation and Control", "Robot Dynamics", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "total-cost-of-ownership", "title": "Total Cost of Ownership", "domain": "economics", "domain_name": "Economics", "definition": "The comprehensive financial assessment of an AI system's lifecycle costs, including inference, maintenance, integration, and opportunity costs, rather than just per-unit pricing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:total-cost-of-ownership", "labels": [ "Total Cost of Ownership" ], "is_subclass_of": [ "AI Economics" ], "wikilinks": [] }, { "id": "total-ionising-dose", "title": "Total Ionising Dose", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:total-ionising-dose", "labels": [ "Total Ionising Dose" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "total-order-broadcast", "title": "Total Order Broadcast", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Total order broadcast, also called atomic broadcast, is a communication primitive in distributed systems that guarantees all correct processes deliver the same set of messages in exactly the same order. It strengthens reliable broadcast with a total ordering property, ensuring that every replica observes an identical sequence of events. Total order broadcast is provably equivalent to consensus and forms the foundation of state machine replication.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:total-order-broadcast", "labels": [ "Total Order Broadcast" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [] }, { "id": "total-supply", "title": "Total Supply", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Total Supply is the count of all tokens that have ever been created on a blockchain network, encompassing circulating tokens, locked or vested tokens, tokens held in treasury reserves, and any tokens that have been minted but not yet distributed\u2014but excluding permanently destroyed (burned) tokens. It differs from the maximum supply (the hard cap set by the protocol) and from circulating supply (tokens freely tradeable on secondary markets). Total supply is a fundamental parameter in tokenomics analysis used to assess inflation, dilution, and long-term value dynamics.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:total-supply", "labels": [ "Total Supply" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "Economic Mechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "total-value-locked", "title": "Total Value Locked", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Total Value Locked (TVL) is a metric that aggregates the market value of all assets deposited in a decentralised finance protocol or across an ecosystem at a given time. It serves as a proxy for adoption, liquidity depth and the economic weight of a protocol's smart contracts. Because it is denominated in volatile assets, TVL can shift with both deposit flows and underlying price movements, so it is interpreted alongside other indicators.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:total-value-locked", "labels": [ "Total Value Locked" ], "is_subclass_of": [ "Blockchain", "DeFi and Economics" ], "wikilinks": [] }, { "id": "toucan-protocol", "title": "Toucan Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain protocol that brings voluntary carbon market credits on-chain as tokens, intended to make carbon offsets tradeable and composable within decentralised finance.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:toucan-protocol", "labels": [ "Toucan Protocol" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [ "Smart Contract", "Carbon Credits", "Voluntary Carbon Market", "Carbon Accounting", "Carbon Markets" ] }, { "id": "tourism-industry", "title": "Tourism Industry", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The Tourism Industry encompasses the businesses, infrastructure, and services facilitating recreational and business travel, including accommodation, transport, attraction management, and destination marketing. In spatial computing contexts, it is a primary adopter of XR, virtual tourism, and location-based immersive experiences to enhance visitor engagement and extend reach beyond physical access.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:tourism-industry", "labels": [ "Tourism Industry" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "tourism-metaverse", "title": "Tourism Metaverse", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A virtual platform enabling users to explore, preview, and experience tourist destinations, cultural sites, and travel experiences through immersive digital environments, supporting sustainable tourism and accessibility to remote or restricted locations.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:tourism-metaverse", "labels": [ "Tourism Metaverse" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "360 Video", "Accessibility Enhancement", "Content Management System", "Cultural Exhibit", "Cultural Heritage Preservation", "ETSI GS MEC", "Geographic Information System", "Tour Guide System", "Translation Service", "Travel Planner", "3D Rendering Engine", "ApplicationLayer", "Avatar System", "CreativeMediaDomain", "Destination Marketing", "Geospatial Engine", "Metaverse Platform", "Photogrammetry", "Spatial Audio", "Virtual Destination" ] }, { "id": "toxic-content-counter-narrative-system", "title": "Toxic Content Counter-Narrative System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Project BroBots is a multi-agent research initiative to identify, classify, and counter toxic online content using NLP-based harm detection and counter-narrative generation. It employs fine-tuned large language models on social media corpora, the Agentic Alliance tech stack, and is motivated by the harms of automated bot-driven misinformation and harassment across internet platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:toxic-content-counter-narrative-system", "labels": [ "Toxic Content Counter-Narrative System", "Project BroBots" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Agentic Alliance", "Agents", "Death of the Internet", "Digital Society Harms" ] }, { "id": "toxicity-detection", "title": "Toxicity Detection", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Toxicity Detection is a machine learning discipline that automatically identifies harmful, abusive, or offensive language in user-generated content, typically using classifiers trained on annotated corpora of hate speech, threats, and harassment. It forms a core component of content moderation pipelines and operates at scale across social platforms, forums, and messaging systems. Toxicity detection systems must balance recall against false-positive rates to avoid over-censorship while protecting users from harm.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:toxicity-detection", "labels": [ "Toxicity Detection" ], "is_subclass_of": [ "AI Application", "Content Moderation" ], "wikilinks": [ "Artificial Intelligence", "Content Moderation" ] }, { "id": "traceability-mechanism", "title": "Traceability Mechanism", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Traceability Mechanism is a systematic approach for recording, maintaining, and retrieving comprehensive documentation of an AI system's development process, data lineage, decision-making logic, and operational history to enable accountability, auditability, and debugging. It encompasses data provenance tracking, model versioning, decision logging, and tamper-evident audit trails that allow stakeholders to reconstruct the causal chain from system outputs back to training data and design choices. Regulatory frameworks including the EU AI Act and GDPR increasingly mandate specific traceability capabilities for high-risk AI deployments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:traceability-mechanism", "labels": [ "Traceability Mechanism" ], "is_subclass_of": [ "AI Governance and Ethics", "Ai Governance Principle" ], "wikilinks": [ "AI Governance Principle", "Change Management", "Decision Logging", "IEEE 2801 Recommended Practice", "ISO/IEC 23053 AI Framework", "Lineage Tracking", "Model Versioning", "NIST AI Risk Management Framework", "Tamper-Evident Logging", "AIEthicsDomain", "Audit Trail", "ConceptualLayer", "Data Provenance", "EU AI Act", "Metadata Management" ] }, { "id": "traceability", "title": "Traceability", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Traceability is the ability to verify the history, location, or application of an item or piece of information by means of recorded identification. It requires that each unit or event be uniquely identified and that the linkages between successive states be persistently recorded, enabling both forward tracking from origin to destination and backward tracing from any point to its provenance. Traceability is foundational to supply-chain integrity, regulatory compliance, food and pharmaceutical safety, and software and data provenance, and it is increasingly underpinned by tamper-evident ledgers.", "entityType": "Class", "qualityScore": 0.78, "maturity": "established", "iri": "urn:ngm:class:traceability", "labels": [ "Traceability", "Material Traceability", "Product Traceability", "Requirements Traceability", "Sustainability Traceability", "Traceability Chain" ], "is_subclass_of": [ "Process" ], "wikilinks": [] }, { "id": "tracked-robot", "title": "Tracked Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Tracked Robot is a ground mobile robot that uses continuous loop tracks\u2014analogous to those on military tanks\u2014rather than wheels to achieve locomotion. The large contact surface area of the track distributes the robot's weight, providing superior traction and stability on uneven, soft, or obstacle-dense terrain such as rubble, mud, stairs, and gravel. Tracked robots are widely deployed in search-and-rescue, military reconnaissance, inspection, and agricultural automation where wheeled platforms would lose grip or become immobilised.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:tracked-robot", "labels": [ "Tracked Robot" ], "is_subclass_of": [ "Ground Robot" ], "wikilinks": [ "Ground Robot", "Robotics" ] }, { "id": "tracking-hardware", "title": "Tracking Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Tracking Hardware comprises the physical sensors and devices used to determine the position and orientation of users, controllers, and objects within spatial computing environments. This includes inertial measurement units, optical trackers, hand-tracking cameras, eye-tracking modules, and SLAM-based inside-out tracking systems that together provide the 6-DoF pose data essential for immersive VR/AR experiences.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:tracking-hardware", "labels": [ "Tracking Hardware" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [ "owl:Thing" ] }, { "id": "tracking-system", "title": "Tracking System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A Tracking System is a hardware and software assembly that continuously determines the position, orientation, and motion of one or more objects or agents within a defined reference frame, using sensing technologies such as optical cameras, inertial measurement units, electromagnetic emitters, ultrasound, GPS, or LiDAR. Tracking systems are foundational components of augmented and virtual reality, robotics, surgical navigation, sports analytics, logistics, and autonomous vehicles, where precise real-time knowledge of spatial state is essential for interaction, control, or safety.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:tracking-system", "labels": [ "Tracking System", "Inside-Out Tracking System", "Tracking Subsystem", "Usage Tracking System" ], "is_subclass_of": [ "Spatial Tracking System" ], "wikilinks": [] }, { "id": "tracking-technology", "title": "Tracking Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Tracking Technology refers to the hardware and software systems that determine the real-time position, orientation, and motion of a user or device in physical space for extended reality applications. Approaches include inside-out tracking using onboard cameras and SLAM, outside-in tracking using fixed base stations, eye tracking for gaze-based interaction, and hand/finger tracking for controller-free input.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:tracking-technology", "labels": [ "Tracking Technology" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "trade-execution", "title": "Trade Execution", "domain": "finance", "domain_name": "Finance", "definition": "Trade execution is the process of converting an investment decision into a completed transaction in a market, encompassing how, when, and where an order is routed and filled. It seeks the best achievable outcome for the trader by managing factors such as price, speed, market impact, and slippage against the order book. Execution quality is a core determinant of realised returns and is governed by best-execution obligations in many jurisdictions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trade-execution", "labels": [ "Trade Execution" ], "is_subclass_of": [ "Market Microstructure" ], "wikilinks": [] }, { "id": "trade-finance-automation", "title": "Trade Finance Automation", "domain": "finance", "domain_name": "Finance", "definition": "Trade Finance Automation is the application of digital technologies\u2014including blockchain-based smart contracts, AI-driven document processing, and electronic data interchange\u2014to streamline and accelerate the complex financial instruments and workflows that facilitate international commerce. Traditional trade finance relies on paper-intensive instruments such as letters of credit, bills of lading, and documentary collections, which are slow, error-prone, and costly to process. Automation replaces manual document verification with machine-readable structured data, triggers payment obligations automatically upon verified delivery conditions, and reduces counterparty risk through distributed ledger transparency. The result is faster settlement, lower operational costs, and improved access to working capital for exporters and importers globally.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trade-finance-automation", "labels": [ "Trade Finance Automation" ], "is_subclass_of": [ "Trade Finance" ], "wikilinks": [] }, { "id": "trade-finance", "title": "Trade Finance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trade finance encompasses the financial instruments, products, and mechanisms that facilitate domestic and international commercial transactions by managing the payment, credit, and risk gap between exporter shipment and importer receipt. Core instruments include letters of credit, bank guarantees, documentary collections, supply chain financing, factoring, and forfaiting, all structured to give exporters payment certainty and importers time to generate revenue from goods before settling. Trade finance is a foundational segment of global banking, representing approximately $9 trillion in annual transaction volume, and is undergoing significant digitisation through blockchain-based platforms, electronic bills of lading, and smart contract automation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trade-finance", "labels": [ "Trade Finance", "International Trade Finance", "Komgo Trade Finance", "SME Trade Finance Access", "Trade Finance Instrument" ], "is_subclass_of": [ "Financial Services" ], "wikilinks": [] }, { "id": "trade-lens", "title": "TradeLens", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "A blockchain-based supply chain platform for the shipping industry, jointly developed by IBM and Maersk, that aimed to digitise and share trade documentation among participants. It was discontinued in 2023.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:trade-lens", "labels": [ "TradeLens" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [ "Distributed Ledger", "Supply Chain", "Blockchain" ] }, { "id": "traditional-banking", "title": "Traditional Banking", "domain": "finance", "domain_name": "Finance", "definition": "Traditional banking is the long-established system by which licensed deposit-taking institutions accept customer funds, extend credit through loans and mortgages, and operate payment and settlement services under prudential regulatory oversight. Banks act as financial intermediaries, transforming short-term deposits into longer-term loans and profiting from the interest-rate spread while managing liquidity and credit risk. The sector is governed by capital-adequacy frameworks such as Basel III/IV, deposit-insurance schemes, and central-bank lender-of-last-resort facilities. Traditional banking is distinguished from shadow banking and fintech alternatives by its formal licensing, balance-sheet-based intermediation, and full integration with national payment and clearing infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:traditional-banking", "labels": [ "Traditional Banking", "Commercial Banking", "Traditional Banking System", "Traditional Banking Transfer" ], "is_subclass_of": [ "Traditional Finance" ], "wikilinks": [ "Financial Regulation", "Payment System", "Traditional Finance" ] }, { "id": "traditional-carbon-registry", "title": "Traditional Carbon Registry", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A traditional carbon registry is a centralised database that records the issuance, ownership, transfer and retirement of carbon credits to prevent double counting in carbon markets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:traditional-carbon-registry", "labels": [ "Traditional Carbon Registry" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [ "Carbon Registry", "Regulatory Compliance", "Verra VCS Standard", "Sustainability Domain", "Carbon Markets" ] }, { "id": "traditional-corporation", "title": "Traditional Corporation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A traditional corporation is a centralised legal entity owned by shareholders and directed by a board, granting limited liability and a hierarchical management structure under company law. Decision-making authority and capital are concentrated, with accountability mediated through fiduciary duties and regulatory oversight. It is the conventional organisational form against which decentralised autonomous organisations are contrasted.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:traditional-corporation", "labels": [ "Traditional Corporation" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "traditional-finance", "title": "Traditional Finance", "domain": "finance", "domain_name": "Finance", "definition": "Traditional Finance (TradFi) denotes the incumbent financial system comprising central banks, commercial and investment banks, securities exchanges, clearing houses, custodians, and regulated intermediaries that operate under sovereign legal frameworks. It relies on trusted third parties to record ownership, settle transactions, and extend credit, in contrast to decentralised or blockchain-native alternatives. Its architecture encompasses fractional-reserve banking, fiat currency issuance, and multilateral settlement rails such as SWIFT and RTGS systems. TradFi institutions are subject to prudential regulation, capital adequacy requirements (Basel III/IV), and conduct oversight by bodies such as the SEC, FCA, and ECB.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:traditional-finance", "labels": [ "Traditional Finance", "Traditional Financial Accounting" ], "is_subclass_of": [ "Financial Infrastructure" ], "wikilinks": [ "DeFi", "Institutional Adoption", "Financial Infrastructure" ] }, { "id": "traditional-mass-media-institution", "title": "Traditional Mass Media Institution", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Legacy media refers to established, traditional forms of mass communication \u2014 including print, broadcast television, and radio \u2014 that preceded the digital internet era. These institutions hold significant cultural authority and are undergoing structural disruption as AI-driven platforms alter content distribution and audience economics.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:traditional-mass-media-institution", "labels": [ "Traditional Mass Media Institution", "legacy media" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "traditional-securities", "title": "Traditional Securities", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Traditional Securities are conventional financial instruments such as equities, bonds and fund units that are issued, held and transferred through established centralised market infrastructure including exchanges, central securities depositories and custodians. They represent legal claims on assets or income and are governed by long-standing securities regulation. In blockchain discourse they form the baseline against which tokenised and on-chain security tokens are compared.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:traditional-securities", "labels": [ "Traditional Securities" ], "is_subclass_of": [ "Blockchain", "Financial Instruments" ], "wikilinks": [] }, { "id": "traffic-management", "title": "Traffic Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Traffic management is the set of techniques for controlling the flow of data across a network to optimise performance, fairness, and reliability. It encompasses traffic shaping, rate limiting, prioritisation, congestion control, and load distribution, applied at routers, gateways, and application proxies. By regulating how bandwidth is allocated and how bursts are smoothed, traffic management upholds quality-of-service guarantees, protects services from overload, and improves overall utilisation of finite network capacity.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:traffic-management", "labels": [ "Traffic Management" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "traffic-shaping", "title": "Traffic Shaping", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A network traffic management technique that regulates the rate and burstiness of outgoing packet flows by buffering packets and releasing them according to a configured profile, classically implemented with token bucket or leaky bucket algorithms. By smoothing bursts and holding flows to contracted rates, shaping delays rather than drops excess traffic, enforcing bandwidth allocations and protecting latency-sensitive classes as a core mechanism of quality-of-service policy at network edges.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:traffic-shaping", "labels": [ "Traffic Shaping" ], "is_subclass_of": [ "Traffic Management" ], "wikilinks": [ "Traffic Management", "Quality Of Service", "Congestion Control", "Rate Limiting", "Bandwidth" ] }, { "id": "trained-data-state", "title": "Trained Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:trained-data-state", "labels": [ "Trained Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "training-corpus", "title": "Training Corpus", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A training corpus is the body of text or other data used to fit the parameters of a machine learning model, whose scale, diversity and quality directly shape what the resulting model can learn. It is the input from which subword vocabularies are derived by algorithms such as byte pair encoding during tokenisation, prior to any model training taking place. Curation choices around a training corpus, including deduplication and filtering, materially affect downstream model behaviour.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:training-corpus", "labels": [ "Training Corpus" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "training-data-distribution", "title": "Training Data Distribution", "domain": "ai", "domain_name": "Ai", "definition": "Training data distribution refers to the statistical properties and compositional characteristics of the dataset used to train a machine learning model, including the relative frequencies of classes, the coverage of input feature space, the presence of rare or tail events, and the demographic or domain balance of examples. The training data distribution determines what patterns a model can learn, what it will generalise from, and where it will fail: a model trained on a distribution that differs from the deployment distribution will exhibit degraded performance due to covariate shift or prior probability shift. Deliberate control of training data distribution \u2014 through curation, resampling, augmentation, and synthetic data generation \u2014 is a primary lever for improving model robustness, fairness, and safety.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:training-data-distribution", "labels": [ "Training Data Distribution", "Training Distribution" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "training-data-repository", "title": "Training Data Repository", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A centralized or distributed storage system for collecting, organizing, versioning, and managing datasets used to train AI and machine learning models, including data provenance tracking, quality assurance, and access control.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:training-data-repository", "labels": [ "Training Data Repository" ], "is_subclass_of": [ "AI Infrastructure", "Data Management" ], "wikilinks": [ "Data Management", "metaverse" ] }, { "id": "training-data", "title": "Training Data", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Training data encompasses all curated, collected, and pre-processed corpora of examples \u2014 text, images, audio, video, structured records, code, and synthetic artefacts \u2014 ingested during the learning phase of Machine Learning and Foundation Models to optimise model parameters via gradient-...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:training-data", "labels": [ "Training Data", "Computer Vision Training Data", "Emotional Training Data", "Labeled Training Data", "Licensed Training Data", "Paired Training Data", "Training Data Corpus", "Training Data Information", "Training Data Pairs", "TrainingData" ], "is_subclass_of": [ "AI Infrastructure", "Data Governance", "Dataset", "Machine Learning Discipline", "Data Management", "Knowledge Representation" ], "wikilinks": [ "AlgorithmLayer", "Annotation", "Annotation Standards", "CDLA", "Code Corpus", "CommonCrawl", "Creative Commons", "Data Cards", "Data Deduplication", "DataGovernanceDomain", "Data Quality Assurance", "Data Splits", "Dataset", "Dataset Bias", "Distribution Shift", "Features", "Image Dataset", "Labels", "Language Identification", "Licensing" ] }, { "id": "training-dataset-metadata", "title": "Training Dataset Metadata", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Training dataset metadata is structured descriptive information about the data used to train a machine learning model, including its provenance, size, collection method, class distribution, known biases, and licensing terms. This metadata is a core component of AI model cards and datasheets, enabling reproducibility, fairness auditing, and regulatory accountability. Thorough dataset metadata underpins model transparency and responsible AI governance.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:training-dataset-metadata", "labels": [ "Training Dataset Metadata" ], "is_subclass_of": [ "AI Infrastructure", "ModelArchitecture" ], "wikilinks": [ "ModelArchitecture" ] }, { "id": "training-dataset", "title": "Training Dataset", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Training Dataset is a curated collection of labelled or unlabelled data used to fit the parameters of a machine learning model, establishing the empirical foundation from which the model generalises to unseen inputs. Its composition\u2014size, diversity, label quality, representational balance, and provenance\u2014fundamentally determines the capabilities and failure modes of the resulting model. Training datasets range from manually annotated corpora (ImageNet, SQuAD) to web-scraped large-scale collections (Common Crawl, LAION-5B) and synthetically generated data. Questions of copyright, consent, bias, and traceability in training datasets have become central to AI governance and legal disputes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:training-dataset", "labels": [ "Training Dataset", "DROID Dataset", "Demonstration Dataset", "Labeled Dataset", "Seed Dataset" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "training-hardware", "title": "Training Hardware", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Training Hardware is a artificial intelligence concept and a type of High-Performance Computing. that enables Neural Network Training.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:training-hardware", "labels": [ "Training Hardware" ], "is_subclass_of": [ "AI Infrastructure", "High-Performance Computing" ], "wikilinks": [ "High-Performance Computing", "Neural Network Training" ] }, { "id": "training-instability", "title": "Training Instability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A set of pathological behaviours during neural network optimisation \u2014 including exploding gradients, vanishing gradients, loss divergence, and oscillating loss curves \u2014 that prevent a model from converging to a useful solution. Training instability arises from interactions between architecture depth, learning rate, batch size, and numerical precision, and is mitigated through techniques such as gradient clipping, mixed-precision training, careful initialisation, and adaptive optimisers.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:training-instability", "labels": [ "Training Instability" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "training-layer", "title": "Training Layer", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Training Layer is the stratum that fits model parameters from data using optimisation procedures. It sits above the Compute and Data strata it consumes and below the Model and Foundation Model strata that hold its results. It contains training loops, optimisers, loss functions, and the orchestration of large-scale learning runs.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:training-layer", "labels": [ "Training Layer" ], "is_subclass_of": [ "AI Technique", "owl:Thing" ], "wikilinks": [ "Compute Layer", "Data Layer", "Model Layer", "Foundation Model Layer", "Gradient Descent", "Backpropagation", "owl:Thing" ] }, { "id": "training-loop", "title": "Training Loop", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Training Loop is the iterative control structure that drives model training, repeatedly performing a forward pass, loss computation, backward pass, and parameter update over batches of data until a stopping condition is reached. It coordinates gradient descent steps with logging, checkpointing, and evaluation at regular intervals. Its structure is common across frameworks even though the specific optimiser and schedule vary.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:training-loop", "labels": [ "Training Loop" ], "is_subclass_of": [ "Gradient Descent" ], "wikilinks": [] }, { "id": "training-method", "title": "Training Method", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Training Method is a systematic algorithm or procedure used to optimise the parameters of a machine learning model by minimising a loss function through iterative updates over labelled or unlabelled data. Training methods span the full spectrum from supervised and unsupervised learning to reinforcement learning and self-supervised pre-training, each with distinct update rules, convergence properties, and data requirements. Specific optimisers such as gradient descent, Adam, and RMSprop, together with regularisation strategies like dropout and batch normalisation, are key components within training methods.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:training-method", "labels": [ "Training Method" ], "is_subclass_of": [ "AI Technique", "MachineLearning" ], "wikilinks": [ "Artificial Intelligence", "MachineLearning" ] }, { "id": "training-methods", "title": "Training Methods", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Training Methods encompass the algorithms and procedures used to optimise machine learning model parameters against training data, including supervised, unsupervised, semi-supervised, reinforcement, transfer, curriculum, and contrastive learning paradigms. The choice of method fundamentally determines model capability, data efficiency, and generalisation behaviour.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:training-methods", "labels": [ "Training Methods" ], "is_subclass_of": [ "AI Technique", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "training-pipeline", "title": "Training Pipeline", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The orchestrated, repeatable sequence of stages that turns raw data into a trained machine-learning model: ingestion and validation, preprocessing and augmentation, batching, the optimisation loop itself, evaluation against held-out data, and registration of versioned artefacts. Treating training as a pipeline rather than a script makes runs reproducible, resumable, and automatable, and is the precondition for continuous retraining and reliable model deployment within MLOps practice.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:training-pipeline", "labels": [ "Training Pipeline" ], "is_subclass_of": [ "Data Pipeline" ], "wikilinks": [ "Data Pipeline", "MLOps", "Data Augmentation", "Hyperparameter Tuning", "Gradient Descent", "Distributed Training", "Model Deployment" ] }, { "id": "training-and-fine-tuning", "title": "Training and fine tuning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Fine-tuning is the post-pre-training adaptation of a large neural network \u2014 typically a transformer-based Foundation Models or Large Language Models \u2014 to narrower task distributions or behavioural objectives by continuing gradient-based optimisation on comparatively small, curated dataset...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:training-and-fine-tuning", "labels": [ "Training and fine tuning", "Training and Fine Tuning" ], "is_subclass_of": [ "AI Technique", "Model Training", "Transfer Learning", "Machine Learning Discipline", "Deep Learning", "Large-Scale Pretrained Foundation Model" ], "wikilinks": [ "AISI", "AISI Frontier AI Safety Framework", "AlgorithmLayer", "AlignmentLayer", "Causal Language Modelling", "Hugging Face", "In-Context Learning", "Low-Rank Adaptation", "MachineLearningDomain", "Pre-trained Model", "Scaling Laws", "Transfer Learning", "Active Learning", "Agents", "AI Alignment", "AI-GroundedDomain", "Anthropic Claude", "ApplicationLayer", "Attention", "Backpropagation" ] }, { "id": "training", "title": "Training", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Training is the supervised or self-supervised process of iteratively adjusting the parameters of a machine learning model to minimise a loss function over a labelled or unlabelled dataset. It encompasses forward passes, backpropagation, gradient descent optimisation, and regularisation techniques such as dropout and weight decay. The output of training is a fitted model whose learned weights encode patterns from the training data, ready for inference on unseen inputs.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:training", "labels": [ "Training", "AI Training", "Parallel Training", "Skills Training" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "trajectory-control", "title": "Trajectory Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Trajectory Control is a robotics control methodology that tracks desired time-varying position, velocity, and acceleration profiles along planned paths. It coordinates joint-space and task-space motion to ensure smooth, precise end-effector movement while respecting dynamic and kinematic constraints.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trajectory-control", "labels": [ "Trajectory Control" ], "is_subclass_of": [ "Actuation and Control", "Motion Control" ], "wikilinks": [ "Motion Control", "Robotics" ] }, { "id": "trajectory-generation", "title": "Trajectory Generation", "domain": "robotics", "domain_name": "Robotics", "definition": "The process of computing a time-parameterized path that specifies the position, velocity, and acceleration of a robot's joints or end-effector as functions of time, enabling smooth motion from start to goal configurations while respecting kinematic and dynamic constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trajectory-generation", "labels": [ "Trajectory Generation", "Compliant Trajectory Generation" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [ "Autonomous Vehicles", "Continuity", "Dynamic Constraints", "Goal Configuration", "Industrial Automation", "Kinematic Constraints", "RB-1003-optimal-control", "RB-1005-forward-kinematics", "RB-1006-inverse-kinematics", "RB-1016-path-planning", "RB-1019-obstacle-avoidance", "Robot Motion", "Smoothness", "Start Configuration", "Task Execution", "Trajectory", "Motion Planning", "Robotics" ] }, { "id": "trajectory-optimisation", "title": "Trajectory Optimisation", "domain": "robotics", "domain_name": "Robotics", "definition": "Trajectory optimisation is the computation of a path and control profile for a dynamic system that minimises a cost subject to physical and operational constraints. It is widely used in robotics, aerospace and autonomous vehicles.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:trajectory-optimisation", "labels": [ "Trajectory Optimisation" ], "is_subclass_of": [ "Motion Planning" ], "wikilinks": [ "Optimisation", "Control Theory", "Path Planning", "Robotics", "Motion Planning" ] }, { "id": "trajectory-planning", "title": "Trajectory Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "The process of computing a time-parameterised path for a robot or autonomous system that satisfies kinematic constraints, avoids obstacles, and achieves a target configuration smoothly and efficiently. Trajectory planning bridges high-level path planning with low-level motion control, incorporating velocity and acceleration profiles, inverse kinematics, and real-time replanning for dynamic environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:trajectory-planning", "labels": [ "Trajectory Planning" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "trajectory-tracking", "title": "Trajectory Tracking", "domain": "robotics", "domain_name": "Robotics", "definition": "Trajectory tracking is the control problem of causing a dynamical system\u2014typically a robot, vehicle, or aerial platform\u2014to follow a prescribed time-parameterised path through configuration space with minimal deviation. It couples a reference trajectory generated by a planner with a feedback controller that corrects errors arising from disturbances, model mismatch, and actuation limits. Trajectory tracking controllers range from classical PID and linear quadratic regulators to model predictive controllers and learning-based approaches. It is a central capability in mobile robotics, autonomous driving, drone flight, and industrial manipulator control.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trajectory-tracking", "labels": [ "Trajectory Tracking", "Arc Trajectory Following", "TrajectoryTracking" ], "is_subclass_of": [ "Trajectory Control" ], "wikilinks": [] }, { "id": "trans-neptunian-object", "title": "Trans-Neptunian Object", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:trans-neptunian-object", "labels": [ "Trans-Neptunian Object" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "transaction-authorisation", "title": "Transaction Authorisation", "domain": "security", "domain_name": "Security", "definition": "Transaction authorisation is the process of verifying that a party has the right to initiate a given transaction and that the transaction has been approved according to defined rules before it is executed. It typically relies on cryptographic proofs such as digital signatures to bind approval to a specific transaction payload, preventing tampering or replay. In financial and blockchain systems, transaction authorisation is a prerequisite gate that must pass before settlement or state change occurs.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:transaction-authorisation", "labels": [ "Transaction Authorisation" ], "is_subclass_of": [ "Authorization" ], "wikilinks": [] }, { "id": "transaction-censorship", "title": "Transaction Censorship", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction censorship is the selective exclusion or delay of specific transactions from a blockchain by miners, validators, or block builders, typically to comply with sanctions lists, extract value, or suppress particular users or protocols. It undermines a blockchain's censorship-resistance guarantees and can arise from regulatory pressure on centralised infrastructure such as relays or MEV builders, or from validator cartels acting in concert. Its severity depends on the degree of validator or mining centralisation, since a sufficiently decentralised set of block producers makes sustained censorship costly to coordinate. Mitigations include inclusion lists, proposer-builder separation, and encrypted mempools.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:transaction-censorship", "labels": [ "Transaction Censorship" ], "is_subclass_of": [ "Censorship Resistance" ], "wikilinks": [] }, { "id": "transaction-confirmation", "title": "Transaction Confirmation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The process by which a blockchain transaction achieves inclusion in a mined or validated block and subsequently accumulates additional blocks on top of it, reducing the probability of reversal. The number of confirmations required for sufficient security depends on consensus mechanism and transaction value; Bitcoin typically requires 6 confirmations while Proof-of-Stake systems with deterministic finality may confirm in a single block.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transaction-confirmation", "labels": [ "Transaction Confirmation", "Transaction Confirmation Time" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "transaction-fee", "title": "Transaction Fee", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction Fee is an economic mechanism in blockchain systems whereby originators of transactions pay a fee \u2014 denominated in the network's native cryptocurrency \u2014 to validators or miners in exchange for including and processing their transaction in a block. Fees simultaneously compensate network participants and provide a spam-prevention signal.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transaction-fee", "labels": [ "Transaction Fee" ], "is_subclass_of": [ "DeFi and Economics", "Blockchain Entity", "EconomicMechanism" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "EconomicLayer", "EconomicMechanism", "TokenEconomicsDomain" ] }, { "id": "transaction-finality", "title": "Transaction Finality", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The point at which a blockchain transaction becomes irreversible and cannot be altered or removed, providing settlement certainty for participants. Finality types include probabilistic (increasing confidence with confirmations), deterministic (explicit protocol guarantee), and economic (cost to revert exceeds benefit).", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transaction-finality", "labels": [ "Transaction Finality", "Absolute Transaction Guarantee" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain" ], "wikilinks": [ "Settlement", "Transaction Confirmation Time", "Blockchain", "Consensus Mechanism", "Deterministic Finality", "Probabilistic Finality", "Transaction", "Transaction Confirmation" ] }, { "id": "transaction-history", "title": "Transaction History", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction history is the ordered, append-only record of all value transfers, state changes, or operations executed within a financial or data system, providing a tamper-evident chronological account of who transacted what, with whom, and when. In blockchain systems, transaction history is the fundamental data structure underpinning the distributed ledger \u2014 every confirmed block links to its predecessor through cryptographic hashing, making retroactive alteration of any transaction computationally prohibitive. Transaction history serves audit, compliance, forensic, tax, and portfolio analytics functions across both traditional and decentralised financial systems.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transaction-history", "labels": [ "Transaction History" ], "is_subclass_of": [ "Blockchain Transaction", "Transaction Ledger" ], "wikilinks": [] }, { "id": "transaction-ledger", "title": "Transaction Ledger", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A transaction ledger is an ordered, append-only record of value transfers or state changes that serves as the authoritative history of activity within a financial or blockchain system. Each entry captures the parties, amounts, and timing of a transaction, and in distributed settings the ledger is cryptographically chained and replicated to guarantee integrity and auditability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transaction-ledger", "labels": [ "Transaction Ledger" ], "is_subclass_of": [ "Distributed Data Structure" ], "wikilinks": [] }, { "id": "transaction-manager", "title": "Transaction Manager", "domain": "data", "domain_name": "Data", "definition": "A transaction manager is a software component that coordinates the atomic execution of operations against a database or distributed system, enforcing the ACID properties of atomicity, consistency, isolation, and durability. It governs commit and rollback, manages concurrency control and locking, and in distributed settings orchestrates multi-resource protocols such as two-phase commit.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transaction-manager", "labels": [ "Transaction Manager", "Private Transaction Manager", "Transaction Coordinator", "Transaction Management" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "transaction-monitoring", "title": "Transaction Monitoring", "domain": "finance", "domain_name": "Finance", "definition": "Transaction monitoring is the automated surveillance of financial transactions \u2014 payments, transfers, trade executions, and account activity \u2014 to identify patterns indicative of money laundering, terrorist financing, fraud, sanctions evasion, or other financial crimes, generating alerts for investigation and mandatory Suspicious Activity Reports (SARs) to regulators. It is a core component of Anti-Money Laundering compliance programmes required by FATF recommendations and national implementing legislation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transaction-monitoring", "labels": [ "Transaction Monitoring", "Transaction Monitor" ], "is_subclass_of": [ "Compliance Monitoring" ], "wikilinks": [] }, { "id": "transaction-output", "title": "Transaction Output", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A transaction output is a component of a blockchain transaction that specifies an amount of value and the conditions under which it can later be spent. In UTXO-based systems each output records a value and a locking script that defines who may consume it, and an unspent output becomes the input to a future transaction. Transaction outputs are the fundamental units of ownership and value transfer in such ledgers, and the set of all unspent outputs constitutes the current state of holdings.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transaction-output", "labels": [ "Transaction Output" ], "is_subclass_of": [ "Blockchain Transaction" ], "wikilinks": [] }, { "id": "transaction-parameter", "title": "Transaction Parameter", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction Parameters are the configuration values that define the execution characteristics of a blockchain transaction, including gas limit, gas price (or priority fee under EIP-1559), nonce, value, and calldata. These parameters determine transaction priority, cost, computational resources consumed, and ordering within blocks, directly influencing DeFi strategies such as MEV extraction and Layer 2 batching efficiency.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transaction-parameter", "labels": [ "Transaction Parameter" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Transaction" ], "wikilinks": [ "Blockchain Transaction" ] }, { "id": "transaction-pool", "title": "Transaction Pool", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Transaction Pool (mempool) is the in-memory distributed data structure held by each full node that stores validated but as-yet-unconfirmed transactions awaiting inclusion in a block. Nodes propagate transactions through the pool via gossip, miners and validators select transactions (typically by fee priority), and the pool is cleared as blocks confirm or transactions expire. Pool size, fee dynamics, and congestion directly determine user-experienced confirmation latency and cost.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transaction-pool", "labels": [ "Transaction Pool" ], "is_subclass_of": [ "Network Component", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure" ] }, { "id": "transaction-privacy", "title": "Transaction Privacy", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction privacy is the property of a payment or ledger system that conceals sensitive details of a transaction, such as sender, receiver and amount, from third parties while preserving verifiability of correctness. On public blockchains it is achieved through cryptographic techniques that prove a transaction is valid without revealing its contents. It addresses the inherent transparency of open ledgers, trading off auditability and regulatory visibility against confidentiality.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:transaction-privacy", "labels": [ "Transaction Privacy" ], "is_subclass_of": [ "Privacy Preserving Computation" ], "wikilinks": [] }, { "id": "transaction-processing", "title": "Transaction Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Transaction Processing encompasses the end-to-end lifecycle of blockchain transactions \u2014 from cryptographic signing and mempool propagation through validation, deterministic state execution, and finality confirmation. It includes parallelisation strategies, layer-2 scaling (state channels, rollups), and cross-chain atomic operations, with MEV mitigation and encrypted mempools addressing ordering fairness.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transaction-processing", "labels": [ "Transaction Processing", "Transaction Engine", "Transaction Processor" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "transaction-propagation", "title": "Transaction Propagation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction propagation is the process by which a newly broadcast transaction spreads across a peer-to-peer blockchain network from its originating node to the rest of the participants. Nodes relay valid transactions to their peers using a gossip protocol, typically announcing availability before transferring full data, so that the transaction reaches miners and validators for inclusion in a block while resisting spam through validation and fee policies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transaction-propagation", "labels": [ "Transaction Propagation" ], "is_subclass_of": [ "Gossip Protocol" ], "wikilinks": [] }, { "id": "transaction-signing", "title": "Transaction Signing", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction signing is the cryptographic process by which the holder of a private key authorises a blockchain transaction by producing a digital signature over its contents. The signature proves that the legitimate key holder approved the exact transaction without revealing the private key, and it binds the transaction to a specific account so that the network can verify authenticity and integrity. It is fundamental to self-custody, wallet security and the trustless validation of transactions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transaction-signing", "labels": [ "Transaction Signing" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "transaction-standard", "title": "Transaction Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A protocol defining secure exchange of digital assets and services within virtual economies, specifying message formats, authentication mechanisms, settlement procedures, and integrity guarantees.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:transaction-standard", "labels": [ "Transaction Standard" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Atomic Swaps", "Authentication Mechanism", "Data Persistence", "Economic Interoperability", "ETSI GR ARF 010", "Integrity Verification", "ISO 20022", "Message Format", "Secure Asset Transfer", "Settlement Protocol", "Transaction Auditability", "Transaction Ledger", "Wallet System", "Consensus Mechanism", "Cryptographic Key Management", "Digital Identity", "Middleware Layer", "Multi-Party Transactions", "Network Protocol", "Payment System" ] }, { "id": "transaction-throughput", "title": "Transaction Throughput", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction throughput is the rate at which a system processes and finalises transactions, commonly expressed as transactions per second, and is a primary measure of a blockchain network's capacity and scalability. It is shaped by block size, block interval, consensus mechanism, and execution efficiency, and it trades off against decentralisation and security in protocol design. Improving throughput is a central goal of layer-2 solutions and sharding.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transaction-throughput", "labels": [ "Transaction Throughput" ], "is_subclass_of": [ "Scalability" ], "wikilinks": [] }, { "id": "transaction-validation", "title": "Transaction Validation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction validation is the process by which network participants check that a blockchain transaction conforms to the protocol rules before accepting it. It ensures only valid transactions enter the ledger by verifying signatures, input availability, and compliance with consensus rules.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transaction-validation", "labels": [ "Transaction Validation", "Parallel Transaction Validation", "Transaction Validation Engine" ], "is_subclass_of": [ "Consensus" ], "wikilinks": [ "Cryptography", "Bitcoin Network", "Transaction", "Consensus", "https://developer.bitcoin.org/devguide/transactions.html", "https://en.bitcoin.it/wiki/Protocol_rules" ] }, { "id": "transaction", "title": "Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "State-changing operation on blockchain within blockchain systems, providing essential functionality for distributed ledger technology operations and properties.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transaction", "labels": [ "Transaction", "Atomic Transaction", "Confirmed Transaction", "Database Transaction", "Transaction Filtering", "Transaction Malleability", "Transaction Proof", "Transaction Queue" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "Distributed Data Structure", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure", "Virtual Economy" ] }, { "id": "transfer-function", "title": "Transfer Function", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A transfer function is the Laplace-domain (continuous-time) or Z-domain (discrete-time) ratio of output to input for a linear time-invariant (LTI) system with zero initial conditions, expressed as a ratio of polynomials whose roots yield the poles and zeros that determine the system's frequency response, stability, and transient behaviour. Transfer functions provide a frequency-domain characterisation of systems ranging from electronic filters and mechanical actuators to feedback control loops.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:transfer-function", "labels": [ "Transfer Function", "Modulation Transfer Function" ], "is_subclass_of": [ "Control Theory" ], "wikilinks": [] }, { "id": "transfer-impact-assessment", "title": "Transfer Impact Assessment", "domain": "governance", "domain_name": "Governance", "definition": "A Transfer Impact Assessment (TIA) is a documented evaluation organisations conduct before transferring personal data across borders to determine whether the destination jurisdiction provides protection essentially equivalent to that of the originating data-protection regime. Required in the EU following the Schrems II ruling, a TIA examines the destination's surveillance laws, the legal remedies available to data subjects, and any supplementary technical and organisational measures needed to make the transfer lawful. It is a core compliance artefact for international data flows under the GDPR.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:transfer-impact-assessment", "labels": [ "Transfer Impact Assessment" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "transfer-learning", "title": "Transfer Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Transfer learning is a machine learning paradigm in which knowledge encoded in a model trained on a source task or domain is systematically reused to improve learning efficiency and performance on a different but related target task or domain. By exploiting shared representations \u2014 such as low-level feature detectors, syntactic structures, or visual hierarchies \u2014 the technique drastically reduces the labelled data, compute, and training time required for downstream tasks. It is foundational to modern deep learning practice, underpinning pre-trained large language models, vision transformers, and multi-modal systems that are subsequently adapted via fine-tuning, prompt tuning, or adapter layers. The paradigm bridges the gap between data-rich source domains and data-scarce target settings, enabling deployment in low-resource clinical, scientific, and industrial contexts.", "entityType": "Class", "qualityScore": 0.76, "maturity": "mature", "iri": "urn:ngm:class:transfer-learning", "labels": [ "Transfer Learning", "Cross-Lingual Transfer Learning" ], "is_subclass_of": [ "Machine Learning", "Deep Learning" ], "wikilinks": [] }, { "id": "transfer-of-funds-regulation", "title": "Transfer Of Funds Regulation", "domain": "governance", "domain_name": "Governance", "definition": "The Transfer of Funds Regulation is a body of financial regulation requiring that information about the payer and payee accompany transfers of funds and certain crypto-asset transfers, so that transactions remain traceable for anti-money-laundering and counter-terrorist-financing purposes. It implements the Financial Action Task Force travel rule within a regulatory framework, obliging payment service providers and crypto-asset service providers to collect, transmit, and screen originator and beneficiary data. It is a central pillar of payment-chain transparency.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transfer-of-funds-regulation", "labels": [ "Transfer Of Funds Regulation", "Transfer of Funds Regulation" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "transfer-orbit", "title": "Transfer Orbit", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:transfer-orbit", "labels": [ "Transfer Orbit" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "transfer-restriction", "title": "Transfer Restriction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A transfer restriction is logic embedded in a tokenised asset that conditions or blocks transfers based on the identity, eligibility, lock-up status or jurisdiction of the parties involved. It allows regulated securities to be represented on a blockchain while enforcing legal constraints automatically at the protocol level. Security token standards such as ERC-1400 expose a check function that returns whether a proposed transfer is permitted, with reason codes for rejected transfers to support auditability.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:transfer-restriction", "labels": [ "Transfer Restriction" ], "is_subclass_of": [ "Security Token" ], "wikilinks": [] }, { "id": "transferable-right", "title": "Transferable Right", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Transferable Right is a legally or technically encoded entitlement that can be assigned, sold, or delegated from one party to another, typically enforced via smart contracts on a blockchain. Examples include tokenised intellectual property licences, digital asset ownership tokens, and royalty claims encoded in NFT standards.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:transferable-right", "labels": [ "Transferable Right" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "owl:Thing" ] }, { "id": "transform-matrix", "title": "Transform Matrix", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A transform matrix is a matrix, typically 4x4 in homogeneous coordinates for 3D graphics, that encodes a composed translation, rotation and scale applied to geometry. Graphics pipelines multiply vertex positions by a chain of transform matrices to move objects from local model space through world, view and projection spaces to screen coordinates. Scene graphs attach a transform matrix to each node so that hierarchical transformations propagate correctly from parent nodes to their children.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:transform-matrix", "labels": [ "Transform Matrix" ], "is_subclass_of": [ "Linear Algebra" ], "wikilinks": [] }, { "id": "transformative-ai", "title": "Transformative AI", "domain": "ai", "domain_name": "Ai", "definition": "Transformative AI refers to artificial intelligence advanced enough to precipitate change on the scale of the agricultural or industrial revolutions, fundamentally reshaping the economy, society, and the trajectory of civilisation. The term, used prominently in AI strategy and safety research, deliberately focuses on societal impact rather than on a specific architecture or a threshold of human-level cognition.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:transformative-ai", "labels": [ "Transformative AI" ], "is_subclass_of": [ "Artificial Intelligence", "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "transformer-architecture", "title": "Transformer Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network architecture based solely on self-attention mechanisms, dispensing with recurrence and convolutions entirely, enabling parallel sequence processing and underpinning modern large language models and multimodal AI systems.", "entityType": "Class", "qualityScore": 0.65, "maturity": "established", "iri": "urn:ngm:class:transformer-architecture", "labels": [ "Transformer Architecture" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "Stacker News", "ArtificialIntelligenceDomain", "Decentralised Web", "Lightning and Similar L2" ] }, { "id": "transformer", "title": "Transformer", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A neural network architecture that relies entirely on self-attention mechanisms rather than recurrence or convolution to process sequential data in parallel, serving as the foundation for modern large language models, vision models, and multimodal systems including GPT, BERT, and their successors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:transformer", "labels": [ "Transformer", "Switch Transformer", "Transformer Model", "Transformer Models", "Transformer Networks", "Transformer Policies" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "AIGroundedDomain", "Foundation Model", "Foundation Model", "Hugging Face", "ISO/IEC 22989", "Large Language Model", "Large Language Model", "NIST AI 600-1", "PyTorch", "TensorFlow", "Vaswani et al. 2017", "Attention Mechanism", "BERT", "Convolutional Neural Network", "EU AI Act", "GPT", "Recurrent Neural Network", "Self Attention" ] }, { "id": "transformers-library", "title": "Transformers Library", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Transformers library is an open-source software framework that provides unified access to thousands of pretrained transformer-based models for natural language processing, computer vision, audio and multimodal tasks. Maintained primarily by Hugging Face, it exposes a consistent interface for loading, fine-tuning and running inference across architectures, and integrates with deep-learning backends such as PyTorch. It has become a de facto standard toolkit for working with large pretrained models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:transformers-library", "labels": [ "Transformers Library" ], "is_subclass_of": [ "Deep Learning Framework" ], "wikilinks": [] }, { "id": "transformers", "title": "Transformers", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The transformer is a neural network architecture introduced by Vaswani et al. in 'Attention Is All You Need' (2017). It replaces recurrence and convolution with multi-head self-attention and position-wise feed-forward layers, enabling fully parallel sequence processing, and underpins large language models and modern vision, speech, and protein-structure systems.", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:transformers", "labels": [ "Transformers", "Efficient Transformers", "Hugging Face Transformers" ], "is_subclass_of": [ "AI Model Architecture", "Neural Network", "Attention Mechanism", "Deep Learning", "Sequence Modelling", "Large-Scale Pretrained Foundation Model" ], "wikilinks": [ "Adam Optimizer", "AIArchitectureDomain", "AlphaFold", "AlphaFold 2", "Audio Processing", "BLOOM", "CUDA", "DeBERTa", "DeepLearningDomain", "Drug Discovery", "Embedding Layer", "Falcon", "Flash Attention", "Gemma", "Google Brain", "GPT-NeoX", "GPU Compute", "Grouped Query Attention", "Hugging Face Transformers", "InferenceLayer" ] }, { "id": "translation", "title": "Translation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine Translation (MT) is the automated conversion of natural language text or speech from a source language into semantically and pragmatically equivalent target language output, evolving from rule-based symbolic approaches (RBMT, 1950s-1990s) through statistical phrase-based models (SMT, Mose...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:translation", "labels": [ "Translation", "Human Translation" ], "is_subclass_of": [ "AI Application", "Natural Language Processing", "Sequence-to-Sequence Learning", "Cross-Lingual Transfer", "Language Technology", "Computational Linguistics" ], "wikilinks": [ "AlgorithmLayer", "Back-Translation", "Beam Search", "Beam Search Decoding", "Bilingual Evaluation Data", "BLEU Score", "BPE Vocabulary", "COMET Benchmark", "COMET Metric", "Computational Linguistics", "Cross-Border Commerce", "Cross-Lingual Information Retrieval", "Cross-Lingual Transfer", "Dictionary-Based Translation", "Diplomatic Communication", "FLORES-200 Evaluation Set", "Global Content Delivery", "GPU Compute", "Healthcare Communication", "Human Post-Editing" ] }, { "id": "transmission-control-protocol", "title": "Transmission Control Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Transmission Control Protocol (TCP) is a connection-oriented transport-layer protocol that provides reliable, ordered, and error-checked delivery of byte streams between applications over an IP network. It establishes connections through a handshake and uses acknowledgements, retransmission, and flow and congestion control to ensure data arrives intact. TCP is the dominant reliable transport underlying most internet applications.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:transmission-control-protocol", "labels": [ "Transmission Control Protocol" ], "is_subclass_of": [ "Transport Layer" ], "wikilinks": [] }, { "id": "transmission-delay", "title": "Transmission Delay", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The time required to push all of a packet's bits onto the transmission medium, calculated as packet size divided by the link's transmission rate. It is one of the four canonical components of network latency alongside propagation, processing, and queueing delay, dominates end-to-end delay on low-bandwidth links, and shrinks proportionally as link capacity increases, which is why upgrading link speed reduces this component but leaves propagation delay unchanged.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:transmission-delay", "labels": [ "Transmission Delay" ], "is_subclass_of": [ "Network Latency" ], "wikilinks": [ "Network Latency", "Latency", "Propagation Delay", "Bandwidth" ] }, { "id": "transmission-network", "title": "Transmission Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A transmission network is the high-voltage infrastructure that carries bulk electrical power over long distances from generation sites to distribution substations near consumers. Comprising transmission lines, transformers, and switching stations, it minimises resistive losses through high voltages and is operated as a synchronised grid balancing supply and demand in real time.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transmission-network", "labels": [ "Transmission Network", "Network Transmission" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "transparency-oecd", "title": "Transparency (OECD)", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI actors should commit to transparency and responsible disclosure regarding AI systems, providing sufficient information to enable people to understand AI outcomes, challenge decisions and participate meaningfully in AI-influenced processes.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:transparency-oecd", "labels": [ "Transparency (OECD)" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "contestability", "Informed decision-making", "trust", "MetaverseDomain" ] }, { "id": "transparency-metrics", "title": "Transparency Metrics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Quantitative and qualitative measures used to assess and communicate the openness, accountability, and visibility of governance, operations, and decision-making processes in digital platforms, DAOs, and metaverse environments. Frameworks such as GRI, SASB, and the T-Index provide standardised scoring across data disclosure, financial reporting, and decision visibility dimensions. In blockchain contexts, on-chain voting records and treasury analytics serve as primary evidence sources.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transparency-metrics", "labels": [ "Transparency Metrics" ], "is_subclass_of": [ "AI Governance and Ethics", "Governance" ], "wikilinks": [ "Governance", "metaverse" ] }, { "id": "transparency-notice", "title": "Transparency Notice", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A formal disclosure document or notification that informs users about data collection practices, processing purposes, privacy policies, and their rights regarding personal information handling in digital platforms and metaverse environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transparency-notice", "labels": [ "Transparency Notice" ], "is_subclass_of": [ "AI Governance and Ethics", "Privacy Disclosure" ], "wikilinks": [ "metaverse", "Privacy Disclosure" ] }, { "id": "transparency-obligation", "title": "Transparency Obligation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A transparency obligation is a regulatory duty to disclose how a system works or that it is in use, such as informing people when they interact with an AI system or with generated content.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:transparency-obligation", "labels": [ "Transparency Obligation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [ "AI Regulation", "Regulatory Compliance", "EU AI Act", "Privacy" ] }, { "id": "transparency-reporting", "title": "Transparency Reporting", "domain": "governance", "domain_name": "Governance", "definition": "Transparency reporting is the periodic, structured disclosure by an organisation of metrics about its handling of user data, content moderation actions, government requests and enforcement decisions. It is a core accountability mechanism in platform governance, enabling regulators, researchers and the public to scrutinise how power is exercised over information and users. Reports typically quantify takedowns, account actions, legal demands and appeal outcomes over a defined period.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:transparency-reporting", "labels": [ "Transparency Reporting" ], "is_subclass_of": [ "Accountability" ], "wikilinks": [] }, { "id": "transparency-and-explainability", "title": "Transparency and Explainability", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Transparency and Explainability is a core AI trustworthiness dimension requiring that AI systems disclose sufficient information about their operation, decision logic, data provenance, and limitations to enable appropriate understanding and oversight by developers, deployers, and affected users. It encompasses traceability (audit trails, dataset documentation), explainability (global and local model explanations, counterfactuals), and communication transparency (disclosing AI involvement and system boundaries). Regulatory frameworks such as the EU AI Act Article 13 mandate transparency obligations for high-risk AI systems, driving demand for methods such as SHAP, LIME, and AI model cards.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:transparency-and-explainability", "labels": [ "Transparency and Explainability" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU AI Act Article 13", "LIME", "Model Cards", "SHAP", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "transparency", "title": "Transparency", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The property of an AI system whereby relevant information about the system's design, operation, capabilities, limitations, and decision-making processes is accessible and understandable to appropriate stakeholders, enabling informed oversight, trust, and accountability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "draft", "iri": "urn:ngm:class:transparency", "labels": [ "Transparency", "Supervisory Transparency", "Transparency (AI-0062)", "Transparency Mandate" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Accountability (AI-0068)", "Explainability (AI-0064)", "MetaverseDomain" ] }, { "id": "transparent-data-processing", "title": "Transparent Data Processing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Data handling practices that are openly disclosed to users, specifying what data is collected, how it is processed, who has access, and for what purposes, so as to enable informed consent and meaningful accountability. GDPR codifies this as the principle of lawfulness, fairness, and transparency; in metaverse and XR contexts it extends to immersive data streams including gaze tracking, movement patterns, and biometric signals. Implementation relies on audit trails, data flow mapping, privacy notices, and increasingly on zero-knowledge proofs for privacy-preserving verification.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transparent-data-processing", "labels": [ "Transparent Data Processing" ], "is_subclass_of": [ "Data Management", "Data Processing" ], "wikilinks": [ "Data Processing", "metaverse" ] }, { "id": "transparent-decision-making", "title": "Transparent Decision Making", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Governance processes where decisions, their rationale, and the factors influencing outcomes are openly visible and accessible to stakeholders, often implemented through on-chain voting, public proposals, and documented deliberations in DAOs and digital platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transparent-decision-making", "labels": [ "Transparent Decision Making", "TransparentDecisionMaking" ], "is_subclass_of": [ "Governance and Safety", "Governance" ], "wikilinks": [ "Governance", "metaverse" ] }, { "id": "transparent-governance", "title": "Transparent Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A governance model where rules, processes, decisions, and their enforcement are openly visible, verifiable, and accessible to all stakeholders, typically implemented through blockchain technology, public smart contracts, and open data practices.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:transparent-governance", "labels": [ "Transparent Governance", "TransparentGovernance" ], "is_subclass_of": [ "AI Governance and Ethics", "Governance" ], "wikilinks": [ "Governance", "metaverse" ] }, { "id": "transparent-revenue-sharing", "title": "Transparent Revenue Sharing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A financial distribution model where revenue allocation percentages, payment calculations, and disbursement records are openly visible and independently verifiable by all stakeholders. It is typically implemented through smart contracts that automatically distribute earnings according to immutable predefined rules, with all transactions recorded on-chain. Common applications include NFT creator royalties, DAO treasury allocations, gaming guild earnings, and metaverse land rental income.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:transparent-revenue-sharing", "labels": [ "Transparent Revenue Sharing" ], "is_subclass_of": [ "Governance and Safety", "Revenue Distribution" ], "wikilinks": [ "metaverse", "Revenue Distribution" ] }, { "id": "transport-layer-security", "title": "Transport Layer Security", "domain": "security", "domain_name": "Security", "definition": "A cryptographic protocol standardised by the IETF that provides mutual authentication, confidentiality and data integrity for communications over a computer network, succeeding the deprecated Secure Sockets Layer and currently at version 1.3 (RFC 8446).", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transport-layer-security", "labels": [ "Transport Layer Security", "TLS (Transport Layer Security)" ], "is_subclass_of": [ "Cryptographic Protocol" ], "wikilinks": [ "Public Key Infrastructure", "Certificate Authority", "Encryption", "Authentication", "TLS", "Cryptographic Protocol" ] }, { "id": "transport-layer", "title": "Transport Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Transport Layer is the stratum that provides end-to-end delivery of data between endpoints over a network. It sits above the Network Layer that routes packets and below the protocol and integration strata that rely on reliable channels. It contains segmentation, flow and congestion control, and connection management.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:transport-layer", "labels": [ "Transport Layer", "Network Transport Layer", "Transport Layer Protocol" ], "is_subclass_of": [ "Network and Communication", "owl:Thing" ], "wikilinks": [ "Network Layer", "Protocol Layer", "Integration Layer", "Transmission Control Protocol", "Congestion Control", "owl:Thing", "IETF (Internet Engineering Task Force)" ] }, { "id": "transport-network", "title": "Transport Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A transport network is the underlying backhaul or fronthaul infrastructure that interconnects the access and core segments of a communications system, carrying aggregated traffic between sites over fibre, microwave or leased links. In mobile networks it links radio sites to the core network and determines achievable latency and capacity. The term also covers physical transportation infrastructure, such as roads, rail and transit systems, that a region depends on for the movement of people and goods.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:transport-network", "labels": [ "Transport Network" ], "is_subclass_of": [ "Telecommunications" ], "wikilinks": [] }, { "id": "transport-protocol", "title": "Transport Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A transport protocol is a network-layer specification that governs end-to-end delivery of data between processes, handling concerns such as multiplexing, reliability, ordering, and flow and congestion control. Examples including TCP, UDP, and QUIC sit above the network layer and provide the communication substrate over which higher-level application and agent protocols operate.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:transport-protocol", "labels": [ "Transport Protocol", "HTTPS Transport", "Network Transport Protocol", "Real-Time Transport Protocol", "Real-time Transport Protocol", "Secure Real-Time Transport Protocol" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "trapdoor-function", "title": "Trapdoor Function", "domain": "security", "domain_name": "Security", "definition": "A trapdoor function is a one-way function that is easy to compute in the forward direction but computationally infeasible to invert, except for a party holding a secret piece of information called the trapdoor. This asymmetry between forward computation and inversion underpins public-key cryptography, where the public key enables encryption or verification and the private trapdoor enables decryption or signing. Candidate trapdoor functions rest on conjectured hard problems such as integer factorisation and the discrete logarithm.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trapdoor-function", "labels": [ "Trapdoor Function" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "travel-rule-protocol", "title": "Travel Rule Protocol", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A travel rule protocol is a technical standard that lets virtual asset service providers securely exchange the originator and beneficiary information required by the FATF Travel Rule when transferring crypto-assets. Implementations such as the IVMS101 data model and the TRP and OpenVASP messaging schemes define how counterparty identity data is formatted, transmitted, and verified between regulated entities.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:travel-rule-protocol", "labels": [ "Travel Rule Protocol" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "travel-rule", "title": "Travel Rule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The Travel Rule (formally Financial Action Task Force|FATF Recommendation 16) is an anti-money-laundering regulatory requirement mandating that Virtual Asset Service Provider|Virtual Asset Service Providers (VASPs) and custodial financial institutions collect, verify, and transmit origina...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:travel-rule", "labels": [ "Travel Rule", "BC-0485-travel-rule", "Cryptocurrency Travel Rule", "Travel Rule Compliance" ], "is_subclass_of": [ "Governance and Regulation", "AML KYC Compliance", "Financial Regulation", "Counter-Terrorist Financing", "Payment Transparency Standard", "FATF Recommendation", "Anti-Money Laundering" ], "wikilinks": [ "Address Attribution", "AML Enforcement", "Anonymous Transfers", "Anti-Money Laundering", "Bank Secrecy Act", "Beneficiary Information Requirement", "Blockchain Address Attribution", "Blockchain Analytics", "Chainalysis", "Counter-Terrorist Financing", "Cross-Border Payment Transparency", "CryptoUK", "De-risking Prevention", "Decentralised Finance", "Decentralised Identity", "EBA", "EBA Travel Rule Guidelines", "FATF Recommendation", "FATF Recommendation 16", "FCA" ] }, { "id": "treasury-analytics", "title": "Treasury Analytics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Treasury analytics is the analysis of an organisation's or DAO's financial holdings, cash flows, and asset allocation to inform liquidity, runway, and risk-management decisions. In the on-chain context it aggregates wallet balances, token positions, and protocol revenues into dashboards and metrics that support transparent, data-driven treasury governance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:treasury-analytics", "labels": [ "Treasury Analytics" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "treasury-diversification", "title": "Treasury Diversification", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Treasury diversification is the practice of spreading an organisation's or DAO's reserves across multiple asset types, such as stablecoins, native tokens, blue-chip crypto, and yield-bearing positions, to reduce concentration risk and stabilise runway. It mitigates the volatility and correlation exposure that arises when a treasury holds predominantly its own governance token.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:treasury-diversification", "labels": [ "Treasury Diversification" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "treasury-management", "title": "Treasury Management", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Treasury management in Blockchain Network|blockchain and DAO|decentralised autonomous organisation (DAO) contexts encompasses the governance-directed custody, allocation, diversification, yield generation, and reporting of organisational funds held in Smart Contract|smart contract-con...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:treasury-management", "labels": [ "Treasury Management", "BC-0464-treasury-management", "Corporate Treasury Management", "DAO Treasury Management", "On-Chain Treasury Management", "Treasury Management System", "TreasuryManagement" ], "is_subclass_of": [ "Governance and Regulation", "Financial Management", "DAO Governance", "Corporate Finance", "Digital Asset Management", "Protocol Economics", "Institutional Asset Management" ], "wikilinks": [ "Aave", "AccountingLayer", "AICPA Digital Asset Practice Aid 2022", "Arbitrum", "Audit Function", "Bitcoin Standard", "Block Explorer", "Central Bank Reserve Management", "Coinshift", "Community Grant Funding", "ComplianceLayer", "Compound", "Corporate Finance", "CorporateFinanceDomain", "Custody Infrastructure", "CustodyLayer", "DAO Legal Frameworks", "DAO Sustainability", "DeFiDomain", "DeFi Ecosystem" ] }, { "id": "treatment-planning-ai", "title": "Treatment Planning AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Artificial intelligence systems that optimise therapeutic interventions by integrating patient-specific clinical data, treatment guidelines, outcome predictions, and resource constraints to generate evidence-based, individualised care plans. Core capabilities include automated treatment protocol selection, dosage optimisation, radiotherapy dose distribution planning, surgical simulation, and adverse event risk stratification.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:treatment-planning-ai", "labels": [ "Treatment Planning AI" ], "is_subclass_of": [ "AI Application", "Medical AI" ], "wikilinks": [ "Radiation Therapy", "Medical AI", "Medical Diagnosis AI", "MetaverseDomain" ] }, { "id": "treatment-planning", "title": "Treatment Planning", "domain": "ai", "domain_name": "Ai", "definition": "The clinical process of deciding how a patient's condition will be treated: integrating diagnosis, imaging, patient history, and evidence-based guidelines into a specific course of therapy with defined goals, doses, sequencing, and review points; in domains such as radiotherapy and surgery it is a quantitative, imaging-driven activity that medical image analysis and medical imaging AI systems support but do not replace.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:treatment-planning", "labels": [ "Treatment Planning" ], "is_subclass_of": [ "Healthcare" ], "wikilinks": [ "Healthcare", "Medical Imaging AI", "Clinical Decision Support" ] }, { "id": "tree-of-thoughts", "title": "tree of thoughts", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Tree of Thoughts (ToT) is a deliberate reasoning framework for large language models that generalises linear chain-of-thought prompting into a tree-structured search over intermediate thought steps. At each node in the tree the model generates multiple candidate next-thoughts, evaluates their promise using an LLM-based heuristic, and selects branches to expand via breadth-first or depth-first search with backtracking. This enables systematic exploration of alternative reasoning paths and is particularly effective for combinatorial planning, mathematical proof construction, or multi-step problem-solving where greedy left-to-right decoding is insufficient. ToT was formalised by Yao et al. (2023) and represents the foundational instance of inference-time compute scaling through structured search.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:tree-of-thoughts", "labels": [ "Tree of Thoughts", "Tree of Thought" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "trezor", "title": "Trezor", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trezor is a line of hardware wallets developed by SatoshiLabs that store cryptocurrency private keys in a tamper-resistant, air-gapped physical device, preventing key exposure to internet-connected hosts. Transactions are signed entirely within the secure element of the device, meaning the private key never leaves the hardware even during active use. Trezor devices implement BIP32 hierarchical deterministic (HD) wallet derivation, BIP39 mnemonic seed phrases, and BIP44 multi-account structures, making them a foundational reference implementation for open-source hardware wallet design. As a mature consumer product launched in 2014, Trezor represents the established category of cold-storage self-custody devices that underpin non-custodial asset management in decentralised finance.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:trezor", "labels": [ "Trezor" ], "is_subclass_of": [ "Cold Storage" ], "wikilinks": [ "Private Key", "Self-Custody", "Key Management", "Cold Storage" ] }, { "id": "triangulation", "title": "Triangulation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Triangulation is the process of determining the 3D position of a point by intersecting lines of sight from two or more known viewpoints. Given calibrated cameras and corresponding image observations, it recovers depth and structure by solving for the point that best explains the rays. Triangulation is a core operation in stereo vision, photogrammetry, structure-from-motion, and positioning systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:triangulation", "labels": [ "Triangulation" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "triple-store", "title": "Triple Store", "domain": "data", "domain_name": "Data", "definition": "A triple store is a database purpose-built to store and query RDF data as subject-predicate-object triples, forming the storage layer for semantic-web and knowledge-graph applications. It supports the SPARQL query language and often provides reasoning and inference over ontologies, enabling expressive graph traversal and entailment beyond what relational stores offer.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:triple-store", "labels": [ "Triple Store" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "triton-inference-server", "title": "Triton Inference Server", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Triton Inference Server is NVIDIA's open-source platform for serving machine-learning models in production across CPUs and GPUs. It supports multiple frameworks through a common interface, batches and schedules concurrent requests, and exposes models over HTTP and gRPC. Triton is a standard component of GPU-accelerated inference stacks, often paired with TensorRT-optimised models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:triton-inference-server", "labels": [ "Triton Inference Server" ], "is_subclass_of": [ "Inference Serving", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [] }, { "id": "trojan-attack", "title": "Trojan Attack", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Trojan Attack is a supply-chain-oriented backdoor attack on AI models in which a pre-trained model or training pipeline is maliciously modified to embed hidden triggers; the model performs normally on clean inputs but produces attacker-defined outputs when specific trigger patterns are present. Unlike general backdoor attacks, Trojan attacks emphasise persistence through fine-tuning and distribution via public model repositories.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:trojan-attack", "labels": [ "Trojan Attack" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "tron-blockchain", "title": "Tron Blockchain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Tron is a public, high-throughput layer-1 blockchain launched in 2018 that uses a delegated proof-of-stake consensus mechanism to achieve low transaction fees and fast block times, positioning itself as an alternative settlement layer for content and payments. It has become one of the largest hosts of USD-pegged stablecoin transfers, particularly Tether (USDT), owing to its low fees relative to Ethereum. Its EVM-compatible virtual machine allows Ethereum smart contracts to be ported to the network with minimal modification.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:tron-blockchain", "labels": [ "Tron Blockchain" ], "is_subclass_of": [ "Layer 1 Blockchain" ], "wikilinks": [] }, { "id": "tropopause", "title": "Tropopause", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tropopause", "labels": [ "Tropopause" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "troposphere", "title": "Troposphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:troposphere", "labels": [ "Troposphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "tropospheric-delay", "title": "Tropospheric Delay", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tropospheric-delay", "labels": [ "Tropospheric Delay" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "true-anomaly", "title": "True Anomaly", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:true-anomaly", "labels": [ "True Anomaly" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "true-negative", "title": "True Negative", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A true negative is an outcome in binary classification where the model correctly predicts the negative class for an instance that is genuinely negative. It is one of the four cells of the confusion matrix and contributes to metrics such as specificity and accuracy. Counting true negatives is essential for evaluating how well a classifier avoids false alarms on negative cases.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:true-negative", "labels": [ "True Negative" ], "is_subclass_of": [ "Confusion Matrix", "AI Technique" ], "wikilinks": [] }, { "id": "true-positive-rate", "title": "True Positive Rate", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "True positive rate is a classification metric measuring the proportion of actual positive cases that a model correctly identifies, also known as sensitivity or recall.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:true-positive-rate", "labels": [ "True Positive Rate" ], "is_subclass_of": [ "Evaluation Metric", "Model Performance" ], "wikilinks": [ "Confusion Matrix", "Sensitivity", "Recall", "Evaluation Metric" ] }, { "id": "true-positive", "title": "True Positive", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A True Positive is an outcome in binary classification where the model correctly predicts the positive class for an instance that genuinely belongs to that class. It is one of the four cells of the confusion matrix, alongside false positives, true negatives and false negatives. Counts of true positives are central to evaluation metrics such as precision, recall and the F1 score.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:true-positive", "labels": [ "True Positive" ], "is_subclass_of": [ "Confusion Matrix", "Model Performance" ], "wikilinks": [] }, { "id": "trust-anchor", "title": "Trust Anchor", "domain": "security", "domain_name": "Security", "definition": "A trust anchor is an authoritative entity or cryptographic artifact that serves as the root of a chain of trust, providing the foundational level of trust from which all subsequent trust assertions in a system are derived. In public key infrastructure (PKI), the trust anchor is typically a root certificate authority (CA) whose self-signed certificate is pre-installed in operating systems and browsers as inherently trusted, enabling the validation of all subordinate certificate chains. In decentralised identity systems, trust anchors may be governance-designated issuers, DID-based root entities, or on-chain registries whose cryptographic material is considered authoritative for a particular trust domain. Trust anchors define the scope and boundaries of a trust framework and are the entities that must be trusted by all relying parties in the system.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trust-anchor", "labels": [ "Trust Anchor" ], "is_subclass_of": [ "Trust Framework" ], "wikilinks": [] }, { "id": "trust-architecture", "title": "Trust Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The structural framework of protocols, technologies, and governance mechanisms that establish, verify, and maintain trust relationships between users, platforms, and services in decentralized digital environments and metaverse ecosystems. It encompasses identity verification through Decentralised Identifiers (DIDs), cryptographically signed Verifiable Credentials, smart-contract-enforced rules, and consensus mechanisms, supporting both zero-trust security models and reputation-based trust scoring across cross-platform interactions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:trust-architecture", "labels": [ "Trust Architecture" ], "is_subclass_of": [ "Governance and Safety", "Security Architecture" ], "wikilinks": [ "metaverse", "Security Architecture" ] }, { "id": "trust-building", "title": "Trust Building", "domain": "governance", "domain_name": "Governance", "definition": "Trust building is the deliberate process of establishing confidence among participants in a digital system through transparency, accountability, reliable behaviour, and clear communication. In online communities and AI-mediated platforms it is fostered by practices such as content provenance, disclosure of automated generation, and enforced community standards that make actors and outputs verifiable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trust-building", "labels": [ "Trust Building", "Trust Formation", "User Trust Building" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "trust-establishment", "title": "Trust Establishment", "domain": "security", "domain_name": "Security", "definition": "Trust Establishment is the process by which parties in a digital or physical system form justified confidence in each other's identities, capabilities, intentions, and assertions prior to exchanging sensitive information or delegating authority. It encompasses cryptographic mechanisms such as certificate chain validation, attestation, and verifiable credential presentation, as well as organisational mechanisms including trust frameworks, legal agreements, and reputation systems. In decentralised and multi-stakeholder environments, trust establishment must operate without relying on a single trusted authority, requiring distributed protocols such as web-of-trust models, blockchain-anchored attestations, and federated identity systems. Trust establishment is a foundational prerequisite for secure communication, authorisation, and coordination across organisational and jurisdictional boundaries.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trust-establishment", "labels": [ "Trust Establishment", "Trust Chain Establishment" ], "is_subclass_of": [ "Trust Architecture" ], "wikilinks": [] }, { "id": "trust-framework-policy", "title": "Trust Framework Policy", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Set of rules and requirements governing participant behavior, accountability, and interoperability in federated digital identity ecosystems within metaverse environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:trust-framework-policy", "labels": [ "Trust Framework Policy" ], "is_subclass_of": [ "Governance and Safety", "Metaverse governance and safeguarding" ], "wikilinks": [ "Accountability Framework", "Audit Mechanism", "Authentication Protocol", "Authorization Framework", "Certification Criteria", "EU eIDAS 2.0", "Federated Identity System", "Interoperable Authentication", "OECD AI Governance", "OpenID Foundation", "Policy Rule Set", "Trust Anchor", "Trust Federation", "Cross-Platform Identity", "DataLayer", "Digital Identity Standards", "Governance Structure", "Legal Framework", "MiddlewareLayer", "Regulatory Compliance" ] }, { "id": "trust-framework", "title": "Trust Framework", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Structured normative architecture that defines the policies, technical standards, legal rules, certification requirements, and governance mechanisms enabling organisations and individuals to establish, assert, verify, and maintain digital trust relationships across participating entities in an ec...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:trust-framework", "labels": [ "Trust Framework" ], "is_subclass_of": [ "Governance and Regulation", "Framework", "Governance Framework", "Policy Framework", "Certification Scheme" ], "wikilinks": [ "Accreditation Body", "Ad Hoc Identity", "Assurance Level", "Authentication", "CertificationLayer", "Certification Scheme", "Conformity Assessment", "Conformity Assessment Body", "Credential Issuer", "Credential Portability", "Cross-Border Authentication", "Cryptographic Proof", "Data Protection Law", "Decentralised Identifiers", "Decentralised Identity", "Digital Credentials", "DigitalTrustDomain", "DSIT", "eIDAS 2", "ETSI" ] }, { "id": "trust-infrastructure", "title": "Trust Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Foundational systems providing authentication, authorisation, encryption, and trust establishment mechanisms. Enables secure interaction between participants in distributed systems through cryptographic protocols, identity verification, and access control policies. Comprises the hardware, software, and procedural controls that collectively guarantee confidentiality, integrity, and availability of trusted communications.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:trust-infrastructure", "labels": [ "Trust Infrastructure" ], "is_subclass_of": [ "Security and Identity", "Infrastructure" ], "wikilinks": [ "Authentication", "Authorisation", "Certificate Authority", "Data Confidentiality", "Encryption Service", "Key Management", "Narrative Gold Mine", "Access Control", "Identity Provider", "Infrastructure", "Network Layer", "Non-Repudiation", "Physical Layer", "Policy Enforcement", "Trust Framework" ] }, { "id": "trust-management", "title": "Trust Management", "domain": "security", "domain_name": "Security", "definition": "Trust Management is the systematic process of establishing, evaluating, and maintaining trust relationships between autonomous agents, services, or identities within a distributed system. It encompasses credential issuance, reputation scoring, revocation, and policy enforcement that let participants decide whether to rely on another party's claims or actions. In multi-agent and decentralised-identity systems it underpins authorisation decisions without requiring a central authority.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:trust-management", "labels": [ "Trust Management" ], "is_subclass_of": [ "Trust Framework" ], "wikilinks": [] }, { "id": "trust-mechanism", "title": "Trust Mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A trust mechanism is a system, protocol, or institution that allows parties who lack prior knowledge of one another to transact or cooperate with reasonable assurance of honest behaviour. Such mechanisms range from intermediaries and reputation systems to cryptographic and consensus protocols that replace interpersonal trust with verifiable, incentive-aligned guarantees.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trust-mechanism", "labels": [ "Trust Mechanism", "Trust Mechanisms" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "trust-minimisation", "title": "Trust Minimisation", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Trust minimisation is a design principle for distributed and decentralised systems that aims to reduce the degree to which participants must rely on any single party's honesty or competence for the system to function correctly. It is typically achieved through cryptographic proofs that let one party verify a claim without trusting the claimant, and through decentralised architectures that remove single points of control. It is a foundational goal of blockchain and decentralised web design.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:trust-minimisation", "labels": [ "Trust Minimisation" ], "is_subclass_of": [ "Decentralisation" ], "wikilinks": [] }, { "id": "trust-model", "title": "Trust Model", "domain": "security", "domain_name": "Security", "definition": "A trust model is a formal description of which entities are trusted, for what, and on what basis within a security or identity system. It defines the roots of trust, the relationships through which trust is delegated or transitively established, and the assumptions an adversary cannot violate. Trust models range from centralised certificate hierarchies to decentralised webs of trust and zero-trust architectures, and they directly shape how authentication and authorisation are designed.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trust-model", "labels": [ "Trust Model" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "trust-over-ip-foundation", "title": "trust over ip foundation", "domain": "governance", "domain_name": "Governance", "definition": "The Trust over IP (ToIP) Foundation is a Linux Foundation project established in 2020 that defines a dual-stack architecture combining cryptographic machine trust at the technical layer with human, organisational, and legal trust governance frameworks for decentralised identity. The lower technical stack encompasses Decentralised Identifiers (DIDs), Verifiable Credentials, cryptographic key management, and peer-to-peer messaging protocols across four numbered layers, whilst the upper governance stack provides trust registries, policy schemas, legal agreements, and assurance level definitions enabling parties to establish authoritative trust relationships without a central intermediary. ToIP produces the governance meta-model and interoperability specifications needed to make self-sovereign identity deployments composable and trustworthy at ecosystem scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:trust-over-ip-foundation", "labels": [ "Trust Over IP Foundation" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "trust-registry", "title": "Trust Registry", "domain": "security", "domain_name": "Security", "definition": "A trust registry is an authoritative, queryable record that lists which issuers, verifiers, and credential types are recognised as trustworthy within a given governance framework. It allows a relying party to programmatically determine whether a presented verifiable credential comes from an accredited source and remains valid. Trust registries underpin scalable decentralised identity ecosystems by providing the machine-readable basis for trust decisions.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:trust-registry", "labels": [ "Trust Registry" ], "is_subclass_of": [ "Trust Framework" ], "wikilinks": [] }, { "id": "trust-score-metric", "title": "Trust Score Metric", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A quantitative measurement representing an entity's trustworthiness, credibility, or risk level, expressed as a numerical value with associated confidence intervals and time validity, used to inform authorization decisions, transaction approvals, and access control policies.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:trust-score-metric", "labels": [ "Trust Score Metric", "Trust Score" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Access Control Decisions", "Behavioral Data", "Calculation Timestamp", "Confidence Interval", "Data Quality Metrics", "ETSI GS MEC", "Score Value", "Scoring Methodology Reference", "Statistical Models", "Transaction Approval", "Validation Rules", "Validity Period", "Calculation Parameters", "Identity Verification", "MiddlewareLayer", "Reputation Scoring Model", "Risk Assessment", "TrustAndGovernanceDomain", "Trust Infrastructure" ] }, { "id": "trust-service-provider", "title": "Trust Service Provider", "domain": "security", "domain_name": "Security", "definition": "A trust service provider (TSP) is an entity that issues and manages trust services such as electronic signatures, seals, timestamps and website authentication certificates. Under regimes like the EU eIDAS regulation, qualified TSPs meet stringent audit and security requirements so that the services they provide carry defined legal effect. A TSP operates the cryptographic infrastructure \u2014 certificate issuance, timestamping authorities and validation services \u2014 that lets relying parties trust the authenticity and integrity of electronic transactions. It is a cornerstone of digital identity and electronic trust frameworks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trust-service-provider", "labels": [ "Trust Service Provider" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "trust-and-safety", "title": "Trust and Safety", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Trust and Safety (T&S) is the professional discipline, operational infrastructure, and regulatory framework that governs the detection, review, enforcement, and remediation of harmful content and abusive behaviour across digital platforms, encompassing the full spectrum from child sexual abuse ma...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:trust-and-safety", "labels": [ "Trust and Safety", "Platform Trust and Safety" ], "is_subclass_of": [ "AI Governance and Ethics", "AI Safety", "Digital Society Harms", "Platform Governance", "Online Safety", "Content Moderation", "AI Ethics", "Data Protection" ], "wikilinks": [ "Classifier Model", "CSAM Detection", "DigitalRightsDomain", "EthicsAndSocietyDomain", "Hash Matching", "Human Review Pipeline", "Incident Response", "Online Safety", "OperationsLayer", "PlatformGovernanceDomain", "Platform Integrity", "PolicyLayer", "RegulatoryComplianceLayer", "TechnicalInfrastructureLayer", "Access Control System", "Agentic Internet", "AI Adoption", "AI Ethics", "AI Liability", "AI Risks" ] }, { "id": "trust-in-automation", "title": "Trust in Automation", "domain": "robotics", "domain_name": "Robotics", "definition": "The psychological state in which a human user accepts vulnerability to an automated system's actions based on expectations that the system will perform appropriately to achieve the user's goals, despite uncertainty and the possibility of negative consequences. It represents the user's confidence in the system's reliability, competence, and integrity, and when miscalibrated manifests as over-trust (complacency) or under-trust (disuse).", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trust-in-automation", "labels": [ "Trust in Automation", "RB-1012-trust-in-automation" ], "is_subclass_of": [ "Human-Robot Interaction", "Human Factors" ], "wikilinks": [ "AI System", "Automated System", "Automation", "Automation Psychology", "Complacency", "Confidence", "Expectation", "Human-Robot Collaboration", "RB-1009-social-robotics", "RB-1011-cobot-safety-levels", "Reliance", "Situation Awareness", "System Performance", "System Reliability", "System Usage", "User Acceptance", "User Experience", "User Interface Design", "Explainability", "Human Factors" ] }, { "id": "trust-in-digital-platforms", "title": "Trust in Digital Platforms", "domain": "governance", "domain_name": "Governance", "definition": "Trust in digital platforms is the confidence users place in online services to handle their data, transactions, and interactions safely, fairly, and reliably. It is shaped by security and privacy guarantees, transparent policies, dispute resolution, and consumer-protection measures, and is a prerequisite for sustained participation and economic activity in digital markets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:trust-in-digital-platforms", "labels": [ "Trust in Digital Platforms" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "trust-over-ip-stack", "title": "Trust over IP Stack", "domain": "governance", "domain_name": "Governance", "definition": "The Trust over IP (ToIP) Stack is a four-layer architecture developed by the Trust Over IP Foundation that provides a complete framework for decentralised digital trust, combining technical protocols (verifiable credentials, DID-based peer-to-peer messaging, verifiable data registries) with governance frameworks at each layer to enable interoperable, privacy-preserving identity and credential exchange across organisational and jurisdictional boundaries. The stack mirrors the TCP/IP model by separating technical interoperability concerns from governance and policy concerns at corresponding layers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:trust-over-ip-stack", "labels": [ "Trust over IP Stack", "Trust over IP", "Trust over IP Technology Stack" ], "is_subclass_of": [ "Trust Framework" ], "wikilinks": [] }, { "id": "trust", "title": "Trust", "domain": "governance", "domain_name": "Governance", "definition": "Trust is a cognitive, relational, and institutional disposition in which an agent accepts vulnerability to the actions of another party based on a positive expectation that the trusted party will act competently, honestly, and benevolently toward the trusting party's interests. In sociotechnical and AI contexts, trust extends beyond interpersonal relationships to encompass institutional trust in organisations, systems trust in technological artefacts, and algorithmic trust in AI decision-making processes. Trust is simultaneously a psychological state, a social institution, and a design property of engineered systems; its calibration \u2014 whether systems are trusted appropriately, excessively, or insufficiently \u2014 carries profound implications for human welfare, democratic function, and the responsible deployment of AI. Across distributed and decentralised architectures, trust is increasingly operationalised via cryptographic attestation, verifiable credentials, and reputation mechanisms that replace reliance on a single central authority.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:trust", "labels": [ "Trust", "Chain of Trust", "Trust Distribution", "Trust Indicator", "Trust Relationships", "Trust Scoring" ], "is_subclass_of": [ "Social Contract" ], "wikilinks": [] }, { "id": "trusted-execution-environment", "title": "trusted execution environment", "domain": "security", "domain_name": "Security", "definition": "A Trusted Execution Environment (TEE) is a hardware-enforced, isolated processing domain within a processor that guarantees confidentiality and integrity of code and data even when the host operating system, hypervisor, or privileged firmware is compromised or untrusted. TEEs are instantiated via technologies such as Intel SGX, AMD SEV-SNP, and Arm TrustZone, each providing hardware-rooted attestation mechanisms that allow a relying party to cryptographically verify the identity, configuration, and integrity of an enclave before exchanging sensitive material. By encrypting enclave memory pages in DRAM and enforcing access control at the silicon level, TEEs form the hardware foundation for confidential computing \u2014 enabling private AI inference, secure key management, privacy-preserving multi-party collaboration, and verifiable off-chain computation in blockchain systems.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:trusted-execution-environment", "labels": [ "Trusted Execution Environment" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "trusted-execution-environments", "title": "Trusted Execution Environments", "domain": "security", "domain_name": "Security", "definition": "Trusted Execution Environments (TEEs) are hardware-enforced isolated regions of memory and compute within a processor where code and data are protected against inspection or tampering by any software outside the enclave, including privileged system software such as the operating system, hypervisor, or firmware. They use processor-level memory encryption, access-control registers, and cryptographic attestation to establish a root of trust anchored in silicon, allowing remote parties to verify the identity and integrity of the code running inside before sharing sensitive data with it. TEEs are the foundational enabling technology for confidential computing, enabling privacy-preserving computation on sensitive workloads in untrusted cloud or edge environments. Prominent implementations include Intel SGX, Intel TDX, AMD SEV-SNP, and ARM TrustZone.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:trusted-execution-environments", "labels": [ "Trusted Execution Environments" ], "is_subclass_of": [ "Confidential Computing" ], "wikilinks": [ "Confidential Computing", "Verifiable Computation", "Secure Enclave", "ARM TrustZone" ] }, { "id": "trusted-execution-pbft", "title": "Trusted Execution PBFT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trusted Execution PBFT is a variant of Practical Byzantine Fault Tolerance that leverages Trusted Execution Environments (TEEs), specifically Intel SGX, to implement a Unique Sequential Identifier Generator (USIG) that reduces the protocol to two message phases rather than the standard three. By anchoring sequence number generation inside a tamper-resistant enclave, it eliminates the view-change overhead of classical PBFT while preserving Byzantine fault safety up to f < n/3 faulty replicas.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:trusted-execution-pbft", "labels": [ "Trusted Execution PBFT" ], "is_subclass_of": [ "Protocol and Consensus", "Practical Byzantine Fault Tolerance" ], "wikilinks": [ "Blockchain", "Practical Byzantine Fault Tolerance" ] }, { "id": "trusted-platform-module", "title": "Trusted Platform Module", "domain": "security", "domain_name": "Security", "definition": "A Trusted Platform Module (TPM) is a dedicated, tamper-resistant hardware component that provides cryptographic functions and secure storage of keys and platform measurements. It generates and protects keys that never leave the chip in plaintext, records integrity measurements in platform configuration registers, and supports operations such as secure boot and remote attestation. Defined by an open Trusted Computing Group specification, the TPM acts as a hardware root of trust on personal computers, servers and embedded devices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:trusted-platform-module", "labels": [ "Trusted Platform Module" ], "is_subclass_of": [ "Hardware Security Module" ], "wikilinks": [] }, { "id": "trusted-setup-ceremony", "title": "Trusted Setup Ceremony", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A trusted setup ceremony is a one-time, multi-party procedure used to generate the public parameters required by certain zero-knowledge proof systems, such as zk-SNARKs, in which participants jointly compute the parameters while each contributes and then destroys a private random value, known as toxic waste. The scheme remains secure provided at least one participant honestly destroys their contribution, so ceremonies are typically run with many independent participants across jurisdictions to make full collusion implausible. Protocols such as Zcash have run public, auditable trusted setup ceremonies to generate the parameters underlying their privacy-preserving transactions.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:trusted-setup-ceremony", "labels": [ "Trusted Setup Ceremony" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [] }, { "id": "trusted-setup", "title": "Trusted Setup", "domain": "security", "domain_name": "Security", "definition": "A trusted setup is a one-time procedure that generates the public parameters (a common reference string) required by certain cryptographic protocols, notably succinct zero-knowledge proof systems. The procedure produces secret randomness, often called toxic waste, that must be irrecoverably destroyed; if it leaks, an adversary can forge proofs. Multi-party ceremonies distribute trust so that the setup remains sound as long as a single participant behaves honestly.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trusted-setup", "labels": [ "Trusted Setup" ], "is_subclass_of": [ "Cryptography" ], "wikilinks": [] }, { "id": "trusted-third-party", "title": "Trusted Third Party", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A trusted third party (TTP) is an intermediary that two or more parties rely upon to facilitate, witness, or settle an interaction without each party having to trust the other directly. Examples include certificate authorities, escrow agents, custodians, and clearing houses. Blockchain systems are largely motivated by the goal of minimising or eliminating trusted third parties, replacing institutional trust with cryptographic verification and decentralised consensus.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:trusted-third-party", "labels": [ "Trusted Third Party" ], "is_subclass_of": [ "Security" ], "wikilinks": [] }, { "id": "trusted-timestamping", "title": "Trusted Timestamping", "domain": "security", "domain_name": "Security", "definition": "Trusted timestamping is the process of securely recording the time at which a piece of data existed, using a trusted third party that cryptographically binds a hash of the data to an authoritative time and signs the result. Standardised by RFC 3161, it produces a timestamp token that proves data integrity and existence-by-time without revealing the data, and is used in digital signatures, legal record-keeping, intellectual-property protection, and regulatory compliance. It establishes a verifiable temporal ordering that survives even after signing keys expire.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trusted-timestamping", "labels": [ "Trusted Timestamping", "Timestamping", "Trusted Time Source" ], "is_subclass_of": [ "Timestamping Service" ], "wikilinks": [] }, { "id": "trustless-coordination", "title": "Trustless Coordination", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trustless coordination is the achievement of agreement and joint action among parties who do not trust one another, enforced by protocol rules and cryptography rather than a trusted intermediary.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:trustless-coordination", "labels": [ "Trustless Coordination" ], "is_subclass_of": [ "Blockchain", "Consensus Mechanism" ], "wikilinks": [ "Cryptography", "Smart Contract", "Interoperability", "Blockchain" ] }, { "id": "trustless-execution", "title": "Trustless Execution", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trustless execution is the property of a computational system whereby parties can rely on the correct execution of agreed logic without trusting any single operator, intermediary or counterparty. Blockchains achieve it by combining deterministic smart-contract code, replicated execution across many independent nodes, consensus mechanisms that make history tamper-evident, and cryptographic verification, so that outcomes \u2014 payments, settlements, state transitions \u2014 follow from code and verifiable data rather than institutional promises.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trustless-execution", "labels": [ "Trustless Execution" ], "is_subclass_of": [ "Decentralisation" ], "wikilinks": [ "Decentralisation", "Smart Contract", "Consensus Mechanism", "Decentralized Application" ] }, { "id": "trustless-settlement", "title": "Trustless Settlement", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Trustless settlement is a mechanism for finalising financial or asset-transfer transactions in which parties need not extend personal or institutional trust to a counterparty or intermediary because correctness is guaranteed by cryptographic protocols and verifiable distributed consensus. The term reflects the displacement of counterparty trust with algorithmic certainty: smart contracts hold and release assets conditionally on cryptographically verifiable proofs, and immutable ledger state provides objective post-settlement finality. Trustless settlement underpins DeFi protocols, atomic swaps, and cross-chain bridges, enabling transaction finality without custodial banks, clearinghouses, or escrow agents.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trustless-settlement", "labels": [ "Trustless Settlement" ], "is_subclass_of": [ "Atomic Settlement" ], "wikilinks": [] }, { "id": "trustless-transaction", "title": "Trustless Transaction", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A trustless transaction is an exchange of value or commitments that completes correctly without either party having to trust the other or a central intermediary. Its guarantees come instead from cryptography, consensus and protocol design that make cheating detectable or impossible. Trustless transactions are a defining capability of blockchains and underpin atomic swaps, payment channels and smart-contract settlement.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:trustless-transaction", "labels": [ "Trustless Transaction" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "trustworthy-ai-framework", "title": "Trustworthy AI Framework", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Comprehensive governance and standards framework establishing principles, requirements, and assessment processes to ensure AI systems are lawful, ethical, and technically robust throughout their lifecycle. Defined by the EU High-Level Expert Group on AI and formalised in the EU AI Act, it implements a risk-based approach with seven trustworthiness dimensions covering human oversight, safety, privacy, transparency, fairness, societal wellbeing, and accountability.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:trustworthy-ai-framework", "labels": [ "Trustworthy AI Framework" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "EU HLEG AI", "ISO/IEC 42001:2023", "AIEthicsDomain", "ConceptualLayer", "EU AI Act" ] }, { "id": "trustworthy-ai-systems", "title": "Trustworthy AI Systems", "domain": "ai", "domain_name": "Ai", "definition": "Trustworthy AI systems are artificial-intelligence systems designed and operated to be reliable, safe, transparent, fair, accountable, and respectful of privacy throughout their lifecycle. The concept, codified in frameworks such as the NIST AI Risk Management Framework and the EU's trustworthy-AI guidelines, integrates technical robustness with governance so that stakeholders can justifiably rely on the system's behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:trustworthy-ai-systems", "labels": [ "Trustworthy AI Systems" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "trustworthy-ai", "title": "trustworthy ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Trustworthy AI denotes artificial intelligence systems designed, developed, and operated in accordance with a multi-criteria framework of lawfulness, ethical soundness, technical robustness, fairness, transparency, accountability, and societal wellbeing, ensuring they produce reliable and safe outcomes throughout their full lifecycle. The concept is most concretely codified in the EU High-Level Expert Group on AI's seven requirements for trustworthy AI and operationalised through the EU AI Act's risk-tiered conformity obligations and the NIST AI Risk Management Framework. Achieving trustworthy AI demands convergent technical measures\u2014such as explainability, robustness testing, and bias mitigation\u2014alongside governance instruments including impact assessments, model documentation, algorithmic audits, and meaningful human oversight mechanisms. It functions as an integrating standard that bridges AI safety, AI ethics, and AI regulation into a coherent assurance regime applicable across the full AI system lifecycle.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:trustworthy-ai", "labels": [ "Trustworthy AI" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "tundra", "title": "Tundra", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:tundra", "labels": [ "Tundra" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "turing-machine", "title": "Turing Machine", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Turing machine is an abstract model of computation consisting of an infinite tape, a read-write head, a finite set of states and a transition function that, given the current state and the symbol under the head, prescribes a symbol to write, a direction to move and a next state. Introduced by Alan Turing, it formalises the notion of an effective procedure and serves as the canonical definition of what is computable. The Church-Turing thesis holds that any function computable by any reasonable mechanism is computable by a Turing machine.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:turing-machine", "labels": [ "Turing Machine" ], "is_subclass_of": [ "Automata Theory" ], "wikilinks": [] }, { "id": "turtle", "title": "Turtle", "domain": "data", "domain_name": "Data", "definition": "Turtle (Terse RDF Triple Language) is a W3C-standardised serialisation syntax for expressing RDF graphs in a compact, human-readable plain-text form. It extends Notation3 (N3) by providing a concise prefix-based shorthand for IRIs, support for blank nodes, and literal datatypes, enabling structured linked-data documents to be authored and exchanged without the verbosity of RDF/XML.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:turtle", "labels": [ "Turtle", "Turtle Serialisation", "Turtle Syntax" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [] }, { "id": "twap-oracle", "title": "Twap Oracle", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A TWAP oracle is a price oracle that reports the time-weighted average price of an asset over a chosen window rather than its instantaneous spot price. By accumulating price-time observations and dividing by elapsed time, it produces a smoothed figure that is expensive to manipulate within a single block or short interval. TWAP oracles are widely deployed by on-chain automated market makers to supply manipulation-resistant price feeds to lending, derivatives, and liquidation systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:twap-oracle", "labels": [ "Twap Oracle", "TWAP Oracle" ], "is_subclass_of": [ "Price Oracle" ], "wikilinks": [] }, { "id": "twelve-factor-app", "title": "Twelve Factor App", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Twelve-Factor App is a methodology for building software-as-a-service applications that are portable, resilient, and suitable for deployment on modern cloud platforms. It defines twelve guidelines covering codebase management, declared dependencies, configuration in the environment, backing services, build-release-run separation, stateless processes, port binding, concurrency, disposability, dev/prod parity, logs as event streams, and admin tasks. Adherence yields applications that scale horizontally, deploy continuously, and integrate cleanly with containerised and orchestrated infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:twelve-factor-app", "labels": [ "Twelve Factor App", "Twelve-Factor App" ], "is_subclass_of": [ "Cloud-Native Applications" ], "wikilinks": [] }, { "id": "twisted-edwards-curve", "title": "Twisted Edwards Curve", "domain": "security", "domain_name": "Security", "definition": "A twisted Edwards curve is a form of elliptic curve, defined by the equation ax squared plus y squared equals 1 plus dx squared y squared, whose complete addition law has no exceptional cases, making implementations naturally resistant to certain side-channel and invalid-curve attacks. Curve25519 in its twisted Edwards form, Ed25519, is the most widely deployed instance, chosen for its combination of speed and misuse resistance. It is the underlying curve group over which the EdDSA signature scheme performs its arithmetic.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:twisted-edwards-curve", "labels": [ "Twisted Edwards Curve" ], "is_subclass_of": [ "Elliptic Curve Cryptography" ], "wikilinks": [] }, { "id": "two-way-peg", "title": "Two Way Peg", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A two-way peg (2WP) is a blockchain mechanism that enables assets to be transferred bidirectionally between a parent chain and a sidechain, with the asset supply conserved across both chains. When an asset is locked on the parent chain, an equivalent representation is minted on the sidechain; when returned to the parent chain, the sidechain representation is burned and the original asset is unlocked. Two-way pegs are the foundational primitive for sidechain interoperability, enabling specialised execution environments whilst tethering their native asset to a more secure base layer.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:two-way-peg", "labels": [ "Two Way Peg", "Two-Way Peg" ], "is_subclass_of": [ "Blockchain", "Sidechain" ], "wikilinks": [] }, { "id": "two-phase-commit", "title": "Two-Phase Commit", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "Two-Phase Commit (2PC) is a distributed transaction coordination protocol that ensures atomic commitment across multiple participant nodes: either all participants commit a transaction or all abort it, with no partial updates persisted. In the prepare phase, a coordinator polls all participants for their readiness to commit; in the commit phase, it broadcasts the final decision based on unanimous consensus from the prepare phase. 2PC is the foundational protocol for distributed ACID transactions but is blocking in the presence of coordinator failure, a limitation addressed by Three-Phase Commit and Paxos-based variants.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:two-phase-commit", "labels": [ "Two-Phase Commit", "Prepare-Commit Phase", "Two Phase Commit" ], "is_subclass_of": [ "Distributed System Protocol" ], "wikilinks": [] }, { "id": "two-tier-distribution-model", "title": "Two-Tier Distribution Model", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The two-tier distribution model is the architecture commonly proposed for central bank digital currencies in which the central bank issues the digital currency but commercial banks and licensed intermediaries handle distribution, wallets, and customer service. This preserves the central bank's monetary authority while leveraging existing financial institutions for onboarding, compliance, and the user-facing layer.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:two-tier-distribution-model", "labels": [ "Two-Tier Distribution Model" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "two-body-problem", "title": "Two-body Problem", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:two-body-problem", "labels": [ "Two-body Problem" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "type-system", "title": "Type System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A type system is a formal component of a programming language that classifies values and expressions into types and defines rules constraining how they may be combined. It is enforced by a checker that rejects programs violating these rules, preventing classes of errors before or during execution. Type systems vary along axes such as static versus dynamic checking, strength, and expressiveness, trading safety guarantees against flexibility.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:type-system", "labels": [ "Type System" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "type-theory", "title": "Type Theory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Type theory is a branch of mathematical logic and theoretical computer science in which every term has an associated type, and well-formedness is governed by typing rules rather than by raw set membership. It serves both as a foundation for mathematics and as the formal basis for type systems in programming languages, where types constrain valid expressions. Through the propositions-as-types correspondence, proofs become programs and types become specifications, linking logic, computation and formal verification.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:type-theory", "labels": [ "Type Theory" ], "is_subclass_of": [ "Mathematical Logic" ], "wikilinks": [] }, { "id": "type-script", "title": "TypeScript", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "TypeScript is a statically typed superset of JavaScript developed and maintained by Microsoft that adds optional type annotations, interfaces, generics, and compile-time type checking to the JavaScript language. Source code is transpiled by the TypeScript compiler (tsc) to plain JavaScript, making it compatible with any JavaScript runtime or browser without requiring runtime changes. The type system enables large-scale application development by surfacing errors at development time, improving editor tooling such as intelligent autocompletion, safe refactoring, and inline documentation. TypeScript has become the dominant choice for enterprise front-end and Node.js back-end development, underpinning major frameworks such as Angular, NestJS, and Deno.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:type-script", "labels": [ "TypeScript" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [ "Programming Language", "Software Engineering" ] }, { "id": "typing-indicators", "title": "Typing Indicators", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Typing indicators are real-time signals that inform participants when another user is actively composing a message in a shared conversation thread. They create a sense of conversational presence and reduce the likelihood of duplicate or crossed messages in synchronous exchanges. As a social cue, they contribute to turn-taking norms similar to those in face-to-face dialogue.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:typing-indicators", "labels": [ "Typing Indicators" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "u-net", "title": "U-Net", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A convolutional neural network architecture with a symmetric encoder-decoder structure and skip connections, originally designed for biomedical image segmentation and widely adopted for dense prediction tasks including diffusion model denoising.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:u-net", "labels": [ "U-Net", "Trainable Encoder Copy" ], "is_subclass_of": [ "Convolutional Neural Network" ], "wikilinks": [ "Convolution", "Encoder", "Decoder", "Image Segmentation", "Semantic Segmentation", "Convolutional Neural Network" ] }, { "id": "ucl", "title": "UCL", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The common abbreviation for University College London, a major research university in London active in computing and artificial intelligence research.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:ucl", "labels": [ "UCL" ], "is_subclass_of": [ "University College London" ], "wikilinks": [ "Machine Learning", "University College London", "Alan Turing Institute" ] }, { "id": "udp", "title": "UDP", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The User Datagram Protocol (UDP) is a connectionless transport-layer protocol that sends discrete datagrams without establishing a session, handshaking, ordering or guaranteed delivery. By omitting the reliability and congestion-control machinery of connection-oriented protocols, UDP achieves low latency and minimal overhead, leaving any required reliability to the application layer. It is the foundation for real-time and high-throughput workloads such as voice, video, gaming, DNS and modern transport protocols built atop it.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:udp", "labels": [ "UDP" ], "is_subclass_of": [ "Transport Protocol" ], "wikilinks": [] }, { "id": "ui-code-generation", "title": "UI Code Generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The automated process of translating visual design specifications or mockups into functional user interface code using AI models.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:ui-code-generation", "labels": [ "UI Code Generation" ], "is_subclass_of": [ "Code Generation" ], "wikilinks": [] }, { "id": "uk-ai-opportunities-action-plan", "title": "UK AI Opportunities Action Plan", "domain": "governance", "domain_name": "Governance", "definition": "The UK AI Opportunities Action Plan is a 2025 UK Government strategy setting out recommendations to accelerate national adoption of artificial intelligence across the public and private sectors. It covers building sovereign compute and data infrastructure, expanding AI skills, establishing AI Growth Zones, and embedding AI in public services to drive productivity and economic growth.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:uk-ai-opportunities-action-plan", "labels": [ "UK AI Opportunities Action Plan" ], "is_subclass_of": [ "AI Governance Framework" ], "wikilinks": [] }, { "id": "uk-ai-safety-institute", "title": "uk ai safety institute", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The UK AI Safety Institute (AISI) is a UK government body established in November 2023 within the Department for Science, Innovation and Technology to evaluate the safety properties of frontier AI models both before and after deployment. It conducts empirical research into emergent AI risks\u2014including dangerous capability elicitation, deceptive alignment, and autonomous replication\u2014and coordinates internationally with counterpart bodies such as the US AI Safety Institute on shared evaluation methodologies and protocols. AISI operates Inspect, an open-source AI evaluation framework, publishes model evaluation reports to inform UK regulatory and procurement decisions, and serves as the primary technical interface between the UK government and leading AI developers for pre-deployment safety assessments. Its work is grounded in the Bletchley Declaration and feeds into broader multilateral AI governance processes including OECD and UN-level deliberations.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:uk-ai-safety-institute", "labels": [ "UK AI Safety Institute", "UK AI Security Institute" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "uk-company-financial-filing-obligations", "title": "UK Company Financial Filing Obligations", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Accounts denotes the financial reporting and filing obligations of a limited company under UK law, encompassing annual accounts preparation, Companies House submission deadlines, confirmation statements, and the penalties regime for late filing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:uk-company-financial-filing-obligations", "labels": [ "UK Company Financial Filing Obligations", "Accounts" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [ "Blockchain" ] }, { "id": "uk-diatf", "title": "UK DIATF", "domain": "security", "domain_name": "Security", "definition": "The UK Digital Identity and Attributes Trust Framework (DIATF) is a government-published set of rules and standards that digital-identity providers must meet to be certified as trustworthy. It defines requirements for security, privacy, fraud management, and inclusion so that certified identity and attribute services can be relied upon across public and private sectors.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:uk-diatf", "labels": [ "UK DIATF" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "uk-data-protection-act-2018", "title": "UK Data Protection Act 2018", "domain": "governance", "domain_name": "Governance", "definition": "The UK Data Protection Act 2018 is the primary UK legislation governing the processing of personal data, implementing and supplementing the UK GDPR and providing the legal framework for data subjects' rights and controllers' obligations. It sets lawful bases for processing, principles such as data minimisation and purpose limitation, and is enforced by the Information Commissioner's Office.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:uk-data-protection-act-2018", "labels": [ "UK Data Protection Act 2018", "Data Protection Act 2018" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "uk-gdpr", "title": "UK GDPR", "domain": "governance", "domain_name": "Governance", "definition": "UK GDPR is the retained EU law version of the General Data Protection Regulation that came into force in the United Kingdom on 1 January 2021 following the end of the Brexit transition period, implemented alongside the Data Protection Act 2018. It preserves the core principles of the EU GDPR \u2014 lawfulness, fairness, transparency, purpose limitation, data minimisation, accuracy, storage limitation, integrity, and accountability \u2014 while allowing the UK government to diverge from EU rules through domestic legislation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:uk-gdpr", "labels": [ "UK GDPR" ], "is_subclass_of": [ "Data Protection Regulation" ], "wikilinks": [] }, { "id": "uk-mlr-2017", "title": "UK MLR 2017", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The UK Money Laundering Regulations 2017 (MLR 2017) are the principal UK regulations transposing anti-money-laundering and counter-terrorist-financing requirements onto regulated firms, including financial institutions and crypto-asset businesses. They mandate customer due diligence, risk assessment, record-keeping, and reporting obligations, and bring virtual-asset service providers under supervision by the FCA.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:uk-mlr-2017", "labels": [ "UK MLR 2017" ], "is_subclass_of": [ "Financial Regulation" ], "wikilinks": [] }, { "id": "uk-national-ai-strategy", "title": "UK National AI Strategy", "domain": "governance", "domain_name": "Governance", "definition": "The UK government's ten-year plan for artificial intelligence, published in September 2021, organised around three pillars: investing in the long-term needs of the AI ecosystem, ensuring AI benefits all sectors and regions, and governing AI effectively. It committed to skills programmes, compute and research investment through bodies including the Alan Turing Institute, and a pro-innovation regulatory approach, and was subsequently extended by the AI Opportunities Action Plan.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:uk-national-ai-strategy", "labels": [ "UK National AI Strategy" ], "is_subclass_of": [ "National AI Strategy" ], "wikilinks": [ "National Ai Strategy", "Alan Turing Institute", "AI Investment", "Industrial Strategy" ] }, { "id": "uk-online-safety-act", "title": "UK Online Safety Act", "domain": "governance", "domain_name": "Governance", "definition": "The UK Online Safety Act 2023 is landmark primary legislation that imposes statutory duties of care on providers of user-to-user services and search engines operating in or accessible from the United Kingdom, compelling them to assess and mitigate risks of illegal content and content harmful to children. It designates Ofcom as the UK's online safety regulator with powers to set binding codes of practice, conduct compliance audits, and impose fines of up to ten per cent of global annual turnover for non-compliance. The Act introduces mandatory age assurance obligations for services hosting pornographic or other age-restricted content, transparency reporting requirements, and a suite of user empowerment tools designed to give adults greater control over the content they encounter. A contested set of provisions concerning end-to-end encrypted messaging \u2014 deferring client-side scanning requirements until Ofcom deems them technically feasible \u2014 reflect the Act's tension between child protection goals and the integrity of secure communications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:uk-online-safety-act", "labels": [ "UK Online Safety Act", "UK Online Safety Act 2023" ], "is_subclass_of": [ "Digital Regulation" ], "wikilinks": [] }, { "id": "uk-tech-ecosystem", "title": "UK Tech Ecosystem", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The United Kingdom's technology industry landscape, ranked as Europe's leading tech ecosystem with a combined market valuation of $1.2 trillion and over 17,000 VC-backed startups; the third-largest AI market globally after the US and China, encompassing unicorns, regional clusters (London, Cambridge), venture capital flows, and national policy frameworks for digital innovation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:uk-tech-ecosystem", "labels": [ "UK Tech Ecosystem" ], "is_subclass_of": [ "Computing and Cloud", "Software Engineering", "Technology Industry" ], "wikilinks": [ "Innovation", "Economic Development", "Innovation", "Tech Entrepreneurship", "Technology", "Technology Industry", "AI Development", "Digital Twin" ] }, { "id": "ulmfi-t", "title": "ULMFiT", "domain": "ai", "domain_name": "Ai", "definition": "ULMFiT (Universal Language Model Fine-tuning) is a transfer-learning method introduced by Howard and Ruder in 2018 that adapts a language model pre-trained on a large general corpus to downstream NLP tasks such as text classification. It popularised techniques including discriminative learning rates, slanted triangular learning rates, and gradual unfreezing, demonstrating that language-model pre-training transfers effectively to many tasks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ulmfi-t", "labels": [ "ULMFiT" ], "is_subclass_of": [ "Natural Language Processing" ], "wikilinks": [] }, { "id": "uma", "title": "UMA", "domain": "blockchain", "domain_name": "Blockchain", "definition": "UMA is a decentralised finance protocol on Ethereum that provides an optimistic oracle for reporting data to smart contracts and tooling for creating synthetic assets.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:uma", "labels": [ "UMA" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Smart Contract", "Ethereum", "DeFi" ] }, { "id": "un-global-compact", "title": "UN Global Compact", "domain": "governance", "domain_name": "Governance", "definition": "The UN Global Compact is a voluntary United Nations initiative encouraging businesses worldwide to adopt sustainable and socially responsible policies aligned with ten principles covering human rights, labour, environment, and anti-corruption. Signatory companies commit to embedding these principles in their operations and to reporting annual progress, making it the world's largest corporate sustainability initiative.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:un-global-compact", "labels": [ "UN Global Compact" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "un-guiding-principles-on-business-and-human-rights", "title": "UN Guiding Principles on Business and Human Rights", "domain": "governance", "domain_name": "Governance", "definition": "The UN Guiding Principles on Business and Human Rights (UNGPs) are a set of 31 principles endorsed by the UN Human Rights Council in 2011 that establish a global standard for preventing and addressing adverse human rights impacts linked to business activity. They rest on three pillars: the state duty to protect, the corporate responsibility to respect, and access to remedy. The framework underpins human-rights due diligence across supply chains, including ethical sourcing of materials and labour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:un-guiding-principles-on-business-and-human-rights", "labels": [ "UN Guiding Principles on Business and Human Rights" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [] }, { "id": "unesco-recommendation-on-the-ethics-of-ai", "title": "UNESCO Recommendation on the Ethics of AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The UNESCO Recommendation on the Ethics of Artificial Intelligence is the first global normative framework adopted unanimously by UNESCO's 193 Member States in November 2021, establishing shared ethical principles, values, and actionable policy guidance for responsible AI development and deployment. It articulates ten foundational principles \u2014 including proportionality, safety, fairness, sustainability, privacy, human oversight, transparency, accountability, AI literacy, and multi-stakeholder governance \u2014 and four core values centred on human dignity, peaceful societies, diversity, and environmental flourishing. As a non-binding but globally authoritative instrument, it shapes national legislation, institutional policy, and international cooperation on AI governance.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:unesco-recommendation-on-the-ethics-of-ai", "labels": [ "UNESCO Recommendation on the Ethics of AI", "UNESCO AI Ethics Recommendation" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "UNESCO Recommendation on the Ethics of AI (2021)", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "urdf-robot-model", "title": "URDF Robot Model", "domain": "robotics", "domain_name": "Robotics", "definition": "A URDF (Unified Robot Description Format) robot model is an XML specification that describes a robot's kinematic and dynamic properties, including links, joints, inertial parameters, visual meshes and collision geometry. It is the canonical model format in the ROS ecosystem, consumed by simulators, motion planners and visualisation tools to reason about a robot's physical structure. URDF files enable consistent representation of articulated robots across simulation and control software.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:urdf-robot-model", "labels": [ "URDF Robot Model" ], "is_subclass_of": [ "Robotics", "Kinematics Model" ], "wikilinks": [] }, { "id": "urdf", "title": "urdf", "domain": "robotics", "domain_name": "Robotics", "definition": "Unified Robot Description Format (URDF) is an XML-based schema used within the Robot Operating System (ROS) ecosystem to fully specify a robot's physical structure, including its kinematic chain of links and joints, collision geometries, inertial properties, and sensor placements. URDF files serve as the authoritative model consumed by simulation environments such as Gazebo, motion planners, and visualisation tools like RViz, enabling consistent robot representation across software components. Each joint element declares its type (fixed, revolute, prismatic, continuous) along with axis, limits, and damping parameters, while link elements reference mesh geometry and material definitions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:urdf", "labels": [ "URDF", "URDF Hardware Description", "URDF Specification" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "uri-canonicaliser", "title": "URI Canonicaliser", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The canonical URI minting and resolution engine implementing the VisionClaw Agentic Container|VisionClaw urn:visionclaw: grammar (ADR-013, R1\u2013R3 rules) for stable, content-addressed, and scope-bearing identifiers, enabling deterministic roundtrip serialisation and cryptographic verification o...", "entityType": "Class", "qualityScore": 0.89, "maturity": "established", "iri": "urn:ngm:class:uri-canonicaliser", "labels": [ "URI Canonicaliser" ], "is_subclass_of": [ "Network and Communication", "Identity Systems" ], "wikilinks": [ "ADR-013", "AgenticSystemsDomain", "BIP-340", "BIP-340 Pubkey", "BIP-340 Schnorr Keypair", "Blockchain Hash Function", "Content-Addressed Storage", "DataGovernanceDomain", "Decentralised Resolution", "Deterministic Serialisation", "Hash Computation", "IETF Content-Addressable Architecture", "JSON Stringification", "RFC 8141", "RFC 8141 URN Syntax", "Scope Bearer", "SemanticWebDomain", "SHA-256 Hash Function", "Slug Derivation", "Tamper Detection" ] }, { "id": "uri", "title": "URI", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A URI (Uniform Resource Identifier) is a compact sequence of characters that uniquely identifies an abstract or physical resource on a network or within a namespace. It generalises both locators, which describe how to access a resource, and names, which identify a resource independently of access, under a single syntactic framework of scheme, authority, path, query and fragment. URIs are foundational to the architecture of the World Wide Web and the Semantic Web, where they serve as globally unique identifiers for documents, data and entities.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:uri", "labels": [ "URI" ], "is_subclass_of": [ "Web Standard" ], "wikilinks": [] }, { "id": "us-ai-safety-institute", "title": "US AI Safety Institute", "domain": "governance", "domain_name": "Governance", "definition": "The US AI Safety Institute (US AISI) is the United States government body, established within the National Institute of Standards and Technology (NIST), tasked with developing the science, evaluations, and guidelines for safe and trustworthy artificial intelligence. It conducts pre-deployment testing of frontier models, produces measurement methods and red-teaming guidance, and advances standards for AI safety. As a distinct organisation from the UK AI Safety Institute, it operates under US jurisdiction and NIST's standards mandate; the two institutes coordinate on evaluations through a bilateral partnership while remaining separate national bodies with their own remits, funding, and legal contexts. In June 2025 the Trump administration renamed the body the Center for AI Standards and Innovation (CAISI), dropping 'safety' from its title and narrowing its mandate to demonstrable national-security risks while keeping it within NIST.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:us-ai-safety-institute", "labels": [ "US AI Safety Institute" ], "is_subclass_of": [ "AI Safety Institute" ], "wikilinks": [ "AI Safety Institute", "UK AI Safety Institute", "NIST" ] }, { "id": "us-regulatory-framework", "title": "US Regulatory Framework", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The US Regulatory Framework for cryptocurrency operates through a fragmented multi-agency structure in which the SEC applies securities law to tokens, FinCEN supervises money transmission and anti-money laundering compliance, the CFTC regulates commodity-classified digital assets and derivatives, and the OCC addresses banking integration. This jurisdictional overlap creates significant compliance complexity for exchanges, DeFi protocols, and digital asset issuers absent comprehensive Congressional legislation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:us-regulatory-framework", "labels": [ "US Regulatory Framework", "US Healthcare Regulation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [ "AML", "AMLKYCCompliance", "Bank Secrecy Act", "BC-0456-virtual-asset-service-providers", "BC-0457-aml-kyc-compliance", "BC-0481-fatf-recommendations", "BC-0482-eu-mica-regulation", "BitLicense", "CBDC", "CFTC", "CFTC", "Cryptocurrency Exchange", "Decentralised Finance", "ETF", "FATF", "FinCEN", "FinCEN", "Howey Test", "Initial Coin Offerings", "IRS" ] }, { "id": "us-china-ai-competition", "title": "US-China AI Competition", "domain": "governance", "domain_name": "Governance", "definition": "The geopolitical and technological rivalry between the United States and China in the development, deployment, and standardization of artificial intelligence capabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:us-china-ai-competition", "labels": [ "US-China AI Competition" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "us-china-chip-export-policy", "title": "US-China Chip Export Policy", "domain": "governance", "domain_name": "Governance", "definition": "The set of governmental regulations and trade strategies employed by the United States to restrict the export of advanced semiconductor technologies and equipment to China.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:us-china-chip-export-policy", "labels": [ "US-China Chip Export Policy" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [] }, { "id": "usb-interface", "title": "USB Interface", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A USB interface is a standardised serial connection point implementing the Universal Serial Bus specification for data transfer and power delivery between a host and a peripheral device. It provides a common physical and protocol layer that hardware wallets, storage devices, and countless other peripherals use to communicate with host computers. USB interfaces are widely chosen for security-sensitive devices because they require a physical connection, limiting remote attack surface compared with wireless alternatives.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:usb-interface", "labels": [ "USB Interface" ], "is_subclass_of": [ "Communication Interface" ], "wikilinks": [] }, { "id": "usd-pipeline", "title": "USD Pipeline", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A USD pipeline is a content production workflow built around Pixar's Universal Scene Description (USD) format, which composes, layers and exchanges complex 3D scene data across digital-content-creation tools. USD enables non-destructive collaboration through composition arcs, variants and references, making it the interchange backbone for film, games and metaverse asset production. It standardises how geometry, materials, animation and motion-capture data flow between applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usd-pipeline", "labels": [ "USD Pipeline" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [] }, { "id": "usd", "title": "USD", "domain": "finance", "domain_name": "Finance", "definition": "The United States dollar, the sovereign fiat currency issued by the United States Federal Reserve, serving as the world's primary reserve currency and the dominant unit of account against which digital assets, stablecoins, commodities, and international trade flows are quoted and settled.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usd", "labels": [ "USD", "Universal Scene Description" ], "is_subclass_of": [ "Fiat Currency", "Money" ], "wikilinks": [ "USDC", "USDT", "Store of Value", "Digital Currency", "Money" ] }, { "id": "usdc", "title": "USDC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fiat-backed stablecoin pegged to the United States dollar, issued against reserves of cash and short-term government securities and redeemable on a one-for-one basis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usdc", "labels": [ "USDC" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [ "USD", "Custody Infrastructure", "USDT", "Payment System", "Stablecoin" ] }, { "id": "usdt", "title": "USDT", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A fiat-backed stablecoin pegged to the United States dollar, issued across multiple ledgers and intended to be redeemable for dollar value held in reserve.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usdt", "labels": [ "USDT", "Tether USDT" ], "is_subclass_of": [ "Stablecoin" ], "wikilinks": [ "USD", "Custody Infrastructure", "USDC", "Payment System", "Stablecoin" ] }, { "id": "utxo-model", "title": "UTXO Model", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Unspent Transaction Output accounting model used in Bitcoin and related blockchains, where each coin is represented as a discrete unspent output that must be fully consumed and re-created by a transaction. The model enables straightforward parallel validation, eliminates double-spend via simple output-state queries, and underpins script-based programmability.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:utxo-model", "labels": [ "UTXO Model", "Extended UTXO Model", "UTXO Accounting Model" ], "is_subclass_of": [ "Protocol and Consensus", "Blockchain Entity", "DistributedDataStructure" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "BlockchainDomain", "Blockchain Entity", "ConceptualLayer", "DistributedDataStructure", "Virtual Economy" ] }, { "id": "utxo", "title": "UTXO", "domain": "blockchain", "domain_name": "Blockchain", "definition": "UTXO (Unspent Transaction Output) is an accounting model used by Bitcoin and several other blockchains in which the ledger state consists of discrete unspent outputs rather than account balances. Each transaction consumes one or more existing unspent outputs as inputs and creates new outputs, and a coin is simply an output that has not yet been spent. Ownership is established by satisfying the locking script attached to an output, typically by providing a valid signature. The model contrasts with the account-based approach used by Ethereum and supports straightforward parallel validation and privacy techniques.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:utxo", "labels": [ "UTXO", "Bitcoin UTXO", "UTXO Commitments", "UTXO Set" ], "is_subclass_of": [ "Blockchain", "Blockchain Entity", "Blockchain Domain" ], "wikilinks": [ "Cryptographic Hash Function", "Digital Signature", "Bitcoin", "Cardano", "Blockchain Domain" ] }, { "id": "uv-unwrapping", "title": "UV Unwrapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "UV unwrapping is the process of flattening a 3D mesh's surface into a 2D coordinate space, called UV space, so that 2D textures can be accurately mapped onto the 3D geometry. It is a prerequisite step for texture mapping and is essential when preparing a character model for texturing, since seams and distortion introduced during unwrapping directly affect final texture quality. Automatic and manual unwrapping tools both aim to minimise stretching while placing seams in inconspicuous locations.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:uv-unwrapping", "labels": [ "UV Unwrapping" ], "is_subclass_of": [ "Texture Mapping" ], "wikilinks": [] }, { "id": "ubiquitous-computing", "title": "Ubiquitous Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ubiquitous Computing is a paradigm in which computational capability is embedded pervasively throughout the physical environment, making computers effectively invisible by integrating them into everyday objects and spaces. Coined by Mark Weiser, the concept encompasses smart devices, ambient intelligence, and the Internet of Things, aiming for seamless and context-aware interaction between people and technology. It contrasts with desktop computing by distributing computation across many heterogeneous, networked devices that operate without demanding explicit user attention.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:ubiquitous-computing", "labels": [ "Ubiquitous Computing" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "uk-industrial-strategy", "title": "Uk Industrial Strategy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "UK Industrial Strategy is the United Kingdom government's framework of coordinated public policy aimed at raising productivity, supporting strategic sectors, and steering long-term economic development through targeted investment, skills, and innovation funding. It identifies priority sectors and missions\u2014such as advanced manufacturing, clean energy, and digital technologies\u2014and aligns regulation, infrastructure, and research support behind them. The strategy seeks to balance market dynamics with directed state intervention to address regional and sectoral imbalances.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:uk-industrial-strategy", "labels": [ "Uk Industrial Strategy", "UK Industrial Strategy" ], "is_subclass_of": [ "Infrastructure", "Legal and Regulatory" ], "wikilinks": [] }, { "id": "uk-research-and-innovation", "title": "Uk Research And Innovation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "UK Research and Innovation (UKRI) is the United Kingdom's national funding agency for research and innovation, established in 2018 as a non-departmental public body bringing together the seven research councils, Innovate UK, and Research England. It allocates public funding across the disciplines, supports universities and businesses, and shapes national research strategy, including substantial investment in artificial intelligence and data science. UKRI underpins much of the UK's academic research and innovation ecosystem.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:uk-research-and-innovation", "labels": [ "Uk Research And Innovation", "UK Research and Innovation" ], "is_subclass_of": [ "Research and Development" ], "wikilinks": [] }, { "id": "ultra-wideband", "title": "Ultra Wideband", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ultra Wideband (UWB) is a short-range radio technology that transmits data using pulses spread across a very wide frequency spectrum (typically 3.1\u201310.6 GHz, bandwidth exceeding 500 MHz), enabling precise time-of-flight ranging and centimetre-accurate indoor positioning. Unlike narrowband technologies, UWB's broad spectrum allocation provides high resistance to multipath interference and coexistence with other radio systems. It is standardised under IEEE 802.15.4z and used in applications from secure device pairing to spatial-computing anchor systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:ultra-wideband", "labels": [ "Ultra Wideband", "Ultra-Wideband" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "ultra-low-latency", "title": "Ultra-Low Latency", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Ultra-low latency describes network or compute paths engineered to keep end-to-end response times to a few milliseconds or less, well below what is achievable with standard best-effort infrastructure. It is achieved through techniques such as edge deployment, dedicated fibre paths, and mobile edge computing that shorten the physical and logical distance between request and response. Ultra-low latency is a requirement for applications such as real-time control, augmented reality, and high-frequency trading.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:ultra-low-latency", "labels": [ "Ultra-Low Latency" ], "is_subclass_of": [ "Latency" ], "wikilinks": [] }, { "id": "ultrasonic-sensor", "title": "Ultrasonic Sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An electronic sensor that emits ultrasonic sound waves (typically ~40 kHz) and measures the time-of-flight of reflected echoes to determine the distance to objects, widely used in robotics for obstacle detection, proximity sensing, and autonomous navigation.", "entityType": "Class", "qualityScore": 0.68, "maturity": "draft", "iri": "urn:ngm:class:ultrasonic-sensor", "labels": [ "Ultrasonic Sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "ultraviolet-astronomy", "title": "Ultraviolet Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:ultraviolet-astronomy", "labels": [ "Ultraviolet Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "uncertainty-propagation", "title": "Uncertainty Propagation", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:uncertainty-propagation", "labels": [ "Uncertainty Propagation" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "uncertainty-quantification", "title": "Uncertainty Quantification", "domain": "ai", "domain_name": "Ai", "definition": "Uncertainty Quantification (UQ) is a discipline concerned with characterising, propagating, and communicating the uncertainties inherent in computational models, predictions, and measurements. It distinguishes between aleatoric uncertainty (irreducible randomness in the data or system) and epistemic uncertainty (reducible uncertainty arising from limited knowledge or data), providing principled methods \u2014 including Bayesian inference, Monte Carlo sampling, conformal prediction, and ensemble methods \u2014 for producing calibrated probability estimates rather than point predictions. UQ is foundational to trustworthy AI, safety-critical systems engineering, and scientific computing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:uncertainty-quantification", "labels": [ "Uncertainty Quantification" ], "is_subclass_of": [ "Probabilistic Model" ], "wikilinks": [] }, { "id": "uncertainty", "title": "Uncertainty", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The condition of incomplete or imperfect knowledge about the state of a system, the outcome of a process, or the truth of a proposition, formalised in probability theory as a distribution over possible values rather than a single determinate answer. Uncertainty is conventionally divided into aleatoric uncertainty, arising from irreducible randomness in the world, and epistemic uncertainty, arising from limited data or model inadequacy and reducible in principle by gathering more evidence. Representing, propagating, and acting under uncertainty is central to probabilistic reasoning, robot localisation, risk assessment, and modern machine learning.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:uncertainty", "labels": [ "Uncertainty" ], "is_subclass_of": [ "Probability Theory" ], "wikilinks": [ "Probability Theory", "Probabilistic Reasoning", "Risk", "Uncertainty Quantification" ] }, { "id": "unconfined-aquifer", "title": "Unconfined Aquifer", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:unconfined-aquifer", "labels": [ "Unconfined Aquifer", "WaterTableAquifer" ], "is_subclass_of": [ "Aquifer" ], "wikilinks": [] }, { "id": "uncrewed-spacecraft", "title": "Uncrewed Spacecraft", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:uncrewed-spacecraft", "labels": [ "Uncrewed Spacecraft" ], "is_subclass_of": [ "Spacecraft" ], "wikilinks": [] }, { "id": "underfitting", "title": "Underfitting", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Underfitting occurs when a machine learning model is insufficiently complex or inadequately trained to capture the underlying structure of its training data, resulting in high bias, low variance, and poor predictive performance on both training and unseen datasets. It is the converse of overfitting and arises from under-parameterisation, insufficient training duration, or excessive regularisation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:underfitting", "labels": [ "Underfitting" ], "is_subclass_of": [ "AI Technique", "Machine Learning Discipline" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "underground-water", "title": "Underground Water", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:underground-water", "labels": [ "Underground Water" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "underwater-robot", "title": "Underwater Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "An Underwater Robot is a robotic platform designed to operate in aquatic environments, encompassing Remotely Operated Vehicles (ROVs) tethered for real-time control and Autonomous Underwater Vehicles (AUVs) executing pre-programmed or AI-guided missions. Applications include deep-sea scientific survey, offshore infrastructure inspection, marine conservation, and defence, with platforms rated from hundreds to thousands of metres depth.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:underwater-robot", "labels": [ "Underwater Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "unfccc", "title": "Unfccc", "domain": "governance", "domain_name": "Governance", "definition": "The United Nations Framework Convention on Climate Change (UNFCCC) is an international environmental treaty, adopted in 1992, that establishes the institutional framework for intergovernmental cooperation to limit greenhouse gas emissions and address climate change. It convenes the annual Conference of the Parties and provides the legal and procedural basis for subsequent agreements including the Kyoto Protocol and the Paris Agreement. The convention defines reporting, review and finance mechanisms that govern global climate action.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:unfccc", "labels": [ "Unfccc", "UNFCCC" ], "is_subclass_of": [ "Governance" ], "wikilinks": [] }, { "id": "unicode", "title": "Unicode", "domain": "standards", "domain_name": "Standards", "definition": "Unicode is an international character encoding standard that assigns a unique code point to every character across the world's writing systems, symbols, and control codes, enabling consistent text representation across platforms and languages. It is implemented through encoding forms such as UTF-8, UTF-16, and UTF-32, which map code points to byte sequences of varying width. Unicode underlies virtually all modern text-based data interchange formats, including JSON and Turtle/RDF serialisations.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:unicode", "labels": [ "Unicode" ], "is_subclass_of": [ "Data Format" ], "wikilinks": [] }, { "id": "unified-communications", "title": "Unified Communications", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "The integration of enterprise communication channels \u2014 voice telephony, video conferencing, instant messaging, presence, voicemail, and content sharing \u2014 into a single coherent platform with a consistent user experience across devices, so that a conversation can move fluidly between modalities; delivered today predominantly as cloud services (UCaaS) such as Microsoft Teams, Zoom, and Cisco Webex.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:unified-communications", "labels": [ "Unified Communications" ], "is_subclass_of": [ "Real-Time Communication" ], "wikilinks": [ "Video Conferencing", "Presence Technology", "Digital Workplace Platform" ] }, { "id": "unified-hardware-access", "title": "Unified Hardware Access", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A standardised abstraction layer that provides consistent programmatic interfaces for accessing diverse XR hardware devices \u2014 including VR headsets, AR glasses, haptic controllers, eye-tracking modules, and six-DoF tracking systems \u2014 enabling cross-platform application development without device-specific code paths. The WebXR Device API (W3C) and OpenXR (Khronos Group) are the primary open standards implementing this abstraction for web and native runtimes respectively.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:unified-hardware-access", "labels": [ "Unified Hardware Access" ], "is_subclass_of": [ "Standards and Interoperability", "Hardware Abstraction" ], "wikilinks": [ "Hardware Abstraction", "metaverse" ] }, { "id": "unique-content-variation", "title": "Unique Content Variation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Procedurally generated or algorithmically created variations of digital content that produce distinct, individualised versions of assets, experiences, or environments, often used in NFTs and generative art to ensure scarcity and uniqueness.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:unique-content-variation", "labels": [ "Unique Content Variation" ], "is_subclass_of": [ "Content and Assets", "Generative Content" ], "wikilinks": [ "Generative Content", "metaverse" ] }, { "id": "unique-identifier", "title": "Unique Identifier", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "A unique identifier is a value assigned to an object, entity, or record so that it can be distinguished unambiguously from all others within a defined scope. Identifiers may be globally unique, such as universally unique identifiers, or unique within a namespace, such as a serial number combined with an issuer prefix. In supply chains, unique identifiers underpin serialisation, traceability, and data exchange across organisations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:unique-identifier", "labels": [ "Unique Identifier" ], "is_subclass_of": [ "Identification" ], "wikilinks": [] }, { "id": "uniswap-governance", "title": "Uniswap Governance", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Uniswap Governance is the on-chain and off-chain decision-making system by which UNI token holders collectively control the parameters, treasury, and development direction of the Uniswap decentralised exchange protocol. Governance proposals traverse a structured lifecycle of temperature checks, consensus checks, and on-chain votes executed through the Governor Bravo contract, with a quorum threshold and timelock delay before execution. UNI holders may delegate their voting power to representatives, enabling liquid democracy patterns in protocol management. Uniswap Governance controls protocol fee switches, liquidity mining programs, grant allocations from the UNI treasury, and deployment of the protocol to new networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:uniswap-governance", "labels": [ "Uniswap Governance" ], "is_subclass_of": [ "On-chain Governance" ], "wikilinks": [] }, { "id": "uniswap", "title": "uniswap", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Uniswap is a decentralised exchange (DEX) protocol deployed on Ethereum and EVM-compatible blockchains that enables permissionless, non-custodial token swaps through an Automated Market Maker (AMM) mechanism. Rather than maintaining an order book, it uses liquidity pools governed by the constant-product invariant (x \u00d7 y = k), where liquidity providers deposit token pairs and earn fees proportional to their pool share. Successive protocol versions have introduced concentrated liquidity (v3), multiple fee tiers, multi-hop routing, and hook-based extensibility (v4), making Uniswap a foundational primitive of decentralised finance. Governance is managed by holders of the UNI governance token through on-chain voting.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:uniswap", "labels": [ "Uniswap" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "unit-testing", "title": "Unit Testing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Unit testing is a software testing practice in which individual units of source code \u2014 typically functions, methods, or classes \u2014 are exercised in isolation to verify that each behaves as specified. Tests are written as small, deterministic, automated checks that assert expected outputs for given inputs and run quickly as part of the development loop. The practice underpins refactoring confidence, regression protection, and continuous integration pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:unit-testing", "labels": [ "Unit Testing" ], "is_subclass_of": [ "Software Testing" ], "wikilinks": [] }, { "id": "united-kingdom", "title": "United Kingdom", "domain": "governance", "domain_name": "Governance", "definition": "The United Kingdom is a sovereign state in north-west Europe comprising England, Scotland, Wales, and Northern Ireland, with a parliamentary system of government and a significant role in global finance, research, and technology policy. Its regulatory bodies and cities, including Manchester and Newcastle, participate in national initiatives spanning industrial strategy, financial services, and digital infrastructure. The UK's legal and regulatory framework governs data protection, financial services, and emerging technology standards within its jurisdiction.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:united-kingdom", "labels": [ "United Kingdom" ], "is_subclass_of": [ "Legal and Regulatory" ], "wikilinks": [] }, { "id": "united-states", "title": "United States", "domain": "governance", "domain_name": "Governance", "definition": "A federal republic in North America and one of the world's largest economies, with a leading role in technology, finance, and research. It comprises fifty states and a federal district.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:united-states", "labels": [ "United States" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "China", "owl:Thing" ] }, { "id": "unity", "title": "Unity", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Unity is a cross-platform real-time 3D engine and editor used to build games, simulations and XR applications across desktop, mobile, console and head-mounted display targets.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:unity", "labels": [ "Unity", "Unity Engine", "Unity ML-Agents", "Unity Platform", "Unity XRI" ], "is_subclass_of": [ "Game Engine" ], "wikilinks": [ "Graphics Pipeline", "Virtual Reality", "Augmented Reality", "OpenXR", "FBX", "Game Engine" ] }, { "id": "universal-access", "title": "Universal Access", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Universal access is the design principle and policy goal that products, services and digital experiences be usable by the widest possible range of people regardless of ability, device, connectivity or context. In immersive and metaverse settings it requires accommodating diverse sensory, motor and cognitive capabilities so that no user is excluded from participation. It is the foundation on which accessible experiences and accessibility guidelines are built.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:universal-access", "labels": [ "Universal Access", "Universal XR Access" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [] }, { "id": "universal-approximation", "title": "Universal Approximation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Universal approximation refers to the theoretical property, formalised in the universal approximation theorem, that a feedforward neural network with at least one hidden layer of sufficient width and a suitable non-linear activation function can approximate any continuous function on a compact domain to arbitrary precision. It provides the mathematical justification for using neural networks as general-purpose function approximators. The property depends on the activation function being non-polynomial and does not by itself guarantee that such a network can be learned efficiently from data.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:universal-approximation", "labels": [ "Universal Approximation" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "universal-avatar", "title": "Universal Avatar", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A cross-platform digital representation of a user that maintains consistent identity, appearance, and asset customisations across multiple metaverse applications, games, and virtual environments. Universal avatars rely on interoperability standards such as glTF and VRM to enable a single avatar to function across thousands of compatible platforms without requiring recreation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:universal-avatar", "labels": [ "Universal Avatar" ], "is_subclass_of": [ "Content and Assets", "Avatar" ], "wikilinks": [ "Avatar" ] }, { "id": "universal-basic-income", "title": "Universal Basic Income", "domain": "governance", "domain_name": "Governance", "definition": "Universal basic income (UBI) is an economic policy under which all members of a population receive a regular, unconditional cash payment from the state or another institution, independent of employment status or means. It is debated as a mechanism to reduce poverty, buffer labour displacement from automation and AI, and address structural inequality. Proponents view it as a redistributive instrument while critics question its fiscal sustainability and incentive effects.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:universal-basic-income", "labels": [ "Universal Basic Income" ], "is_subclass_of": [ "Economic Mechanism" ], "wikilinks": [] }, { "id": "universal-design", "title": "Universal Design", "domain": "standards", "domain_name": "Standards", "definition": "Universal Design is the architectural, product, and communication design philosophy asserting that environments, systems, and products should be designed from the outset to be usable by all people \u2014 regardless of age, disability, or circumstance \u2014 to the greatest extent possible without the need for adaptation or specialised design. Codified by architect Ronald Mace at NC State University in the 1990s across seven principles (equitable use, flexibility, simple and intuitive use, perceptible information, tolerance for error, low physical effort, size and space for approach and use), Universal Design goes beyond legal accessibility compliance to treat inclusive design as a quality standard that benefits all users. It has been extended to digital products, extended reality experiences, and AI systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:universal-design", "labels": [ "Universal Design" ], "is_subclass_of": [ "Design Thinking" ], "wikilinks": [] }, { "id": "universal-manifest", "title": "Universal Manifest", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A standardized metadata document describing identifiers, permissions, relationships, and provenance of a user's digital assets and identities across platforms, enabling cross-platform portability and interoperability.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:universal-manifest", "labels": [ "Universal Manifest" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Asset Interoperability", "Asset Management System", "Cryptographic Signature", "Decentralized Identifier", "Decentralized Ownership", "ETSI GR ARF 010", "Identity Credentials", "Interoperability Domain", "MSF Use Case Register", "Permission Grants", "Permissioned Access", "Provenance Record", "Relationship Graph", "Trust Registry", "Verifiable Credential", "Asset Registry", "Avatar Portability", "Cross-Platform Identity", "Data Format Standard", "Data Layer" ] }, { "id": "universal-scene-description", "title": "Universal Scene Description", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Universal Scene Description is an open-source, extensible framework originally developed by Pixar Animation Studios and released as open source under the Modified Apache 2.0 licence, providing a unified file format family and programmatic scene-graph API for describing, composing, simulating, and...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:universal-scene-description", "labels": [ "Universal Scene Description", "Scene Description", "Scene Description Language", "USD Scene Description", "USD Universal Scene Description" ], "is_subclass_of": [ "Standards and Interoperability", "3D File Format", "Asset Pipeline", "Scene Graph", "Open Standard", "Interoperability" ], "wikilinks": [ "3DGraphicsDomain", "Alembic", "Alliance for OpenUSD", "AOUSD", "Apple Vision Pro", "AR Quick Look", "Asset Pipeline", "Asset Resolution", "Autodesk Maya", "C++ API", "Collaborative Scene Assembly", "COLLADA", "Composition Arcs", "Crate Format", "CreativeToolsDomain", "Cross-DCC Interoperability", "Digital Twins", "FBX", "Film and Animation Pipeline", "FormatLayer" ] }, { "id": "universal-time-1", "title": "Universal Time 1", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:universal-time-1", "labels": [ "Universal Time 1" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "university-college-london", "title": "University College London", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "University College London (UCL) is a public research university in Bloomsbury, London, England, and the founding college of the federal University of London. It is one of the world's leading multidisciplinary universities, ranked consistently among the global top ten, with particular strength in computer science, artificial intelligence, machine learning, neuroscience, and biomedical engineering. UCL is a founding partner of the Alan Turing Institute \u2014 the UK national institute for data science and AI \u2014 and hosts major research groups in deep learning, natural language processing, reinforcement learning, and computational statistics. Its Computer Science and Engineering faculties supply significant research output and graduate talent to the UK and global AI ecosystem.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:university-college-london", "labels": [ "University College London" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Machine Learning", "Artificial Intelligence", "Alan Turing Institute", "Entity" ] }, { "id": "university-of-cambridge", "title": "University of Cambridge", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The University of Cambridge is a collegiate public research university in Cambridge, England, founded in 1209, making it one of the oldest continuously operating universities in the world and the second-oldest in the English-speaking world. Organised as a federation of 31 autonomous colleges and over 150 academic departments grouped into six schools, it is a founding member of the Russell Group and a member of the Coimbra Group and League of European Research Universities. Cambridge has produced foundational contributions to computing, mathematics, artificial intelligence, natural language processing, and computer vision through institutions such as the Computer Laboratory, the Machine Intelligence Laboratory, and the Cavendish Laboratory. Its Cambridge Cluster (Silicon Fen) represents one of Europe's most significant technology and deep-tech startup ecosystems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:university-of-cambridge", "labels": [ "University of Cambridge" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Imperial College London", "University of Edinburgh", "University of Manchester", "owl:Thing" ] }, { "id": "university-of-edinburgh", "title": "University of Edinburgh", "domain": "ai", "domain_name": "Ai", "definition": "The University of Edinburgh is a public research university in Edinburgh, Scotland, founded by royal charter in 1582, making it one of the oldest universities in the English-speaking world and the sixth-oldest in the United Kingdom. It is a member of the Russell Group and is internationally recognised for research in informatics, artificial intelligence, machine learning, natural language processing, medicine, and the humanities. Its School of Informatics is among the largest and most influential computer science and AI research centres in Europe, with foundational contributions to neural networks, probabilistic programming, and language modelling. The university's research ecosystem spans industry partnerships, spin-out companies, and national research institutes including the Alan Turing Institute.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:university-of-edinburgh", "labels": [ "University of Edinburgh" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "University of Cambridge", "Imperial College London", "University of Manchester", "owl:Thing" ] }, { "id": "university-of-london", "title": "University of London", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The University of London is a federal public research university in London, founded in 1836, comprising a group of largely autonomous constituent colleges and research institutes that award degrees under its charter. King's College London and University College London are among its best-known constituent colleges, each operating substantial independent research programmes, including in artificial intelligence and machine learning. The federal structure allows constituent colleges to share some central services while retaining separate admissions, governance and research identities.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:university-of-london", "labels": [ "University of London" ], "is_subclass_of": [ "Research University" ], "wikilinks": [] }, { "id": "university-of-manchester", "title": "University of Manchester", "domain": "ai", "domain_name": "Ai", "definition": "The University of Manchester is a public research university in Manchester, England, formed in 2004 by the merger of the Victoria University of Manchester and UMIST, with institutional roots tracing to 1824. It is a founding member of the Russell Group and a global leader in artificial intelligence, machine learning, materials science, and computer science, housing the Alan Turing Institute affiliate node and the Department of Computer Science where early stored-program computing was pioneered. The university produced the Manchester Baby (1948), the first operational stored-program electronic computer, and its researchers isolated graphene in 2004. It is consistently ranked among the world's top 30 research universities and has produced over 25 Nobel laureates across physics, chemistry, economics, and medicine.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:university-of-manchester", "labels": [ "University of Manchester" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "University of Cambridge", "University of Edinburgh", "Imperial College London", "owl:Thing" ] }, { "id": "university-of-oxford", "title": "University of Oxford", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The University of Oxford is a collegiate research university located in Oxford, England, widely regarded as one of the world's foremost institutions of higher education and research. Founded in the twelfth century, it comprises forty-four autonomous colleges and a network of academic departments spanning the humanities, sciences, social sciences, and engineering. Oxford is a leading site of research in artificial intelligence, machine learning, computer vision, natural language processing, and computational biology, and plays a central role in UK and global AI policy, ethics, and governance discourse. Its Future of Humanity Institute and Oxford Internet Institute are internationally recognised centres for the study of transformative technologies and their societal implications.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:university-of-oxford", "labels": [ "University of Oxford" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "Artificial Intelligence", "Machine Learning", "Alan Turing Institute", "Entity" ] }, { "id": "university-of-sheffield", "title": "University of Sheffield", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A public research university in Sheffield, England, and a member of the Russell Group of research-intensive universities. It is known for engineering, materials science, and natural language processing research.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:university-of-sheffield", "labels": [ "University of Sheffield" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Natural Language Processing", "United States", "owl:Thing" ] }, { "id": "unlabeled-data", "title": "Unlabeled Data", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Unlabeled data is a collection of raw observations, such as images, text, or sensor readings, that lacks the target annotations or ground-truth labels needed for supervised learning. It is typically abundant and inexpensive to collect relative to labelled data, since it requires no manual annotation effort. Unlabeled data is the input substrate for unsupervised learning, self-supervised pretraining, and active learning, which selectively queries labels for the most informative examples.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:unlabeled-data", "labels": [ "Unlabeled Data" ], "is_subclass_of": [ "Training Data" ], "wikilinks": [] }, { "id": "unlinkability", "title": "Unlinkability", "domain": "security", "domain_name": "Security", "definition": "Unlinkability is a privacy property that prevents an adversary from determining whether two or more items of interest, such as messages, transactions or sessions, are related to the same entity. It is a core goal of privacy-enhancing technologies and is achieved through techniques such as pseudonym rotation, mixing, blinding and zero-knowledge constructions. Unlinkability protects against profiling and correlation while still permitting legitimate use of a system.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:unlinkability", "labels": [ "Unlinkability" ], "is_subclass_of": [ "Privacy-Enhancing Technologies" ], "wikilinks": [] }, { "id": "unlocking-script", "title": "Unlocking Script", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The script supplied in a Bitcoin transaction input, historically called scriptSig, that satisfies the spending conditions imposed by the locking script of the unspent output it references, typically by providing digital signatures and public keys; the input is valid only if executing the unlocking script followed by the locking script leaves a true value on the stack, proving authorisation to spend the UTXO.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:unlocking-script", "labels": [ "Unlocking Script" ], "is_subclass_of": [ "Script" ], "wikilinks": [ "Script", "Locking Script", "UTXO" ] }, { "id": "unmanned-aerial-vehicle", "title": "Unmanned Aerial Vehicle", "domain": "robotics", "domain_name": "Robotics", "definition": "An Unmanned Aerial Vehicle (UAV), commonly called a drone, is an aircraft that operates without an onboard human pilot, controlled either remotely or autonomously by onboard flight software. UAVs integrate flight control, navigation, sensing and communication subsystems to perform tasks ranging from aerial imaging to delivery and inspection. They span scales from small quadcopters to fixed-wing platforms and increasingly rely on autonomy for navigation and obstacle avoidance.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:unmanned-aerial-vehicle", "labels": [ "Unmanned Aerial Vehicle" ], "is_subclass_of": [ "Robotics", "Robot Type" ], "wikilinks": [] }, { "id": "unreal-engine", "title": "Unreal Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Unreal Engine is a real-time three-dimensional creation tool and game engine developed by Epic Games. It provides rendering, physics, animation, audio and scripting systems used to build games, virtual production environments, architectural visualisations and simulations. The engine is widely adopted for its high-fidelity rendering, its visual scripting system Blueprints, and its use in film and television virtual production with LED volume stages.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:unreal-engine", "labels": [ "Unreal Engine", "Unreal Engine XR", "Unreal PCG" ], "is_subclass_of": [ "Game Engine", "Creative Media Domain" ], "wikilinks": [ "Nanite", "Lumen", "GPU Rendering", "Virtual Production", "Digital Twin", "Metaverse Domain", "Computer Vision Domain", "3D Reconstruction", "Creative Media Domain" ] }, { "id": "unspent-transaction-output", "title": "Unspent Transaction Output", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An Unspent Transaction Output (UTXO) is a discrete amount of cryptocurrency that has been received by an address and not yet spent, forming the fundamental accounting unit in UTXO-based blockchains such as Bitcoin. Each transaction consumes one or more existing UTXOs as inputs and creates new UTXOs as outputs, with the global set of all UTXOs representing the current ledger state. Validating a transaction requires confirming that its referenced inputs exist in the UTXO set and have not already been spent, which is central to preventing double spending.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:unspent-transaction-output", "labels": [ "Unspent Transaction Output" ], "is_subclass_of": [ "Blockchain Transaction", "Transaction" ], "wikilinks": [] }, { "id": "unsupervised-learning", "title": "Unsupervised Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Unsupervised Learning discovers hidden patterns, structures, and representations in unlabeled data without explicit supervision. Key techniques include clustering, dimensionality reduction, density estimation, anomaly detection, and generative modelling, enabling exploratory data analysis and latent feature extraction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:unsupervised-learning", "labels": [ "Unsupervised Learning", "UnsupervisedLearning" ], "is_subclass_of": [ "Machine Learning", "AI Technique" ], "wikilinks": [ "Autoencoders", "Clustering", "Generative Models", "Dimensionality Reduction" ] }, { "id": "upper-atmosphere", "title": "Upper Atmosphere", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:upper-atmosphere", "labels": [ "Upper Atmosphere" ], "is_subclass_of": [ "Atmosphere Layer" ], "wikilinks": [] }, { "id": "upper-ontology", "title": "Upper Ontology", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "An upper ontology, also called a foundational or top-level ontology, defines very general categories such as object, process, quality and relation that are common across all subject domains. It provides a shared semantic backbone onto which domain ontologies can be aligned, improving interoperability, reuse and reasoning. By committing to clear distinctions such as endurants versus perdurants, it grounds more specific vocabularies in a consistent conceptual framework.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:upper-ontology", "labels": [ "Upper Ontology" ], "is_subclass_of": [ "Ontology" ], "wikilinks": [] }, { "id": "urban-computing", "title": "Urban Computing", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Urban Computing is the acquisition, integration, and analysis of large, heterogeneous data generated by sensors, devices, vehicles, and people in cities to understand and improve urban systems. It applies data analytics, machine learning, and pervasive sensing to problems such as traffic management, energy use, environmental monitoring, and public services. Drawing on Internet of Things infrastructure and geospatial data, urban computing closes the loop between sensing the city and acting on it. It is closely related to the smart cities agenda and to ubiquitous and pervasive computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:urban-computing", "labels": [ "Urban Computing" ], "is_subclass_of": [ "Ubiquitous Computing" ], "wikilinks": [] }, { "id": "urban-data-platform", "title": "Urban Data Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "An urban data platform is a shared digital infrastructure layer that ingests, integrates and exposes data from across a city's sensors, services and administrative systems through common APIs and data models. It underpins city-scale applications such as traffic management, environmental monitoring and citizen services by giving disparate systems a common data substrate rather than isolated silos. Urban data platforms are a foundational requirement for digital twins of society and are deployed by city administrations such as Manchester's.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:urban-data-platform", "labels": [ "Urban Data Platform" ], "is_subclass_of": [ "Digital Infrastructure" ], "wikilinks": [] }, { "id": "urban-heat-monitoring", "title": "Urban Heat Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:urban-heat-monitoring", "labels": [ "Urban Heat Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "urban-planning", "title": "Urban Planning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Urban planning is the discipline concerned with the design, regulation and long-term development of land use, infrastructure and public space within cities and regions. Contemporary practice increasingly draws on geospatial data and digital twins of urban systems to model traffic, utilities, housing and environmental impact before physical intervention. It sits at the intersection of policy, civil engineering and spatial computing tools that visualise and simulate proposed changes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:urban-planning", "labels": [ "Urban Planning" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "uri-scheme", "title": "Uri Scheme", "domain": "security", "domain_name": "Security", "definition": "A URI scheme is the leading component of a Uniform Resource Identifier that names the namespace, protocol, or resolution mechanism by which the remainder of the identifier is to be interpreted, appearing before the colon delimiter (for example http, https, mailto, did, urn, or ipfs). Schemes are registered with IANA under provisional or permanent status and define the syntax and semantics of the scheme-specific part. Custom and decentralised schemes such as did: and ipfs: extend the URI model to identity and content-addressed systems, making the scheme a foundational element of how distributed resources are addressed and trusted.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:uri-scheme", "labels": [ "Uri Scheme", "URI Scheme" ], "is_subclass_of": [ "Protocol" ], "wikilinks": [] }, { "id": "usability-testing", "title": "Usability Testing", "domain": "data", "domain_name": "Data", "definition": "Usability testing is an empirical user research method in which representative users are observed attempting to complete realistic tasks with a product, system, or prototype, while the evaluator records errors, task completion times, help-seeking behaviour, and verbal commentary to identify usability problems and inform design improvements. Unlike expert-based heuristic evaluation, usability testing generates direct evidence of how real people interact with an interface under ecologically valid conditions. It is a core practice in human-computer interaction, product design, and user experience research, conducted through moderated in-person sessions, remote think-aloud protocols, or automated unmoderated testing platforms.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usability-testing", "labels": [ "Usability Testing", "UsabilityTesting" ], "is_subclass_of": [ "User Research" ], "wikilinks": [] }, { "id": "usability", "title": "Usability", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Usability is the extent to which a product, system, or service can be used by specified users to achieve specified goals with effectiveness, efficiency, and satisfaction in a specified context of use, as defined by ISO 9241-11. It is a core quality attribute in human-computer interaction and user-centred design that encompasses learnability, memorability, error prevention, and user satisfaction. Usability is measured empirically through user testing, heuristic evaluation, and cognitive walkthroughs, and is distinct from\u2014yet prerequisite to\u2014broader user experience concerns. Poor usability results in task failure, user frustration, abandonment, and safety risks in safety-critical systems.", "entityType": "Class", "qualityScore": 0.75, "maturity": "mature", "iri": "urn:ngm:class:usability", "labels": [ "Usability", "Universal Usability" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [ "HRI", "User Experience", "https://www.nngroup.com/articles/usability-101-introduction-to-usability/", "https://www.iso.org/standard/77520.html" ] }, { "id": "usage-analytics", "title": "Usage Analytics", "domain": "data", "domain_name": "Data", "definition": "Usage analytics is the collection, measurement and interpretation of data describing how users interact with a product, service or platform, including session frequency, feature adoption, retention and engagement patterns. It transforms raw event telemetry into actionable insight for product, design and governance decisions. When applied responsibly it can inform well-being metrics while raising privacy considerations about behavioural tracking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:usage-analytics", "labels": [ "Usage Analytics" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "usage-based-consumption", "title": "Usage-Based Consumption", "domain": "economics", "domain_name": "Economics", "definition": "A pricing and consumption model where users pay for AI services based on actual usage (e.g., tokens) rather than fixed seats or subscriptions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:usage-based-consumption", "labels": [ "Usage-Based Consumption" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "usage-based-pricing", "title": "Usage-Based Pricing", "domain": "economics", "domain_name": "Economics", "definition": "A pricing strategy where customers are charged based on the actual consumption of a service or resource, such as compute time, API calls, or data tokens, rather than a fixed periodic fee.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:usage-based-pricing", "labels": [ "Usage-Based Pricing" ], "is_subclass_of": [ "SaaS Business Model" ], "wikilinks": [] }, { "id": "user-agreement-compliance", "title": "User Agreement Compliance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Process ensuring user actions within a metaverse platform adhere to declared policies, terms of service, and acceptable use guidelines.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:user-agreement-compliance", "labels": [ "User Agreement Compliance" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "Behavior Analytics", "Enforcement Mechanisms", "ETSI GR ARF 010", "Monitoring System", "MSF Use Cases", "Platform Management", "Platform Safety", "Policy Adherence", "Remediation Process", "User Accountability", "User Agreement", "User Monitoring", "Violation Detection", "Access Control", "Audit Trail", "Governance Framework", "Identity Management", "MiddlewareLayer", "Policy Enforcement", "Risk Mitigation" ] }, { "id": "user-authentication-mechanism", "title": "User Authentication Mechanism", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A User Authentication Mechanism is a protocol or system component that verifies the claimed identity of a user before granting access to a resource or service. Mechanisms include password-based credentials, OAuth 2.0 federated login, biometric binding, multi-factor authentication, and decentralised identity using DIDs, each offering different trade-offs between security, usability, and privacy.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-authentication-mechanism", "labels": [ "User Authentication Mechanism" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "user-authentication", "title": "User Authentication", "domain": "security", "domain_name": "Security", "definition": "User authentication is the process of verifying that a person claiming a digital identity is who they assert to be, by validating one or more authentication factors: something they know (password, PIN), something they have (hardware token, mobile device), or something they are (biometric). It is the gateway control between unauthenticated network access and authorised use of a system or resource, and its assurance level \u2014 defined by standards such as NIST SP 800-63 \u2014 must be calibrated to the sensitivity of protected resources. Modern implementations favour phishing-resistant factors (passkeys, hardware security keys) over knowledge-based authentication.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:user-authentication", "labels": [ "User Authentication" ], "is_subclass_of": [ "Authentication" ], "wikilinks": [] }, { "id": "user-awareness", "title": "User Awareness", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The level of understanding and knowledge that users possess about digital systems, privacy practices, security risks, and their rights when interacting with metaverse platforms and virtual environments. Effective user awareness encompasses privacy literacy, recognition of security threats, platform policy comprehension, and knowledge of available remedies, forming a prerequisite for informed consent and safe participation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-awareness", "labels": [ "User Awareness" ], "is_subclass_of": [ "Governance and Safety", "User Education" ], "wikilinks": [ "metaverse", "User Education" ] }, { "id": "user-behaviour-data", "title": "User Behaviour Data", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "User behaviour data is the record of actions a user takes while interacting with a system, such as clicks, views, purchases, dwell time and navigation paths. It is collected through logging and instrumentation and forms the primary training signal for collaborative filtering, recommendation and personalisation systems. Its quality and volume directly determine how accurately downstream models can infer user preferences and predict future actions.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:user-behaviour-data", "labels": [ "User Behaviour Data" ], "is_subclass_of": [ "Dataset" ], "wikilinks": [] }, { "id": "user-consent-token", "title": "User Consent Token", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A cryptographically verifiable digital token that represents and enforces user consent for data processing, collection, sharing, or participation in virtual environments with granular permission controls and revocation mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:user-consent-token", "labels": [ "User Consent Token" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Automated Privacy Compliance", "Blockchain Ledger", "Consent Audit Trail", "Consent Management Framework", "Consent Payload", "Consent Registry", "Consent Revocation", "Cryptographic Key", "Cryptographic Signature", "Data Schema", "Decentralized Identifier (DID)", "GDPR", "ISO 29184", "Privacy Policy", "Revocation Mechanism", "Scope Definition", "Time Oracle", "User Data Sovereignty", "Verifiable Credential", "W3C DID Core" ] }, { "id": "user-context-awareness", "title": "User Context Awareness", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User Context Awareness is the capability of digital systems to sense, model, and respond to a user's current situation\u2014encompassing location, activity, device state, social context, and environmental factors\u2014in order to deliver personalised, timely, and relevant experiences. It underpins adaptive XR interfaces, smart environment switching, and context-driven content recommendation in spatial computing.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-context-awareness", "labels": [ "User Context Awareness" ], "is_subclass_of": [ "Interaction Technology", "Context Aware Computing" ], "wikilinks": [ "Context-Aware Computing", "metaverse" ] }, { "id": "user-control", "title": "User Control", "domain": "governance", "domain_name": "Governance", "definition": "User control is the principle and set of mechanisms that give individuals meaningful authority over how their data, identity and experience are managed, including the ability to grant, review and revoke permissions. It is central to data-protection and consent frameworks, embodying user autonomy and self-determination over personal information. Strong user control is a prerequisite for trustworthy, ethically governed digital systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-control", "labels": [ "User Control" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "user-directory", "title": "User Directory", "domain": "security", "domain_name": "Security", "definition": "A user directory is a centralised repository that stores and organises identity records, attributes and credentials for the principals of a system, supporting authentication and authorisation queries. Common implementations include LDAP directories, Active Directory and cloud identity stores, which identity providers consume to validate and describe users. It is a core component of identity and access management infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-directory", "labels": [ "User Directory" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "user-education", "title": "User Education", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Structured programmes, tutorials, and onboarding materials that build user competencies in operating spatial computing platforms, XR devices, and metaverse applications, including safety guidance, digital literacy, and skill development delivered through interactive or immersive pedagogic methods.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-education", "labels": [ "User Education" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "user-engagement", "title": "User Engagement", "domain": "metaverse", "domain_name": "Metaverse", "definition": "User Engagement is the set of qualitative and quantitative dimensions that characterise the depth, frequency, and quality of a person's interaction with a digital product, service, or community. High engagement reflects meaningful value exchange: users return voluntarily, invest attention and effort, and develop lasting behaviours around the product. Engagement metrics\u2014session duration, interaction depth, return rate, and social sharing\u2014serve as proxies for value delivered and are key inputs to product development prioritisation and business model viability. Ethical engagement design balances compelling interaction patterns against the risk of exploiting psychological vulnerabilities through addictive mechanics or manipulative nudging.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:user-engagement", "labels": [ "User Engagement" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [] }, { "id": "user-experience-design", "title": "User Experience Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User Experience Design is the discipline of shaping the overall perception, ease and satisfaction a person derives from interacting with a product, system or service. It synthesises research into user needs, information architecture, interaction flows and visual presentation into coherent experiences. In spatial computing it extends to embodied, three-dimensional and multimodal interactions across mixed reality. The goal is to make systems useful, usable and desirable across their full lifecycle.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:user-experience-design", "labels": [ "User Experience Design" ], "is_subclass_of": [ "Human-Computer Interaction" ], "wikilinks": [] }, { "id": "user-experience-layer", "title": "User Experience Layer", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The User Experience Layer is the topmost human-facing stratum concerned with how effective, accessible, and satisfying interaction with a system is. It sits above the Presentation Layer that renders the interface and has no technical layer above it. It contains interaction design, usability criteria, accessibility requirements, and user research findings.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:user-experience-layer", "labels": [ "User Experience Layer" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Presentation Layer", "Social Layer", "Usability", "Accessibility", "owl:Thing", "ISO (International Organization for Standardization)" ] }, { "id": "user-experience", "title": "user experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User Experience (UX) is the totality of perceptions, responses, and outcomes that a person encounters before, during, and after interacting with a product, service, or system \u2014 encompassing functional usability, aesthetic appeal, emotional resonance, accessibility, and long-term satisfaction. In spatial and immersive computing contexts, UX additionally addresses physical comfort, motion sickness mitigation, spatial audio fidelity, haptic feedback, and the cognitive load imposed by three-dimensional interaction paradigms. UX design is an iterative, evidence-based discipline drawing on human-computer interaction research, cognitive psychology, and ergonomics, and is formalised in ISO 9241-210 as encompassing all perceptions and responses resulting from actual or anticipated use. Empirical evaluation methods \u2014 including heuristic analysis, usability testing, eye tracking, and physiological measurement \u2014 ground UX practice in observable user behaviour.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:user-experience", "labels": [ "User Experience", "Consistent User Experience", "User Experience Assessment", "User Query Experience", "UserExperience" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "user-identity-management", "title": "User Identity Management", "domain": "security", "domain_name": "Security", "definition": "User identity management is the set of processes and systems that create, maintain, authenticate and deprovision the digital identities of users across a platform's lifecycle. It governs how users register, prove who they are, manage profiles and link accounts, and is foundational to access control and personalised experiences. In collaborative and metaverse platforms it also coordinates presence, avatars and cross-session continuity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-identity-management", "labels": [ "User Identity Management", "User Identity System" ], "is_subclass_of": [ "Identity Management" ], "wikilinks": [] }, { "id": "user-interface-architecture", "title": "User Interface Architecture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The structural design and organization of user interface components, interaction patterns, and navigation systems that enable users to interact with metaverse platforms and virtual environments across different devices and modalities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-interface-architecture", "labels": [ "User Interface Architecture" ], "is_subclass_of": [ "Interaction Technology", "System Architecture" ], "wikilinks": [ "metaverse", "System Architecture" ] }, { "id": "user-interface-design", "title": "User Interface Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User interface design is the discipline of shaping the visual and interactive surfaces through which people operate digital products, determining layout, controls, typography, colour and feedback. It translates user goals and information structure into coherent, learnable and aesthetically consistent screens and interactions. As a sibling of interaction design and a contributor to overall user experience, it balances usability, accessibility and brand expression.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:user-interface-design", "labels": [ "User Interface Design" ], "is_subclass_of": [ "Interface Design", "Interaction Technology" ], "wikilinks": [] }, { "id": "user-interface-standard", "title": "User Interface Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specifications and guidelines that define consistent patterns, components, and interactions for user interfaces in metaverse and XR environments, ensuring usability, accessibility, and cross-platform consistency.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:user-interface-standard", "labels": [ "User Interface Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "Sensor Input", "metaverse", "Standards" ] }, { "id": "user-interface", "title": "User Interface", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A User Interface (UI) is the aggregate of visual, auditory, tactile, and interactive components through which a human perceives and controls a software system, forming the boundary between human cognition and computational logic. UIs span a spectrum from two-dimensional graphical desktop and web environments to three-dimensional spatial interfaces rendered in augmented and virtual reality, encompassing gesture, voice, gaze, and haptic modalities. Effective UI design integrates human-computer interaction principles \u2014 affordance, feedback, constraints, and error prevention \u2014 with visual design, accessibility requirements, and rendering performance budgets. In complex sociotechnical systems, the UI layer mediates trust, cognitive load, and task efficiency, making it a critical determinant of system adoption and usability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:user-interface", "labels": [ "User Interface" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "user-monitoring", "title": "User Monitoring", "domain": "governance", "domain_name": "Governance", "definition": "User monitoring is the observation and logging of user behaviour and interactions within a platform to detect policy violations, abuse, safety risks or anomalous activity. In metaverse and online environments it supports trust-and-safety enforcement and agreement compliance, balancing protective oversight against privacy and consent. It typically feeds moderation, incident response and compliance reporting workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-monitoring", "labels": [ "User Monitoring" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [] }, { "id": "user-navigation", "title": "User Navigation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "User navigation is the set of interaction techniques and interface affordances that let people move through and orient themselves within a digital or spatial environment, from menus and links to locomotion in 3D space. In immersive contexts it covers teleportation, smooth locomotion and wayfinding cues, while in 2D it covers information architecture and discovery flows. Effective navigation reduces cognitive load and supports search and content discovery.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-navigation", "labels": [ "User Navigation" ], "is_subclass_of": [ "Interaction Technology" ], "wikilinks": [] }, { "id": "user-privacy-control", "title": "User Privacy Control", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "User privacy controls are interface and systemic mechanisms that allow individuals to view, modify, restrict, or delete their personal data held by an AI or data-driven system. They operationalise data subject rights such as the right to be forgotten, consent withdrawal, and data portability mandated by regulations like GDPR. Effective privacy controls combine technical enforcement \u2014 such as deletion pipelines and audit trails \u2014 with accessible user interfaces.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:user-privacy-control", "labels": [ "User Privacy Control" ], "is_subclass_of": [ "AI Governance and Ethics", "Privacy Mechanism" ], "wikilinks": [ "Privacy Mechanism" ] }, { "id": "user-privacy", "title": "User Privacy", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "User Privacy encompasses the rights, technical mechanisms, and regulatory obligations protecting individuals from unwanted disclosure of their identity, transactions, and behavioural patterns within digital and blockchain systems. Protections span cryptographic techniques (zero-knowledge proofs, ring signatures, confidential transactions), network-layer obfuscation (Dandelion++, Tor), and compliance frameworks (GDPR, data minimisation principles).", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:user-privacy", "labels": [ "User Privacy" ], "is_subclass_of": [ "Legal and Regulatory", "Privacy Framework" ], "wikilinks": [] }, { "id": "user-profiling", "title": "User Profiling", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "User profiling is the construction of a structured model of an individual or segment from observed attributes, behaviours and interaction history, used to predict preferences, intent or risk. Profiles aggregate explicit data such as stated preferences with implicit signals such as clicks, dwell time and purchases, and may be updated continuously as new behaviour is observed. Because profiles concern people, their construction raises consent, fairness and privacy obligations that constrain what may be collected and inferred.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:user-profiling", "labels": [ "User Profiling" ], "is_subclass_of": [ "Behavioural Analytics" ], "wikilinks": [] }, { "id": "user-protection", "title": "User Protection", "domain": "governance", "domain_name": "Governance", "definition": "User Protection encompasses the legal frameworks, platform policies, and technical mechanisms designed to safeguard individuals from harm when using digital services and online platforms. It spans data protection rights, content moderation obligations, anti-manipulation requirements, and consumer protection rules applicable to platform operators. Regulation such as the UK Online Safety Act and the EU Digital Services Act impose positive duties on platforms to assess and mitigate risks to users, particularly minors and vulnerable populations. Technical controls include content filtering, safety settings, transparency mechanisms, and redress systems that give users agency over their online experience.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:user-protection", "labels": [ "User Protection", "User Protection Measure" ], "is_subclass_of": [ "Consumer Protection" ], "wikilinks": [] }, { "id": "user-provisioning", "title": "User Provisioning", "domain": "security", "domain_name": "Security", "definition": "User provisioning is the identity-and-access-management process of creating, configuring, maintaining and eventually removing user accounts and their associated access entitlements across systems and applications. It encompasses the full account lifecycle \u2014 from initial onboarding through entitlement changes to deprovisioning at offboarding \u2014 and is increasingly automated through directory services and standards such as SCIM. Effective provisioning ensures that each identity holds exactly the access required, supporting least-privilege and timely revocation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:user-provisioning", "labels": [ "User Provisioning" ], "is_subclass_of": [ "Identity and Access Management" ], "wikilinks": [] }, { "id": "user-representation", "title": "User Representation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "User representation is the visual and behavioural embodiment of a person within a virtual environment, encompassing avatars, presence indicators and the mapping of real movement onto a digital persona. It conveys identity, expression and social presence, allowing others to perceive and interact with the user in shared spaces. It is a defining element of avatar systems and virtual worlds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-representation", "labels": [ "User Representation" ], "is_subclass_of": [ "Virtual Environment" ], "wikilinks": [] }, { "id": "user-research", "title": "User Research", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Systematic investigation of user behaviors, needs, preferences, and experiences in metaverse environments through qualitative and quantitative methods to inform design decisions and improve virtual world experiences.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:user-research", "labels": [ "User Research", "User Research with Disabled Participants", "User Research with Disabled Users" ], "is_subclass_of": [ "Research Methods" ], "wikilinks": [ "Research Methods" ] }, { "id": "user-safety", "title": "User Safety", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User Safety encompasses the technical, design, policy, and governance measures deployed by platforms and application developers to protect users from harm across digital and immersive environments, including harassment, predatory behaviour, unwanted exposure to harmful content, physical discomfort from extended use, and violations of privacy or consent. In spatially embodied contexts such as extended reality and the metaverse, abusive interactions carry heightened psychological impact because spatial audio, avatar proximity, and embodied presence amplify the subjective experience of harm beyond that typical of text-based social media. Effective user safety frameworks integrate proactive content moderation, privacy-preserving reporting, age-appropriate design, graduated enforcement mechanisms, and platform governance structures accountable to regulators and communities alike. The discipline draws on human factors research, risk management, digital rights law, and AI-driven trust-and-safety tooling to balance protection against harm with preservation of free expression and user autonomy.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:user-safety", "labels": [ "User Safety" ], "is_subclass_of": [ "Governance Framework" ], "wikilinks": [ "owl:Thing" ] }, { "id": "user-segment", "title": "User Segment", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:user-segment", "labels": [ "User Segment" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "user-sovereignty", "title": "User Sovereignty", "domain": "governance", "domain_name": "Governance", "definition": "User sovereignty is the principle that individuals should control their own data, identity, and assets rather than depending on intermediaries. It is associated with decentralised systems and privacy advocacy.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:user-sovereignty", "labels": [ "User Sovereignty" ], "is_subclass_of": [ "Digital Rights" ], "wikilinks": [ "Cryptography", "Privacy", "Web3", "Digital Rights", "https://en.wikipedia.org/wiki/Self-sovereign_identity", "https://www.w3.org/TR/did-core/" ] }, { "id": "user-trust", "title": "User Trust", "domain": "governance", "domain_name": "Governance", "definition": "User trust is the confidence individuals place in a system to handle their data, decisions and interactions reliably, securely and in their interests. It is built through transparency, predictable behaviour, privacy safeguards and demonstrable accountability, and is a key outcome of privacy and data-protection practices. Eroded trust undermines adoption, while sustained trust is a competitive and ethical asset.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:user-trust", "labels": [ "User Trust" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "user-centred-design", "title": "User-Centred Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User-centred design is an iterative design methodology that grounds every stage of a system's development in the needs, behaviours and feedback of its intended users, typically through research, prototyping and usability testing. It treats user needs as the primary design constraint rather than an afterthought, in contrast to technology-first approaches. In interactive and spatial systems it shapes affordances, interaction patterns and accessibility provisions so that interfaces remain intuitive across diverse users and contexts.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:user-centred-design", "labels": [ "User-Centred Design", "User Centered Design", "User Centred Design", "User-Centered Design" ], "is_subclass_of": [ "User Experience Design" ], "wikilinks": [] }, { "id": "user-generated-content", "title": "User-Generated Content", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "User-generated content (UGC) is media and information created and published by the users of a platform rather than by its operators or professional producers. It encompasses text, images, video, 3D assets, reviews, and virtual-world artefacts contributed by a community. UGC drives engagement and network effects on social and metaverse platforms while creating moderation, rights, and quality-control obligations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:user-generated-content", "labels": [ "User-Generated Content", "User Generated Content" ], "is_subclass_of": [ "Digital Content" ], "wikilinks": [] }, { "id": "utility-ai", "title": "Utility AI", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A game artificial intelligence architecture in which an agent scores every available action by evaluating weighted utility functions over the current world state \u2014 factors such as health, distance, threat, ammunition, and needs \u2014 and selects the highest-scoring option. Response curves map raw game variables onto normalised utilities that are combined multiplicatively or additively, yielding nuanced, context-sensitive behaviour that degrades gracefully and is easier to tune than rigid rule-based state machines or behaviour trees.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:utility-ai", "labels": [ "Utility AI" ], "is_subclass_of": [ "Game AI" ], "wikilinks": [ "Game AI", "Behaviour Tree", "Finite State Machine", "Decision Making", "AI Game Agent" ] }, { "id": "utility-function", "title": "Utility Function", "domain": "ai", "domain_name": "Ai", "definition": "A utility function is a mathematical mapping from outcomes or states to a real-valued measure of preference, used to formalise an agent's goals so that decisions can be compared and optimised. In game theory and reinforcement learning it provides the quantity an agent seeks to maximise, underpinning concepts such as expected utility and reward. Utility functions are a core construction of utility theory, applied to model rational choice under uncertainty in both single-agent and multi-agent systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:utility-function", "labels": [ "Utility Function" ], "is_subclass_of": [ "Utility Theory" ], "wikilinks": [] }, { "id": "utility-theory", "title": "Utility Theory", "domain": "ai", "domain_name": "Ai", "definition": "Utility Theory is the formal framework for representing an agent's preferences over outcomes as a numerical utility function, such that the agent's rational behaviour can be modelled as the maximisation of expected utility. It provides the axiomatic foundation \u2014 completeness, transitivity, continuity, and independence \u2014 under which preferences admit a utility representation. In artificial intelligence it grounds rational-agent design, decision-making under uncertainty, and the objective functions of planning and reinforcement-learning systems. It is closely tied to decision theory, game theory, and economic models of choice.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:utility-theory", "labels": [ "Utility Theory" ], "is_subclass_of": [ "Decision Theory" ], "wikilinks": [] }, { "id": "utility-token", "title": "Utility Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A blockchain token that confers its holder a right to access, use, or consume a specific product, service, or platform resource rather than representing equity or a claim on profits. Utility tokens derive value from the utility of the underlying service and are a primary means of bootstrapping decentralised protocol economies.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:utility-token", "labels": [ "Utility Token" ], "is_subclass_of": [ "Token and Asset", "Cryptocurrency Token" ], "wikilinks": [ "Blockchain", "Cryptocurrency Token" ] }, { "id": "uv-mapping", "title": "Uv Mapping", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "UV mapping is the process of projecting a 3D model's surface onto a two-dimensional coordinate space so that texture images can be applied accurately to its geometry. The letters U and V denote the axes of this 2D texture space, distinct from the X, Y, and Z axes of the model. By unwrapping the mesh into UV islands, artists control how textures, normal maps, and other surface data wrap around the object with minimal stretching or seams.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:uv-mapping", "labels": [ "Uv Mapping", "UV Mapping" ], "is_subclass_of": [ "3D Modelling" ], "wikilinks": [] }, { "id": "v2x-communication", "title": "V2X Communication", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Vehicle-to-everything (V2X) communication is a wireless technology that lets vehicles exchange real-time information with other vehicles, roadside infrastructure, pedestrians, and networks. It extends the perception of automated and connected vehicles beyond their onboard sensors by sharing position, speed, intent, and hazard data, improving safety and traffic efficiency. Implemented through cellular C-V2X or dedicated short-range radio standards, it is a key enabler of cooperative driving and intelligent transport systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:v2x-communication", "labels": [ "V2X Communication" ], "is_subclass_of": [ "Wireless Communication" ], "wikilinks": [] }, { "id": "vae", "title": "vae", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Variational Autoencoder (VAE) is a deep generative model that learns a probabilistic, continuous latent-space representation of data by jointly optimising a reconstruction loss and a Kullback-Leibler divergence regularisation term, using amortised variational inference to make the intractable posterior distribution tractable. The encoder network (recognition model) maps input data to the parameters of a Gaussian posterior over latent codes, while the decoder network maps samples drawn from that posterior back to the data space; the reparameterisation trick renders the sampling step differentiable, enabling end-to-end gradient-based learning. VAEs underpin latent diffusion models, representation learning, disentanglement research, and multimodal generative systems, and have been extended by hierarchical, vector-quantised, and conditional variants that dramatically improve fidelity and controllability.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:vae", "labels": [ "VAE", "3D Causal VAE", "VAE Image Encoding", "Variational Autoencoder", "Video VAE" ], "is_subclass_of": [ "Deep Generative Model" ], "wikilinks": [] }, { "id": "validation-report", "title": "VALIDATION_REPORT", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Validation Report is a systematic assessment document verifying ontology compliance, data quality, and structural integrity across knowledge graphs and semantic systems. It checks that class hierarchies, relation targets, definition coverage, and maturity fields meet specified quality thresholds, producing actionable findings for enrichment pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:validation-report", "labels": [ "VALIDATION_REPORT" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain", "QualityAssurance" ] }, { "id": "vc-dimension", "title": "VC Dimension", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The Vapnik-Chervonenkis (VC) dimension is a measure of the capacity of a hypothesis class, defined as the size of the largest set of points that the class can shatter \u2014 classify correctly under every possible binary labelling. Introduced by Vapnik and Chervonenkis in 1971, it yields distribution-free generalisation bounds in statistical learning theory: a finite VC dimension guarantees that empirical risk converges uniformly to true risk as sample size grows, linking model complexity to the amount of data needed to learn reliably.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:vc-dimension", "labels": [ "VC Dimension" ], "is_subclass_of": [ "Model Complexity" ], "wikilinks": [ "Model Complexity", "Statistical Learning Theory", "Bias-Variance Tradeoff", "Overfitting" ] }, { "id": "verification-report", "title": "VERIFICATION REPORT", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A verification report is a structured document recording the outcomes of systematic assessment activities that confirm ontologies, knowledge graphs, and metaverse infrastructure meet specified quality standards and functional requirements. It captures defect categorisation, schema compliance checks, semantic consistency validation, RDF triple counts, OWL axiom correctness, and remediation progress to provide transparent evidence of production readiness.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:verification-report", "labels": [ "VERIFICATION REPORT" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "DataIntegrity", "DataQualityMetrics", "OntologyAssets", "OntologyStandards", "OWLAxioms", "QualityMetrics", "RDFTriples", "SchemaCompliance", "SemanticConsistency", "KnowledgeGraph", "MetaverseDomain" ] }, { "id": "verification", "title": "VERIFICATION", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Verification is the process of confirming that information, claims, and digital artefacts meet specified standards for accuracy, completeness, and compliance with constraints through cryptographic proofs, rule-based validation, consensus mechanisms, and user acceptance testing. Within metaverse and blockchain contexts it encompasses transaction authenticity, smart contract correctness, ontology schema conformance, and digital asset provenance, establishing trust between stakeholders in decentralised systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:verification", "labels": [ "VERIFICATION", "Asset Verification", "Proof Verification", "User Verification", "Verification", "Verification Mechanism", "Verification Procedure", "Verification and Assurance" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "BlockchainConsensus", "CryptographicProofs", "DigitalAssetAuthenticity", "MetaversePlatforms", "OntologyValidation", "RuleBaseValidation", "SmartContractAudit", "TransactionVerification", "UserAcceptanceTesting", "DigitalSignature", "MetaverseDomain" ] }, { "id": "vfx-tools", "title": "VFX Tools", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "VFX Tools are software applications and pipelines used to create, composite, and render visual effects for film, television, real-time engines, and immersive media. They encompass particle simulation, fluid and cloth dynamics, compositing, colour grading, and procedural generation capabilities, and increasingly interface with real-time rendering engines for deployment in spatial computing contexts.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:vfx-tools", "labels": [ "VFX Tools" ], "is_subclass_of": [ "Display and Rendering", "Visual Effects" ], "wikilinks": [ "owl:Thing" ] }, { "id": "vp9-codec", "title": "VP9 Codec", "domain": "metaverse", "domain_name": "Metaverse", "definition": "VP9 is an open, royalty-free video compression codec developed by Google as the successor to VP8, achieving roughly half the bitrate of H.264 at comparable quality. It supports resolutions up to 8K, 10/12-bit colour depth, and is widely deployed for streaming on YouTube and within WebRTC real-time communication. VP9's open licensing made it a key building block for browser-native video before AV1 succeeded it.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vp9-codec", "labels": [ "VP9 Codec", "VP9" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "vpn", "title": "VPN", "domain": "security", "domain_name": "Security", "definition": "A Virtual Private Network (VPN) is a security technology that establishes an encrypted tunnel between a user's device and a remote server, routing traffic through it so that data confidentiality and integrity are preserved across untrusted networks. By encapsulating and encrypting packets, a VPN conceals the user's originating address and protects communications from interception on shared or public infrastructure. VPNs are widely used for secure remote access to private networks, for privacy on public Wi-Fi, and for circumventing network-level filtering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vpn", "labels": [ "VPN" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "vr-controllers", "title": "VR Controllers", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Handheld input devices designed for virtual reality systems that enable users to interact with virtual environments through motion tracking, buttons, triggers, and haptic feedback, providing intuitive manipulation of virtual objects and navigation through immersive spaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:vr-controllers", "labels": [ "VR Controllers" ], "is_subclass_of": [ "Interaction Technology", "XR Hardware" ], "wikilinks": [ "metaverse", "XR Hardware" ] }, { "id": "vr-experiences", "title": "VR Experiences", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Designed interactive episodes delivered through fully immersive virtual reality, in which a head-mounted display replaces the user's visual field with a rendered or captured environment they can look around, move through, and act within. VR experiences span entertainment, training, virtual tours, and destination previews; unlike AR experiences, which overlay content on the physical world, they substitute the world entirely, trading environmental awareness for presence and total authorial control of the scene.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:vr-experiences", "labels": [ "VR Experiences" ], "is_subclass_of": [ "Immersive Experience" ], "wikilinks": [ "Immersive Experience", "Virtual Reality", "Destination Marketing" ] }, { "id": "vr-psychology", "title": "VR Psychology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "VR Psychology is the scientific study of how virtual reality environments affect human cognition, perception, emotion, behaviour, and wellbeing. It encompasses research into presence, embodiment, cybersickness, avatar effects, and therapeutic applications of immersive virtual environments, informing the design of XR systems that are psychologically safe and effective.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:vr-psychology", "labels": [ "VR Psychology" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "vr-rendering-engine", "title": "VR Rendering Engine", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specialised real-time graphics software that generates stereoscopic imagery for virtual reality headsets, optimising frame rates (minimum 90 fps), motion-to-photon latency (sub-20 ms), and visual fidelity while managing VR-specific pipeline requirements including foveated rendering, asynchronous reprojection, and lens distortion correction. Leading implementations include Unreal Engine 5 (Lumen, Nanite) and Unity (URP/HDRP), both supporting major XR hardware via OpenXR and platform-specific SDKs.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:vr-rendering-engine", "labels": [ "VR Rendering Engine" ], "is_subclass_of": [ "Display and Rendering", "Game Engine" ], "wikilinks": [ "Game Engine", "metaverse" ] }, { "id": "vrchat", "title": "VRChat", "domain": "metaverse", "domain_name": "Metaverse", "definition": "VRChat is a social virtual reality platform where users interact as custom avatars in user-created 3D worlds, with support for both VR headsets and desktop access.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:vrchat", "labels": [ "VRChat" ], "is_subclass_of": [ "Metaverse Platform" ], "wikilinks": [ "Avatar System", "Virtual Reality", "Social VR", "Avatar Customization", "Avatar Interoperability", "Metaverse Platform" ] }, { "id": "vrm-format", "title": "VRM Format", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An open file format standard for 3D humanoid avatars built on glTF, designed for cross-platform interoperability in VR, AR, and metaverse applications. It incorporates standardised humanoid rigging, facial blend-shape expressions, first-person gaze controls, and embedded licensing metadata, enabling seamless avatar portability across different virtual environments. In 2024 the VRM Consortium began collaboration with the Khronos Group to advance VRM specifications as official glTF extensions for international standardisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:vrm-format", "labels": [ "VRM Format", "VRM Avatar Format" ], "is_subclass_of": [ "Standards and Interoperability", "3D File Format" ], "wikilinks": [ "3D File Format", "metaverse" ] }, { "id": "vacuum-arc-thruster", "title": "Vacuum Arc Thruster", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vacuum-arc-thruster", "labels": [ "Vacuum Arc Thruster" ], "is_subclass_of": [ "Thruster" ], "wikilinks": [] }, { "id": "vadose-zone", "title": "Vadose Zone", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vadose-zone", "labels": [ "Vadose Zone", "UnsaturatedZone" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "validated-data-state", "title": "Validated Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:validated-data-state", "labels": [ "Validated Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "validation-constraints", "title": "Validation Constraints", "domain": "data", "domain_name": "Data", "definition": "Validation constraints are the formally specified rules that data or metadata must satisfy to be considered valid against a schema, covering datatypes, cardinality, value ranges, required fields and structural relationships. Expressed through languages such as SHACL, JSON Schema or XSD, they let systems automatically verify conformance and reject malformed input. They are an integral part of metadata standards and ontology schemas.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:validation-constraints", "labels": [ "Validation Constraints" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "validation-dataset", "title": "Validation Dataset", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:validation-dataset", "labels": [ "Validation Dataset" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "validation-evidence", "title": "Validation Evidence", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:validation-evidence", "labels": [ "Validation Evidence" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "validation-process", "title": "Validation Process", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Activity of systematically checking wher systems, components, or implementations satisfy specified requirements, standards, and compliance criteria through verification testing and quality assurance procedures.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:validation-process", "labels": [ "Validation Process", "Validation" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "Compliance Testing", "IEEE P2048-9", "ISO 9001", "Quality Assurance Workflows", "Quality Certification", "Quality Metrics", "Regulatory Approval", "Requirement Verification", "Requirements Specification", "Results Analysis", "System Acceptance", "Test Data", "Test Execution", "Test Framework", "Testing Tools", "Validation Criteria", "Validation Rules", "Compliance Standards", "Compliance Verification", "Data Layer" ] }, { "id": "validation-rules", "title": "Validation Rules", "domain": "data", "domain_name": "Data", "definition": "Defined constraints that data or input must satisfy to be accepted, used to enforce correctness, consistency and integrity before processing or storage.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:validation-rules", "labels": [ "Validation Rules" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [ "Constraint", "Data Management", "Rule-Based Systems" ] }, { "id": "validation-set", "title": "Validation Set", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A Validation Set is a held-out partition of data, distinct from the training and test sets, used to tune hyperparameters and make model-selection decisions during machine learning development. By evaluating candidate models on data not used for fitting, it provides an unbiased signal for choices such as architecture, regularisation strength and early stopping. The test set is reserved for a final, untouched estimate of generalisation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:validation-set", "labels": [ "Validation Set" ], "is_subclass_of": [ "Model Evaluation", "Dataset" ], "wikilinks": [] }, { "id": "validation-tools", "title": "Validation Tools", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Validation tools are software utilities that verify whether data, documents or implementations conform to a defined specification, schema or interoperability profile. They automate conformance checking, report violations, and support certification and compatibility testing across standards-based ecosystems. By catching errors early they reduce integration risk and ensure data and formats interoperate as intended.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:validation-tools", "labels": [ "Validation Tools" ], "is_subclass_of": [ "Software Development" ], "wikilinks": [] }, { "id": "validator-economics", "title": "Validator Economics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Validator economics describes the incentive structures, cost-reward equilibria, and rational behaviour models governing participants who operate validator nodes in proof-of-stake and delegated consensus blockchain networks. It encompasses staking rewards (block rewards, transaction fees, MEV capture), slashing penalties for Byzantine behaviour (double-signing, downtime), the competitive market for delegated stake, operational costs (hardware, bandwidth, maintenance), and the resulting equilibrium between security budget, validator profitability, and network decentralisation. Well-designed validator economics align individual rational self-interest with network security and liveness.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:validator-economics", "labels": [ "Validator Economics" ], "is_subclass_of": [ "Blockchain Economics" ], "wikilinks": [] }, { "id": "validator-network", "title": "Validator Network", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A validator network is the collective set of nodes in a proof-of-stake or similar blockchain that are responsible for proposing blocks, attesting to their validity, and participating in the consensus process that finalises the chain's state. Individual validators stake capital as a bond against misbehaviour and are rewarded or penalised according to their participation and honesty. The size, geographic distribution, and stake concentration of a validator network are key determinants of a blockchain's decentralisation and censorship resistance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:validator-network", "labels": [ "Validator Network" ], "is_subclass_of": [ "Validator" ], "wikilinks": [] }, { "id": "validator-node", "title": "Validator Node", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A network participant in a Proof of Stake or similar blockchain that locks collateral (stake) to earn the right to propose and attest to blocks, and is subject to slashing penalties for equivocation or liveness failures. Validator nodes form the security backbone of staking-based consensus, replacing the hash-rate competition of mining.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:validator-node", "labels": [ "Validator Node", "Validator Nodes" ], "is_subclass_of": [ "Blockchain Entity", "NetworkComponent" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "NetworkComponent", "SecurityLayer" ] }, { "id": "validator-selection", "title": "Validator Selection", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Validator selection is the consensus mechanism by which a proof-of-stake blockchain chooses which staked participants are eligible to propose and attest to blocks in a given slot or epoch. It typically uses stake-weighted, often pseudo-randomised sampling to assign block production while preserving security and decentralisation. The design directly affects fairness, finality, energy efficiency and resistance to manipulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:validator-selection", "labels": [ "Validator Selection", "Committee Selection" ], "is_subclass_of": [ "Proof of Stake" ], "wikilinks": [] }, { "id": "validator-set", "title": "validator set", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A validator set is the finite, dynamically managed collection of nodes in a proof-of-stake or delegated consensus blockchain that are authorised to propose blocks, attest to their validity, and participate in the finality process during each epoch. Membership is gated by a minimum stake bond whose collateral is subject to slashing penalties for protocol violations such as equivocation or surround voting, aligning validator incentives with network security. The set is recomposed at epoch boundaries according to delegation weights, unbonding queues, and slashing events, and its cardinality governs the fundamental trade-off between decentralisation, Byzantine fault tolerance, and consensus latency. In sharded or layered architectures, validator sets may be further partitioned into sub-committees to scale throughput while preserving cryptographic security guarantees.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:validator-set", "labels": [ "Validator Set", "Permissioned Validator Set" ], "is_subclass_of": [ "Consensus Mechanism" ], "wikilinks": [] }, { "id": "validator", "title": "Validator", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A validator is a network participant in a proof-of-stake blockchain that is responsible for proposing new blocks, attesting to the validity of blocks proposed by others, and participating in the finality mechanism of the network. Validators commit a quantity of cryptocurrency as collateral (stake), which is subject to slashing \u2014 partial or total confiscation \u2014 if the validator behaves dishonestly or fails liveness requirements. Through their collective attestations and proposals, validators form the active set that drives consensus and maintains the canonical chain history.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:validator", "labels": [ "Validator", "Blockchain Validator", "Federated Validator", "Schema Validator", "Transaction Validator" ], "is_subclass_of": [ "Consensus Protocol" ], "wikilinks": [ "Proof of Stake", "Consensus Mechanism", "Block", "Consensus Protocol" ] }, { "id": "validity-proof", "title": "Validity Proof", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A validity proof is a cryptographic proof, typically a succinct zero-knowledge proof, that attests that a batch of state transitions was computed correctly according to the rules of a system. In layer-2 rollups it allows a base chain to accept a compressed state update after verifying a single proof, without re-executing the underlying transactions. Validity proofs give immediate, trustless finality in contrast to optimistic schemes that rely on fraud challenges.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:validity-proof", "labels": [ "Validity Proof" ], "is_subclass_of": [ "Cryptographic Proof System" ], "wikilinks": [] }, { "id": "valuation-multiples", "title": "Valuation Multiples", "domain": "economics", "domain_name": "Economics", "definition": "A financial metric representing the ratio of a company's stock price or market capitalization to a fundamental value such as earnings or revenue, used to assess relative worth.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:valuation-multiples", "labels": [ "Valuation Multiples" ], "is_subclass_of": [ "Financial Metrics" ], "wikilinks": [] }, { "id": "value-alignment", "title": "Value Alignment", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The challenge and process of ensuring AI systems pursue objectives that align with human values, even as those systems become more capable and autonomous. Value alignment addresses both technical and philosophical questions about encoding human preferences into AI behaviour.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:value-alignment", "labels": [ "Value Alignment", "Value Alignment Problem" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [ "barsky1987fisher", "Bhatia2021", "borio2017fx", "cagan1958demand", "davies2010history", "hall2009inflation", "homer1996history", "maurel2012keynesian", "ponzi2021alden", "selgin1996defense", "stroukal2018can", "szabo2002shelling", "white1914fiat", "Bitcoin As Money", "MetaverseDomain", "Money" ] }, { "id": "value-function", "title": "Value Function", "domain": "ai", "domain_name": "Ai", "definition": "A value function in reinforcement learning estimates the expected cumulative future reward obtainable from a given state (state-value) or state-action pair (action-value) under a particular policy. It captures the long-term desirability of situations rather than immediate reward, and satisfies the recursive Bellman equation that relates the value of a state to the values of its successors. Value functions are central to dynamic programming, temporal-difference learning, and actor-critic methods, providing the signal that guides an agent toward reward-maximising behaviour.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:value-function", "labels": [ "Value Function", "ValueFunction" ], "is_subclass_of": [ "Reinforcement Learning", "AI Technique" ], "wikilinks": [] }, { "id": "value-transfer", "title": "Value Transfer", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The mechanism and process by which economic value, rights, or utility are exchanged between parties across physical, digital, and virtual domains, encompassing monetary systems, token-based systems, resource allocation, and rights transfer protocols.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:value-transfer", "labels": [ "Value Transfer", "Inter-Chain Value Transfer", "Value Exchange" ], "is_subclass_of": [ "DeFi and Economics", "Cryptocurrency" ], "wikilinks": [ "1inch", "Aave", "ACM Digital Library", "AI Agent Economic Interaction", "Aptos", "Arbitrum", "Argent", "AutoGPT", "Avalanche", "Axelar", "Axie Infinity", "Aztec", "Aztec Network", "Balancer", "Bancor", "Barclays", "Base", "BBS+ Signatures", "Binance Smart Chain", "BIP-341" ] }, { "id": "value-vector", "title": "Value Vector", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Value Vector is a structured numerical representation that encodes the relative importance or utility of different outcomes, objectives, or features within an AI system, enabling multi-objective optimisation and preference-guided decision-making. It aggregates scalar values across dimensions such as safety, performance, cost, and alignment to express composite desiderata, and is foundational to reward modelling, preference learning, and value alignment research.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:value-vector", "labels": [ "Value Vector" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "value-sensitive-design", "title": "Value-Sensitive Design", "domain": "governance", "domain_name": "Governance", "definition": "Value-sensitive design is a methodology for building technology that accounts for human values \u2014 such as privacy, autonomy, fairness and wellbeing \u2014 throughout the design process rather than as an afterthought. It combines conceptual investigation of stakeholder values, empirical study of how people are actually affected, and technical work to embed the resulting requirements into system design. The approach is widely referenced in AI ethics frameworks and standards, including IEEE 7000, as a structured way to surface and reconcile competing stakeholder values before deployment.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:value-sensitive-design", "labels": [ "Value-Sensitive Design" ], "is_subclass_of": [ "AI Ethics" ], "wikilinks": [] }, { "id": "value", "title": "Value", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "In the AI domain, a principle, preference, or objective that guides the behaviour and optimisation targets of an intelligent system. Values encode what outcomes a system should pursue or avoid, underpinning alignment research, reward function design, and ethical AI. They range from measurable utility metrics to abstract normative principles such as fairness, human dignity, and autonomy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:value", "labels": [ "Value", "Value Capture", "Value Framework" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "van-allen-radiation-belt", "title": "Van Allen Radiation Belt", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:van-allen-radiation-belt", "labels": [ "Van Allen Radiation Belt" ], "is_subclass_of": [ "Radiation Belt" ], "wikilinks": [] }, { "id": "vanishing-gradient-problem", "title": "Vanishing Gradient Problem", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The vanishing gradient problem is a difficulty in training deep or recurrent neural networks in which error gradients shrink exponentially as they are propagated backwards through many layers or time steps, leaving early parameters with negligible updates. It arises because repeated multiplication by small derivative terms, characteristic of saturating activation functions, drives the gradient towards zero. The effect severely impedes learning of long-range dependencies and was a central obstacle to deep learning before mitigations emerged. Remedies include non-saturating activations, careful weight initialisation, normalisation, residual connections and gated recurrent architectures such as the LSTM.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vanishing-gradient-problem", "labels": [ "Vanishing Gradient Problem" ], "is_subclass_of": [ "Machine Learning", "Deep Learning" ], "wikilinks": [] }, { "id": "variable-impedance-control", "title": "Variable Impedance Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Variable Impedance Control is a robot control strategy that modulates stiffness, damping, and inertia parameters online during task execution, adapting mechanical impedance in response to sensed interaction forces, task phase, or environmental uncertainty rather than maintaining fixed impedance. This contrasts with conventional impedance control, which prescribes constant mechanical properties, by enabling robots to behave compliantly during contact-rich or uncertain phases and rigidly during free-space precision movements. Key implementations include learning-based approaches that infer optimal impedance trajectories from demonstrations, model predictive formulations that optimise impedance over a receding horizon, and biomimetic strategies that replicate the variable stiffness observed in human neuromuscular systems to achieve safe, dexterous physical interaction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:variable-impedance-control", "labels": [ "Variable Impedance Control" ], "is_subclass_of": [ "Actuation and Control", "Impedance Control" ], "wikilinks": [ "Impedance Control", "Robotics" ] }, { "id": "variable-star", "title": "Variable Star", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:variable-star", "labels": [ "Variable Star" ], "is_subclass_of": [ "Star" ], "wikilinks": [] }, { "id": "variable-stiffness-actuator", "title": "Variable Stiffness Actuator", "domain": "robotics", "domain_name": "Robotics", "definition": "A variable stiffness actuator (VSA) is a compliant robotic actuator whose mechanical stiffness can be adjusted independently of its position, typically using antagonistic springs or adjustable elastic elements. This adjustable compliance lets robots store and release energy, absorb impacts and interact safely with humans and uncertain environments. VSAs are central to safe, energy-efficient force control in physical human-robot interaction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:variable-stiffness-actuator", "labels": [ "Variable Stiffness Actuator" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "variance-reduction", "title": "Variance Reduction", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Variance Reduction is a family of techniques that lower the statistical variance of Monte Carlo and stochastic estimators so that fewer samples are needed to reach a target accuracy. Methods such as importance sampling, control variates, antithetic variates, and stratification reshape how samples are drawn or combined without introducing bias. In machine learning, variance reduction stabilises gradient estimates in stochastic optimisation and accelerates convergence of simulation-based methods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:variance-reduction", "labels": [ "Variance Reduction" ], "is_subclass_of": [ "Monte Carlo Integration" ], "wikilinks": [] }, { "id": "variance", "title": "Variance", "domain": "data", "domain_name": "Data", "definition": "The fundamental statistical measure of dispersion, defined as the expected squared deviation of a random variable from its mean, Var(X) = E[(X \u2212 E[X])\u00b2]. Its square root is the standard deviation; its multivariate generalisation is the covariance matrix; and it parameterises the spread of the Gaussian distribution. Variance governs the reliability of sample-based estimates, decomposes prediction error in the bias-variance trade-off, and is the quantity that variance-reduction techniques in Monte Carlo estimation and stochastic optimisation exist to control.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:variance", "labels": [ "Variance" ], "is_subclass_of": [ "Statistics" ], "wikilinks": [ "Statistics", "Covariance Matrix", "Gaussian Distribution", "Probability Distribution", "Random Sampling" ] }, { "id": "variational-autoencoder", "title": "Variational Autoencoder", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A generative model that pairs an encoder mapping inputs to a probability distribution over a latent space with a decoder that reconstructs inputs, trained to maximise a variational lower bound (ELBO) on the data likelihood via the reparameterisation trick.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:variational-autoencoder", "labels": [ "Variational Autoencoder" ], "is_subclass_of": [ "Autoencoder" ], "wikilinks": [ "Variational Inference", "Backpropagation", "Image Generation", "Generative Model", "Autoencoder" ] }, { "id": "variational-autoencoders", "title": "Variational Autoencoders", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Generative neural network architecture combining autoencoders with variational inference, learning a probabilistic latent space via an encoder that outputs distribution parameters and a decoder that reconstructs data by sampling from that distribution. Trained by maximising the Evidence Lower BOund (ELBO), which balances reconstruction fidelity and KL-divergence regularisation.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:variational-autoencoders", "labels": [ "Variational Autoencoders" ], "is_subclass_of": [ "AI Model Architecture", "Generative Model" ], "wikilinks": [ "3D Convolutional Network", "Active Units", "Adam Optimizer", "AdamW", "Adversarial Autoencoder", "Alireza Makhzani", "AlphaFold", "AlphaFold2", "Amortized Inference", "Anomaly Detection", "AR", "Attention Mechanisms", "Audio Generation", "Audio Synthesis", "Autonomous Robots", "Autoregressive", "Autoregressive Model", "Batch Normalization", "Bayesian Optimization", "Berkeley" ] }, { "id": "variational-inference", "title": "variational inference", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Variational inference (VI) is a family of algorithms in Bayesian machine learning that approximates intractable posterior distributions p(z|x) by positing a simpler, tractable family of distributions q(z; \u03c6) and optimising its parameters to minimise the Kullback-Leibler divergence from the true posterior, equivalently maximising the Evidence Lower BOund (ELBO) on the log marginal likelihood. By recasting probabilistic inference as an optimisation problem rather than a sampling problem, VI achieves scalability to large datasets and high-dimensional latent spaces that Markov Chain Monte Carlo methods cannot easily reach. The reparameterisation trick enables gradient-based ELBO optimisation through stochastic estimates, making VI the foundational inference engine for Variational Autoencoders, hierarchical generative models, and probabilistic programming systems. VI trades posterior exactness for computational tractability, and is the preferred method wherever fast, amortised, or online Bayesian inference is required.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:variational-inference", "labels": [ "Variational Inference" ], "is_subclass_of": [ "Bayesian Inference" ], "wikilinks": [] }, { "id": "ve-chain", "title": "VeChain", "domain": "blockchain", "domain_name": "Blockchain", "definition": "VeChain is a public blockchain platform oriented toward supply-chain management and enterprise applications. It uses a proof-of-authority consensus model in which a limited set of authorised validators produce blocks, prioritising throughput and predictability for business use. The platform employs a dual-token design, separating the asset used to hold value from the token used to pay for transaction costs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:ve-chain", "labels": [ "VeChain" ], "is_subclass_of": [ "Blockchain", "Layer 1 Blockchain", "Blockchain Domain" ], "wikilinks": [ "Proof of Authority", "Smart Contract", "Supply Chain Tracking", "Product Provenance", "Internet of Things", "Digital Asset Domain", "Blockchain Domain" ] }, { "id": "vector-clock", "title": "Vector Clock", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A vector clock is a data structure used in distributed systems to capture causal relationships between events across multiple nodes without relying on synchronised physical time. Each node maintains a counter for every other node in the system, incrementing its own counter on each local event and merging received counters on communication. By comparing vector timestamps, systems can determine whether events are causally related, concurrent, or ordered, which is fundamental for conflict detection and resolution in collaborative editing.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:vector-clock", "labels": [ "Vector Clock" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "vector-clocks", "title": "Vector Clocks", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "A mechanism for ordering events in a distributed system by assigning each process a vector of counters. Comparing vectors determines whether one event causally precedes another or whether they are concurrent.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:vector-clocks", "labels": [ "Vector Clocks", "Hybrid Logical Clocks" ], "is_subclass_of": [ "Clock Synchronization" ], "wikilinks": [ "Distributed Computing", "Eventual Consistency", "Clock Synchronization" ] }, { "id": "vector-commitment", "title": "Vector Commitment", "domain": "security", "domain_name": "Security", "definition": "A vector commitment is a cryptographic primitive that produces a short, binding commitment to an ordered sequence of values such that the committer can later open any individual position with a compact proof of its value. It is position-binding, meaning one cannot produce valid openings of two different values at the same index, and it supports succinct membership and update proofs. Vector commitments generalise Merkle trees and underpin verifiable databases, stateless blockchains and proof systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:vector-commitment", "labels": [ "Vector Commitment" ], "is_subclass_of": [ "Commitment Scheme" ], "wikilinks": [] }, { "id": "vector-database", "title": "vector database", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A vector database is a specialised data storage and retrieval system optimised for indexing and querying high-dimensional dense vectors (embeddings) using approximate nearest-neighbour (ANN) search algorithms such as HNSW, IVF-PQ, and ScaNN. Unlike traditional relational databases, vector databases solve the problem of finding the k most similar vectors to a query in Euclidean or cosine space over millions to billions of embeddings representing text, images, audio, or multimodal content. They combine ANN indexing with structured metadata filtering, persistence, CRUD operations, and access control, enabling retrieval-augmented generation, semantic search, and large-scale recommendation at production quality. The category encompasses dedicated systems (Pinecone, Qdrant, Weaviate, Milvus, Chroma) and vector-search extensions added to existing databases (pgvector for PostgreSQL, Redis Stack).", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:vector-database", "labels": [ "Vector Database" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "vector-databases", "title": "Vector Databases", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Database systems specialised for storing high-dimensional vector embeddings and performing similarity search over them, typically using approximate nearest neighbour indexing algorithms such as HNSW and IVF.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vector-databases", "labels": [ "Vector Databases" ], "is_subclass_of": [ "Vector Database" ], "wikilinks": [ "Embeddings", "Vector Search", "Retrieval-Augmented Generation", "Semantic Search", "Vector Database" ] }, { "id": "vector-embedding", "title": "Vector Embedding", "domain": "ai", "domain_name": "Ai", "definition": "A vector embedding is a dense, fixed-length numerical representation of a discrete object \u2014 a word, sentence, image, or user \u2014 in a continuous high-dimensional space, learned such that semantically similar objects map to nearby points. Embeddings convert unstructured data into a form amenable to mathematical operations like distance and dot product, enabling similarity search, clustering, recommendation, and retrieval. They are produced by neural encoders and are the representational substrate beneath modern search, retrieval-augmented generation, and multimodal AI.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:vector-embedding", "labels": [ "Vector Embedding" ], "is_subclass_of": [ "Embedding" ], "wikilinks": [] }, { "id": "vector-index", "title": "Vector Index", "domain": "ai", "domain_name": "Ai", "definition": "A vector index is a data structure that organises high-dimensional embedding vectors to enable fast approximate nearest-neighbour search over large collections. Using methods such as HNSW graphs, IVF partitioning or product quantisation, it trades a small amount of recall for large gains in query latency and scalability. Vector indexes are the retrieval engine behind semantic search and retrieval-augmented generation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vector-index", "labels": [ "Vector Index" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "vector-search", "title": "Vector Search", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Vector search is a retrieval paradigm that identifies the items in a corpus whose high-dimensional vector representations are most similar to a query vector, using distance or similarity metrics such as cosine similarity, dot product, or Euclidean distance. Because exhaustive pairwise comparison scales as O(n\u00b7d) and becomes intractable for large corpora, practical systems use approximate nearest neighbour (ANN) algorithms \u2014 including HNSW, IVF-PQ, and LSH \u2014 that trade a small, configurable recall loss for orders-of-magnitude latency improvement. Vectors are typically produced by embedding models that encode semantic, visual, or multimodal meaning into dense float arrays, making vector search inherently meaning-sensitive rather than lexical. The paradigm underpins semantic search, recommendation systems, retrieval-augmented generation, duplicate detection, and cross-modal retrieval across text, image, audio, and structured data.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vector-search", "labels": [ "Vector Search" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [ "Embeddings", "Semantic Search", "Vector Database", "Information Retrieval" ] }, { "id": "vector-space-model", "title": "Vector Space Model", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The vector space model represents documents, queries or tokens as vectors in a high-dimensional space, typically weighted by term frequency, so that similarity between items can be computed geometrically. Cosine similarity between vectors is the standard measure of relatedness in this representation, underpinning classical information retrieval ranking. Modern token and word embeddings generalise the vector space model by learning dense, continuous representations rather than sparse term-frequency vectors.", "entityType": "Class", "qualityScore": 0.55, "maturity": "mature", "iri": "urn:ngm:class:vector-space-model", "labels": [ "Vector Space Model" ], "is_subclass_of": [ "Information Retrieval" ], "wikilinks": [] }, { "id": "vector-store", "title": "Vector Store", "domain": "data", "domain_name": "Data", "definition": "A storage system optimised for holding high-dimensional vector embeddings and retrieving them by similarity. It supports nearest-neighbour search used in semantic retrieval and machine learning applications.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:vector-store", "labels": [ "Vector Store" ], "is_subclass_of": [ "Vector Database" ], "wikilinks": [ "Vector Database", "Natural Language Processing", "Knowledge Representation" ] }, { "id": "vegetation-index", "title": "Vegetation Index", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vegetation-index", "labels": [ "Vegetation Index" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vehicle-to-grid", "title": "Vehicle to Grid", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Vehicle-to-grid (V2G) is a technology and operating model that allows an electric vehicle's battery to discharge stored energy back into the electricity grid, turning parked EVs into a distributed, dispatchable energy resource. It requires bidirectional charging hardware and grid-integration software that coordinates charge and discharge cycles with grid operator signals and electricity prices. V2G is a key enabling technology for demand response programmes and for smoothing the intermittency of renewable generation on the grid.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:vehicle-to-grid", "labels": [ "Vehicle to Grid", "Vehicle-to-Grid" ], "is_subclass_of": [ "Smart Grid" ], "wikilinks": [] }, { "id": "velocity", "title": "Velocity", "domain": "robotics", "domain_name": "Robotics", "definition": "Velocity is the vector quantity expressing rate of change of position with respect to time, formally defined as v = dx/dt. In robotics and autonomous systems, it is a fundamental parameter governing motion planning, collision avoidance, and control, with algorithms such as Velocity Obstacle (VO) and Reciprocal Velocity Obstacle (RVO) built directly upon it.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:velocity", "labels": [ "Velocity" ], "is_subclass_of": [ "Actuation and Control", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "velocity-control", "title": "VelocityControl", "domain": "robotics", "domain_name": "Robotics", "definition": "A closed-loop feedback control strategy that regulates the rate of change of position (linear velocity v in m/s or angular velocity \u03c9 in rad/s) of a robotic actuator, motor shaft, conveyor system, or mobile robot platform by continuously measuring actual velocity via tachometers (\u00b1full-scale accu...", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:velocity-control", "labels": [ "VelocityControl", "Velocity Control" ], "is_subclass_of": [ "Actuation and Control", "ControlAlgorithm", "FeedbackControl", "MotionControl", "ServoControl" ], "wikilinks": [ "ANSI/RIA R15.06 Industrial Robot Safety", "BandwidthDesign", "CascadeControl", "ConstantThroughput", "ControllerTuning", "CurrentAmplifier", "DigitalController", "EnergyEfficiency", "ErrorCalculation", "Feedback Control of Dynamic Systems - Franklin et al.", "FeedforwardCompensation", "FeedforwardControl", "IEC 61800 Adjustable Speed Electrical Power Drive Systems", "IEEE Std 1547 Distributed Energy Resources", "ISO 8373 Robotics Vocabulary", "MotorDriver", "NoiseFiltering", "ObserverDesign", "PIControl", "PIController" ] }, { "id": "vendor-lock-in", "title": "Vendor Lock-in", "domain": "standards", "domain_name": "Standards", "definition": "Vendor lock-in is a situation in which a customer becomes dependent on a particular supplier's products or services and cannot switch to an alternative without substantial cost, effort or disruption. It arises from proprietary formats, non-portable data, integration dependencies and incompatible interfaces that raise switching costs. Open standards, data portability and interoperability are the principal countermeasures that preserve customer choice.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vendor-lock-in", "labels": [ "Vendor Lock-in", "Vendor Lock-In" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "vendor-neutrality", "title": "Vendor Neutrality", "domain": "standards", "domain_name": "Standards", "definition": "Vendor Neutrality is the principle that systems, standards and procurement should not favour or depend upon any single supplier's proprietary technology. It is achieved through open standards, well-defined interfaces and portable data formats that allow components from different vendors to interoperate and be substituted. Vendor neutrality reduces lock-in, preserves bargaining power and protects long-term access to data and capabilities.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vendor-neutrality", "labels": [ "Vendor Neutrality" ], "is_subclass_of": [ "Open Standards" ], "wikilinks": [] }, { "id": "venture-capital", "title": "Venture Capital", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Venture capital is a form of private financing in which investors provide funding to early-stage, high-growth companies in exchange for equity or, in crypto markets, for token allocations. It accepts elevated risk in pursuit of outsized returns, typically deploying capital in staged rounds with active governance and mentorship. In the blockchain sector, venture capital frequently coexists with and contrasts against token-sale mechanisms as a route to capitalising protocols.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:venture-capital", "labels": [ "Venture Capital" ], "is_subclass_of": [ "Finance" ], "wikilinks": [] }, { "id": "venue-tethered-immersive-experience", "title": "Venue-Tethered Immersive Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An immersive or interactive experience tied to a specific physical venue or geographic location, combining real-world presence with digital or spatial computing overlays. Location-based experiences leverage spatial mapping, real-time content delivery, and user proximity to deliver contextually relevant narrative or interactive content that cannot be replicated remotely.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:venue-tethered-immersive-experience", "labels": [ "Venue-Tethered Immersive Experience", "Location Based Experience" ], "is_subclass_of": [ "Platform and Environment", "Immersive Experience" ], "wikilinks": [ "Immersive" ] }, { "id": "vercel", "title": "Vercel", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Vercel is a company that provides a cloud platform for deploying and hosting web applications, with an emphasis on frontend frameworks.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:vercel", "labels": [ "Vercel" ], "is_subclass_of": [ "Cloud Platform" ], "wikilinks": [ "Cloud Computing", "Cloud Platform", "Scalability" ] }, { "id": "verifiable-computation", "title": "Verifiable Computation", "domain": "security", "domain_name": "Security", "definition": "Techniques that let a party outsource a computation and receive a proof that the result is correct, allowing efficient verification without redoing the work. Proof systems such as SNARKs and STARKs make verification exponentially cheaper than re-execution, enabling trust-minimised delegation at scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:verifiable-computation", "labels": [ "Verifiable Computation", "Provable Computation" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [ "Cryptographic Proof", "Zero-Knowledge Proof", "Scalability", "Trusted Execution Environments" ] }, { "id": "verifiable-credential-vc", "title": "Verifiable Credential (VC)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A W3C standard for tamper-evident credentials that can be cryptographically verified, containing claims made by an issuer about a subject, enabling trustable digital attestations without requiring direct communication with the issuer.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:verifiable-credential-vc", "labels": [ "Verifiable Credential (VC)", "Verifiable Credential" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Claim", "Credential Metadata", "Credential Schema", "Credential Status Registry", "Cryptographic Proof", "Identity Wallet", "Issuer Signature", "JSON-LD", "Linked Data Signatures", "Privacy-Preserving Verification", "Selective Disclosure", "Self-Sovereign Identity (SSI)", "Trustable Attestations", "Verifiable Presentations", "W3C VC Data Model", "W3C Verifiable Credentials Data Model", "Decentralized Identity (DID)", "Digital Signature", "MiddlewareLayer", "Public Key Infrastructure" ] }, { "id": "verifiable-credential-standard", "title": "Verifiable Credential Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The W3C Verifiable Credentials Data Model specification defines a structured, cryptographically verifiable format for expressing claims about subjects. Credentials are issued by a trusted issuer, held by a subject, and verified by relying parties using digital signatures and Decentralized Identifiers (DIDs), enabling portable, privacy-preserving identity across metaverse and web platforms without centralised databases.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:verifiable-credential-standard", "labels": [ "Verifiable Credential Standard", "Credential" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "verifiable-credential-surface", "title": "Verifiable Credential Surface", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A standardised JSON-LD 1.1 surface (S3) for issuing and storing W3C Verifiable Credentials 2.0|W3C VC 2.0 credentials signed by agents' DID Nostr Identity|did:nostr DIDs using Schnorr Signature|Schnorr signatures over JCS Canonicalisation|JCS-canonicalised payloads, enabling v...", "entityType": "Class", "qualityScore": 0.9, "maturity": "established", "iri": "urn:ngm:class:verifiable-credential-surface", "labels": [ "Verifiable Credential Surface" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "AgenticSystemsDomain", "BIP-340", "Blockchain Verification", "Capability Attestation", "Compliance Audit", "Credential Indexing", "Credential Issuance", "Credential Verification", "CredentialsDomain", "CredentialsLayer", "Decentralised Trust", "DID Resolution", "JCS Canonicalisation", "JCS Canonicalisation", "JCS Canonicalisation", "JSON-LD 1.1", "PRD-006", "RFC 8785", "RFC 8785 Canonical JSON", "Schnorr Proof" ] }, { "id": "verifiable-credentials", "title": "Verifiable Credentials", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographically secured, tamper-evident digital credentials standardised by the W3C Verifiable Credentials Data Model v2.0 (May 2025 REC) that encode machine-verifiable claims about subjects \u2014 enabling a holder/issuer/verifier triangle in which issuers sign claims with DIDs, holders selectively...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:verifiable-credentials", "labels": [ "Verifiable Credentials", "BC-0458-verifiable-credentials", "Verifiable Credential", "Verifiable Organizational Credentials", "VerifiableCredentials" ], "is_subclass_of": [ "Cryptographic Primitive", "Digital Identity", "Cryptographic Proof", "Distributed Identity", "Self-Sovereign Identity", "W3C Standard" ], "wikilinks": [ "BBS+ Signature", "Bitstring Status List", "CBOR", "Centralised Identity Provider", "Credential Schema", "Credential Status", "Credential Subject", "Cross-Border Identity Recognition", "Cryptographic Proof", "CryptographyDomain", "Decentralised Identifiers", "DID Document", "DIDComm", "EdDSA", "EUDI Wallet", "European Commission eIDAS", "Federated Identity", "GOV.UK One Login", "Healthcare Data Interoperability", "IdentityDomain" ] }, { "id": "verifiable-data-registry", "title": "Verifiable Data Registry", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A system that mediates the creation, verification, and resolution of decentralized identifiers and verifiable credentials, typically implemented as a blockchain, distributed ledger, or decentralized network for storing public key and revocation information.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:verifiable-data-registry", "labels": [ "Verifiable Data Registry", "Verifiable Data" ], "is_subclass_of": [ "Data Registry" ], "wikilinks": [ "Data Registry", "metaverse" ] }, { "id": "verifiable-inference", "title": "Verifiable Inference", "domain": "ai", "domain_name": "Ai", "definition": "Verifiable inference is the capability to cryptographically prove that a specific machine-learning model produced a given output for a given input without requiring trust in the compute provider. Techniques include zero-knowledge proofs of model execution (zkML), trusted execution environments and optimistic verification, which let third parties audit results. It is essential for decentralised and trust-minimised AI compute markets where inference is outsourced.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:verifiable-inference", "labels": [ "Verifiable Inference" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "verifiable-random-function", "title": "Verifiable Random Function", "domain": "security", "domain_name": "Security", "definition": "A Verifiable Random Function (VRF) is a cryptographic primitive, introduced by Micali, Rabin, and Vadhan (1999), that maps an input to a pseudorandom output and produces a non-interactive proof allowing any third party to verify that the output was computed correctly from a given public key and input without learning the private key. The VRF owner possesses a private key SK and public key PK; given input alpha, they compute output beta and proof pi such that any verifier with PK can confirm that beta = VRF(SK, alpha) without requiring SK. VRFs provide both uniqueness (exactly one valid output per key-input pair) and pseudorandomness (output is indistinguishable from random to anyone without SK). They are deployed in blockchain systems for unpredictable, manipulation-resistant leader election (Algorand, Cardano, Hedera), NFT trait generation, on-chain lotteries, and as the cryptographic core of Chainlink VRF and similar oracle randomness services.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:verifiable-random-function", "labels": [ "Verifiable Random Function" ], "is_subclass_of": [ "Cryptographic Proof" ], "wikilinks": [] }, { "id": "verification-method", "title": "Verification Method", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A verification method is an entry in a decentralised identifier document that specifies the cryptographic material and parameters used to authenticate or authorise actions on behalf of the identifier subject. It typically expresses a public key, key type and controller, and is referenced by verification relationships such as authentication, assertion and key agreement. Verification methods are the mechanism by which proofs presented by a subject are checked against the published DID document.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:verification-method", "labels": [ "Verification Method" ], "is_subclass_of": [ "DID Document" ], "wikilinks": [] }, { "id": "verification-process", "title": "Verification Process", "domain": "governance", "domain_name": "Governance", "definition": "A verification process is the structured procedure by which claims or records are independently checked against evidence and methodology before being accepted as valid. In carbon markets it confirms that emission reductions or removals are real, additional, measurable and permanent prior to credit issuance and registry entry. Verification provides the assurance and integrity on which accounting and trading systems depend.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:verification-process", "labels": [ "Verification Process" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "verification-standard", "title": "Verification Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Specifications and protocols that define how digital identities, credentials, and claims are validated and authenticated in metaverse environments, ensuring trust and security through cryptographic proofs and standardized verification processes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:verification-standard", "labels": [ "Verification Standard" ], "is_subclass_of": [ "Standards and Interoperability", "Standards" ], "wikilinks": [ "metaverse", "Standards" ] }, { "id": "verified-carbon-standard", "title": "Verified Carbon Standard", "domain": "finance", "domain_name": "Finance", "definition": "The Verified Carbon Standard (VCS) is the most widely used voluntary greenhouse-gas crediting programme, administered by the non-profit Verra. It defines methodologies, validation and verification requirements, and a registry under which emission-reduction and removal projects can issue tradable Verified Carbon Units. By certifying that claimed reductions are real, additional, measurable and permanent, VCS provides the integrity framework that underpins much of the voluntary carbon market.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:verified-carbon-standard", "labels": [ "Verified Carbon Standard" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [] }, { "id": "verified-data-state", "title": "Verified Data State", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:verified-data-state", "labels": [ "Verified Data State" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "verra-vcs-methodology", "title": "Verra VCS Methodology", "domain": "governance", "domain_name": "Governance", "definition": "A Verra VCS methodology is an approved technical protocol under Verra's Verified Carbon Standard that specifies how a particular project type must quantify, monitor and report greenhouse-gas emission reductions or removals. Each methodology defines baselines, additionality tests, measurement procedures and monitoring requirements so that issued Verified Carbon Units are credible and comparable. It is the rule set against which carbon-credit projects are validated and verified.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:verra-vcs-methodology", "labels": [ "Verra VCS Methodology" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "verra-vcs-standard", "title": "Verra VCS Standard", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The Verra Verified Carbon Standard is a widely used programme for certifying voluntary carbon credits, setting rules for quantifying, verifying and issuing emission reductions and removals.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:verra-vcs-standard", "labels": [ "Verra VCS Standard", "VCS Standard", "Verra Verified Carbon Standard" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [ "Carbon Markets", "Regulatory Compliance", "Carbon Registry", "Sustainability Domain" ] }, { "id": "version-control", "title": "version control", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Version control is a methodology and toolset for systematically tracking, managing, and auditing changes to files and artefacts over time, enabling multiple contributors to develop concurrently on isolated branches and integrate their work through well-defined merge strategies. Distributed version control systems such as Git maintain a full directed-acyclic-graph (DAG) history locally on every node, supporting offline operation, cryptographically signed commits, and fine-grained blame and bisect operations. Version control underpins modern software delivery practices including continuous integration, infrastructure-as-code, dataset lineage, and MLOps pipelines by providing immutable, content-addressed snapshots of any evolving asset.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:version-control", "labels": [ "Version Control", "Asset Version Control", "Distributed Version Control", "Multi-Version Concurrency Control", "Version Control System" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "version-history", "title": "Version History", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Version History is a chronological record of all changes made to a shared document or workspace, capturing who made each change and when. It enables collaborators to review past states, compare diffs, and restore previous versions if an edit introduces errors. This audit trail is critical for accountability and safe experimentation in distributed teams.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:version-history", "labels": [ "Version History" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "versioning-system", "title": "Versioning System", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A versioning system is a mechanism that tracks, identifies and manages successive states of an artefact such as code, data, documents or taxonomy entries over time. It records changes, supports retrieval of historical versions, and resolves concurrent edits, enabling reproducibility, auditability and rollback. In curation and registry contexts it ensures controlled evolution of records while preserving provenance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:versioning-system", "labels": [ "Versioning System", "Versioning" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "vertex-attribute", "title": "Vertex Attribute", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A per-vertex data channel attached to the vertices of a polygon mesh \u2014 position, normal, tangent, texture coordinates, colour, skinning joint indices and weights \u2014 supplied to the vertex shader as its input. Attributes are stored in vertex buffers with a declared layout (location, format, offset, stride), interpolated across triangles during rasterisation for per-pixel shading, and standardised by interchange formats such as glTF, making them the fundamental unit of geometry data throughout the graphics pipeline.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:vertex-attribute", "labels": [ "Vertex Attribute" ], "is_subclass_of": [ "Data Structure" ], "wikilinks": [ "Graphics Pipeline", "Vertex Shader", "Polygon Mesh" ] }, { "id": "vertex-buffer", "title": "Vertex Buffer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A vertex buffer is a region of GPU-accessible memory that stores the attributes of a mesh's vertices, such as positions, normals, texture coordinates and colours, in a contiguous layout for the graphics pipeline to consume. By uploading geometry once and referencing it across many draw calls, vertex buffers minimise CPU-to-GPU transfer and enable high-throughput rendering. They are bound to the input-assembler stage and described by a vertex layout that maps buffer bytes to shader inputs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:vertex-buffer", "labels": [ "Vertex Buffer" ], "is_subclass_of": [ "Rendering Pipeline" ], "wikilinks": [] }, { "id": "vertex-processing", "title": "Vertex Processing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Vertex processing is the stage of the graphics pipeline that operates on each vertex of geometry, transforming its position from model space through world, view, and clip space, and computing or passing per-vertex attributes such as normals, texture coordinates, and colours. Executed largely by programmable vertex shaders on the GPU, it prepares primitives for clipping, perspective division, and rasterisation, and may feed subsequent geometry, tessellation, and fragment stages. As a foundational step in real-time rendering it determines screen-space placement and the interpolated inputs available to downstream shading.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vertex-processing", "labels": [ "Vertex Processing" ], "is_subclass_of": [ "Graphics Pipeline" ], "wikilinks": [] }, { "id": "vertex-shader", "title": "Vertex Shader", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A mandatory, programmable GPU stage that processes individual vertices within the graphics pipeline, transforming 3D coordinates through model, view, and projection matrices into clip space whilst computing per-vertex attributes such as normals, texture coordinates, and lighting terms that are subsequently interpolated across primitives for the fragment shader.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:vertex-shader", "labels": [ "Vertex Shader" ], "is_subclass_of": [ "Display and Rendering", "Metaverse" ], "wikilinks": [ "GPU Programming", "Graphics Pipeline", "Compute Shader", "Metaverse", "Pixel Shader", "Rasterization" ] }, { "id": "vertical-ai-models", "title": "Vertical AI Models", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Machine learning models specifically trained and optimized for a particular industry or narrow domain, such as customer service or finance, to outperform general-purpose models on domain-specific tasks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:vertical-ai-models", "labels": [ "Vertical AI Models" ], "is_subclass_of": [ "Model" ], "wikilinks": [] }, { "id": "vertical-coordinate-reference-system", "title": "Vertical Coordinate Reference System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vertical-coordinate-reference-system", "labels": [ "Vertical Coordinate Reference System" ], "is_subclass_of": [ "Coordinate Reference System" ], "wikilinks": [] }, { "id": "vertical-coordinate-system", "title": "Vertical Coordinate System", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vertical-coordinate-system", "labels": [ "Vertical Coordinate System" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vertical-coordinate", "title": "Vertical Coordinate", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vertical-coordinate", "labels": [ "Vertical Coordinate", "VerticalCoordinate" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vertical-datum", "title": "Vertical Datum", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vertical-datum", "labels": [ "Vertical Datum" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vertical-dilution-of-precision", "title": "Vertical Dilution of Precision", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vertical-dilution-of-precision", "labels": [ "Vertical Dilution of Precision" ], "is_subclass_of": [ "Dilution of Precision" ], "wikilinks": [] }, { "id": "vesting-schedule", "title": "Vesting Schedule", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A vesting schedule is a contractually or programmatically enforced timetable that releases allocated digital assets or equity to recipients incrementally over a defined period, typically beginning after an initial cliff during which no release occurs. In blockchain and token-based systems, schedules are commonly encoded in smart contracts that autonomously unlock portions of a token allocation at fixed intervals or milestones, aligning the long-term incentives of founders, contributors, and investors with project health. The schedule's parameters \u2014 cliff length, total duration, release cadence, and revocation conditions \u2014 are a principal instrument of tokenomics design and governance risk management.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:vesting-schedule", "labels": [ "Vesting Schedule" ], "is_subclass_of": [ "Token and Asset", "Token Distribution" ], "wikilinks": [ "Smart Contract", "Tokenomics Governance", "Governance Token", "Token" ] }, { "id": "vibe-coding", "title": "Vibe Coding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A software development approach where AI agents generate code based on high-level natural language intent rather than explicit instructions.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:vibe-coding", "labels": [ "Vibe Coding" ], "is_subclass_of": [ "Enterprise Ai" ], "wikilinks": [] }, { "id": "vibration-isolation", "title": "Vibration Isolation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Vibration isolation is an engineering technique that mechanically decouples sensitive equipment from ambient structural or acoustic vibration, using damped mounts, pneumatic isolators or active feedback systems to attenuate disturbance across a target frequency band. It is essential wherever nanometre- or sub-wavelength-scale positional stability is required, since even small vibrations can blur measurements or misalign optical paths. Precision manufacturing tooling and quantum network nodes both depend on vibration isolation to hold tolerances that ambient building or seismic vibration would otherwise violate.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:vibration-isolation", "labels": [ "Vibration Isolation" ], "is_subclass_of": [ "Precision Manufacturing" ], "wikilinks": [] }, { "id": "vibration-testing", "title": "Vibration Testing", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:vibration-testing", "labels": [ "Vibration Testing" ], "is_subclass_of": [ "Spacecraft Environmental Testing" ], "wikilinks": [] }, { "id": "video-async", "title": "Video Async", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Asynchronous video communication is a mode of distributed collaboration in which participants record, share, and consume video messages at times of their own choosing rather than in a live session, combining the visual richness and non-verbal expressiveness of video with the time-zone flexibility and cognitive benefits of asynchronous working. Tools such as screen-and-webcam recorders reduce dependence on synchronous meetings, enable clearer knowledge sharing across distributed teams, and produce persistent, referenceable communication artefacts.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:video-async", "labels": [ "Video Async" ], "is_subclass_of": [ "Communication Technology", "Asynchronous Collaboration" ], "wikilinks": [ "Collaboration Tools", "Loom", "Asynchronous Collaboration", "TelecollaborationDomain" ] }, { "id": "video-codec", "title": "Video Codec", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A video codec (coder-decoder) is an algorithm or hardware implementation that compresses and decompresses digital video by exploiting spatial redundancy within frames (intra-prediction), temporal redundancy across frames (inter-prediction with motion compensation), and transform coding of residuals, enabling practical storage and transmission of video at bitrates orders of magnitude lower than uncompressed formats whilst maintaining perceptual quality.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:video-codec", "labels": [ "Video Codec" ], "is_subclass_of": [ "Encoder Decoder Architecture" ], "wikilinks": [] }, { "id": "video-compression", "title": "Video Compression", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Video compression is the process of encoding digital video data to reduce file size and bit rate for efficient storage and transmission, while maintaining perceptually acceptable quality. It exploits three principal forms of redundancy: spatial redundancy within individual frames, temporal redundancy across consecutive frames, and perceptual redundancy based on limitations of human visual perception. Modern video codecs such as H.264/AVC, H.265/HEVC, VP9, and AV1 implement intra-frame prediction, inter-frame motion compensation, discrete cosine or wavelet transforms, quantisation, and entropy coding to achieve compression ratios that make streaming, broadcast, and real-time communication practical over bandwidth-limited networks.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:video-compression", "labels": [ "Video Compression" ], "is_subclass_of": [ "Data Compression" ], "wikilinks": [ "Content Delivery", "Real-Time Communication", "Content Delivery Network", "owl:Thing" ] }, { "id": "video-conferencing", "title": "Video Conferencing", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Real-time audio-visual communication technology enabling multiple geographically distributed participants to see and hear each other simultaneously through internet-connected devices, supporting face-to-face interaction across distances via compressed media streams, network protocols, and centralised or peer-to-peer media routing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:video-conferencing", "labels": [ "Video Conferencing", "Multilingual Video Conferencing", "Video Conferencing Platform" ], "is_subclass_of": [ "Synchronous Collaboration", "TC-0010-Synchronous-Collaboration" ], "wikilinks": [ "Face-to-Face Interaction", "Real-Time Communication Protocols", "TC-0010-Synchronous-Collaboration", "TC-0012-Screen-Sharing", "TC-0013-Virtual-Meeting", "TC-0016-Remote-Pair-Programming", "TELE-020-virtual-reality-telepresence", "TELE-021-augmented-reality-collaboration", "TELE-100-ai-avatars", "TELE-105-real-time-language-translation", "TELE-107-ai-meeting-assistants", "TELE-150-webrtc", "TELE-153-5g-telepresence" ] }, { "id": "video-encoding", "title": "Video Encoding", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Video encoding is the process of compressing raw or lightly-compressed video frames into a deliverable bitstream using a video codec, encompassing decisions about encoding parameters (resolution, frame rate, bitrate mode, keyframe interval, codec profile and level), rate control algorithms, and hardware or software encoder selection to balance output quality, file size, and encoding speed for a given delivery target.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:video-encoding", "labels": [ "Video Encoding", "Hardware Video Encoding", "MP4 Encoding" ], "is_subclass_of": [ "Video Codec" ], "wikilinks": [] }, { "id": "video-games", "title": "Video Games", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Video games are interactive electronic entertainment software in which players engage with a virtual environment through input devices, receiving real-time audiovisual feedback rendered by a game engine. They span genres from narrative role-playing games to competitive multiplayer shooters, sports simulations, and puzzle games, delivered across platforms including dedicated consoles, personal computers, mobile devices, and cloud streaming services. Video games are the largest entertainment sector by revenue globally, intersecting with AI, virtual reality, blockchain-based asset ownership, and esports as a spectator sport.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:video-games", "labels": [ "Video Games", "Games" ], "is_subclass_of": [ "Digital Entertainment" ], "wikilinks": [] }, { "id": "video-generation", "title": "video generation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Video generation is the AI-driven synthesis of temporally coherent video sequences from text prompts, reference images, or other conditioning signals using generative models such as latent diffusion, autoregressive transformers, or flow-matching networks. Unlike image generation, video synthesis must maintain temporal consistency across frames, model object trajectories and camera motion, and produce physically plausible dynamics across variable-length sequences. Contemporary approaches encode video into a compressed spatiotemporal latent space via a 3D variational autoencoder, apply a diffusion or flow-matching process in that space conditioned on text or visual embeddings, then decode to pixel space. Scaling laws, large curated training datasets, and architectural advances such as full spatiotemporal attention have driven rapid capability growth, enabling cinematic-quality outputs at multiple seconds of duration.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:video-generation", "labels": [ "Video Generation", "AI Video Generation" ], "is_subclass_of": [ "Generative AI" ], "wikilinks": [] }, { "id": "video-object-segmentation", "title": "Video Object Segmentation", "domain": "ai", "domain_name": "Ai", "definition": "Video object segmentation is the computer-vision task of delineating and tracking the pixel-level boundaries of one or more objects across the frames of a video sequence. It extends single-image segmentation with temporal coherence, propagating masks while handling motion, occlusion and appearance change. It underpins video editing, autonomous perception, surveillance and content analysis.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:video-object-segmentation", "labels": [ "Video Object Segmentation" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "video-production-pipeline", "title": "Video Production Pipeline", "domain": "ai", "domain_name": "Ai", "definition": "A video production pipeline is the end-to-end sequence of stages through which video content moves from capture or generation through editing, effects, colour, restoration and final delivery. Increasingly it incorporates AI stages such as upscaling, denoising and restoration alongside traditional editorial and compositing steps. The pipeline coordinates assets, tools and review across teams to produce polished output efficiently.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:video-production-pipeline", "labels": [ "Video Production Pipeline", "Video Post-Production" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "video-streaming", "title": "Video Streaming", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Video Streaming is the continuous, real-time transmission of encoded video data over a network to a client device that decodes and renders the content progressively, without requiring the complete file to be downloaded before playback begins. It relies on transport protocols such as RTMP, HLS, DASH, and WebRTC, combined with adaptive bitrate (ABR) algorithms that dynamically adjust resolution and compression to match available network conditions. Modern deployments encompass live streaming, video-on-demand (VoD), 360-degree video, volumetric video, and cloud-rendered remote-display streaming, each with distinct latency, bandwidth, and codec trade-offs. Content delivery networks (CDNs), edge computing nodes, and specialised video codecs such as H.264, H.265/HEVC, AV1, and VP9 are the principal technical pillars enabling scalable, low-latency distribution at global scale.", "entityType": "Class", "qualityScore": 0.0, "maturity": "mature", "iri": "urn:ngm:class:video-streaming", "labels": [ "Video Streaming", "Resilient Video Streaming" ], "is_subclass_of": [ "Network Infrastructure" ], "wikilinks": [] }, { "id": "video-synthesis", "title": "Video Synthesis", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Video synthesis is the generation of novel video sequences from text, image, or other conditioning signals using generative models, most commonly diffusion models extended across a temporal dimension to maintain coherence between frames. It is used in creative tools for content generation, simulation, and film pre-visualisation. Key challenges include maintaining temporal consistency, object permanence, and physically plausible motion across generated frames.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:video-synthesis", "labels": [ "Video Synthesis" ], "is_subclass_of": [ "Diffusion Models" ], "wikilinks": [] }, { "id": "video-understanding", "title": "Video Understanding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Video understanding is the field of artificial intelligence concerned with extracting semantic meaning from video, including recognising objects, actions, events, and their temporal relationships across frames. Unlike single-image analysis, it must model motion, temporal context, and long-range dependencies to interpret what is happening over time. Modern approaches combine spatial feature extraction with temporal modelling using recurrent, 3D-convolutional, and transformer-based architectures, increasingly fused with language for captioning, retrieval, and question answering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:video-understanding", "labels": [ "Video Understanding" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "viewing-geometry", "title": "Viewing Geometry", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:viewing-geometry", "labels": [ "Viewing Geometry" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vircadia", "title": "Vircadia", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Vircadia is an open-source, community-governed, self-hostable Metaverse platform forked from High Fidelity's codebase under the Apache 2.0 licence, providing a federated architecture for persistent, spatially-aware virtual worlds with support for desktop and VR Headsets clients, posit...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:vircadia", "labels": [ "Vircadia" ], "is_subclass_of": [ "Platform and Environment", "Social VR", "Virtual World", "Metaverse", "Spatial Computing Paradigm", "Open Source Software", "Distributed Systems" ], "wikilinks": [ "AltspaceVR", "Apache 2.0 Licence", "Apple Vision Pro", "Assignment Client", "Assignment Client Architecture", "Audio Mixer", "Avatar Customisation", "Avatar Physics", "Bullet Physics", "C++ Runtime", "Cloud Hosting", "Desktop Clients", "Domain Access Control", "Domain Federation", "Domain Server", "Entity Server", "Federated Virtual Worlds", "Frustum Culling", "Full-Body Tracking", "glTF Format" ] }, { "id": "virtual-asset-trading", "title": "Virtual Asset Trading", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The exchange, purchase, and sale of digital assets \u2014 including cryptocurrencies, NFTs, virtual land parcels, in-game items, and other tokenised value \u2014 through centralised exchanges, decentralised exchanges (DEXs), NFT marketplaces, and peer-to-peer transactions. Trading is governed by AML/KYC requirements, FATF Travel Rule obligations, and emerging securities classification frameworks that vary by jurisdiction.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-asset-trading", "labels": [ "Virtual Asset Trading", "Virtual Asset Exchange" ], "is_subclass_of": [ "Platform and Environment", "Virtual Economy" ], "wikilinks": [ "metaverse", "Virtual Economy" ] }, { "id": "virtual-asset", "title": "Virtual Asset", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital items with economic value within metaverse environments, spanning non-fungible tokens representing unique digital collectibles, avatars and avatar customisations, virtual real estate, and in-game items with market-tradeable value. Ownership and transfer of virtual assets rely on blockchain smart contracts or centralised platform ledgers, with interoperability standards enabling asset portability across platforms and markets. Economic significance spans consumer entertainment and investment speculation, with regulatory frameworks increasingly addressing taxation and classification under securities laws.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:virtual-asset", "labels": [ "Virtual Asset", "VirtualAsset" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "BlockchainTechnology", "Collectible", "createdBy", "DEX", "dt:generatedBy", "dt:storedOn", "dt:tokenizedAs", "dt:tradedOn", "dt:verifiedBy", "hasMetadata", "hasOwner", "InGameItems", "IPFS", "MetaversePlatforms", "NFT", "NFT", "usedInWorld", "VirtualAssetTaxation", "VirtualRealEstate", "VirtualRealEstate" ] }, { "id": "virtual-assistant", "title": "Virtual Assistant", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A virtual assistant is a software agent that interprets natural-language requests and performs tasks or retrieves information on behalf of a user through conversational interaction. It combines speech recognition or text understanding, intent recognition, dialogue management and task execution to mediate between the user and underlying services, devices or knowledge sources. Virtual assistants range from voice-driven smart-speaker agents to text-based assistants embedded in applications, increasingly powered by large language models for open-ended reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:virtual-assistant", "labels": [ "Virtual Assistant", "Virtual Assistants" ], "is_subclass_of": [ "Conversational AI" ], "wikilinks": [] }, { "id": "virtual-background", "title": "Virtual Background", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A Virtual Background is a real-time image or video composited behind a video-call participant using chroma-key or AI-based background segmentation, replacing the physical environment visible to the camera. It allows remote workers to project a professional or branded appearance regardless of their physical location, protecting privacy and maintaining visual consistency across distributed teams. Virtual backgrounds are natively supported in Zoom, Microsoft Teams, and Google Meet.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:virtual-background", "labels": [ "Virtual Background" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "virtual-camera", "title": "Virtual Camera", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A simulated camera system in virtual production environments that captures real-time virtual scenes, enabling filmmakers to visualise and shoot CG environments as though using a physical camera with real-world lens characteristics \u2014 including focal length, aperture, and depth-of-field \u2014 and six-DoF movement tracking. It integrates with motion capture rigs, LED volume stages, and game engines to deliver live compositing preview and Genlock-synchronised output.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-camera", "labels": [ "Virtual Camera" ], "is_subclass_of": [ "Display and Rendering", "Virtual Production" ], "wikilinks": [ "metaverse", "Virtual Production" ] }, { "id": "virtual-city-model", "title": "Virtual City Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital representation of an urban environment including buildings, infrastructure, terrain, and dynamic elements, used for urban planning, simulation, gaming, and metaverse world-building based on real or imagined cities. Data standards such as CityGML (OGC), 3D Tiles, and BIM/IFC underpin interoperable city model exchange, while capture techniques including LiDAR scanning and photogrammetry feed real-world geometry into the model.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-city-model", "labels": [ "Virtual City Model" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin" ], "wikilinks": [ "Digital Twin", "metaverse" ] }, { "id": "virtual-classroom", "title": "Virtual Classroom", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An online learning environment that replicates a traditional classroom's interactive experience without physical boundaries, leveraging education technology to connect students and instructors through real-time engagement, collaboration tools, and shared learning materials.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-classroom", "labels": [ "Virtual Classroom" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "metaverse", "Virtual Environment" ] }, { "id": "virtual-clinic", "title": "Virtual Clinic", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A telemedicine platform that enables clinicians to provide health services by connecting with patients through virtual means, offering remote consultations, diagnosis, treatment recommendations, and follow-up care using HIPAA-compliant video and digital communication technologies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-clinic", "labels": [ "Virtual Clinic" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "metaverse", "Virtual Environment" ] }, { "id": "virtual-collaboration", "title": "Virtual Collaboration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A 3D immersive environment enabling geographically distributed teams to interact via avatar identities, perform work tasks in real time, and access shared artefacts through metaverse-enabled workspaces. Capabilities include interactive whiteboards, document co-editing, spatial audio, and persistent virtual rooms accessible via desktop, mobile, and VR/AR devices.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:virtual-collaboration", "labels": [ "Virtual Collaboration" ], "is_subclass_of": [ "Virtual Workspace" ], "wikilinks": [ "metaverse", "Virtual Workspace" ] }, { "id": "virtual-commerce", "title": "Virtual Commerce", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An emerging form of e-commerce (V-commerce) that enables buying and selling of goods and services within virtual reality environments, including virtual storefronts, enhanced product visualization, social shopping experiences, and immersive customer-product interactions in metaverse platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:virtual-commerce", "labels": [ "Virtual Commerce" ], "is_subclass_of": [ "Platform and Environment", "Virtual Economy" ], "wikilinks": [ "metaverse", "Virtual Economy" ] }, { "id": "virtual-commissioning", "title": "Virtual Commissioning", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The simulation and testing of manufacturing systems in a virtual environment before physical implementation, using digital twins and real-time simulation to validate control algorithms, robot motion, and process sequences, reducing commissioning time and costs during development.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-commissioning", "labels": [ "Virtual Commissioning" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin" ], "wikilinks": [ "Digital Twin", "metaverse" ] }, { "id": "virtual-community-platform", "title": "Virtual Community Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Spatial computing platforms that enable users to create and interact within immersive 3D virtual environments as digital avatars, supporting social interaction, community building, collaborative activities, and shared experiences across geographic boundaries.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-community-platform", "labels": [ "Virtual Community Platform" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "metaverse", "Metaverse" ] }, { "id": "virtual-community-practice", "title": "Virtual Community Practice", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The methodologies, norms, and behaviours that govern how members interact, collaborate, and build relationships within virtual community platforms. This encompasses governance structures, social conventions, moderation policies, and collective practices that shape online community culture in metaverse and social VR environments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-community-practice", "labels": [ "Virtual Community Practice" ], "is_subclass_of": [ "Governance and Safety", "Community Governance" ], "wikilinks": [ "Community Governance", "metaverse" ] }, { "id": "virtual-concerts", "title": "Virtual Concerts", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Live musical performances conducted in metaverse or virtual reality environments where virtual avatars perform on virtual stages synced to music, enabling immersive concert experiences accessible globally through VR headsets, gaming platforms, or streaming services.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-concerts", "labels": [ "Virtual Concerts", "Virtual Concert" ], "is_subclass_of": [ "Platform and Environment", "Virtual Event" ], "wikilinks": [ "metaverse", "Virtual Event" ] }, { "id": "virtual-currency", "title": "Virtual Currency", "domain": "finance", "domain_name": "Finance", "definition": "A digital representation of value issued and controlled by platform developers, used and accepted electronically within a specific virtual community. Virtual currencies include closed in-game currencies (e.g., World of Warcraft gold), hybrid purchasable tokens (e.g., Robux), and convertible currencies exchangeable for fiat money, each with distinct regulatory and economic implications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-currency", "labels": [ "Virtual Currency", "VirtualCurrency" ], "is_subclass_of": [ "Digital Asset" ], "wikilinks": [ "Digital Asset", "metaverse" ] }, { "id": "virtual-destination", "title": "Virtual Destination", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital location or space within a virtual environment that serves as a point of interest or travel objective, encompassing photogrammetry-reconstructed heritage sites, procedurally generated fantasy worlds, brand experience spaces, and interactive event venues accessible through metaverse platforms. Virtual destinations underpin virtual tourism, educational field trips, and social gathering use-cases, and are created using real-time rendering engines together with photogrammetry and 3D modelling pipelines.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-destination", "labels": [ "Virtual Destination" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "metaverse", "Virtual Environment" ] }, { "id": "virtual-economy-infrastructure", "title": "Virtual Economy Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The collection of technologies, platforms, and systems essential for creating, managing, and sustaining economic activities within metaverse environments, including blockchain networks, payment systems, marketplaces, and decentralized finance protocols.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-economy-infrastructure", "labels": [ "Virtual Economy Infrastructure" ], "is_subclass_of": [ "Computing and Cloud", "Virtual Economy" ], "wikilinks": [ "metaverse", "Virtual Economy" ] }, { "id": "virtual-economy-market-framing", "title": "Virtual Economy Market Framing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A conceptual framing of metaverse platforms as economic ecosystems in which virtual goods, services, experiences, and digital assets are produced, traded, and consumed. This view foregrounds creator economies, digital ownership, interoperable asset standards, and market dynamics including adoption curves, pricing models, and regulatory challenges specific to virtual worlds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-economy-market-framing", "labels": [ "Virtual Economy Market Framing", "Metaverse as Markets" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Update Cycle" ] }, { "id": "virtual-economy", "title": "Virtual Economy", "domain": "metaverse", "domain_name": "Metaverse", "definition": "A virtual economy is a system of economic activity that takes place within a simulated or digital environment, encompassing the production, exchange, and consumption of virtual goods, services, and currencies. These economies exhibit emergent market dynamics \u2014 including supply-demand equilibria, price discovery, and capital accumulation \u2014 that parallel real-world economic phenomena while remaining governed by platform rules rather than national jurisdictions. Virtual economies range from closed in-game markets with proprietary currencies to open metaverse ecosystems where blockchain-backed assets and decentralised finance protocols enable interoperable, user-owned value. Their study bridges game theory, monetary economics, and digital rights, making them a critical lens through which to understand the financialisation of participatory digital spaces.", "entityType": "Class", "qualityScore": 0.73, "maturity": "emerging", "iri": "urn:ngm:class:virtual-economy", "labels": [ "Virtual Economy", "VirtualEconomy" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "virtual-environment-connectivity", "title": "Virtual Environment Connectivity", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The networking technologies and protocols that enable seamless communication, data synchronisation, and user interaction across distributed virtual environments, including real-time streaming, peer-to-peer networks, and cross-platform interoperability systems. Achieving sub-20ms latency for VR experiences requires a combination of 5G/6G mobile networks, edge computing, WebRTC for browser-to-browser communication, and software-defined networking for traffic management.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-environment-connectivity", "labels": [ "Virtual Environment Connectivity" ], "is_subclass_of": [ "Platform and Environment", "Network Infrastructure" ], "wikilinks": [ "metaverse", "Network Infrastructure" ] }, { "id": "virtual-environment-creation", "title": "Virtual Environment Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of designing, building, and deploying immersive 3D virtual worlds using game engines, modelling software, and AI-powered tools, encompassing static and dynamic assets, physics simulation, and interactive experiences for metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-environment-creation", "labels": [ "Virtual Environment Creation" ], "is_subclass_of": [ "Platform and Environment", "3D Development" ], "wikilinks": [ "3D Development", "metaverse" ] }, { "id": "virtual-environment-design", "title": "Virtual Environment Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual Environment Design is the multidisciplinary practice of creating the spatial, visual, auditory, and interactive properties of computer-generated environments intended for exploration or habitation through virtual reality, augmented reality, or mixed reality systems. It integrates principles from architecture, industrial design, game design, cognitive psychology, and human-computer interaction to produce environments that are usable, compelling, and safe, balancing perceptual fidelity, performance constraints, and the unique ergonomics of immersive display hardware.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:virtual-environment-design", "labels": [ "Virtual Environment Design" ], "is_subclass_of": [ "Virtual Environment Creation" ], "wikilinks": [] }, { "id": "virtual-environment-specification", "title": "Virtual Environment Specification", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical standards and requirements documents that define the parameters, capabilities, and constraints of virtual environments, including rendering requirements, interaction models, physics simulation rules, and interoperability protocols.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-environment-specification", "labels": [ "Virtual Environment Specification" ], "is_subclass_of": [ "Platform and Environment", "Technical Standards" ], "wikilinks": [ "metaverse", "Technical Standards" ] }, { "id": "virtual-environment", "title": "Virtual Environment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A simulated three-dimensional (3D) digital space created with computer hardware and software that enables users to explore, interact, and experience an immersive surrounding approximating reality, accessed through devices such as VR headsets, AR glasses, or standard displays.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:virtual-environment", "labels": [ "Virtual Environment", "Cave Automatic Virtual Environment", "Collaborative Virtual Environment", "VirtualEnvironment" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "metaverse", "Metaverse" ] }, { "id": "virtual-event-platform", "title": "Virtual Event Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software platforms that enable hosting digital events in metaverse environments, supporting conferences, exhibitions, concerts, and networking through immersive 3D spaces with avatar-based attendance, interactive features, and real-time collaboration tools.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-event-platform", "labels": [ "Virtual Event Platform", "Event Platform" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "metaverse", "Metaverse" ] }, { "id": "virtual-event", "title": "Virtual Event", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Virtual Event is a digitally mediated gathering in which geographically distributed participants convene within a shared virtual environment \u2014 rendered through web browsers, desktop clients, or immersive XR headsets \u2014 to engage in social, educational, commercial, or cultural activities in real or near-real time. The class encompasses a spectrum from simple webinar-style presentations and live-streamed concerts, through avatar-populated social spaces, to fully embodied multi-user XR experiences featuring spatial audio, persistent virtual objects, and interactive exhibits. Virtual Events differ from asynchronous digital content by requiring simultaneous presence and bidirectional communication channels, and they differ from purely textual virtual spaces by incorporating rich media and, optionally, three-dimensional spatial representation. Their design draws on principles from event management, human-computer interaction, streaming media infrastructure, and spatial computing.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:virtual-event", "labels": [ "Virtual Event" ], "is_subclass_of": [ "Distributed Collaboration", "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "virtual-experience", "title": "Virtual Experience", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A user-facing encounter or activity delivered within a virtual environment, encompassing immersive simulations, interactive narratives, social engagements, and entertainment events that are accessed through XR devices, gaming platforms, or web-based spatial interfaces.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-experience", "labels": [ "Virtual Experience", "Unified Virtual Experience" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "virtual-factory", "title": "Virtual Factory", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A physically accurate digital twin representation of a manufacturing facility that enables modelling, simulation, analysis, and optimisation of production processes, resources, and operations without physical prototypes or pilot plants. It integrates IoT sensor data, AI-driven analytics, and real-time bidirectional synchronisation; platforms such as NVIDIA Omniverse (OpenUSD) and Siemens Tecnomatix are primary implementation environments. Virtual factories are a core Industry 4.0 concept supporting predictive maintenance, layout optimisation, and safe robotics training.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-factory", "labels": [ "Virtual Factory" ], "is_subclass_of": [ "Platform and Environment", "Digital Twin" ], "wikilinks": [ "Digital Twin", "metaverse" ] }, { "id": "virtual-field-trip", "title": "Virtual Field Trip", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An educational experience using VR, AR, or MR technologies that transports students to locations they could not otherwise visit, enabling immersive learning through 360-degree environments, interactive simulations, and experiential content without leaving the classroom. Research indicates VR can improve knowledge acquisition and retention compared to traditional instruction; platforms such as Google Arts and Culture and Google Expeditions have demonstrated feasibility at scale across subjects including science, history, and geography.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-field-trip", "labels": [ "Virtual Field Trip" ], "is_subclass_of": [ "Platform and Environment", "Educational Technology" ], "wikilinks": [ "Educational Technology", "metaverse" ] }, { "id": "virtual-group-dynamics", "title": "Virtual Group Dynamics", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The psychological and behavioural patterns of interaction, role formation, norm development, decision-making, conflict, and cohesion within groups operating in virtual environments, shaped by unique affordances of digital mediation such as anonymity, persistence, and spatial distribution. Key phenomena include the online disinhibition effect, avatar-mediated identity, emergent leadership hierarchies, and scalable broadcast communication structures.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-group-dynamics", "labels": [ "Virtual Group Dynamics" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "Collective Intelligence", "Online Disinhibition Effect", "Virtual Community", "Metaverse", "Proteus Effect", "Social Presence" ] }, { "id": "virtual-identity", "title": "Virtual Identity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A digital representation encompassing behaviours, preferences, movements, actions, and decisions made in digital spaces, extending beyond 3D avatars to include representation, data, and identification across AR, VR, MR, and web platforms using decentralized identity technologies.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-identity", "labels": [ "Virtual Identity", "VirtualIdentity" ], "is_subclass_of": [ "Security and Identity", "Digital Identity" ], "wikilinks": [ "Digital Identity", "metaverse" ] }, { "id": "virtual-interaction-logging", "title": "Virtual Interaction Logging", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The systematic recording and analysis of user interactions, behaviours, and activities within virtual environments, capturing data on avatar movements, communications, transactions, and engagement patterns for analytics, compliance, and experience optimisation. Logging pipelines must balance comprehensive telemetry collection with data minimisation, user consent, and retention policies required under privacy regulations such as GDPR.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-interaction-logging", "labels": [ "Virtual Interaction Logging" ], "is_subclass_of": [ "Governance and Safety", "Data Analytics" ], "wikilinks": [ "Data Analytics", "metaverse" ] }, { "id": "virtual-labor", "title": "Virtual Labor", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Work performed in virtual environments that generates economic value, spanning content creation, service provision, virtual construction, platform governance, and social labour. Virtual labour is often platform-dependent, contract-based, and remunerated via virtual currency or revenue-sharing models, raising significant questions around labour rights, taxation of cross-border virtual income, and platform exploitation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-labor", "labels": [ "Virtual Labor" ], "is_subclass_of": [ "Platform and Environment", "Metaverse" ], "wikilinks": [ "User-Generated Content", "Digital Asset", "Land Economics", "Metaverse", "Virtual Currency", "Virtual Economy" ] }, { "id": "virtual-land-rights", "title": "Virtual Land Rights", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Blockchain-based ownership and property rights for digital land parcels in metaverse platforms, represented as NFTs that serve as digital deeds providing proof of ownership, development rights, and transferability of virtual real estate assets.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-land-rights", "labels": [ "Virtual Land Rights", "Virtual Land Ownership" ], "is_subclass_of": [ "Governance and Safety", "Digital Property Rights" ], "wikilinks": [ "Digital Property Rights", "metaverse" ] }, { "id": "virtual-lighting-model", "title": "Virtual Lighting Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Mathematical description of light behavior for rendering realistic illumination in 3D scenes, simulating light emission, transport, and surface interaction.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-lighting-model", "labels": [ "Virtual Lighting Model", "Lighting Model" ], "is_subclass_of": [ "Display and Rendering", "Photorealistic Rendering" ], "wikilinks": [ "Ambient Occlusion", "BRDF Function", "Dynamic Lighting", "Graphics Processing Unit", "Light Source Model", "Material Properties", "Mood and Atmosphere", "Realistic Illumination", "Shader Program", "Shading System", "Shadow Computation", "SIGGRAPH Standards", "Surface Normals", "ComputeLayer", "CreativeMediaDomain", "Global Illumination", "Light Parameters", "Photorealistic Rendering", "Rasterization", "Ray Tracing" ] }, { "id": "virtual-location", "title": "Virtual Location", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A defined spatial coordinate or addressable place within a virtual environment, representing specific positions, areas, or destinations that users can navigate to, reference, and interact with in metaverse platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-location", "labels": [ "Virtual Location" ], "is_subclass_of": [ "Platform and Environment", "Virtual Environment" ], "wikilinks": [ "metaverse", "Virtual Environment" ] }, { "id": "virtual-machine", "title": "Virtual Machine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A sandboxed runtime environment that executes smart contract bytecode deterministically across distributed nodes, translating high-level contract code into low-level operations whilst metering computational resource consumption and enforcing state-transition rules. The Ethereum Virtual Machine is the canonical reference implementation, with alternatives including WebAssembly-based runtimes for improved performance.", "entityType": "Class", "qualityScore": 0.65, "maturity": "emerging", "iri": "urn:ngm:class:virtual-machine", "labels": [ "Virtual Machine", "Stack-Based Virtual Machine" ], "is_subclass_of": [ "Computing Infrastructure" ], "wikilinks": [ "CairoVM", "EVM", "MoveVM", "SolanaSVM", "Solidity", "WASM", "MetaverseDomain", "SmartContract" ] }, { "id": "virtual-meeting", "title": "Virtual Meeting", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A synchronous gathering of geographically distributed participants conducted through video conferencing, spatial computing, or immersive VR/AR technologies to enable remote collaboration. Virtual meetings span traditional video-call platforms with screen sharing and virtual whiteboards through to persistent 3D avatar-based spaces such as Meta Horizon Workrooms and Microsoft Mesh, which replicate physical presence cues including spatial audio and non-verbal body language.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-meeting", "labels": [ "Virtual Meeting", "Virtual Meeting Platforms", "Virtual Meeting Space", "VirtualMeeting" ], "is_subclass_of": [ "Communication Technology", "Telecollaboration" ], "wikilinks": [ "Telecollaboration" ] }, { "id": "virtual-model", "title": "Virtual Model", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital 3D representation of objects, characters, environments, or systems created using computer graphics software for use in metaverse platforms, simulations, games, and virtual production, including static assets, animated models, and procedurally generated content. Models are authored in formats such as glTF/GLB, FBX, and USD, and are consumed by game engines, real-time renderers, and XR runtimes.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-model", "labels": [ "Virtual Model" ], "is_subclass_of": [ "Content and Assets", "3D Asset" ], "wikilinks": [ "3D Asset", "metaverse" ] }, { "id": "virtual-museum-tour", "title": "Virtual Museum Tour", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An immersive digital experience that enables users to explore museum collections, exhibitions, and cultural heritage sites through VR, AR, 360-degree imagery, or interactive 3D environments, providing global access to art and artefacts regardless of physical location. Delivery formats range from photogrammetry-reconstructed 3D galleries (Smithsonian, British Museum) to fully navigable VR spaces, with platforms such as Google Arts and Culture and Matterport enabling broad public access across device types.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-museum-tour", "labels": [ "Virtual Museum Tour" ], "is_subclass_of": [ "Platform and Environment", "Virtual Tour" ], "wikilinks": [ "metaverse", "Virtual Tour" ] }, { "id": "virtual-nation-state", "title": "Virtual Nation State", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital sovereign entity existing within metaverse platforms that offers forms of virtual citizenship, governance structures, economic systems, and legal frameworks, potentially providing e-residency, digital identity, and participation in virtual economies independent of geographic territory.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-nation-state", "labels": [ "Virtual Nation State" ], "is_subclass_of": [ "Governance and Safety", "Digital Governance" ], "wikilinks": [ "Digital Governance", "metaverse" ] }, { "id": "virtual-network", "title": "Virtual Network", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A software-defined network infrastructure that creates logical network segments over physical hardware, enabling isolated, configurable communication channels for metaverse platforms, cloud services, and distributed applications through technologies such as VLANs, VPNs, Software Defined Networking (SDN), and 5G network slicing. Virtual networks provide Quality of Service guarantees, traffic isolation, and bandwidth allocation essential for latency-sensitive XR workloads.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-network", "labels": [ "Virtual Network" ], "is_subclass_of": [ "Platform and Environment", "Network Infrastructure" ], "wikilinks": [ "metaverse", "Network Infrastructure" ] }, { "id": "virtual-notary-service", "title": "Virtual Notary Service", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Autonomous agent providing cryptographic attestation, timestamping, and verification services for digital documents and transactions through distributed ledger anchoring and automated certification protocols.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-notary-service", "labels": [ "Virtual Notary Service" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Audit Trail Generator", "Blockchain Anchoring", "Certificate Authority Interface", "Cryptographic Algorithm", "Cryptographic Hash Function", "Document Authentication", "eIDAS Regulation", "ETSI TS 119 312", "Hash Function Module", "ISO 27001", "Legal Compliance", "Tamper Evidence", "Timestamp Authority", "Timestamping Service", "Trusted Timestamping", "Verification Protocol", "Audit Trail", "Blockchain", "Consensus Protocol", "Digital Certificate" ] }, { "id": "virtual-object-pose", "title": "Virtual Object Pose", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The position, orientation, and scale of a 3D object within a virtual environment, represented as a combined transform: translation (X, Y, Z world coordinates), rotation (quaternion or Euler angles), and scale factors. Accurate pose determination underpins AR object anchoring, motion capture replay, physics simulation, and hand-object interaction in XR systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-object-pose", "labels": [ "Virtual Object Pose" ], "is_subclass_of": [ "Platform and Environment", "Physics Simulation", "Spatial Computing Paradigm" ], "wikilinks": [ "metaverse", "Spatial Computing" ] }, { "id": "virtual-objects", "title": "Virtual Objects", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital 3D assets existing within virtual environments that users can interact with, own, trade, and customise, including avatars, wearables, furniture, vehicles, and environmental elements. NFT smart contracts on blockchain networks provide verifiable, transferable ownership records, while interoperable formats such as glTF and VRM support cross-platform portability of virtual objects.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-objects", "labels": [ "Virtual Objects" ], "is_subclass_of": [ "Content and Assets", "Digital Asset" ], "wikilinks": [ "Digital Asset", "metaverse" ] }, { "id": "virtual-office-spaces", "title": "Virtual Office Spaces", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Virtual office spaces are persistent 3D environments hosted on VR or metaverse platforms where distributed teams gather as avatars to collaborate, attend meetings, use shared whiteboards, and navigate spatial office layouts, replicating physical workplace social dynamics whilst enabling global remote participation through immersive telepresence. Unlike transient video calls, these environments preserve spatial context across sessions \u2014 desks remain assigned, whiteboards retain content, and colleagues can be discovered by proximity cues. Leading implementations include Microsoft Mesh, Meta Horizon Workrooms, and Spatial.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:virtual-office-spaces", "labels": [ "Virtual Office Spaces", "TELE-301-virtual-office-spaces", "Virtual Office" ], "is_subclass_of": [ "Workspace Tools", "Virtual Environment" ], "wikilinks": [ "TELE-002-telecollaboration", "TELE-020-virtual-reality-telepresence", "TELE-026-microsoft-mesh", "TELE-027-spatial-platform", "TELE-028-horizon-workrooms", "TELE-300-digital-twin-collaboration", "TELE-302-shared-whiteboards", "Virtual Environment" ] }, { "id": "virtual-performance-space", "title": "Virtual Performance Space", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive virtual venue environment designed for hosting live performances, concerts, theater productions, events, and social gatherings with real-time audience interaction and multimedia presentation capabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:virtual-performance-space", "labels": [ "Virtual Performance Space" ], "is_subclass_of": [ "Platform and Environment", "Virtual Reality" ], "wikilinks": [ "Audience Seating", "Content Delivery Network", "Decentraland", "Fortnite Concerts", "Hybrid Performances", "Lighting System", "Live Events", "Real-Time Streaming", "Social Gatherings", "Social Interaction Features", "Social Presence System", "Wave XR", "3D Rendering Engine", "ApplicationLayer", "Audio System", "Avatar System", "CreativeMediaDomain", "Metaverse Venue", "Network Infrastructure", "Physics Engine" ] }, { "id": "virtual-physical-collision", "title": "Virtual Physical Collision", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The detection and simulation of contact events between virtual objects and real-world physical elements in mixed and augmented reality environments. Virtual-physical collision systems generate haptic alerts, visual boundary warnings, and passthrough camera activations to prevent user injury and to produce physically plausible interactions between digital content and the real world.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-physical-collision", "labels": [ "Virtual Physical Collision", "Virtual-Physical Collision" ], "is_subclass_of": [ "Platform and Environment", "Physics Simulation" ], "wikilinks": [ "metaverse", "Physics Simulation" ] }, { "id": "virtual-presence", "title": "Virtual Presence", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The psychological experience of being there within a virtual place or situation, encompassing spatial presence (sense of location), social presence (connection with others), and self-presence (embodiment in virtual form), achieved through telepresence technologies and immersive environments.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:virtual-presence", "labels": [ "Virtual Presence" ], "is_subclass_of": [ "Platform and Environment", "Telepresence" ], "wikilinks": [ "metaverse", "Telepresence" ] }, { "id": "virtual-private-network", "title": "Virtual Private Network", "domain": "security", "domain_name": "Security", "definition": "A virtual private network (VPN) is a technology that establishes an encrypted tunnel over a shared or public network, allowing devices to communicate as if they were directly connected to a private network. It authenticates endpoints, encrypts traffic in transit, and encapsulates packets so that data confidentiality and integrity are preserved across untrusted links. VPNs are used for secure remote access to corporate resources, site-to-site connectivity, and privacy-preserving internet use. Common implementations rely on protocols such as IPsec, TLS, and WireGuard, though zero-trust architectures increasingly complement or supersede perimeter VPN models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:virtual-private-network", "labels": [ "Virtual Private Network" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "virtual-production-vp", "title": "Virtual Production (VP)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Production technique blending real and virtual scenes using XR and real-time rendering for film, broadcast, and immersive content creation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-production-vp", "labels": [ "Virtual Production (VP)" ], "is_subclass_of": [ "Display and Rendering", "Rendering Technique" ], "wikilinks": [ "Broadcast Production", "Camera Tracking", "Camera Tracking System", "Color Grading", "Compositing Pipeline", "Film Production Workflow", "In-Camera VFX", "Interactive Filmmaking", "LED Display System", "LED Volume", "Live Compositing", "Previsualization", "Real-Time Graphics", "Render Engine", "SIGGRAPH Production WG", "SMPTE ST 2119", "ApplicationLayer", "ComputeLayer", "CreativeMediaDomain", "Game Engine" ] }, { "id": "virtual-production-pipeline", "title": "Virtual Production Pipeline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The end-to-end workflow for creating film and television content using real-time rendering, LED volumes, motion capture, and game engine technology, integrating pre-visualization, on-set virtual environments, and in-camera visual effects to replace traditional post-production VFX processes.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:virtual-production-pipeline", "labels": [ "Virtual Production Pipeline" ], "is_subclass_of": [ "Display and Rendering", "Virtual Production" ], "wikilinks": [ "metaverse", "Virtual Production" ] }, { "id": "virtual-production-volume", "title": "Virtual Production Volume", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Large-scale physical LED wall or projection stage environment that merges live-action footage with real-time rendered 3D backgrounds, including LED panels, tracking systems, camera infrastructure, and stage hardware.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-production-volume", "labels": [ "Virtual Production Volume", "LED Volume Stage" ], "is_subclass_of": [ "Display and Rendering", "Virtual Production Pipeline" ], "wikilinks": [ "Camera Tracking", "Camera Tracking System", "Color Management System", "Display Processor", "Film Production Studio", "In-Camera VFX", "Interactive Filmmaking", "ISO/IEC 23090-3", "LED Wall", "Lighting Rig", "Physical Stage", "Real-time Background Rendering", "Real-time Rendering Engine", "Rendering Cluster", "SMPTE ST 2117", "ApplicationLayer", "CreativeMediaDomain", "Network Infrastructure", "Virtual Location", "Virtual Production Pipeline" ] }, { "id": "virtual-production-workflow", "title": "Virtual Production Workflow", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The end-to-end process for creating film and television content using virtual production techniques, integrating pre-visualisation, virtual art department asset creation, real-time game engine rendering, LED volume shooting, and post-production refinement in a non-linear pipeline where VFX work begins in pre-production rather than post-production. This approach captures composited digital environments in-camera, enabling creative decisions about lighting and framing to be made on set.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-production-workflow", "labels": [ "Virtual Production Workflow", "Film Production Workflow" ], "is_subclass_of": [ "Display and Rendering", "Production Pipeline" ], "wikilinks": [ "metaverse", "Production Pipeline" ] }, { "id": "virtual-production", "title": "Virtual Production", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Real-time filmmaking technique combining LED Volume stages, game engine rendering, and in-camera visual effects (ICVFX) to create photorealistic virtual environments during live-action production, enabling directors to see final composited imagery on set. It integrates motion capture, photogrammetry, and neural rendering to compress post-production timelines and allow creative decisions to be made on set rather than in post.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:virtual-production", "labels": [ "Virtual Production" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "1899", "3:2 pulldown", "3D modelers", "5G private networks", "7680 Hz", "Absen", "ACES", "ACES 1.2", "ACES 1.3", "ACES AP1", "Adaptive Performance", "Adobe Firefly", "After Effects", "Agisoft Metashape", "AI denoising", "AI-driven", "AI person detection", "AI-powered animation", "AI-powered rotoscoping", "Alembic" ] }, { "id": "virtual-property-right", "title": "Virtual Property Right", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A legally recognized claim to ownership, use, transfer, or exclusion rights over digital assets, virtual goods, or intangible resources within virtual environments, enforced through technical mechanisms, platform policies, or legal frameworks.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:virtual-property-right", "labels": [ "Virtual Property Right" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Authentication Mechanism", "Enforcement System", "Exclusion Right", "IP Protection", "Legal Entity", "Legal Recognition", "Legal System", "NFT", "Ownership Claim", "Property Law Framework", "Transfer Mechanism", "Uniform Commercial Code (UCC) Article 12", "Usage Permission", "World Intellectual Property Organization (WIPO)", "ApplicationLayer", "Asset Registry", "Blockchain", "Digital Identity", "Digital Ownership", "Digital Signature" ] }, { "id": "virtual-reality-vr", "title": "Virtual Reality (VR)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Immersive technology system that combines physical head-mounted display hardware with virtual computer-generated 3D environments to create fully encompassing sensory experiences that replace user perception of the physical world.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-reality-vr", "labels": [ "Virtual Reality (VR)", "VR Headset" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "6DoF Tracking", "GPU", "Head-Mounted Display", "IEEE VR Standards", "ISO/IEC 18039", "Khronos OpenXR", "Low-Latency Display", "Stereoscopic Rendering", "Tracking Sensors", "ApplicationLayer", "Extended Reality (XR)", "Haptic Feedback", "Immersive Gaming", "InteractionDomain", "Motion Tracking", "Real-time Rendering", "Spatial Audio System", "Spatial Computing", "Virtual Presence", "Virtual Tourism" ] }, { "id": "virtual-reality-applications", "title": "Virtual Reality Applications", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Software programmes and use-case deployments that leverage virtual reality technology to deliver immersive, interactive experiences for domains including healthcare, education, enterprise training, entertainment, and social interaction, running on dedicated VR hardware or spatial computing platforms.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-reality-applications", "labels": [ "Virtual Reality Applications" ], "is_subclass_of": [ "Platform and Environment", "Virtual Reality" ], "wikilinks": [ "owl:Thing" ] }, { "id": "virtual-reality-platform", "title": "Virtual Reality Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An integrated hardware and software ecosystem that provides the compute, display, tracking, and runtime services required to deliver immersive virtual reality experiences. A VR platform encompasses the headset, motion controllers, inside-out or external tracking system, operating system, SDK, content store, and developer toolchain. Platforms differ in their openness to third-party applications, support for room-scale interaction, and adherence to interoperability standards such as OpenXR.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-reality-platform", "labels": [ "Virtual Reality Platform" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "virtual-reality-telepresence", "title": "Virtual Reality Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "\"The use of virtual reality head-mounted displays and immersive 3D environments to enable remote participants to experience shared virtual spaces with stereoscopic vision, spatial audio, head tracking, and avatar embodiment, creating a subjective sense of co-location despite geographical separati...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:virtual-reality-telepresence", "labels": [ "Virtual Reality Telepresence", "TELE-020-virtual-reality-telepresence" ], "is_subclass_of": [ "Telepresence", "TELE 001 telepresence" ], "wikilinks": [ "RemoteDesignReview", "TELE-021-augmented-reality-collaboration", "TELE-026-microsoft-mesh", "TELE-027-spatial-platform", "TELE-028-horizon-workrooms", "TELE-058-foveated-rendering", "TELE-100-ai-avatars", "TELE-102-codec-avatars", "TELE-110-spatial-audio-processing", "TELE-150-webrtc", "TELE-154-edge-computing-telepresence", "TELE-157-predictive-tracking", "TELE-001-telepresence" ] }, { "id": "virtual-reality", "title": "Virtual Reality", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual Reality (VR) is a fully immersive, computer-generated simulation technology that replaces the user's physical environment with an interactive three-dimensional world, experienced through head-mounted displays that deliver stereoscopic rendering, spatial audio, and six-degrees-of-freedom motion tracking. VR systems generate a sense of presence by maintaining sub-20 ms motion-to-photon latency, high-refresh-rate displays, and positional tracking of the user's head and hands. As a foundational modality within Extended Reality (XR), VR is architecturally distinguished from Augmented Reality by occluding the real world entirely rather than overlaying digital content upon it. Applications span entertainment, professional training, tele-presence collaboration, clinical therapy, and spatial computing platforms.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:virtual-reality", "labels": [ "Virtual Reality", "VirtualReality" ], "is_subclass_of": [ "Extended Reality (XR)" ], "wikilinks": [] }, { "id": "virtual-replica", "title": "Virtual Replica", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A digital representation of a physical object, system, or environment created through 3D modelling, photogrammetry, or LiDAR scanning, serving as a static or dynamic copy for visualisation, simulation, analysis, and testing without the real-time bidirectional data connectivity that distinguishes a full digital twin. Virtual replicas are typically point-in-time snapshots; they support heritage preservation, product safety testing, urban planning, and manufacturing visualisation workflows.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-replica", "labels": [ "Virtual Replica", "Virtual Replica Creation" ], "is_subclass_of": [ "Platform and Environment", "Digital Model" ], "wikilinks": [ "Digital Model", "metaverse" ] }, { "id": "virtual-scouting", "title": "Virtual Scouting", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of exploring and evaluating potential filming locations or virtual production environments using digital tools \u2014 VR headsets, 360-degree cameras, LiDAR, and game engines \u2014 enabling directors and cinematographers to navigate locations remotely, compose shots, analyse lighting, and make blocking decisions without physical travel. Tools within Unreal Engine provide measurement, interaction, and real-time world-building capabilities compatible with major VR headsets, substantially reducing pre-production travel costs and carbon footprint.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-scouting", "labels": [ "Virtual Scouting" ], "is_subclass_of": [ "Content and Assets", "Pre Production" ], "wikilinks": [ "metaverse", "Pre-Production" ] }, { "id": "virtual-securities-offering-vso", "title": "Virtual Securities Offering (VSO)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A tokenized securities issuance process that leverages blockchain technology to create, distribute, and manage digital representations of traditional securities with embedded regulatory compliance and automated governance mechanisms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:virtual-securities-offering-vso", "labels": [ "Virtual Securities Offering (VSO)" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "24/7 Trading", "Automated Compliance", "Compliance Validation Engine", "Custody Solution", "Distribution Mechanism", "FINRA", "Fractional Ownership", "Global Capital Access", "Investor Registry", "KYC/AML System", "Legal Structure", "MiCA", "Programmable Securities", "SEC", "Securities Law", "Token Issuance Contract", "Transfer Agent", "ApplicationLayer", "Audit Trail", "Blockchain Network" ] }, { "id": "virtual-set-design", "title": "Virtual Set Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The creation of digital environments and backgrounds for film, television, and broadcast production using 3D modelling, game engines, and LED volume wall technology, enabling dynamic digital sets modifiable in real time during filming. Virtual sets replace or augment physical construction, with camera tracking systems providing perspective-correct parallax so actors interact believably with the rendered environment.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-set-design", "labels": [ "Virtual Set Design" ], "is_subclass_of": [ "Content and Assets", "Production Design" ], "wikilinks": [ "Computer Vision", "metaverse", "Production Design" ] }, { "id": "virtual-society-regulations", "title": "Virtual Society Regulations", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The legal frameworks, governance structures, and policy mechanisms designed to regulate behaviour, protect rights, and ensure safety within metaverse platforms and virtual world environments, encompassing intellectual property, data privacy, content moderation, and cross-jurisdictional enforcement challenges. The EU regulatory suite \u2014 GDPR, Digital Services Act, Digital Markets Act, and AI Act \u2014 forms the most comprehensive current framework, with international coordination emerging through the Global Digital Compact adopted in September 2024.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-society-regulations", "labels": [ "Virtual Society Regulations" ], "is_subclass_of": [ "AI Governance and Ethics", "Digital Governance" ], "wikilinks": [ "Digital Governance", "metaverse" ] }, { "id": "virtual-society", "title": "Virtual Society", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An organised social structure emerging within virtual worlds and metaverse platforms, comprising the norms, governance mechanisms, economic systems, communities, and social interactions of persistent digital environments. Virtual societies exhibit properties analogous to physical societies \u2014 ownership, law, culture, and commerce \u2014 mediated through digital platforms, avatars, and blockchain-based economies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:virtual-society", "labels": [ "Virtual Society" ], "is_subclass_of": [ "Platform and Environment", "Metaverse", "Social Structure" ], "wikilinks": [ "Social Structure" ] }, { "id": "virtual-stage", "title": "Virtual Stage", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A production environment surrounded by curved LED walls and floors displaying real-time computer-generated imagery, enabling in-camera visual effects (ICVFX) for film, television, and broadcast production. Digital backgrounds rendered by a game engine react dynamically to tracked camera movement, capturing realistic lighting and reflections directly on-sensor. Epic Games' StageCraft system and Unreal Engine power the majority of commercial installations, with facilities now exceeding 1,700 m\u00b2 of LED surface area.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-stage", "labels": [ "Virtual Stage" ], "is_subclass_of": [ "Platform and Environment", "Production Facility" ], "wikilinks": [ "metaverse", "Production Facility" ] }, { "id": "virtual-theater", "title": "Virtual Theater", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A form of immersive performing arts that uses VR headsets and virtual environments to present theatrical productions, enabling audiences to experience performances as silent observers within the scene, with some productions featuring interactive narratives where story outcomes depend on audience ...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-theater", "labels": [ "Virtual Theater" ], "is_subclass_of": [ "Platform and Environment", "Performing Arts" ], "wikilinks": [ "metaverse", "Performing Arts" ] }, { "id": "virtual-tour", "title": "Virtual Tour", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An interactive 360-degree digital representation of a physical location that enables remote exploration through web browsers or VR headsets, commonly used in real estate, museums, hospitality, and education to provide immersive walkthroughs without physical presence.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-tour", "labels": [ "Virtual Tour", "Virtual Heritage Tour" ], "is_subclass_of": [ "Platform and Environment", "Immersive Experience" ], "wikilinks": [ "Immersive Experience", "metaverse" ] }, { "id": "virtual-tourism", "title": "Virtual Tourism", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The use of VR, AR, and metaverse technologies to explore travel destinations remotely, enabling users to experience locations, hotels, and attractions virtually before booking or as an alternative to physical travel, particularly for those with physical, economic, or accessibility constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:virtual-tourism", "labels": [ "Virtual Tourism" ], "is_subclass_of": [ "Platform and Environment", "Tourism Industry" ], "wikilinks": [ "metaverse", "Tourism Industry" ] }, { "id": "virtual-training", "title": "Virtual Training", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The use of VR, AR, and simulation technologies to create immersive learning experiences for workforce development, enabling employees to practise skills, rehearse high-stakes scenarios, and learn procedures in safe, repeatable virtual environments with measurable outcomes. Empirical evidence \u2014 including Walmart's programme covering over one million employees \u2014 demonstrates 75% material retention and up to 4x focus improvement over e-learning. The global market was valued at USD 380 billion in 2024 and is projected to reach USD 1.42 trillion by 2034.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-training", "labels": [ "Virtual Training", "Virtual Training Simulation" ], "is_subclass_of": [ "AI Application", "Workforce Development" ], "wikilinks": [ "metaverse", "Workforce Development" ] }, { "id": "virtual-transactions", "title": "Virtual Transactions", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Virtual transactions are exchanges of value conducted entirely within digital environments, transferring virtual currency, tokens or digital goods between participants. They power in-world economies in games and metaverse platforms and rely on digital payment systems and currencies for settlement. Their integrity depends on secure accounting, fraud prevention and clear ownership semantics.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:virtual-transactions", "labels": [ "Virtual Transactions" ], "is_subclass_of": [ "Virtual Economy" ], "wikilinks": [] }, { "id": "virtual-try-on", "title": "Virtual Try-On", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Virtual try-on is a computer vision and graphics technique that overlays clothing, accessories, or cosmetics onto an image or video of a person so they can preview how an item would look.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:virtual-try-on", "labels": [ "Virtual Try-On" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [ "Computer Vision", "Image Generation", "E-Commerce" ] }, { "id": "virtual-wedding", "title": "Virtual Wedding", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A marriage ceremony conducted within a metaverse or virtual reality platform in which participants are represented by digital avatars, enabling global attendance without physical travel. Virtual weddings may incorporate NFT wedding gifts, custom virtual venues, and live-streamed officiants, but typically require separate legal proceedings for official recognition under applicable jurisdictional law.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-wedding", "labels": [ "Virtual Wedding" ], "is_subclass_of": [ "Platform and Environment", "Virtual Event" ], "wikilinks": [ "metaverse", "Virtual Event" ] }, { "id": "virtual-workspace", "title": "Virtual Workspace", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Virtual Workspace is a spatially organised digital environment \u2014 typically experienced through XR headsets or desktop 3D interfaces \u2014 that replicates or enhances the collaborative and cognitive functions of a physical office. It supports remote collaboration, persistent shared artefacts, spatial audio, and embodied presence, addressing the limitations of flat video-conferencing for complex knowledge work.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:virtual-workspace", "labels": [ "Virtual Workspace", "Shared Virtual Workspace" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "virtual-world-building", "title": "Virtual World Building", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The process of creating 3D virtual environments for metaverse platforms, games, and simulations using tools ranging from no-code drag-and-drop builders to professional game engines, enabling the construction of immersive digital spaces with customisable assets, terrain, physics, and interactive elements. Generative AI is rapidly lowering barriers through voice and text-prompt environment generation, while Unity and Unreal Engine remain the professional standard for complex, large-scale world creation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-world-building", "labels": [ "Virtual World Building" ], "is_subclass_of": [ "Platform and Environment", "3D Design" ], "wikilinks": [ "3D Design", "metaverse" ] }, { "id": "virtual-world-creation", "title": "Virtual World Creation", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The comprehensive process of designing, developing, and deploying immersive 3D virtual environments using game engines, procedural generation algorithms, and specialized terrain tools, encompassing everything from initial concept to fully realized interactive digital worlds for games, metaverse p...", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-world-creation", "labels": [ "Virtual World Creation" ], "is_subclass_of": [ "Platform and Environment", "Digital Content Creation" ], "wikilinks": [ "Digital Content Creation", "metaverse" ] }, { "id": "virtual-world-infrastructure", "title": "Virtual World Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The underlying technical foundation required to operate metaverse platforms, encompassing cloud computing services, edge networks, distributed systems, real-time rendering capabilities, networking protocols, and blockchain integration that together enable persistent, scalable virtual environments with low-latency user interactions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-world-infrastructure", "labels": [ "Virtual World Infrastructure" ], "is_subclass_of": [ "Computing and Cloud", "Digital Infrastructure" ], "wikilinks": [ "Digital Infrastructure", "metaverse" ] }, { "id": "virtual-world-operation", "title": "Virtual World Operation", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Virtual world operation is the ongoing technical and administrative running of a persistent virtual environment, including server orchestration, state synchronisation, content updates, moderation and economy management. It keeps the world available, consistent and safe for concurrent users across sessions. Operation is the runtime layer that realises the design encoded in a metaverse architecture or stack.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:virtual-world-operation", "labels": [ "Virtual World Operation" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "virtual-world-platform", "title": "Virtual World Platform", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A software ecosystem that enables users to create, explore, and interact within persistent 3D virtual environments. It encompasses both centralised platforms \u2014 such as Roblox (214 million monthly active users) and Fortnite \u2014 controlled by platform operators, and blockchain-based decentralised platforms \u2014 such as Decentraland and The Sandbox \u2014 where users hold genuine ownership of digital assets and participate in on-chain governance. The metaverse market hosting these platforms is valued at USD 103.6 billion and projected to reach USD 507.8 billion by 2030.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:virtual-world-platform", "labels": [ "Virtual World Platform" ], "is_subclass_of": [ "Platform and Environment", "Digital Platform" ], "wikilinks": [ "Digital Platform", "metaverse" ] }, { "id": "virtual-world-traversal", "title": "Virtual World Traversal", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual World Traversal is the set of techniques, interfaces, and protocols that enable users and agents to navigate between and within persistent virtual environments in the metaverse \u2014 including scene transitions, cross-world identity portability, spatial audio transitions, and continuous presence mechanics.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:virtual-world-traversal", "labels": [ "Virtual World Traversal" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Metaverse" ] }, { "id": "virtual-world", "title": "Virtual World", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A self-contained digital environment with persistent state, spatial properties, user interaction capabilities, and internal rules that simulate physical or fantastical worlds, providing a shared space for multiple users to interact with each other and digital objects.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:virtual-world", "labels": [ "Virtual World", "Virtual Worlds", "VirtualWorld" ], "is_subclass_of": [ "Platform and Environment", "Platform and Environment" ], "wikilinks": [ "Client Application", "Collaborative Work", "Content Delivery Network", "Creative Expression", "Economic System", "IEEE VR Standards", "ISO/IEC 23005", "Object Persistence", "Server Infrastructure", "Social Interaction", "Social System", "User Representation", "World Space", "3D Rendering Engine", "ApplicationLayer", "Asset Management", "Authentication Service", "Database System", "Digital Economy", "Digital Real Estate" ] }, { "id": "virtualisation", "title": "Virtualisation", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Virtualisation is the creation of abstracted, software-defined representations of physical computing resources \u2014 processors, memory, storage and networks \u2014 allowing multiple isolated environments to share one set of hardware. A hypervisor or equivalent control layer presents each environment with the illusion of dedicated resources while multiplexing the underlying hardware. It is the foundational technology behind cloud computing, enabling consolidation, isolation, elastic provisioning and efficient utilisation of infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:virtualisation", "labels": [ "Virtualisation", "Virtualization" ], "is_subclass_of": [ "Infrastructure", "Operating System" ], "wikilinks": [] }, { "id": "visa", "title": "Visa", "domain": "finance", "domain_name": "Finance", "definition": "Visa is a multinational payment technology company that operates one of the largest electronic payment networks, connecting cardholders, merchants, and banks. It is headquartered in the United States.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:visa", "labels": [ "Visa" ], "is_subclass_of": [ "Payment Network" ], "wikilinks": [ "Payment System", "Settlement", "Payment Network", "https://www.visa.com", "https://usa.visa.com/about-visa.html" ] }, { "id": "vision-language-model", "title": "Vision Language Model", "domain": "ai", "domain_name": "Ai", "definition": "A Vision Language Model (VLM) is a multimodal neural architecture that jointly processes visual inputs (images or video) and natural language text, learning shared representations that enable cross-modal tasks such as image captioning, visual question answering, visual grounding, and instruction-following based on image context. VLMs typically pair a visual encoder with a large language model backbone connected by a learned projection mechanism.", "entityType": "Class", "qualityScore": 0.85, "maturity": "emerging", "iri": "urn:ngm:class:vision-language-model", "labels": [ "Vision Language Model", "Vision-Language Model" ], "is_subclass_of": [ "Multimodal Models" ], "wikilinks": [] }, { "id": "vision-processing", "title": "Vision Processing", "domain": "ai", "domain_name": "Ai", "definition": "Vision processing is the computational transformation of raw image and video data into structured representations and decisions, spanning low-level operations such as filtering and feature extraction through high-level recognition and interpretation. It is the algorithmic core of computer-vision systems and specialised applications such as medical imaging. Efficient vision processing increasingly runs on dedicated accelerators to meet real-time demands.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:apple-mixed-reality-headsetcessing", "labels": [ "Vision Processing" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "vision-transformer", "title": "Vision Transformer", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The Vision Transformer (ViT) is a neural network architecture that applies the transformer self-attention mechanism directly to sequences of fixed-size image patches, treating each patch embedding as a token analogous to a word in natural-language processing. Introduced by Dosovitskiy et al. (2020), ViT demonstrated that pure attention-based models can match or exceed convolutional networks on image classification benchmarks when pre-trained on sufficiently large datasets.", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:vision-transformer", "labels": [ "Vision Transformer", "Transformer Vision Model" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "vision-transformers", "title": "Vision Transformers", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Transformer-based neural network architectures applied to images by splitting an image into fixed-size patches and treating those patches as a sequence of tokens for self-attention, enabling global context modelling across the full image without relying on local convolutional receptive fields.", "entityType": "Class", "qualityScore": 0.72, "maturity": "growing", "iri": "urn:ngm:class:vision-transformers", "labels": [ "Vision Transformers" ], "is_subclass_of": [ "Vision Transformer" ], "wikilinks": [ "Transformer", "Attention Mechanism", "Image Segmentation", "Computer Vision", "Vision Transformer" ] }, { "id": "vision-language-action-models", "title": "Vision-Language-Action Models", "domain": "ai", "domain_name": "Ai", "definition": "Vision-language-action (VLA) models are foundation models that jointly process visual observations and natural-language instructions to produce executable action sequences for embodied agents such as robots. Extending vision-language models with an action-generation head, VLAs are trained on large datasets pairing perception and instructions with demonstrated behaviour, enabling generalisation across tasks, objects, and embodiments. They represent a convergence of multimodal learning and robotics, aiming for generalist policies that follow open-ended commands rather than executing narrowly scripted skills.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:vision-language-action-models", "labels": [ "Vision-Language-Action Models", "Vision-Language-Action Model" ], "is_subclass_of": [ "Vision Language Model" ], "wikilinks": [] }, { "id": "vision-claw-agentic-container", "title": "VisionClaw Agentic Container", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Sovereign Mesh|sovereign, manifest-driven agentic runtime container that orchestrates DID Nostr Identity|DID Nostr agents across a decentralised Peer-to-Peer Network|peer-to-peer network, enabling autonomous agents to operate with verifiable credentials, pluggable adapters, and cont...", "entityType": "Class", "qualityScore": 0.88, "maturity": "established", "iri": "urn:ngm:class:vision-claw-agentic-container", "labels": [ "VisionClaw Agentic Container" ], "is_subclass_of": [ "AI Application", "AI Agent System" ], "wikilinks": [ "ADR-005", "ADR-006", "ADR-007", "ADR-008", "ADR-009", "ADR-010", "ADR-012", "ADR-013", "Agent Bead", "AgenticSystemsDomain", "Autonomous Agent Operation", "BIP-340 Schnorr Keypair", "BIP-340 Schnorr Keypair", "BIP-340 Schnorr Keypair", "BlockchainIntegrationDomain", "Consultation MCP", "CoordinationLayer", "Cross-Agent Federation", "DDD-004", "Decentralised Coordination" ] }, { "id": "visioning-lab-property-crosswalk", "title": "VisioningLab-Property-Crosswalk", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A mapping framework that aligns VisioningLab's hybrid Logseq page properties with standard RDF, OWL, and SKOS predicates so that locally authored metadata can be published as valid linked data. The crosswalk records each property-to-predicate correspondence, supporting ontology interoperability and round-tripping between the knowledge graph and external semantic web tooling.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:visioning-lab-property-crosswalk", "labels": [ "VisioningLab-Property-Crosswalk" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "Aliases", "MetaverseDomain" ] }, { "id": "visual-design", "title": "Visual Design", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Visual design is the practice of shaping the appearance of interfaces and artefacts through the deliberate arrangement of typography, colour, imagery, space and composition to communicate meaning and guide attention. It establishes visual hierarchy, consistency and aesthetic quality so that products are both legible and emotionally resonant. As a core part of user experience and interaction design, visual design translates structure and intent into perceivable form, increasingly across spatial and immersive media.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:visual-design", "labels": [ "Visual Design" ], "is_subclass_of": [ "User Experience" ], "wikilinks": [] }, { "id": "visual-development", "title": "Visual Development", "domain": "ai", "domain_name": "Ai", "definition": "Visual development is the pre-production creative process of exploring and defining the look, mood, characters, and environments of a visual project through concept sketches, colour studies, and design iteration before final production begins.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:visual-development", "labels": [ "Visual Development" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "visual-effects", "title": "Visual Effects", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Visual Effects (VFX) are the computational and compositing techniques used to create, simulate, or augment imagery that cannot be practically captured in-camera, spanning both offline film/broadcast pipelines and real-time interactive applications. In offline production, VFX encompasses computer-generated imagery (CGI), digital compositing, matte painting, motion capture integration, and physically accurate simulation of fluids, cloth, destruction, and crowds. In real-time and spatial-computing contexts, VFX includes particle systems, post-processing stacks (bloom, depth-of-field, ambient occlusion, screen-space reflections), procedural shaders, and GPU-accelerated simulation executed within physically-based rendering pipelines at interactive frame rates.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:visual-effects", "labels": [ "Visual Effects", "Dynamic Visual Effects", "Film Visual Effects", "Visual Effects Pipeline" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "visual-grounding", "title": "Visual Grounding", "domain": "ai", "domain_name": "Ai", "definition": "Visual grounding is the task of localising the region of an image or scene that corresponds to a natural-language expression, linking words to specific visual entities. It connects language understanding to perception, enabling models to point at, select or act on the object a user refers to. Visual grounding is foundational for vision-language models and for agents that operate graphical interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:visual-grounding", "labels": [ "Visual Grounding" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "visual-marker", "title": "Visual Marker", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A distinctive visual pattern or fiducial placed in physical environments to enable computer vision systems to determine position, orientation, and pose for augmented reality applications. Types include QR codes, AprilTags, ARTags, and custom image targets that serve as reference points for overlaying digital content; detection relies on camera capture, template matching, and pose estimation algorithms, often combined with SLAM or IMU sensor fusion for robust tracking.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:visual-marker", "labels": [ "Visual Marker", "ArUco Marker" ], "is_subclass_of": [ "Interaction Technology", "AR Technology" ], "wikilinks": [ "AR Technology", "metaverse" ] }, { "id": "visual-odometry", "title": "visual odometry", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Visual odometry (VO) is a technique for incrementally estimating the six-degree-of-freedom pose (position and orientation) of a camera-equipped agent by detecting and tracking salient features across consecutive image frames and computing the relative camera motion between them via geometric constraints such as the essential or fundamental matrix. It provides ego-motion estimation without relying on GPS, wheel encoders, or external beacons, making it applicable in GPS-denied environments such as indoor spaces, underground tunnels, and planetary surfaces. Scale ambiguity in monocular configurations is resolved by stereo baselines or depth cameras; drift is bounded by fusing inertial measurements (visual-inertial odometry) or by applying loop-closure detection within a full SLAM pipeline.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:visual-odometry", "labels": [ "Visual Odometry", "VisualOdometry" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "visual-perception", "title": "Visual Perception", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual perception is a robot or agent's capability to interpret its environment from camera and visual sensor data, extracting the objects, surfaces, motion and spatial structure needed to act. It transforms raw imagery into actionable scene understanding that drives behaviours such as gaze control, manipulation and navigation. Visual perception is a core perceptual modality for embodied robotic systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-perception", "labels": [ "Visual Perception" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "visual-place-recognition", "title": "Visual Place Recognition", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual place recognition (VPR) is the task of identifying whether a currently observed scene corresponds to a previously visited location by matching image content against a database of geotagged or topologically indexed images. It underpins loop closure in SLAM and global re-localisation for autonomous systems, relying on appearance- and condition-invariant descriptors. Robustness to viewpoint, illumination, and seasonal change is the central research challenge.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-place-recognition", "labels": [ "Visual Place Recognition", "Place Recognition", "Place Recognition Module" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [] }, { "id": "visual-question-answering", "title": "Visual Question Answering", "domain": "ai", "domain_name": "Ai", "definition": "Visual question answering (VQA) is a multimodal AI task in which a system produces a natural-language answer to a free-form question posed about an image or scene. It requires jointly grounding linguistic semantics in visual content, combining object recognition, spatial reasoning, and language understanding. VQA is a benchmark capability for vision-language models and a building block for assistive and augmented-reality interfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-question-answering", "labels": [ "Visual Question Answering" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "visual-representation", "title": "Visual Representation", "domain": "ai", "domain_name": "Ai", "definition": "A visual representation is a learned encoding of image or video content into a feature vector or embedding that captures semantic and structural properties useful for downstream tasks. Such representations, produced by convolutional or transformer-based encoders, support classification, retrieval, detection, and multimodal alignment. The quality of a visual representation determines transferability and sample efficiency across vision applications.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-representation", "labels": [ "Visual Representation" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "visual-slam", "title": "Visual SLAM", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual simultaneous localisation and mapping (Visual SLAM) is the process of concurrently estimating a camera's trajectory and reconstructing a map of an unknown environment using image data alone or fused with inertial measurements. It combines feature tracking or direct photometric alignment with pose-graph or bundle-adjustment optimisation and loop closure. Visual SLAM is foundational to AR headsets, drones, and mobile robots that lack external positioning.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-slam", "labels": [ "Visual SLAM" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "visual-inertial-odometry", "title": "Visual-Inertial Odometry", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual-inertial odometry (VIO) is a technique that estimates a device's six-degree-of-freedom motion and pose by fusing camera imagery with inertial measurement unit data. The complementary fusion compensates for the drift of inertial sensors and the scale ambiguity of monocular vision, yielding robust, low-latency tracking. VIO underpins inside-out tracking for mixed-reality headsets, drones and mobile robots.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:visual-inertial-odometry", "labels": [ "Visual-Inertial Odometry" ], "is_subclass_of": [ "Navigation and Planning" ], "wikilinks": [] }, { "id": "visual-servoing", "title": "VisualServoing", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual servoing is a robot control paradigm in which camera-derived visual measurements are used as feedback signals within a closed-loop control system to guide the motion of a robotic manipulator or mobile platform towards a goal configuration defined in visual terms. Rather than relying on pre-computed geometric trajectories, visual servoing continuously computes control signals from current image features or image-space error signals, making it inherently adaptive to object pose uncertainty and disturbances. The two primary architectures are image-based visual servoing (IBVS), which minimises error in image feature space, and position-based visual servoing (PBVS), which reconstructs 3D pose and minimises Cartesian error. Visual servoing enables tasks such as precise grasping, assembly alignment, and autonomous navigation without requiring exact geometric calibration.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:visual-servoing", "labels": [ "VisualServoing", "Visual Servoing" ], "is_subclass_of": [ "Robot Control" ], "wikilinks": [] }, { "id": "visualization-layer", "title": "Visualization Layer", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Graphics and rendering systems responsible for displaying virtual environments, objects, and interfaces through advanced rendering pipelines and visual processing.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:visualization-layer", "labels": [ "Visualization Layer" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [ "3D Scene Graph", "Display Manager", "GPU Resources", "Graphics Pipeline", "Lighting System", "Material System", "MSF Taxonomy 2025", "Presentation Infrastructure", "Shader System", "User Interface Rendering", "Visual Output", "CreativeMediaDomain", "Display Hardware", "Graphics API", "Immersive Experiences", "Rendering Engine", "Rendering Pipeline" ] }, { "id": "visualization", "title": "Visualization", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:visualization", "labels": [ "Visualization", "Visualize" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "vocabulary-governance", "title": "Vocabulary Governance", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "The policies, processes, and organizational structures for managing controlled vocabularies, taxonomies, thesauri, and ontologies throughout their lifecycle, including version control, role-based access permissions, quality assurance, and collaborative maintenance to ensure terminology consistenc...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:vocabulary-governance", "labels": [ "Vocabulary Governance" ], "is_subclass_of": [ "AI Governance and Ethics", "Knowledge Management" ], "wikilinks": [ "Knowledge Management", "metaverse" ] }, { "id": "vocabulary-services", "title": "Vocabulary Services", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Technical infrastructure and APIs that provide programmatic access to controlled vocabularies, taxonomies, and ontologies, enabling applications to retrieve concept definitions, navigate broader/narrower hierarchies, validate terminology, and perform cross-vocabulary mapping for consistent semantic interpretation across platforms. Common interfaces include SPARQL endpoints, SKOS APIs, and REST services backed by platforms such as PoolParty or Metaphactory.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:vocabulary-services", "labels": [ "Vocabulary Services" ], "is_subclass_of": [ "Standards and Interoperability", "Knowledge Management" ], "wikilinks": [ "Knowledge Management", "metaverse" ] }, { "id": "vocabulary", "title": "Vocabulary", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The complete set of unique tokens in a language model's tokenisation scheme, typically ranging from 32,000 to 128,000 entries in modern architectures. Vocabulary size directly governs embedding matrix dimensions, output layer size, and token-level granularity, with larger vocabularies improving expressiveness and training efficiency at the cost of increased memory during inference.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:vocabulary", "labels": [ "Vocabulary" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "voice-activity-detection", "title": "Voice Activity Detection", "domain": "ai", "domain_name": "Ai", "definition": "The automated process of classifying audio frames as containing human speech or non-speech (silence, background noise, music) in order to segment an audio stream before downstream processing. Voice activity detection (VAD) reduces computational load on speech-sensitive systems by forwarding only speech-active segments to recognition, enhancement, or analysis modules. Modern VAD systems use neural classifiers trained on diverse acoustic conditions to achieve robust detection under noise, reverberation, and overlapping sounds. It is a foundational pre-processing stage in speech pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:voice-activity-detection", "labels": [ "Voice Activity Detection" ], "is_subclass_of": [ "AI Technique" ], "wikilinks": [] }, { "id": "voice-assistant", "title": "Voice Assistant", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A voice assistant is a software agent that responds to spoken commands and questions using speech recognition and natural language processing. Examples include assistants built into phones and smart speakers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:voice-assistant", "labels": [ "Voice Assistant", "Personalised Voice Assistant" ], "is_subclass_of": [ "Conversational AI" ], "wikilinks": [ "Speech Recognition", "Natural Language Processing", "User Experience", "AI Agent", "Conversational AI", "https://en.wikipedia.org/wiki/Virtual_assistant", "https://developer.amazon.com/en-US/alexa" ] }, { "id": "voice-cloning", "title": "voice cloning", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Voice cloning is a speech synthesis technique that uses a short reference audio recording of a target speaker to condition or fine-tune a neural text-to-speech model so that it reproduces that speaker's vocal characteristics \u2014 including timbre, prosody, accent, and speaking rhythm \u2014 when given arbitrary text input. Modern systems employ a speaker encoder network that extracts a fixed-dimensional speaker embedding from reference audio, which conditions a sequence-to-sequence acoustic model and vocoder to produce personalised synthetic speech. Zero-shot voice cloning generalises this capability to entirely unseen speakers without fine-tuning by leveraging large pre-trained generative models. The technology underpins both beneficial applications such as accessibility aids and audiobook narration, and adversarial uses including audio deepfakes and voice fraud.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:voice-cloning", "labels": [ "Voice Cloning", "Voice Clone", "Voice Cloning Attack" ], "is_subclass_of": [ "SpeechSynthesis" ], "wikilinks": [] }, { "id": "voice-input", "title": "Voice Input", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Voice input is an interaction modality in which spoken language is captured, recognised, and interpreted as commands or content, allowing users to control systems and enter data hands-free. It combines microphone capture, speech recognition, and natural-language understanding to map utterances onto actions or text. In spatial computing, voice input is a primary modality for immersive and accessible interfaces where conventional keyboards and pointers are impractical.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:voice-input", "labels": [ "Voice Input" ], "is_subclass_of": [ "Multimodal Interaction" ], "wikilinks": [] }, { "id": "voice-interaction", "title": "Voice Interaction", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Communication mod enabling control and conversation through speech recognition, natural language understanding, and text-to-speech synthesis.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:voice-interaction", "labels": [ "Voice Interaction", "Voice Chat", "VoiceInteraction" ], "is_subclass_of": [ "Interaction Technology", "Interaction Technology" ], "wikilinks": [ "ACM + ETSI", "Acoustic Environment", "Audio Processing", "Hands-Free Control", "Language Model", "Language Support", "Microphone", "Multimodal Interaction", "Natural Communication", "Natural Language Understanding", "Speech Synthesis", "Voice Assistant", "Voice Commands", "Accessibility", "InteractionDomain", "Network Latency", "NetworkLayer", "Speech Recognition", "Text-to-Speech" ] }, { "id": "voice-interfaces", "title": "Voice Interfaces", "domain": "ai", "domain_name": "Ai", "definition": "Voice interfaces are human-computer interaction systems that accept spoken input and respond with synthesised speech, chaining automatic speech recognition, natural-language understanding, dialogue management, and text-to-speech. They enable hands-free, eyes-free interaction across smart speakers, vehicles, and accessibility tools. Latency, recognition accuracy in noise, and natural turn-taking are the principal usability constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:voice-interfaces", "labels": [ "Voice Interfaces" ], "is_subclass_of": [ "Human Computer Interaction" ], "wikilinks": [] }, { "id": "voice-memo", "title": "Voice Memo", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "An asynchronous audio communication artefact that allows users to record and share spoken messages, conveying prosodic nuance, emotional tone, and contextual richness not available in text. Voice memos occupy a middle tier of communication fidelity between text messages and synchronous video calls, enabling remote teams to maintain natural communication rhythms across time zones without requiring real-time scheduling.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:voice-memo", "labels": [ "Voice Memo" ], "is_subclass_of": [ "Communication Technology", "Asynchronous Collaboration" ], "wikilinks": [ "Collaboration Tools", "Asynchronous Collaboration", "TelecollaborationDomain" ] }, { "id": "voice-over-ip", "title": "Voice Over Ip", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Voice over IP (VoIP) is a family of technologies that digitise, compress, and transmit voice communications as packetised data over Internet Protocol networks rather than through dedicated circuit-switched telephony infrastructure. Audio is sampled, encoded using codecs such as G.711, G.729, or Opus, packetised, and transported using the Real-time Transport Protocol (RTP) over UDP, with session management handled by signalling protocols such as SIP or H.323. VoIP enables cost reduction, feature richness, and integration with unified communications platforms, but introduces quality-of-service sensitivities to packet loss, jitter, and latency. Encryption via SRTP and TLS/DTLS provides confidentiality and integrity for enterprise and consumer deployments.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:voice-over-ip", "labels": [ "Voice Over Ip", "Voice over IP" ], "is_subclass_of": [ "Real-Time Communication" ], "wikilinks": [] }, { "id": "voice-user-interface", "title": "Voice User Interface", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Voice User Interface (VUI) is a human-computer interaction modality in which users issue commands and receive responses through spoken language rather than visual controls. It chains automatic speech recognition, natural-language understanding, dialogue management and speech synthesis to turn utterances into actions and synthesised replies. VUIs power virtual assistants, in-car systems and accessibility tools where hands-free or eyes-free operation is valuable.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:voice-user-interface", "labels": [ "Voice User Interface" ], "is_subclass_of": [ "User Interface" ], "wikilinks": [] }, { "id": "voice-of-customer", "title": "Voice of Customer", "domain": "ai", "domain_name": "Ai", "definition": "Voice of Customer (VoC) is the systematic capture and analysis of customer feedback, from surveys, support transcripts, reviews and social channels, to surface expressed needs, preferences and pain points that inform product and service decisions. Modern VoC programmes increasingly use natural language processing to extract sentiment and themes from unstructured feedback at scale. It is a core input to customer experience management processes.", "entityType": "Class", "qualityScore": 0.55, "maturity": "draft", "iri": "urn:ngm:class:voice-of-customer", "labels": [ "Voice of Customer" ], "is_subclass_of": [ "Customer Experience" ], "wikilinks": [] }, { "id": "volume-rendering", "title": "Volume Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Volume rendering produces images directly from three-dimensional scalar or density fields by integrating colour and opacity along view rays, rather than rendering explicit surfaces.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:volume-rendering", "labels": [ "Volume Rendering", "Volume Rendering Integral" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Graphics Pipeline", "Neural Radiance Fields", "Volumetric Video", "Real-Time Rendering", "Computer Graphics" ] }, { "id": "volumetric-capture", "title": "Volumetric Capture", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Volumetric Capture is the acquisition process that records a subject or environment in three spatial dimensions over time, using arrays of cameras, depth sensors, structured-light projectors, or LiDAR to produce a fully navigable 4D (space + time) representation. The captured data \u2014 typically as dense point-cloud sequences, multi-view video, or neural radiance fields \u2014 can be rendered from any arbitrary viewpoint and replayed freely in time, enabling photorealistic holographic telepresence, immersive sports and entertainment broadcasting, and real-time digital-twin creation. It sits at the convergence of photogrammetry, computer vision, and real-time graphics pipelines, demanding tightly calibrated hardware rigs and specialised compression and streaming infrastructure.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:volumetric-capture", "labels": [ "Volumetric Capture" ], "is_subclass_of": [ "Photogrammetry" ], "wikilinks": [] }, { "id": "volumetric-rendering", "title": "Volumetric Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Volumetric rendering is the process of producing images of three-dimensional density fields where light is absorbed, emitted, and scattered as it travels through a participating medium. Rather than rendering surfaces, it integrates radiance along rays passing through a volume, capturing effects such as smoke, clouds, fog, and translucent materials. The technique underlies medical visualisation, visual effects, and neural scene representations that store the world as a continuous volumetric function.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:volumetric-rendering", "labels": [ "Volumetric Rendering" ], "is_subclass_of": [ "Rendering" ], "wikilinks": [] }, { "id": "volumetric-video", "title": "Volumetric Video", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Volumetric video captures a subject from many viewpoints to reconstruct a moving three-dimensional representation that can be viewed from any angle, rather than a fixed two-dimensional image sequence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:volumetric-video", "labels": [ "Volumetric Video" ], "is_subclass_of": [ "Computer Graphics" ], "wikilinks": [ "Photogrammetry", "Volume Rendering", "Point Cloud", "Performance Capture", "Computer Graphics" ] }, { "id": "voluntary-carbon-market", "title": "Voluntary Carbon Market", "domain": "governance", "domain_name": "Governance", "definition": "The Voluntary Carbon Market (VCM) is a decentralised, private-sector marketplace in which corporations, governments, and individuals voluntarily purchase and retire carbon credits representing certified reductions or removals of greenhouse gas emissions, without being compelled by regulatory compliance frameworks. Participation is driven by corporate net-zero commitments, reputational and ESG pressures, and a desire to mobilise private finance for climate projects ahead of, or beyond, mandatory obligations. Each credit typically represents one metric tonne of CO\u2082-equivalent avoided or sequestered, verified by independent auditors against published standards such as the Verra Verified Carbon Standard (VCS) or the Gold Standard. The market operates through project developers, registries, brokers, and exchanges, with integrity increasingly governed by the Core Carbon Principles issued by the Integrity Council for the Voluntary Carbon Market (ICVCM).", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:voluntary-carbon-market", "labels": [ "Voluntary Carbon Market", "Carbon Credit Market", "Voluntary Carbon Markets" ], "is_subclass_of": [ "Carbon Markets" ], "wikilinks": [] }, { "id": "voronoi-diagram", "title": "Voronoi Diagram", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A Voronoi diagram is a partition of a plane (or higher-dimensional space) into regions based on proximity to a set of seed points, where each region contains all locations closer to its seed than to any other. It is a foundational structure in computational geometry, dual to the Delaunay triangulation, and supports nearest-neighbour queries, spatial interpolation and procedural generation. Voronoi tessellations appear across spatial analysis, graphics and natural sciences.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:voronoi-diagram", "labels": [ "Voronoi Diagram" ], "is_subclass_of": [ "Computational Geometry" ], "wikilinks": [] }, { "id": "vote-delegation", "title": "Vote Delegation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Vote delegation is a governance mechanism by which a token holder assigns their voting power to another address that votes on their behalf, without transferring ownership of the underlying assets. It is central to on-chain DAO governance, enabling passive holders to entrust active, informed delegates and improving participation rates. Delegation can be revoked or reassigned, and may be liquid, allowing chained or topic-specific delegation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vote-delegation", "labels": [ "Vote Delegation", "Rational Delegation", "Transitive Re-delegation" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "vote-escrow", "title": "Vote Escrow", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Vote escrow (ve) is a tokenomics mechanism in which holders lock governance tokens for a chosen duration in exchange for non-transferable, time-decaying voting power and often boosted protocol rewards. Popularised by Curve's veCRV model, it aligns voter incentives with long-term protocol health by rewarding commitment over short-term speculation. Locked positions decay linearly to zero at unlock, requiring periodic re-locking to maintain influence.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vote-escrow", "labels": [ "Vote Escrow", "Vote-Escrow Mechanism" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [] }, { "id": "vote-escrow-model", "title": "Vote-Escrow Model", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The vote-escrow model is a DeFi tokenomics mechanism in which holders of a governance token lock their tokens for a chosen duration \u2014 typically up to four years \u2014 in exchange for a non-transferable vote-escrow token (such as veCRV in Curve Finance) that grants proportional governance voting power and fee-sharing rights, with voting weight decaying linearly as the lock approaches expiry, aligning token holder incentives with long-term protocol health by penalising short-term speculation. The model was pioneered by Curve Finance and subsequently adopted across dozens of DeFi protocols, spawning meta-governance layers such as Convex Finance that aggregate and direct veToken voting power.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:vote-escrow-model", "labels": [ "Vote-Escrow Model", "Vote Escrow Model", "Vote-Escrowed Token" ], "is_subclass_of": [ "Tokenomics" ], "wikilinks": [] }, { "id": "voting-mechanism", "title": "voting mechanism", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A voting mechanism is a formalised procedure by which participants in a collective system \u2014 such as token holders in a decentralised protocol, members of a cooperative, or delegates in a representative body \u2014 express preferences or binding decisions on resource allocation, parameter changes, protocol upgrades, or dispute resolution. The design of a voting mechanism encodes trade-offs between participation breadth, sybil resistance, plutocracy risk, voter apathy, and decisional legitimacy; canonical variants include token-weighted voting, quadratic voting, conviction voting, ranked-choice voting, and delegated liquid democracy. The chosen mechanism directly shapes the security, fairness, and perceived legitimacy of governance outcomes in both on-chain and off-chain contexts.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:voting-mechanism", "labels": [ "Voting Mechanism", "VotingMechanism" ], "is_subclass_of": [ "Governance and Regulation" ], "wikilinks": [] }, { "id": "voting-power", "title": "Voting Power", "domain": "governance", "domain_name": "Governance", "definition": "Voting power is the quantified influence a participant can exert over a collective decision, expressed as the weight their ballot carries relative to all eligible ballots. In token-based governance systems it is typically proportional to the number of governance tokens held, delegated, or staked at a given snapshot block, though alternative schemes such as quadratic voting, reputation weighting, and one-person-one-vote deliberately decouple influence from raw holdings to resist plutocratic capture.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:voting-power", "labels": [ "Voting Power" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance", "Governance Token", "On-chain Governance", "Snapshot Governance", "Quorum" ] }, { "id": "voting-round", "title": "Voting Round", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A voting round is a discrete phase within a Byzantine fault-tolerant consensus protocol during which participating nodes exchange and tally votes on a proposed value or block. Multi-round protocols such as PBFT and Tendermint progress through prepare, pre-commit, and commit rounds to achieve agreement despite faulty or malicious participants. The round abstraction provides liveness via view changes and safety via supermajority thresholds.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:voting-round", "labels": [ "Voting Round" ], "is_subclass_of": [ "Consensus Protocol" ], "wikilinks": [] }, { "id": "voting-system", "title": "Voting System", "domain": "governance", "domain_name": "Governance", "definition": "A voting system is a method for collecting and aggregating participants' preferences into a collective decision, used in governance, elections, and on-chain protocols.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:voting-system", "labels": [ "Voting System" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance Frameworks", "Governance" ] }, { "id": "voting-systems", "title": "Voting Systems", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Blockchain-based electoral systems employing cryptographic verification, end-to-end verifiability, and distributed ledger technology to enable secure voting whilst facing critical security challenges identified by MIT research showing vulnerabilities allowing vote alteration, academic consensus o...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:voting-systems", "labels": [ "Voting Systems", "Voting Strategy" ], "is_subclass_of": [ "Governance and Regulation", "Blockchain Governance" ], "wikilinks": [ "BC-0142-smart-contract", "BC-0456-self-sovereign-identity", "BC-0457-decentralized-identifiers", "BC-0458-verifiable-credentials", "BC-0462-on-chain-voting", "BC-0463-governance-token", "BC-0470-dao-legal-structures", "BlockchainDomain" ] }, { "id": "votium", "title": "Votium", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Votium is a marketplace on Ethereum where protocols pay holders of vote-escrowed CRV to direct Curve gauge emissions. It operates as a bribery market for governance influence.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:votium", "labels": [ "Votium" ], "is_subclass_of": [ "Hidden Hand" ], "wikilinks": [ "Gauge Voting", "Convex Finance", "Tokenomics", "Curve Finance", "Hidden Hand" ] }, { "id": "voxel-grid", "title": "Voxel Grid", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A voxel grid is a regular three-dimensional lattice that partitions space into uniformly sized cubic cells, each cell (voxel) storing occupancy, colour, density, or other attributes of the volume it covers. It provides a structured spatial representation used to discretise point clouds, build occupancy maps, and accelerate spatial queries in robotics and computer graphics. By contrast with continuous point sets, the fixed cell structure trades fine resolution for predictable indexing and constant-time neighbourhood access.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:voxel-grid", "labels": [ "Voxel Grid" ], "is_subclass_of": [ "Spatial Computing", "Point Cloud" ], "wikilinks": [] }, { "id": "voxel", "title": "Voxel", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A voxel is the volumetric analogue of a pixel: a discrete unit of value located on a regular three-dimensional grid that represents a sample of space, encoding attributes such as density, colour, opacity, or material. Voxels underpin volumetric data structures used in medical imaging, scientific simulation, terrain and procedural modelling, and game engines, and they can be stored efficiently in sparse structures such as octrees to skip empty space. Unlike polygon meshes that describe only surfaces, voxels represent the full interior of objects, enabling destructible geometry, fluid simulation, and direct volume rendering.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:voxel", "labels": [ "Voxel" ], "is_subclass_of": [ "Volume Rendering" ], "wikilinks": [] }, { "id": "vulkan", "title": "Vulkan", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Vulkan is a low-overhead, cross-platform graphics and compute API developed by the Khronos Group and released in 2016 as the successor to OpenGL, designed to give developers explicit control over GPU resources including memory allocation, synchronisation, command buffer submission, and render pass configuration in order to minimise CPU overhead and achieve predictable, high-performance rendering across diverse hardware. Vulkan operates closer to metal than its predecessor: applications manage their own memory pools, pipeline state objects, descriptor sets, and queue families, while the driver's role is reduced to translating API calls into hardware commands with minimal hidden magic. Vulkan shaders are compiled to SPIR-V, a portable intermediate representation, enabling shader code authored in GLSL or HLSL to execute on any conforming GPU without driver-side shader compilation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vulkan", "labels": [ "Vulkan", "Khronos Vulkan KHR_ray_tracing", "Vulkan API" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "vulnerability-analysis", "title": "Vulnerability Analysis", "domain": "security", "domain_name": "Security", "definition": "Vulnerability analysis is the systematic process of discovering, characterising, and prioritising weaknesses in software, systems, or networks that could be exploited to compromise confidentiality, integrity, or availability. It combines static and dynamic code inspection, configuration review, and exploitability assessment to feed remediation and risk decisions. Unlike threat modelling, which is attacker-centric and design-stage, vulnerability analysis focuses on concrete flaws in deployed or candidate artefacts.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vulnerability-analysis", "labels": [ "Vulnerability Analysis" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "vulnerability-assessment", "title": "Vulnerability Assessment", "domain": "security", "domain_name": "Security", "definition": "Vulnerability assessment is the structured evaluation of an organisation's systems against known weaknesses, producing a prioritised inventory of exposures and recommended mitigations. It is typically scope-bounded and recurring, using automated scanners and authenticated checks to map findings to severity and asset criticality. Mandated by many cybersecurity standards, it provides the baseline evidence for compliance and continuous risk reduction.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vulnerability-assessment", "labels": [ "Vulnerability Assessment" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "vulnerability-management", "title": "Vulnerability Management", "domain": "security", "domain_name": "Security", "definition": "The continuous process of identifying, assessing, prioritising and remediating security weaknesses across systems and software to reduce the risk of exploitation, integrating asset discovery, severity scoring, threat intelligence and remediation tracking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:vulnerability-management", "labels": [ "Vulnerability Management" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [ "Vulnerability", "Risk Management", "Network Security", "Security", "Cybersecurity" ] }, { "id": "vulnerability-scanner", "title": "Vulnerability Scanner", "domain": "security", "domain_name": "Security", "definition": "A vulnerability scanner is a tool that inspects systems, networks or applications to identify known security weaknesses. It compares discovered software and configurations against databases of disclosed vulnerabilities.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:vulnerability-scanner", "labels": [ "Vulnerability Scanner" ], "is_subclass_of": [ "Vulnerability" ], "wikilinks": [ "Vulnerability", "Penetration Testing", "Cybersecurity" ] }, { "id": "vulnerability-scanning", "title": "Vulnerability Scanning", "domain": "security", "domain_name": "Security", "definition": "Vulnerability scanning is the automated process of inspecting systems, networks, and applications to identify known security weaknesses by comparing observed configurations and software versions against databases of disclosed vulnerabilities. It produces prioritised findings that feed remediation and patch-management workflows, and is typically run on a recurring schedule across an organisation's assets. Scanning is a detective control that complements deeper manual assessment such as penetration testing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:vulnerability-scanning", "labels": [ "Vulnerability Scanning" ], "is_subclass_of": [ "Vulnerability" ], "wikilinks": [] }, { "id": "vulnerability", "title": "Vulnerability", "domain": "security", "domain_name": "Security", "definition": "A Vulnerability is a weakness, flaw, or inadequacy in a system's design, implementation, or configuration that can be exploited to violate security policies or compromise integrity. Vulnerabilities are classified by origin (design flaw, implementation bug, configuration error), severity (CVSS score), and CIA impact (confidentiality, integrity, availability), and are managed through scanning, patching, and compensating controls.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:vulnerability", "labels": [ "Vulnerability", "Vulnerability Research", "Vulnerability Surface" ], "is_subclass_of": [ "Risk", "Blockchain" ], "wikilinks": [ "Attack Vector", "Blockchain", "Resilience", "Risk", "Security", "Threat Actor" ] }, { "id": "vyper", "title": "Vyper", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Vyper is a contract-oriented, statically typed programming language for the Ethereum Virtual Machine (EVM) that prioritises security, simplicity, and auditability over expressive power. Inspired by Python syntax, it deliberately omits class inheritance, function overloading, recursive calling, and inline assembly to eliminate entire categories of smart-contract vulnerabilities. The language enforces bounds checking, explicit integer overflow handling, and strong typing so that the compiled EVM bytecode is tractable for formal verification and manual code review. Vyper was initially specified by Vitalik Buterin and the Ethereum Foundation team as a safer companion language to Solidity.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:vyper", "labels": [ "Vyper" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [ "Ethereum Virtual Machine", "Smart Contract", "Solidity", "Programming Language" ] }, { "id": "w-3-c-2022-did-core-1-0-recommendation", "title": "W3C 2022 DID Core 1.0 Recommendation", "domain": "standards", "domain_name": "Standards", "definition": "The W3C Recommendation for Decentralized Identifiers (DIDs) version 1.0, published in 2022, defining the core DID data model, syntax and resolution architecture. It establishes DIDs as a standard type of verifiable identifier.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-2022-did-core-1-0-recommendation", "labels": [ "W3C 2022 DID Core 1.0 Recommendation" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-cognitive-ai-community-group", "title": "W3C Cognitive AI Community Group", "domain": "standards", "domain_name": "Standards", "definition": "The W3C Cognitive AI Community Group is a W3C community group that explores approaches to cognitive artificial intelligence and related Web technologies.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-cognitive-ai-community-group", "labels": [ "W3C Cognitive AI Community Group" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-did-core-1-0", "title": "W3C DID Core 1.0", "domain": "standards", "domain_name": "Standards", "definition": "W3C DID Core 1.0 specifies decentralised identifiers, a type of identifier that enables verifiable, self-sovereign digital identity independent of centralised registries.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-did-core-1-0", "labels": [ "W3C DID Core 1.0" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-did-core-specification", "title": "W3C DID Core Specification", "domain": "standards", "domain_name": "Standards", "definition": "A W3C specification defining decentralised identifiers (DIDs), a type of identifier that enables verifiable, self-sovereign digital identity independent of any centralised registry. It defines the DID data model and syntax.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-did-core-specification", "labels": [ "W3C DID Core Specification" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c-did-core", "title": "w3c did core", "domain": "blockchain", "domain_name": "Blockchain", "definition": "W3C DID Core is the World Wide Web Consortium Recommendation (published July 2022) defining Decentralised Identifiers (DIDs) \u2014 globally unique, controller-owned URIs of the form did:: that are cryptographically verifiable and resolvable without a centralised registration authority. Each DID resolves to a JSON-LD DID Document containing verification methods (public keys), authentication suites, and service endpoints. The specification is method-agnostic, accommodating distributed ledgers, peer-to-peer networks, and web servers as verifiable data registries, and forms the identity layer of the broader Self-Sovereign Identity and Verifiable Credentials ecosystem.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:w3-c-did-core", "labels": [ "W3C DID Core", "DID Core", "W3C DID Core 1.0" ], "is_subclass_of": [ "Decentralised Identity" ], "wikilinks": [] }, { "id": "w3-c-did-specification", "title": "W3C DID Specification", "domain": "security", "domain_name": "Security", "definition": "The W3C Decentralized Identifiers (DIDs) specification is a World Wide Web Consortium recommendation that defines a new type of globally unique, persistent, cryptographically verifiable identifier that does not require a centralised registration authority. A DID resolves to a DID Document containing cryptographic material, service endpoints, and verification methods, enabling the subject to authenticate and authorise interactions without reliance on any single identity provider. The specification defines a generic DID syntax and data model that is method-agnostic, with specific DID methods implementing the create/read/update/deactivate operations on different verifiable data registries including blockchains, distributed ledgers, and peer-to-peer networks. It became a W3C Recommendation in July 2022.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:w3-c-did-specification", "labels": [ "W3C DID Specification", "W3C DID Spec", "W3C Decentralized Identifier Specification" ], "is_subclass_of": [ "W3C DID" ], "wikilinks": [] }, { "id": "w-3-c-did-working-group", "title": "W3C DID Working Group", "domain": "standards", "domain_name": "Standards", "definition": "The W3C working group responsible for developing the Decentralized Identifiers (DIDs) specifications. It is a standards-development group rather than a single document.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-did-working-group", "labels": [ "W3C DID Working Group" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c-did", "title": "W3C DID", "domain": "security", "domain_name": "Security", "definition": "W3C DID (Decentralised Identifier) is a W3C Recommendation standard (published July 2022) that defines a new type of globally unique, persistent, and cryptographically verifiable identifier that does not require a centralised registration authority. A DID resolves to a DID Document \u2014 a JSON-LD data structure containing public keys, authentication mechanisms, and service endpoints \u2014 enabling the DID subject to prove control and establish secure communication without depending on a third-party identity provider. DIDs are the foundational primitive for self-sovereign identity systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:w3-c-did", "labels": [ "W3C DID", "W3C Decentralized Identifiers" ], "is_subclass_of": [ "W3C DID Core" ], "wikilinks": [] }, { "id": "w-3-c-linked-data-platform", "title": "W3C Linked Data Platform", "domain": "standards", "domain_name": "Standards", "definition": "The W3C Linked Data Platform defines a set of rules for read-write operations on Linked Data resources using HTTP.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-linked-data-platform", "labels": [ "W3C Linked Data Platform" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-owl-2", "title": "W3C OWL 2", "domain": "standards", "domain_name": "Standards", "definition": "W3C OWL 2 is the Web Ontology Language, a knowledge representation language for authoring ontologies that describe classes, properties and individuals on the Web.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-owl-2", "labels": [ "W3C OWL 2", "OWL 2", "OWL 2 Semantics" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-prov-o", "title": "W3C PROV-O", "domain": "standards", "domain_name": "Standards", "definition": "A W3C Recommendation defining the PROV Ontology, an OWL ontology for expressing provenance information using the PROV data model. It represents entities, activities and agents involved in producing data.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-prov-o", "labels": [ "W3C PROV-O" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c-prov", "title": "W3C PROV", "domain": "data", "domain_name": "Data", "definition": "W3C PROV is a family of W3C Recommendations that define a domain-agnostic data model and serialisations for representing provenance, the records of entities, activities, and agents involved in producing or influencing a piece of data. Centred on the PROV-DM data model and PROV-O OWL ontology, it enables interoperable description of how artefacts came to be, supporting trust, reproducibility, and auditing. It is widely used in scientific workflows, data catalogues, and metadata management systems.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:w3-c-prov", "labels": [ "W3C PROV" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "w-3-c-rdf-1-1", "title": "W3C RDF 1.1", "domain": "standards", "domain_name": "Standards", "definition": "W3C RDF 1.1 is the Resource Description Framework, a data model for representing information about resources as subject-predicate-object triples on the Web.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-rdf-1-1", "labels": [ "W3C RDF 1.1" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3c-recommendation", "title": "W3C Recommendation", "domain": "security", "domain_name": "Security", "definition": "A W3C Recommendation is the final, ratified stage of a technical specification produced by the World Wide Web Consortium, signifying that the document has completed the consortium's review process and is endorsed for broad deployment. It carries the strongest standing in the W3C Recommendation Track, having passed through Working Draft, Candidate Recommendation and Proposed Recommendation maturity levels. Many web security and identity specifications, such as WebAuthn and Verifiable Credentials, are published as W3C Recommendations.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:w3c-recommendation", "labels": [ "W3C Recommendation" ], "is_subclass_of": [ "Web Standard" ], "wikilinks": [] }, { "id": "w-3-c-sparql-1-1", "title": "W3C SPARQL 1.1", "domain": "standards", "domain_name": "Standards", "definition": "A W3C Recommendation defining SPARQL 1.1, the query language and protocol for retrieving and manipulating data stored in RDF format. It includes query, update and federation features.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-sparql-1-1", "labels": [ "W3C SPARQL 1.1", "SPARQL 1.1 W3C Recommendation" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-standard", "title": "W3C Standard", "domain": "standards", "domain_name": "Standards", "definition": "A W3C Standard is a specification published as a Recommendation by the World Wide Web Consortium.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-standard", "labels": [ "W3C Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-standards", "title": "W3C Standards", "domain": "standards", "domain_name": "Standards", "definition": "A general reference to standards published by the World Wide Web Consortium (W3C). No specific standard is identified by this label alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-standards", "labels": [ "W3C Standards" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-verifiable-credentials-2-0", "title": "W3C Verifiable Credentials 2.0", "domain": "standards", "domain_name": "Standards", "definition": "A W3C specification, version 2.0, defining a data model for verifiable credentials and presentations that can be cryptographically verified. It standardises the structure of digital credentials and claims.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-verifiable-credentials-2-0", "labels": [ "W3C Verifiable Credentials 2.0", "W3C Verifiable Credential Data Model" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-verifiable-credentials-data-model-2-0", "title": "W3C Verifiable Credentials Data Model 2.0", "domain": "standards", "domain_name": "Standards", "definition": "A W3C specification, version 2.0, defining a data model for verifiable credentials that can be cryptographically verified. It describes the structure of credentials, claims and presentations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-verifiable-credentials-data-model-2-0", "labels": [ "W3C Verifiable Credentials Data Model 2.0" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-verifiable-credentials-data-model-v-2-0", "title": "W3C Verifiable Credentials Data Model v2.0", "domain": "standards", "domain_name": "Standards", "definition": "Version 2.0 of the W3C Verifiable Credentials Data Model specifies a data model for cryptographically verifiable claims exchanged between issuers, holders and verifiers.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-verifiable-credentials-data-model-v-2-0", "labels": [ "W3C Verifiable Credentials Data Model v2.0", "W3C VC 2.0 Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-verifiable-credentials-data-model", "title": "W3C Verifiable Credentials Data Model", "domain": "security", "domain_name": "Security", "definition": "The W3C Verifiable Credentials Data Model (VCDM) is a W3C Recommendation that defines a standardised, machine-readable data model for expressing and cryptographically verifying claims about subjects on the Web. It establishes a three-party trust triangle comprising an issuer that creates and signs credentials, a holder that stores and presents them, and a verifier that checks their authenticity and validity. Credentials are expressed as JSON or JSON-LD documents and may be signed using proof mechanisms such as Data Integrity Proofs or JWT/SD-JWT, enabling selective disclosure and privacy-preserving presentation. The model underpins decentralised identity systems and Self-Sovereign Identity architectures by separating identity assertion from centralised identity providers.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:w-3-c-verifiable-credentials-data-model", "labels": [ "W3C Verifiable Credentials Data Model", "Verifiable Credentials Data Model" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c-verifiable-credentials", "title": "W3C Verifiable Credentials", "domain": "security", "domain_name": "Security", "definition": "W3C Verifiable Credentials (VCs) are a standardised data model and serialisation format published by the World Wide Web Consortium that enables the cryptographic expression of credentials \u2014 such as educational qualifications, identity attributes, and professional licences \u2014 in a tamper-evident, machine-verifiable form. The standard defines three roles: issuer (creates and signs the credential), holder (stores and presents it), and verifier (validates the signature and claims), forming a trust triangle that operates without requiring a centralised credential registry. VCs are designed to interoperate with Decentralised Identifiers (DIDs) to enable self-sovereign identity systems in which individuals and organisations control their own digital identity without dependence on a single provider. The VC Data Model 2.0 became a W3C Recommendation in 2024, adding selective disclosure via SD-JWT and BBS+ signatures, JSON Schema validation, and expanded media-type support.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:w3-c-verifiable-credentials", "labels": [ "W3C Verifiable Credentials", "W3C Verifiable Credentials Data Model", "W3C Verifiable Credentials Working Group" ], "is_subclass_of": [ "Verifiable Credentials" ], "wikilinks": [] }, { "id": "w-3-c-vocabulary", "title": "W3C Vocabulary", "domain": "standards", "domain_name": "Standards", "definition": "A vocabulary published by the World Wide Web Consortium (W3C). The specific vocabulary is not determined from the identifier alone.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-vocabulary", "labels": [ "W3C Vocabulary" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c-web-speech-api", "title": "W3C Web Speech API", "domain": "metaverse", "domain_name": "Metaverse", "definition": "The W3C Web Speech API is a browser interface specification that exposes speech recognition (speech-to-text) and speech synthesis (text-to-speech) to web applications through standardised JavaScript objects. It lets pages capture spoken input and produce spoken output without bespoke plugins, underpinning voice-driven and accessibility features on the web. Implementation depth and recognition backends vary across browsers, with some delegating recognition to cloud services.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:w3-c-web-speech-api", "labels": [ "W3C Web Speech API" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "w-3-c-web-xr-device-api", "title": "W3C WebXR Device API", "domain": "standards", "domain_name": "Standards", "definition": "The W3C WebXR Device API specifies an interface for accessing virtual reality and augmented reality devices, including sensors and head-mounted displays, in Web applications.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-web-xr-device-api", "labels": [ "W3C WebXR Device API", "WebXR Device API" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w-3-c-web-xr", "title": "W3C WebXR", "domain": "standards", "domain_name": "Standards", "definition": "W3C WebXR refers to the WebXR work at W3C, which specifies interfaces for virtual reality and augmented reality experiences delivered through the Web.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:w-3-c-web-xr", "labels": [ "W3C WebXR" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "w3-c", "title": "W3C", "domain": "governance", "domain_name": "Governance", "definition": "The World Wide Web Consortium (W3C) is the principal international standards organisation for the World Wide Web, founded by Tim Berners-Lee in 1994 and operating through a global membership model to develop open, interoperable web standards. W3C produces technical specifications, guidelines, and software that define the core technologies of the web including HTML, CSS, SVG, XML, RDF, and the semantic web stack. Its standards process emphasises consensus, royalty-free licensing, and broad implementability across browsers and platforms.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:w3-c", "labels": [ "W3C", "W3C (World Wide Web Consortium)", "W3C Data Privacy Vocabulary", "W3C SHACL", "W3C Voice Browser Working Group" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "wai-aria", "title": "WAI-ARIA", "domain": "standards", "domain_name": "Standards", "definition": "WAI-ARIA (Web Accessibility Initiative Accessible Rich Internet Applications) is a W3C technical specification that defines a set of HTML attributes for roles, states, and properties, enabling assistive technologies such as screen readers to correctly interpret dynamic web content and custom user-interface widgets. It supplements native HTML semantics where markup alone cannot convey a component's purpose or current state, for example custom sliders, tab panels, and live regions. WAI-ARIA is a foundational standard underpinning modern web accessibility conformance.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:wai-aria", "labels": [ "WAI-ARIA" ], "is_subclass_of": [ "Accessibility" ], "wikilinks": [] }, { "id": "wcag-2-2", "title": "WCAG 2.2", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "WCAG 2.2 is the October 2023 release of the Web Content Accessibility Guidelines published by the W3C Web Accessibility Initiative, introducing nine new success criteria that extend the 2.1 standard with stronger provisions for keyboard navigation, focus appearance, and accessible authentication. It remains structured around four principles \u2014 Perceivable, Operable, Understandable, and Robust \u2014 with conformance levels A, AA, and AAA. The release removes the Parsing criterion (4.1.1) while tightening requirements for pointer gestures, dragging movements, and redundant input.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:wcag-2-2", "labels": [ "WCAG 2.2", "WCAG", "WCAG 2.2 Accessibility" ], "is_subclass_of": [ "Accessibility Standard" ], "wikilinks": [] }, { "id": "wco-safe-framework", "title": "WCO SAFE Framework", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The WCO SAFE Framework of Standards is a World Customs Organization instrument that sets out principles and standards to secure and facilitate global supply chains. It establishes customs-to-customs network arrangements, customs-to-business partnerships through Authorised Economic Operator (AEO) programmes, and risk-based advance electronic cargo data exchange. SAFE balances trade facilitation with security against terrorism and illicit trade.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wco-safe-framework", "labels": [ "WCO SAFE Framework" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "wd14-tagger", "title": "WD14 Tagger", "domain": "ai", "domain_name": "Ai", "definition": "WD14 Tagger is an automatic image-tagging model and tool that predicts Danbooru-style descriptive tags for images, widely used to caption training datasets for diffusion-model fine-tuning. Built on convolutional or transformer backbones trained on large tagged anime/illustration corpora, it outputs ranked tag confidences that captioning pipelines threshold and assemble into prompts. It is a standard preprocessing step in DreamBooth and LoRA training workflows.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:wd14-tagger", "labels": [ "WD14 Tagger" ], "is_subclass_of": [ "Computer Vision" ], "wikilinks": [] }, { "id": "wgsl", "title": "WGSL", "domain": "metaverse", "domain_name": "Metaverse", "definition": "WGSL (WebGPU Shading Language) is the shader programming language of the WebGPU API, used to write vertex, fragment, and compute shaders that run on the GPU from web applications. Designed for safety and portability, it maps cleanly onto native backends (Vulkan, Metal, Direct3D) while avoiding the platform-specific behaviours of older shading languages. WGSL is central to high-performance graphics and GPU compute on the modern web.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:wgsl", "labels": [ "WGSL" ], "is_subclass_of": [ "Programming Language" ], "wikilinks": [] }, { "id": "who-evm", "title": "WHO EVM", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "WHO EVM (Effective Vaccine Management) is a World Health Organization assessment framework and set of standards for evaluating and improving immunisation supply chains, with particular emphasis on cold-chain integrity, storage, and distribution. It defines criteria and scored indicators across receipt, storage, temperature monitoring, and stock management to ensure vaccine potency from manufacturer to point of use. EVM assessments guide investments in cold-chain equipment and monitoring practice.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:who-evm", "labels": [ "WHO EVM" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "wto-trade-facilitation-agreement", "title": "WTO Trade Facilitation Agreement", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "The WTO Trade Facilitation Agreement (TFA) is a multilateral treaty, in force since 2017, that sets binding measures to expedite the movement, release, and clearance of goods across borders. It mandates simplified customs procedures, advance ruling systems, electronic payment, single-window submission, and cooperation between border agencies. The TFA matters because it reduces trade costs and delays, with provisions tied to the capacity of developing-country members.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wto-trade-facilitation-agreement", "labels": [ "WTO Trade Facilitation Agreement" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "wallet-address", "title": "Wallet Address", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A wallet address is a public, shareable identifier on a blockchain network to which assets can be sent and from which ownership is asserted. It is typically derived deterministically from a public key by hashing and encoding, allowing anyone to send funds to it while only the holder of the corresponding private key can authorise outgoing transactions. Addresses are network-specific in format and serve as the destination in transactions and the anchor for on-chain balance accounting.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:wallet-address", "labels": [ "Wallet Address" ], "is_subclass_of": [ "Blockchain", "Public Key" ], "wikilinks": [] }, { "id": "wallet", "title": "Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A software or hardware interface that stores cryptographic private keys and enables users to manage cryptocurrency assets, sign transactions, and interact with blockchain networks securely. Wallets range from hot custodial services to air-gapped hardware devices, and manage key derivation, address generation, UTXO selection, and transaction broadcasting.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:wallet", "labels": [ "Wallet", "Chain Abstraction Wallet", "Cryptographic Wallet", "Ethereum Wallet", "Mobile Wallet", "Payment Wallet", "Phantom Wallet", "Self-Hosted Wallet", "Software Wallet", "Wallet Integration", "Wallet Software" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "12-word seed phrase", "2FA", "2-of-2 multisig", "2-of-3 multisig", "24-word seed phrase", "3-of-5 multisig", "5-of-7 multisig", "Account", "ACINQ", "address index", "address poisoning", "address reuse", "Address verification", "air-gapped", "Air-gapped", "Atomic Finance", "authenticator apps", "avalanche noise source", "Backup redundancy", "bech32" ] }, { "id": "walmart", "title": "Walmart", "domain": "supply-chain", "domain_name": "Supply Chain", "definition": "Walmart is a multinational retail company operating a chain of stores and e-commerce services, and an early adopter of blockchain-based supply chain traceability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:walmart", "labels": [ "Walmart" ], "is_subclass_of": [ "Supply Chain" ], "wikilinks": [ "Supply Chain", "Food Safety", "Distributed Ledger" ] }, { "id": "warehouse-automation", "title": "warehouse automation", "domain": "robotics", "domain_name": "Robotics", "definition": "Warehouse automation is the systematic deployment of robotic systems, autonomous mobile robots (AMRs), automated storage and retrieval systems (AS/RS), conveyor networks, and AI-driven software orchestration to execute goods induction, storage, picking, sorting, packing, and despatch with minimal direct human intervention. It integrates perception subsystems for item identification and collision-free navigation, motion-planning algorithms for physical task execution, and warehouse management system (WMS) integration for real-time order orchestration. Modern architectures layer machine learning for demand forecasting, adaptive task scheduling, and anomaly detection on top of heterogeneous robotic fleets, forming closed-loop feedback systems between physical material flow and digital supply-chain signals.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:warehouse-automation", "labels": [ "Warehouse Automation" ], "is_subclass_of": [ "IndustrialAutomation" ], "wikilinks": [] }, { "id": "warehouse-management-system", "title": "Warehouse Management System", "domain": "robotics", "domain_name": "Robotics", "definition": "A warehouse management system (WMS) is the software platform that orchestrates and optimises the day-to-day operations of a warehouse or distribution centre, controlling receiving, putaway, storage location, inventory tracking, order picking, packing, and despatch. It maintains a real-time digital model of stock and locations, directs labour and equipment via task assignment, and integrates with enterprise resource planning, transport, and automation layers. As the control plane for intralogistics, the WMS underpins inventory accuracy, fulfilment throughput, and coordination with robotic warehouse automation.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:warehouse-management-system", "labels": [ "Warehouse Management System" ], "is_subclass_of": [ "Inventory Management" ], "wikilinks": [] }, { "id": "warehouse-robotics", "title": "Warehouse Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Warehouse robotics is the application of autonomous mobile robots, robotic arms, and coordinated fleets to automate storage, picking, sorting, and transport tasks within fulfilment and distribution centres. Systems integrate navigation, perception, fleet orchestration, and warehouse management software to move goods and people efficiently. It is one of the most commercially mature robotics domains, driven by e-commerce throughput and labour demands.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:warehouse-robotics", "labels": [ "Warehouse Robotics" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [] }, { "id": "warm-pool", "title": "Warm Pool", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:warm-pool", "labels": [ "Warm Pool" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "warmup", "title": "Warmup", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A training technique where the learning rate starts small and gradually increases over a fixed number of initial steps to stabilise optimisation. Warmup prevents large early gradients from destabilising weight updates and is standard practice for large transformer models, typically preceding a cosine or linear decay schedule.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:warmup", "labels": [ "Warmup" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "waste-management", "title": "Waste Management", "domain": "blockchain", "domain_name": "Blockchain", "definition": "The collection, transport, processing, and disposal of waste materials, increasingly augmented by blockchain-based distributed ledgers and smart contracts that create immutable records of waste generation, segregation, and recovery throughout the circular economy lifecycle. Blockchain-enabled systems enable verifiable recycling rates, incentivise responsible disposal, and prevent illegal dumping.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:waste-management", "labels": [ "Waste Management", "Waste Hierarchy" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "BC-0188-self-sovereign-identity", "BC-0202-zero-knowledge-proofs", "BC-0319-micropayments", "BC-0440-blockchain-interoperability", "BC-0441-supply-chain-traceability", "BC-0442-certification-and-compliance", "BC-0445-conflict-mineral-tracking", "BC-0453-ethical-sourcing", "BC-0455-product-recall-management", "BC-0464-carbon-credit-tokenisation", "BC-0496-carbon-offset-verification", "BC-0502-renewable-energy-certificates", "Deposit Return Schemes", "E-Waste", "E-WasteTracking", "ExtendedProducerResponsibility", "Extended Producer Responsibility", "HazardousWaste", "Ocean Plastic", "Pay-As-You-Throw" ] }, { "id": "watchtower", "title": "Watchtower", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A watchtower is a third-party service in payment-channel networks such as the Lightning Network that monitors the blockchain on a client's behalf and reacts to fraudulent channel-closure attempts. When a counterparty broadcasts a revoked, outdated channel state, the watchtower submits a penalty (justice) transaction that claims the cheater's funds, allowing the honest party to remain offline safely. Watchtowers store encrypted justice transactions indexed by transaction hints so they learn nothing about channel contents.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:watchtower", "labels": [ "Watchtower" ], "is_subclass_of": [ "Lightning Network" ], "wikilinks": [] }, { "id": "water-column", "title": "Water Column", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:water-column", "labels": [ "Water Column" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "water-quality-monitoring", "title": "Water Quality Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:water-quality-monitoring", "labels": [ "Water Quality Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "water-table", "title": "Water Table", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:water-table", "labels": [ "Water Table" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "waterfall-model", "title": "Waterfall Model", "domain": "governance", "domain_name": "Governance", "definition": "The waterfall model is a sequential software development methodology in which progress flows downward through distinct phases such as requirements, design, implementation, verification, and maintenance. Each phase is completed and signed off before the next begins, producing extensive documentation at every gate. It offers predictability and clear milestones but assumes requirements are stable and discovered early, which limits its ability to absorb change.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:waterfall-model", "labels": [ "Waterfall Model" ], "is_subclass_of": [ "Project Management" ], "wikilinks": [] }, { "id": "watermarking-service", "title": "Watermarking Service", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Digital infrastructure that embeds imperceptible markers into media content to establish provenance, authenticate ownership, detect tampering, and identify AI-generated content, supporting content credentials standards like C2PA for verifiable digital asset tracking.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:watermarking-service", "labels": [ "Watermarking Service", "Watermarking" ], "is_subclass_of": [ "Governance and Safety", "Content Authentication" ], "wikilinks": [ "Content Authentication", "metaverse" ] }, { "id": "wave-net", "title": "WaveNet", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A deep autoregressive neural network developed by DeepMind for generating raw audio waveforms one sample at a time, using stacks of dilated causal convolutions to capture long-range temporal dependencies in audio.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wave-net", "labels": [ "WaveNet" ], "is_subclass_of": [ "Convolutional Neural Network" ], "wikilinks": [ "Convolution", "Autoregressive Model", "Text-to-Speech", "Speech Recognition", "Convolutional Neural Network" ] }, { "id": "waveguide-optics", "title": "Waveguide Optics", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Waveguide optics is the optical-engineering approach used in augmented- and mixed-reality displays to channel projected light from a microdisplay into the user's eye via total internal reflection in a thin transparent substrate. Diffractive, reflective, or holographic gratings couple light in and out, enabling compact, see-through eyewear with a wide eyebox. It is the dominant optical architecture for consumer AR glasses, trading manufacturing complexity for form factor and transparency.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:waveguide-optics", "labels": [ "Waveguide Optics", "Diffractive Waveguide", "Holographic Waveguide" ], "is_subclass_of": [ "Display and Rendering" ], "wikilinks": [] }, { "id": "wavelet-analysis", "title": "Wavelet Analysis", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:wavelet-analysis", "labels": [ "Wavelet Analysis" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "wavelet-transform", "title": "Wavelet Transform", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "The wavelet transform is a signal-processing technique that represents a signal as a sum of scaled and translated copies of a localised oscillating basis function called a wavelet. Unlike the Fourier transform, which trades all time resolution for frequency resolution, the wavelet transform provides simultaneous time and frequency localisation through multiresolution analysis. It is widely used for compression, denoising and feature extraction in machine-learning pipelines.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:wavelet-transform", "labels": [ "Wavelet Transform" ], "is_subclass_of": [ "Signal Processing" ], "wikilinks": [] }, { "id": "wayve", "title": "Wayve", "domain": "robotics", "domain_name": "Robotics", "definition": "Wayve is a British company developing self-driving technology based on end-to-end machine learning rather than hand-coded rules. It is headquartered in London.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:wayve", "labels": [ "Wayve" ], "is_subclass_of": [ "Autonomous Driving" ], "wikilinks": [ "Machine Learning", "Computer Vision", "Autonomous Vehicle", "Deep Learning", "Autonomous Driving", "https://wayve.ai", "https://wayve.ai/technology" ] }, { "id": "weak-to-strong-generalisation", "title": "Weak-to-Strong Generalisation", "domain": "ai", "domain_name": "Ai", "definition": "Weak-to-strong generalisation is an AI-alignment research paradigm investigating whether a more capable model can be reliably supervised and improved using labels or feedback from a weaker supervisor. It serves as an empirical analogue for the superalignment problem, in which humans must oversee superhuman systems they cannot fully evaluate. Findings explore how strong students recover latent capabilities beyond the noisy weak teacher's own performance.", "entityType": "Class", "qualityScore": 0.72, "maturity": "experimental", "iri": "urn:ngm:class:weak-to-strong-generalisation", "labels": [ "Weak-to-Strong Generalisation" ], "is_subclass_of": [ "AI Safety" ], "wikilinks": [] }, { "id": "wearable-ai", "title": "Wearable AI", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Wearable AI refers to body-worn devices that run or stream machine learning to sense context, recognise speech and imagery, and present information, often through glasses, earbuds or wrist devices.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:wearable-ai", "labels": [ "Wearable AI" ], "is_subclass_of": [ "Spatial Computing", "Wearable Device Integration", "Spatial Computing Domain" ], "wikilinks": [ "Edge Computing", "Augmented Reality", "Computer Vision", "Narrow AI", "Spatial Computing Domain" ] }, { "id": "wearable-computing-platform", "title": "Wearable Computing Platform", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Hardware and software ecosystems for body-worn computing devices including smartwatches, smart glasses, fitness trackers, and XR headsets, providing operating systems, development frameworks, and connectivity infrastructure for continuous personal computing and metaverse interaction.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:wearable-computing-platform", "labels": [ "Wearable Computing Platform" ], "is_subclass_of": [ "Computing and Cloud", "Computing Platform" ], "wikilinks": [ "Computing Platform", "metaverse" ] }, { "id": "wearable-computing", "title": "Wearable Computing", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Wearable computing is the design and use of computing devices worn on or close to the body, integrating sensing, processing and display into garments, accessories and head-mounted units. These devices provide continuous, context-aware interaction and data capture without occupying the hands, enabling hands-free assistance, health monitoring and immersive experiences. It is a foundational layer of spatial and ambient computing.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:wearable-computing", "labels": [ "Wearable Computing" ], "is_subclass_of": [ "Spatial Computing", "Human Interface Device" ], "wikilinks": [] }, { "id": "wearable-device-integration", "title": "Wearable Device Integration", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The technical processes and protocols for connecting wearable computing devices with metaverse platforms, enterprise systems, and cloud services, enabling seamless data exchange, cross-device synchronisation, and coordinated multi-device experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:wearable-device-integration", "labels": [ "Wearable Device Integration", "Wearable Device", "Wearable Devices" ], "is_subclass_of": [ "Interaction Technology", "System Integration" ], "wikilinks": [ "metaverse", "System Integration" ] }, { "id": "wearable-robotics", "title": "Wearable Robotics", "domain": "robotics", "domain_name": "Robotics", "definition": "Wearable robotics encompasses powered mechanical systems worn on or attached to the human body to augment, assist or restore physical capability. These devices, including exoskeletons and powered orthoses, sense the wearer's intent and motion and deliver coordinated actuation in real time to support movement, reduce effort or rehabilitate impaired function. The field integrates biomechanics, control engineering, sensing and human-robot interaction.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:wearable-robotics", "labels": [ "Wearable Robotics" ], "is_subclass_of": [ "Robotics", "Robot Type" ], "wikilinks": [] }, { "id": "weaviate", "title": "Weaviate", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Weaviate is an open-source vector database for storing objects and their vector embeddings to support semantic search and retrieval. It is developed by Weaviate B.V.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:weaviate", "labels": [ "Weaviate" ], "is_subclass_of": [ "Vector Database" ], "wikilinks": [ "Embeddings", "Semantic Search", "Retrieval-Augmented Generation", "Machine Learning", "Vector Database", "https://weaviate.io", "https://weaviate.io/developers/weaviate" ] }, { "id": "web-access-control", "title": "Web Access Control", "domain": "security", "domain_name": "Security", "definition": "Web Access Control (WAC) is a decentralised authorisation system for web resources that uses RDF-based access control lists to specify which agents may read, write, append, or control linked-data resources identified by URIs. It is a core mechanism in the Solid ecosystem, letting individuals govern access to their personal data pods using WebID-based identity. WAC decouples authorisation from any central server, aligning with self-sovereign data principles.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:web-access-control", "labels": [ "Web Access Control", "W3C Web Access Control" ], "is_subclass_of": [ "Security and Identity" ], "wikilinks": [] }, { "id": "web-api", "title": "Web Api", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A web API is an application programming interface exposed over the web using HTTP, allowing programs to request and exchange data with a remote service rather than rendering pages for humans. It defines a contract of endpoints, request and response formats, authentication and error semantics, commonly returning structured data such as JSON. Web APIs are the backbone of integration between applications, mobile clients, microservices and third-party platforms, with REST and GraphQL being two prevailing architectural styles.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:web-api", "labels": [ "Web Api", "Web API" ], "is_subclass_of": [ "API Design" ], "wikilinks": [] }, { "id": "web-application-firewall", "title": "Web Application Firewall", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A web application firewall is a security control that inspects and filters HTTP and HTTPS traffic between clients and a web application to detect and block application-layer attacks. Operating at layer seven, it applies signature, rule and behavioural policies to mitigate threats such as injection, cross-site scripting and automated abuse that traditional network firewalls cannot see. It is commonly deployed as a reverse proxy, an inline appliance or a cloud service in front of the protected application.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:web-application-firewall", "labels": [ "Web Application Firewall" ], "is_subclass_of": [ "Network Security" ], "wikilinks": [] }, { "id": "web-application", "title": "Web Application", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A web application is interactive software delivered over the web and executed primarily within a browser, combining client-side code with server-side services accessed over HTTP. It renders user interfaces, manages state, and communicates with back-end APIs to provide functionality comparable to native applications without installation. Web applications rely on open web standards for portability across devices and platforms.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:web-application", "labels": [ "Web Application" ], "is_subclass_of": [ "Infrastructure", "Software Engineering" ], "wikilinks": [] }, { "id": "web-browser", "title": "Web Browser", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A web browser is a client software application that requests, renders, and executes web content - HTML, CSS, and JavaScript - retrieved over HTTP, providing the runtime environment in which web applications and interactive tools operate. It exposes APIs for graphics rendering, networking, and local storage that higher-level tools such as browser automation frameworks and notebook interfaces depend on. Modern browsers also serve as a general-purpose application host, running everything from productivity software to real-time collaborative tools.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:web-browser", "labels": [ "Web Browser" ], "is_subclass_of": [ "Web Application" ], "wikilinks": [] }, { "id": "web-contracts", "title": "Web Contracts", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Web Contracts is a transport-agnostic smart-contract system that runs verifiable agreements over plain web files instead of a global blockchain. It separates concerns into four layers: an immutable contract.json (rules in any language), a mutable state.json (a JCS-canonicalised, SHA-256 hash-chained sequence of states), a ledger.json (multi-currency balances), and a Trail that anchors the hash chain to Bitcoin via Block Trails for tamper evidence. Contracts declare effects (credit/debit/transfer) that an executor applies atomically; any verifier can replay the state chain from genesis, check the hashes, and confirm the Bitcoin anchoring, so cheating breaks the chain detectably. Identity is did:nostr and authentication is NIP-98 signed HTTP, which lets both humans (via NIP-07) and autonomous agents participate without global consensus or gas.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:web-contracts", "labels": [ "Web Contracts", "Web Contract", "WebContracts", "webcontracts" ], "is_subclass_of": [ "Smart Contract" ], "wikilinks": [] }, { "id": "web-crawler", "title": "Web Crawler", "domain": "data", "domain_name": "Data", "definition": "A web crawler is an automated program that systematically traverses the web by following hyperlinks, fetching pages, and queuing newly discovered URLs for further retrieval. It underpins search engine indexing, SEO analysis, and large-scale dataset construction for training language models, typically respecting robots.txt directives and rate limits to avoid overloading target servers. Crawlers must handle duplicate content detection, URL normalisation, and politeness policies at scale, distinguishing them from targeted scraping of a known set of pages. Modern crawlers increasingly render JavaScript to capture content generated by single-page applications.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:web-crawler", "labels": [ "Web Crawler" ], "is_subclass_of": [ "Search Engine" ], "wikilinks": [] }, { "id": "web-dev-and-consumer-tooling", "title": "Web Dev and Consumer Tooling", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The ecosystem of frameworks, deployment platforms, low-code builders, and developer utilities used to construct and ship web applications and AI-assisted consumer products. Encompasses front-end frameworks (React Three Fiber, Streamlit, FastHTML), deployment pipelines (Vercel), and no-code/low-code AI builders enabling rapid product prototyping.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:web-dev-and-consumer-tooling", "labels": [ "Web Dev and Consumer Tooling", "WebDev and Consumer Tooling" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [ "Image Generation", "SHOULD" ] }, { "id": "web-scraping", "title": "Web Scraping", "domain": "ai", "domain_name": "Ai", "definition": "Web scraping is the automated extraction of structured data from websites by programmatically fetching HTML documents and parsing their content to capture specific fields, tables, or text. It encompasses tools ranging from simple HTTP request libraries to headless browser automation frameworks capable of executing JavaScript and interacting with dynamic single-page applications. Web scraping is widely used to construct training datasets for machine learning models, to aggregate pricing and market intelligence, and to archive public information. Its practice intersects with legal questions around terms of service compliance, copyright, and personal data protection under GDPR and equivalent regulations.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:web-scraping", "labels": [ "Web Scraping" ], "is_subclass_of": [ "Data Pipeline" ], "wikilinks": [] }, { "id": "web-services", "title": "Web Services", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Web services are software components that expose machine-to-machine functionality over a network using standardised protocols and data formats. They enable interoperable application integration across heterogeneous platforms by defining contracts (interface descriptions), message envelopes, and transport bindings independent of the implementing technology. Web services encompass both the older SOAP/WSDL stack and lightweight RESTful styles, and underpin service-oriented and microservice architectures.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:web-services", "labels": [ "Web Services" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "web-standard", "title": "Web Standard", "domain": "standards", "domain_name": "Standards", "definition": "A Web Standard is a formally published technical specification developed by recognised standards bodies\u2014primarily the W3C, WHATWG, IETF, and ECMA International\u2014that defines interoperable behaviours, data formats, APIs, or protocols for the open web. These specifications undergird every browser-rendered application by guaranteeing that HTML, CSS, JavaScript, and associated platform APIs behave consistently across vendor implementations. Web Standards are normative documents that pass through community drafting, multi-stakeholder review, and vendor implementation before reaching Recommendation or Living Standard status, ensuring that the web remains an open, royalty-free platform.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:web-standard", "labels": [ "Web Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "web-standards", "title": "Web Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Web standards are the formal specifications and protocols \u2014 including HTML, CSS, HTTP, and the WebXR Device API \u2014 that define how content and applications are described, transmitted, and rendered on the World Wide Web. Developed by consensus bodies such as the W3C, they ensure interoperability across browsers, devices, and immersive spatial computing platforms.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:web-standards", "labels": [ "Web Standards", "Semantic Web Standards", "W3C Web Standards" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "web-technology", "title": "Web Technology", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Web Technology encompasses the protocols, standards, frameworks, and distributed infrastructure that enable the authoring, delivery, and interaction of resources on the World Wide Web and its decentralised successors. It spans the foundational internet protocols (HTTP/HTTPS, DNS, TLS), client-side and server-side execution environments (browsers, JavaScript runtimes, WebAssembly), and the emerging Web3 stack (content-addressed storage, blockchain naming, decentralised identity) that shifts data sovereignty from centralised operators to individual users. Together these layers define how information is addressed, transmitted, rendered, and secured across heterogeneous networked devices at planetary scale.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-technology", "labels": [ "Web Technology" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "web-of-things", "title": "Web of Things", "domain": "standards", "domain_name": "Standards", "definition": "A W3C standardisation effort that applies web architecture, protocols, and semantics to the Internet of Things, describing physical devices through machine-readable Thing Descriptions based on JSON-LD so that heterogeneous sensors, actuators, and services can be discovered, composed, and controlled through uniform web interfaces regardless of the underlying transport protocol, providing an interoperability layer that bridges fragmented IoT ecosystems and makes device capabilities legible to software agents.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:web-of-things", "labels": [ "Web of Things" ], "is_subclass_of": [ "Web Standards" ], "wikilinks": [ "Web Standards", "Internet of Things", "Semantic Web", "World Wide Web Consortium", "Agent-to-Agent Protocol" ] }, { "id": "web-of-trust", "title": "Web of Trust", "domain": "security", "domain_name": "Security", "definition": "A web of trust is a decentralised trust model in which participants vouch for the authenticity of one another's public keys by signing them, building confidence through chains of peer endorsements rather than a central authority. Trust is transitive and weighted: a key gains credibility as more trusted parties attest to it. Originating with PGP, the model contrasts with the hierarchical certificate-authority approach of public-key infrastructure.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:web-of-trust", "labels": [ "Web of Trust" ], "is_subclass_of": [ "Trust Establishment" ], "wikilinks": [] }, { "id": "web-3-infrastructure", "title": "Web3 Infrastructure", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Web3 infrastructure is the set of protocols, nodes, storage and tooling that supports decentralised applications built on blockchain networks, encompassing peer-to-peer communication layers, decentralised storage, indexing services, cross-chain bridges, and wallet primitives.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-3-infrastructure", "labels": [ "Web3 Infrastructure" ], "is_subclass_of": [ "Blockchain", "Decentralised Web" ], "wikilinks": [ "Distributed Ledger Technology", "Decentralised Identity", "Interoperability", "Blockchain", "https://ethereum.org/en/developers/docs/", "https://en.wikipedia.org/wiki/Web3" ] }, { "id": "web3-wallet", "title": "Web3 Wallet", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Web3 wallet is a software or hardware tool that manages a user's cryptographic key material and lets them hold blockchain assets, sign transactions and authenticate to decentralised applications. Unlike a custodial account, it places control of the private key with the user, who proves ownership and authorises state changes by signing locally. Web3 wallets typically expose an injected provider or connection protocol that bridges a browser or mobile app to one or more blockchain networks.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:web3-wallet", "labels": [ "Web3 Wallet" ], "is_subclass_of": [ "Cryptocurrency Wallet" ], "wikilinks": [] }, { "id": "web3", "title": "Web3", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Web3 is a blockchain-anchored paradigm for the decentralised web, first coined by ereum co-founder Gavin Wood in his April 2014 essay \"DApps: What Web 3.0 Looks Like\", that reconceives the internet as a zero-trust interaction system eliminating the need to trust any individual institution or plat...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:web3", "labels": [ "Web3", "Web3 Ecosystem", "Web3 Protocol" ], "is_subclass_of": [ "DeFi and Economics", "Decentralised Finance", "Peer-to-Peer Network", "Blockchain Network", "Digital Asset", "Distributed Systems" ], "wikilinks": [ "Aave", "Centralised Identity", "Content Addressed Storage", "Crypto Wallet", "DeFi", "Decentralised Application", "Decentralised Finance", "Decentralised Identity", "DEX", "DigitalIdentityDomain", "Digital Sovereignty", "DistributedSystemsDomain", "EigenLayer", "EIP-4361", "EIP-7702", "ENS", "ERC-4337", "Ethereum Foundation", "Filecoin", "IETF" ] }, { "id": "web-assembly", "title": "WebAssembly", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "WebAssembly (Wasm) is a binary instruction format for a stack-based virtual machine, standardised by the W3C, that provides a portable compilation target for high-level languages such as C, C++, Rust, and Go, enabling near-native execution speed inside Web Browser sandboxes and server-side ru...", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-assembly", "labels": [ "WebAssembly", "WebAssembly 2.0 Specification" ], "is_subclass_of": [ "Computing and Cloud", "Binary Instruction Format", "Portable Execution Environment" ], "wikilinks": [ "AI Domain", "Binary Encoding Standard", "Binary Instruction Format", "Bytecode Alliance", "Component Model", "Containerd", "Cross-Language Interoperability", "Deterministic Replay", "Edge Computing Stack", "Edge Inference", "Host Runtime", "JavaScript", "Kubernetes", "Large Language Model", "LLVM Compiler Toolchain", "micro-ROS", "MobileNet", "ONNX Runtime", "PLDI 2017 WebAssembly Paper", "Portable Execution Environment" ] }, { "id": "web-authn", "title": "WebAuthn", "domain": "security", "domain_name": "Security", "definition": "WebAuthn (Web Authentication) is a W3C and FIDO Alliance standard that enables web applications to authenticate users using public-key cryptography rather than passwords, through hardware or software authenticators such as security keys, platform biometrics, and passkeys. The browser exposes the navigator.credentials API, which delegates cryptographic operations to a CTAP-compliant authenticator; the authenticator generates a key pair, stores the private key in a secure enclave, and signs authentication challenges that the relying party verifies using the registered public key. WebAuthn eliminates shared secrets from the authentication path, making phishing, credential stuffing, and replay attacks fundamentally impossible by design. It is the technical foundation of the passkey ecosystem deployed by Apple, Google, and Microsoft.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-authn", "labels": [ "WebAuthn", "W3C WebAuthn" ], "is_subclass_of": [ "Authentication Mechanism" ], "wikilinks": [] }, { "id": "web-gl", "title": "WebGL", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "WebGL (Web Graphics Library) is a royalty-free JavaScript API that exposes a subset of OpenGL ES 2.0 and 3.0 to web browsers, enabling hardware-accelerated 2D and 3D rendering directly inside an HTML canvas element without requiring browser plug-ins. It communicates directly with the GPU through the browser's graphics pipeline, exposing programmable vertex and fragment shaders written in GLSL ES. Standardised by the Khronos Group and supported natively in all major browsers, WebGL underpins interactive visualisations, browser-based games, scientific data rendering, and WebXR spatial computing experiences. Its successor API, WebGPU, offers a more modern GPU abstraction while WebGL remains the dominant, battle-tested standard for cross-platform GPU-accelerated web graphics.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:web-gl", "labels": [ "WebGL" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [ "GPU", "Real-Time Rendering", "Computer Graphics", "Graphics API" ] }, { "id": "webgpu", "title": "WebGPU", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "WebGPU is a modern web standard and API that exposes the capabilities of contemporary graphics processing units to web applications for both rendering and general-purpose computation. It provides a low-overhead, explicit interface modelled on native APIs such as Vulkan, Metal and Direct3D 12, succeeding WebGL. WebGPU enables high-performance graphics, compute shaders, and GPU-accelerated machine learning directly in the browser.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:webgpu", "labels": [ "WebGPU" ], "is_subclass_of": [ "Graphics API" ], "wikilinks": [] }, { "id": "webid-decentralised-identity-uri", "title": "WebID Decentralised Identity URI", "domain": "security", "domain_name": "Security", "definition": "A WebID is a URI that uniquely identifies an agent (person, organisation, or software) on the web and dereferences to an RDF profile document containing structured data about that agent. It underpins decentralised identity and access control across the Solid ecosystem and linked-data applications, allowing any service to authenticate and authorise users without a central identity provider.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:webid-decentralised-identity-uri", "labels": [ "WebID Decentralised Identity URI", "webid" ], "is_subclass_of": [ "Network Component" ], "wikilinks": [ "Solid" ] }, { "id": "web-id-profile", "title": "WebID Profile", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A WebID Profile is an RDF document accessible at a dereferenceable HTTP URI that describes a person or agent, linking their identity to cryptographic keys, social contacts, and access control preferences, thereby enabling decentralised authentication and authorisation on the web without a centralised identity provider. The profile uses vocabularies such as FOAF and vCard to express identity attributes, and the WebID-TLS and WebID-OIDC protocols use it to authenticate agents by verifying control of the URI through certificate or token proofs. WebID Profiles are a foundational component of the Solid decentralised web platform, where they serve as the entry point for discovering a user's data pods and access control rules. They embody the self-sovereign principle that identity should be controlled by the individual rather than delegated to a platform.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:web-id-profile", "labels": [ "WebID Profile" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [] }, { "id": "web-id", "title": "WebID", "domain": "security", "domain_name": "Security", "definition": "WebID is a decentralised identity mechanism in which a person or agent is identified by an HTTP(S) URI that dereferences to a machine-readable profile document describing them in RDF. Combined with authentication methods such as WebID-TLS or WebID-OIDC, it lets users prove control of their identifier without a central identity provider. WebID is a foundational building block of the Solid project and the broader linked-data web, enabling user-controlled identity, profiles, and access control across decentralised applications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "emerging", "iri": "urn:ngm:class:web-id", "labels": [ "WebID" ], "is_subclass_of": [ "Decentralized Identity" ], "wikilinks": [] }, { "id": "web-rtc-w-3-c-specification", "title": "WebRTC W3C Specification", "domain": "standards", "domain_name": "Standards", "definition": "A W3C specification defining WebRTC, an API enabling real-time peer-to-peer audio, video and data communication between browsers. It defines the JavaScript interfaces for media and data transport.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:web-rtc-w-3-c-specification", "labels": [ "WebRTC W3C Specification", "W3C WebRTC 1.0 Specification", "WebRTC W3C Standard" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [ "W3C", "Technical Standard" ] }, { "id": "web-rtc", "title": "WebRTC", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "WebRTC (Web Real-Time Communication) is a W3C and IETF co-standardised open framework that enables peer-to-peer exchange of audio, video, and arbitrary data between web browsers and native applications using a JavaScript API (getUserMedia, RTCPeerConnection, RTCDataChannel), combining", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:web-rtc", "labels": [ "WebRTC", "IETF WebRTC", "TELE-150-webrtc", "WebRTC Simulcast", "WebRTC Standard" ], "is_subclass_of": [ "Protocol and Infrastructure", "Real-Time Communication Protocol", "Peer-to-Peer Network Protocol" ], "wikilinks": [ "Collaborative XR", "DTLS-SRTP", "DTLS-SRTP Encryption", "DTLS-SRTP RFC 5764", "getUserMedia API", "ICE Protocol", "ICE Protocol", "IETF RFC 7478 RTCWEB Overview", "IETF RTCWEB Standards", "libwebrtc Library", "O'Reilly WebRTC Book", "Opus Audio Codec", "Peer-to-Peer Network Protocol", "Peer-to-Peer Video Conferencing", "Real-Time Communication Protocol", "Real-Time Data Transfer", "RTCDataChannel", "RTCPeerConnection", "SDP Offer-Answer", "SDP Session Description" ] }, { "id": "web-socket-protocol", "title": "WebSocket Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The WebSocket Protocol is a standardised full-duplex communication protocol defined in RFC 6455 (2011) that provides a persistent, low-latency bidirectional channel between a client and a server over a single TCP connection. It was designed to overcome the limitations of HTTP polling and long-polling by upgrading an initial HTTP handshake to a persistent framed message channel, enabling servers to push data to clients without client-initiated requests. WebSocket frames carry minimal overhead \u2014 a 2-byte header for small messages \u2014 making the protocol suitable for high-frequency data streams such as financial tickers, collaborative editing, gaming, and real-time AI agent communications. It is supported natively in all major browsers and server-side runtimes, and forms the transport layer for many higher-level protocols including STOMP, MQTT-over-WebSocket, and the OpenAI Realtime API.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-socket-protocol", "labels": [ "WebSocket Protocol", "WebSocket RFC 6455" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "web-socket", "title": "websocket", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "WebSocket is an application-layer communication protocol defined in RFC 6455 that establishes a persistent, full-duplex channel over a single TCP connection, initiated via an HTTP/1.1 upgrade handshake. Unlike the request-response model of HTTP, WebSocket permits the server and client to send data frames independently at any time after connection establishment, enabling low-latency bidirectional communication. The protocol specifies a lightweight framing mechanism with opcodes for text, binary, ping/pong keepalives, and graceful connection close, and is extended by RFC 7692 for per-message DEFLATE compression. It is the de facto standard transport for real-time web applications including collaborative editing, live dashboards, chat systems, and streaming LLM token output.", "entityType": "Class", "qualityScore": 0.72, "maturity": "mature", "iri": "urn:ngm:class:web-socket", "labels": [ "WebSocket", "WebSocket API", "WebSocket Support" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [] }, { "id": "web-sockets", "title": "WebSockets", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "WebSockets is a protocol providing full-duplex communication channels over a single TCP connection between a client and a server. It is widely used for real-time web applications.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:web-sockets", "labels": [ "WebSockets" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "HTTP", "Network Communication", "Web Technology", "Communication Protocol", "https://datatracker.ietf.org/doc/html/rfc6455", "https://developer.mozilla.org/en-US/docs/Web/API/WebSockets_API" ] }, { "id": "web-xr-api", "title": "WebXR API", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A W3C web standard providing browser-native interfaces for rendering stereoscopic 3D content and handling spatial input from XR headsets and controllers, enabling AR/VR experiences without native application installation. WebXR supersedes the earlier WebVR specification and is implemented across Chromium-based browsers and Firefox.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:web-xr-api", "labels": [ "WebXR API" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "EWG/MSF taxonomy", "MetaverseDomain" ] }, { "id": "web-xr", "title": "WebXR", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "WebXR is a W3C web standard and browser API that enables access to virtual reality and augmented reality hardware, exposing device pose tracking, controller input, reference spaces, and an XR-integrated render loop so that immersive spatial experiences can be delivered through ordinary URLs without native installation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:web-xr", "labels": [ "WebXR" ], "is_subclass_of": [ "Spatial Computing", "Standards and Interoperability", "Spatial Computing Domain" ], "wikilinks": [ "Web Standard", "Graphics API", "Virtual Reality", "Augmented Reality", "Spatial Computing", "Spatial Computing Domain" ] }, { "id": "webhook", "title": "Webhook", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A webhook is an event-driven integration mechanism in which a server sends an HTTP request to a pre-registered URL when a specified event occurs, pushing data to consumers instead of requiring them to poll. It enables loosely coupled, near-real-time communication between web services and is a standard pattern for notifications, CI triggers, and payment events. Reliability concerns are addressed through retries, idempotency keys, and signature verification.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:webhook", "labels": [ "Webhook", "Webhook Automation" ], "is_subclass_of": [ "Network and Communication" ], "wikilinks": [] }, { "id": "webinar-broadcast", "title": "Webinar Broadcast", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A webinar broadcast is a one-to-many online event in which one or more presenters stream audio and video to a large remote audience with limited bidirectional interaction. Participants typically engage through moderated Q&A, polls, and chat rather than direct audio or video contribution. Webinars are used for training, product launches, and large-scale knowledge dissemination across geographically distributed audiences.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:webinar-broadcast", "labels": [ "Webinar Broadcast" ], "is_subclass_of": [ "Communication Technology" ], "wikilinks": [] }, { "id": "weight-decay", "title": "Weight Decay", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A regularisation technique that adds a penalty proportional to the L2 norm of model weights to the loss function, discouraging large weight magnitudes and thereby limiting model complexity. Weight decay prevents overfitting, promotes simpler generalisable solutions, and in optimisers such as AdamW is implemented directly in the weight update step rather than via loss augmentation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:weight-decay", "labels": [ "Weight Decay" ], "is_subclass_of": [ "AI Technique", "Machine Learning Technique" ], "wikilinks": [ "ArtificialIntelligenceDomain" ] }, { "id": "weight-initialisation", "title": "Weight Initialisation", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Weight initialisation is the procedure of assigning starting values to the trainable parameters of a neural network before training commences. The choice of initialisation scheme affects gradient flow, convergence speed, and the avoidance of vanishing or exploding activations across deep layers. Common schemes such as Xavier (Glorot) and He initialisation scale the variance of initial weights according to layer fan-in and fan-out.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:weight-initialisation", "labels": [ "Weight Initialisation" ], "is_subclass_of": [ "Neural Network Training" ], "wikilinks": [] }, { "id": "weight-matrix", "title": "Weight Matrix", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A weight matrix is a two-dimensional array of learnable parameters that defines the linear transformation applied between two layers of a neural network. Each element encodes the strength of the connection between an input unit and an output unit, and the matrix is multiplied with the input activation vector to produce the pre-activation output. Weight matrices are initialised, then iteratively updated during training via gradient-based optimisation to minimise a loss function.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:weight-matrix", "labels": [ "Weight Matrix" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "weight-sharing", "title": "Weight Sharing", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Weight sharing is a neural-network design technique in which the same set of learnable parameters is reused across multiple positions, time steps, or model components rather than learning independent parameters for each. By tying parameters together it dramatically reduces model size, encodes structural priors such as translation invariance, and improves data efficiency and generalisation. It is the defining mechanism of convolutional layers, which apply one filter across all spatial locations, and of recurrent networks, which reuse the same transition weights across every time step. Weight sharing also appears in Siamese architectures, in neural architecture search, and in parameter-efficient model designs.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:weight-sharing", "labels": [ "Weight Sharing" ], "is_subclass_of": [ "Neural Network" ], "wikilinks": [] }, { "id": "weight-update", "title": "Weight Update", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "A weight update is the adjustment applied to a neural network's parameters after each training step, computed by scaling the gradient produced by backpropagation by a learning rate and subtracting it from the current weights. Successive weight updates across many epochs move the network toward a configuration that minimises the training loss. The update rule \u2014 plain gradient descent, momentum, or an adaptive optimiser such as Adam \u2014 determines convergence speed and stability.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:weight-update", "labels": [ "Weight Update" ], "is_subclass_of": [ "Backpropagation" ], "wikilinks": [] }, { "id": "weighted-graph", "title": "Weighted Graph", "domain": "data", "domain_name": "Data", "definition": "A weighted graph is a graph in which each edge carries a numeric weight representing cost, distance, capacity or another quantity relevant to the traversal or connection it models. Weights generalise simple adjacency into a richer structure that supports shortest-path, minimum-spanning-tree and flow algorithms. Search algorithms such as A* use edge weights together with a heuristic to find least-cost paths efficiently.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:weighted-graph", "labels": [ "Weighted Graph" ], "is_subclass_of": [ "Graph Theory" ], "wikilinks": [] }, { "id": "weights-and-biases", "title": "Weights and Biases", "domain": "ai", "domain_name": "Ai", "definition": "Weights & Biases (W&B) is a machine-learning experiment-tracking and MLOps platform that logs metrics, hyperparameters, model checkpoints, datasets, and system telemetry to enable reproducible and comparable training runs. It provides dashboards, artifact versioning, hyperparameter sweeps, and model-registry features that integrate with common training frameworks. W&B is widely adopted for managing and visualising the lifecycle of deep-learning experiments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:weights-and-biases", "labels": [ "Weights and Biases" ], "is_subclass_of": [ "AI Infrastructure" ], "wikilinks": [] }, { "id": "welfare-economics", "title": "Welfare Economics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Welfare economics is the branch of economics that evaluates the allocation of resources and the distribution of outcomes in terms of aggregate social well-being. It formalises notions of efficiency, principally Pareto efficiency, alongside frameworks for comparing distributions through social welfare functions, and provides the normative basis for assessing market outcomes, market failures, and policy interventions. In tokenised and blockchain economies it informs mechanism design, public-goods funding, and the analysis of incentive structures that determine whether decentralised systems produce socially desirable equilibria.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:welfare-economics", "labels": [ "Welfare Economics" ], "is_subclass_of": [ "Microeconomics" ], "wikilinks": [] }, { "id": "well-being", "title": "Well Being", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "AI should enhance human and societal well-being by augmenting human capabilities, enriching quality of life, supporting physical and mental health, enabling meaningful work, strengthening social connections, and contributing to flourishing individuals and communities.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:well-being", "labels": [ "Well Being", "User Well-being", "Well-Being" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Cook1977; @Kleinke1986; @Fagel2010", "Kendon1967", "Kleinke1986; @Nguyen2009", "Otsuka2005", "torok2017cascading", "Metaverse and Telecollaboration", "MetaverseDomain" ] }, { "id": "wheel-odometry", "title": "Wheel Odometry", "domain": "robotics", "domain_name": "Robotics", "definition": "Wheel odometry is a method of estimating a mobile robot's change in position and orientation by counting wheel rotations measured with encoders and applying a kinematic motion model. As an instance of dead reckoning, it integrates incremental wheel displacement over time to track pose relative to a starting point. It is simple and low-cost but accumulates drift from wheel slip, uneven terrain and calibration error, so it is typically fused with other sensors.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:wheel-odometry", "labels": [ "Wheel Odometry" ], "is_subclass_of": [ "Odometry" ], "wikilinks": [] }, { "id": "wheeled-mobile-robot", "title": "Wheeled Mobile Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A mobile robot that uses wheels as its primary locomotion mechanism, enabling efficient navigation on flat or structured surfaces. Wheeled mobile robots span applications from industrial logistics and warehouse automation to outdoor infrastructure maintenance, typically combining SLAM-based navigation, sensor fusion, and modular payloads.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:wheeled-mobile-robot", "labels": [ "Wheeled Mobile Robot" ], "is_subclass_of": [ "Robot Type", "Robotics", "Mobile Robot" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "wheeled-robot", "title": "Wheeled Robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A Wheeled Robot is a mobile robot platform that uses wheels as its primary locomotion mechanism. Wheeled robots offer high energy efficiency and speed on flat terrain, making them dominant in warehouse automation, last-mile delivery, and research platforms. Differential drive, omnidirectional, and car-like Ackermann steering configurations each present distinct kinematic constraints that influence navigation algorithm design.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:wheeled-robot", "labels": [ "Wheeled Robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "owl:Thing" ] }, { "id": "whiskbroom-scanner", "title": "Whiskbroom Scanner", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:whiskbroom-scanner", "labels": [ "Whiskbroom Scanner" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "whisper", "title": "Whisper", "domain": "ai", "domain_name": "Ai", "definition": "Whisper is an automatic speech recognition model from OpenAI trained on a large multilingual dataset. It transcribes and translates speech across many languages and is released as open source.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:whisper", "labels": [ "Whisper", "Whisper ASR", "Whisper STT" ], "is_subclass_of": [ "Automatic Speech Recognition" ], "wikilinks": [ "Transformer", "Attention Mechanism", "Speech Recognition", "Translation", "Speech Processing", "OpenAI", "Automatic Speech Recognition" ] }, { "id": "whistleblower-protection", "title": "Whistleblower Protection", "domain": "governance", "domain_name": "Governance", "definition": "Whistleblower protection comprises legal and procedural safeguards that shield individuals who report wrongdoing, safety risks, or legal violations from retaliation such as dismissal, demotion, or harassment. In the AI-governance context, it empowers employees of AI developers to disclose information about unmitigated risks from frontier models to regulators or the public. Such provisions appear in legislation like California's frontier-AI safety bills, creating accountability channels beyond internal controls.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:whistleblower-protection", "labels": [ "Whistleblower Protection", "Whistleblower Mechanism" ], "is_subclass_of": [ "AI Governance and Ethics" ], "wikilinks": [] }, { "id": "white-dwarf", "title": "White Dwarf", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:white-dwarf", "labels": [ "White Dwarf" ], "is_subclass_of": [ "Stellar Remnant" ], "wikilinks": [] }, { "id": "whole-body-control", "title": "whole body control", "domain": "robotics", "domain_name": "Robotics", "definition": "Whole Body Control (WBC) is a control framework for legged and humanoid robots that simultaneously optimises motion tasks, contact forces, and balance constraints across all degrees of freedom by solving a hierarchical quadratic programme or weighted task-space objective at each control cycle. By treating locomotion, manipulation, and postural balance as a unified optimisation problem, WBC avoids the sub-optimality of treating these objectives separately. It typically relies on a rigid-body dynamics model and is often combined with model predictive control or reinforcement learning policies for online adaptation to uneven terrain and external disturbances.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:whole-body-control", "labels": [ "Whole Body Control", "Whole-Body Control", "Whole-Body Controller" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [] }, { "id": "wholesale-cbdc", "title": "Wholesale CBDC", "domain": "finance", "domain_name": "Finance", "definition": "A Wholesale Central Bank Digital Currency (wCBDC) is a form of central bank money issued in digital form and restricted to use by financial institutions\u2014commercial banks, clearing houses, and other regulated entities\u2014for large-value interbank settlement and financial market infrastructure operations, as distinct from retail CBDC which is available to the general public. Wholesale CBDCs are designed to modernise payment system infrastructure by enabling atomic delivery-versus-payment settlement, programmable payment conditions through smart contracts, and 24/7 settlement finality, addressing inefficiencies in legacy correspondent banking and real-time gross settlement systems. Multiple central banks are actively piloting wholesale CBDC, including Project Jura (BIS, Banque de France, SNB), Project Dunbar (BIS, MAS, SARB, RBA, BNM), and the Bank of England's New Payments Architecture.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wholesale-cbdc", "labels": [ "Wholesale CBDC", "Wholesale CBDC Settlement" ], "is_subclass_of": [ "Central Bank Digital Currency" ], "wikilinks": [] }, { "id": "wholesale-power-auction", "title": "Wholesale Power Auction", "domain": "economics", "domain_name": "Economics", "definition": "A market mechanism where electricity generators and large consumers bid to secure long-term contracts for power capacity at competitive prices.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:wholesale-power-auction", "labels": [ "Wholesale Power Auction" ], "is_subclass_of": [ "Energy Policy" ], "wikilinks": [] }, { "id": "wi-fi", "title": "Wi-Fi", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Wi-Fi is a family of wireless networking technologies based on the IEEE 802.11 standards that allow devices to connect to a local area network and the internet without physical cabling. It is governed and certified by the Wi-Fi Alliance and operates primarily in the 2.4 GHz, 5 GHz, and 6 GHz radio bands.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wi-fi", "labels": [ "Wi-Fi" ], "is_subclass_of": [ "Network Protocol" ], "wikilinks": [ "Internet of Things", "Network Communication", "Communication Protocol", "Network Protocol", "https://www.wi-fi.org", "https://standards.ieee.org/ieee/802.11/" ] }, { "id": "wide-area-network", "title": "Wide Area Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A wide area network (WAN) is a telecommunications network that interconnects sites across large geographic distances \u2014 cities, countries, or continents \u2014 typically by carrying traffic over links leased from or operated by carriers rather than infrastructure the user owns end to end. WANs join local area networks into a single reachable whole using technologies ranging from leased lines, MPLS, and carrier Ethernet to broadband internet, cellular, and satellite, with the internet itself the largest example, and they trade the high bandwidth and low latency of the LAN for reach, at recurring circuit cost.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:wide-area-network", "labels": [ "Wide Area Network" ], "is_subclass_of": [ "Networking Infrastructure" ], "wikilinks": [ "Networking Infrastructure", "Local Area Network", "Software-Defined Networking", "Virtual Private Network", "Telecommunications" ] }, { "id": "wiener-process", "title": "Wiener Process", "domain": "data", "domain_name": "Data", "definition": "The Wiener process, also called standard Brownian motion, is a continuous-time stochastic process characterised by independent, stationary, normally distributed increments starting at zero, with continuous but nowhere-differentiable sample paths. It is the canonical mathematical model of random continuous motion and the foundational driver of stochastic calculus and diffusion equations. The Wiener process underpins models in finance, physics, and the noise schedules of diffusion generative models.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wiener-process", "labels": [ "Wiener Process" ], "is_subclass_of": [ "Optimisation" ], "wikilinks": [] }, { "id": "wikidata", "title": "Wikidata", "domain": "data", "domain_name": "Data", "definition": "Wikidata is a free, collaboratively edited knowledge base operated by the Wikimedia Foundation that stores structured data as machine-readable items and statements, serving as a central data repository for Wikipedia and the broader web. Each item has a stable identifier and is described by property-value statements with references and qualifiers, exported as linked data and queryable via SPARQL. Multilingual by design and released under a public-domain licence, Wikidata is one of the largest open knowledge graphs and a key hub in the linked-data web.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:wikidata", "labels": [ "Wikidata" ], "is_subclass_of": [ "Knowledge Graph" ], "wikilinks": [] }, { "id": "wildfire-monitoring", "title": "Wildfire Monitoring", "domain": "earth-observation-and-geospatial-sensing", "domain_name": "Earth Observation And Geospatial Sensing", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:wildfire-monitoring", "labels": [ "Wildfire Monitoring" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "wireless-communication", "title": "Wireless Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Wireless Communication is the transfer of information between two or more points without a physical conductor, using electromagnetic waves such as radio, microwave or infrared. It encompasses the modulation, transmission, propagation and reception of signals across shared spectrum, governed by protocols that manage access, error control and interference. Wireless systems underpin mobile, satellite and short-range networks that connect devices, infrastructure and people.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:wireless-communication", "labels": [ "Wireless Communication" ], "is_subclass_of": [ "Telecommunications" ], "wikilinks": [] }, { "id": "wireless-connectivity", "title": "Wireless Connectivity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Wireless connectivity is the transmission of data between devices over radio frequencies without physical cabling. It encompasses short-range links such as Wi-Fi and Bluetooth, medium-range technologies such as Zigbee and LoRa, and wide-area mobile networks such as 4G LTE and 5G NR, all governed by spectrum allocation and standardised protocol stacks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wireless-connectivity", "labels": [ "Wireless Connectivity", "Wireless Access Point", "Wireless Local Area Network" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [ "Communication Protocol", "Real-Time Communication", "Internet of Things", "5G" ] }, { "id": "wireless-network", "title": "Wireless Network", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A communication network in which nodes exchange data over radio, infrared, or other electromagnetic links rather than physical cabling, spanning technologies from short-range personal-area protocols such as Bluetooth and Zigbee, through Wi-Fi local-area networks, to cellular and satellite systems covering entire regions. Wireless networks depend on regulated spectrum allocation to avoid interference, employ modulation, coding, and medium-access schemes to share the channel, and trade bandwidth, range, power consumption, and mobility against one another. They are the connective substrate for mobile computing, smart-home devices, mesh deployments, and the Internet of Things.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:wireless-network", "labels": [ "Wireless Network" ], "is_subclass_of": [ "Networking" ], "wikilinks": [ "Networking", "Mesh Network", "Spectrum Allocation", "Smart Home" ] }, { "id": "wireless-radio", "title": "Wireless Radio", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Wireless radio refers to the technology and systems that transmit and receive information using electromagnetic waves propagated through free space without physical conductors, encompassing the hardware, protocols, modulation schemes, and spectrum management practices that enable wireless communication across a range of frequencies from kilohertz to millimetre-wave bands. It is the physical-layer foundation of all wireless networking standards including cellular, Wi-Fi, Bluetooth, and satellite communications.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:wireless-radio", "labels": [ "Wireless Radio" ], "is_subclass_of": [ "Telecommunications" ], "wikilinks": [] }, { "id": "wireless-telemetry-module", "title": "Wireless Telemetry Module", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A wireless telemetry module is an embedded hardware subsystem that acquires sensor measurements and transmits them over a radio link to a remote receiver without a wired connection. It combines signal conditioning, analogue-to-digital conversion, a microcontroller, and a low-power radio, and is used where wired tethering is impractical, such as implanted devices or remote field sensors. Power efficiency, antenna design, and link reliability are its central engineering constraints.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wireless-telemetry-module", "labels": [ "Wireless Telemetry Module" ], "is_subclass_of": [ "Sensor" ], "wikilinks": [] }, { "id": "wireless-value-realization", "title": "Wireless Value Realization", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Wireless Value Realization is the strategic process of identifying, measuring, and maximising the business and operational benefits derived from wireless technologies \u2014 including 5G, Wi-Fi 6/7, Bluetooth, and LPWAN \u2014 by aligning connectivity investments with productivity, efficiency, and innovation outcomes. It encompasses analytics-driven optimisation of wireless data flows, integration with edge computing and IoT sensor networks, and governance frameworks that translate connectivity capability into quantifiable organisational value. The concept moves beyond infrastructure provision to treat wireless as an active enabler of digital transformation.", "entityType": "Class", "qualityScore": 0.75, "maturity": "emerging", "iri": "urn:ngm:class:wireless-value-realization", "labels": [ "Wireless Value Realization" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "wireless-hart", "title": "WirelessHART", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "WirelessHART is an open wireless communication standard for industrial process automation, built on the IEEE 802.15.4 physical layer and using time-synchronised, self-organising mesh networking with channel hopping for reliability. It extends the HART field-device protocol to wireless sensor networks in plants, providing deterministic, secure, low-power monitoring and control. Centrally managed by a network manager, it prioritises robustness against interference in harsh RF environments.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wireless-hart", "labels": [ "WirelessHART", "IEC 62591 WirelessHART" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "witness-data", "title": "Witness Data", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Witness data is the portion of a Bitcoin transaction that contains the signatures and scripts proving authorisation to spend inputs, separated from the core transaction body by Segregated Witness. Moving this data into a distinct structure fixes transaction malleability and allows witness bytes to be discounted when computing block weight. Witness data is also the field where inscriptions such as Ordinals embed arbitrary content.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:witness-data", "labels": [ "Witness Data" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [] }, { "id": "word-embedding", "title": "Word Embedding", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A Word Embedding is a dense, continuous vector representation of a word learned such that semantically or syntactically similar words occupy nearby positions in the vector space. Trained from large text corpora using distributional statistics, embeddings capture relationships through geometric structure, enabling analogical reasoning and similarity computation. They transformed natural language processing by replacing sparse one-hot encodings with low-dimensional features that generalise across vocabulary and feed downstream neural models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:word-embedding", "labels": [ "Word Embedding" ], "is_subclass_of": [ "Embedding" ], "wikilinks": [] }, { "id": "word-embeddings", "title": "Word Embeddings", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Word embeddings are dense, real-valued vector representations of words learned so that semantic and syntactic relationships are reflected as geometric relationships in a continuous vector space. Words with similar meanings map to nearby points, and linear offsets often capture analogical structure. They are learned from large text corpora by predicting words from their contexts, and replaced sparse one-hot encodings as the default input representation for natural language processing. Word embeddings underpin downstream tasks from classification to machine translation and serve as the conceptual precursor to contextual representations produced by transformer models.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:word-embeddings", "labels": [ "Word Embeddings" ], "is_subclass_of": [ "Representation Learning" ], "wikilinks": [] }, { "id": "word-piece", "title": "WordPiece", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A subword tokenisation algorithm that iteratively merges character sequences by maximising the likelihood of the training corpus under a unigram language model, rather than merging the most frequent pairs. WordPiece is the default tokeniser for BERT, DistilBERT, and ELECTRA, producing vocabularies of approximately 30,000 tokens that handle rare and compound words via subword splitting.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:word-piece", "labels": [ "WordPiece" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "workflow-automation", "title": "workflow automation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Workflow automation is the systematic encoding of business processes, data-transformation pipelines, or task sequences into executable software so that they proceed reliably and repeatedly with minimal human intervention, using rule-based triggers, conditional branching, state machines, and event-driven messaging. It spans a spectrum from deterministic rule engines and robotic process automation (RPA), which replicate structured human UI interactions, to agentic AI systems in which large language models serve as planning kernels that dynamically compose multi-step tool-call sequences in response to high-level goals. Architecturally, workflow automation systems couple an orchestration layer \u2014 responsible for sequencing, error recovery, and state persistence \u2014 with an integration layer that provides connectors to APIs, databases, messaging queues, and human-approval interfaces. By reducing operational latency, enforcing process compliance, and enabling organisations to scale knowledge-work capacity, workflow automation has become a foundational capability across enterprise IT, scientific data pipelines, and AI agent infrastructure.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:workflow-automation", "labels": [ "Workflow Automation", "Workflow Automation Engine" ], "is_subclass_of": [ "AI Application" ], "wikilinks": [] }, { "id": "workflow-engine", "title": "Workflow Engine", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A workflow engine is a software system that executes, coordinates, and monitors multi-step processes defined as a sequence or graph of tasks, managing state transitions, branching, retries, and human or system handoffs. It separates process definition from execution, enabling durable, observable orchestration of long-running business or computational workflows. Workflow engines underpin enterprise process automation and increasingly the orchestration of AI agent pipelines.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:workflow-engine", "labels": [ "Workflow Engine", "Classical Workflow Engine", "Workflow Builder" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "workflow-orchestration", "title": "Workflow Orchestration", "domain": "ai", "domain_name": "Ai", "definition": "Workflow orchestration is the coordination of multiple interdependent tasks, services, or agents into a coherent end-to-end process, managing ordering, data flow, conditional branching, error handling, and resource allocation. It centralises control logic so that distributed components execute in the correct sequence with the right inputs. In agentic AI systems, orchestration governs how tool calls, sub-agents, and model invocations are sequenced to accomplish complex goals.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:workflow-orchestration", "labels": [ "Workflow Orchestration", "Low-Code Workflow Orchestration" ], "is_subclass_of": [ "AI Agent System" ], "wikilinks": [] }, { "id": "workforce-augmentation", "title": "Workforce Augmentation", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Workforce Augmentation refers to the use of AI tools, agentic systems, and intelligent automation to extend the cognitive and physical capabilities of human workers rather than replacing them. By offloading repetitive tasks to AI assistants and copilots, organisations can redeploy human attention to higher-value judgement and creative work, improving both productivity and job quality.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:workforce-augmentation", "labels": [ "Workforce Augmentation" ], "is_subclass_of": [ "AI Application", "Artificial Intelligence" ], "wikilinks": [ "Artificial Intelligence" ] }, { "id": "workforce-development", "title": "Workforce Development", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Workforce development encompasses structured programmes, training environments, and skills-acquisition systems that equip individuals and organisations with the competencies required by evolving labour markets. In spatial computing contexts, it leverages immersive learning, simulation, and augmented connected workforce platforms to deliver scalable, experiential upskilling.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:workforce-development", "labels": [ "Workforce Development", "Skills Development" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "owl:Thing" ] }, { "id": "workforce-management", "title": "Workforce Management", "domain": "ai", "domain_name": "Ai", "definition": "Workforce Management (WFM) encompasses the integrated set of processes, technologies, and practices that organisations use to optimise employee productivity, schedule labour resources, track attendance and time, manage compliance with labour regulations, and forecast staffing requirements. Modern WFM systems combine real-time data on demand, skill availability, and operational constraints to generate optimised schedules, and increasingly incorporate AI-driven forecasting and adaptive scheduling algorithms. The field is undergoing significant transformation as AI automation shifts skill requirements, gig economy models create more fluid workforce composition, and remote and hybrid working patterns demand new coordination approaches.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:workforce-management", "labels": [ "Workforce Management" ], "is_subclass_of": [ "Workforce Development" ], "wikilinks": [] }, { "id": "working-group", "title": "Working Group", "domain": "standards", "domain_name": "Standards", "definition": "A working group is a chartered subgroup of a standards organisation or collaborative body tasked with developing, reviewing and reaching consensus on a specific technical topic or deliverable. It brings together domain experts who draft specifications, resolve issues and progress documents through review and approval stages. Working groups are the principal unit of work through which formal standards and specifications are produced.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:working-group", "labels": [ "Working Group" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "working-memory", "title": "Working Memory", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Working memory is the short-lived, actively maintained store an agent uses to hold and manipulate the information relevant to its current task. In large-language-model agents it is realised through the context window, intermediate reasoning traces, and scratchpads, augmented by external stores when the active context is insufficient. It is distinguished from long-term memory by its limited capacity, volatility, and tight coupling to ongoing reasoning.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:working-memory", "labels": [ "Working Memory" ], "is_subclass_of": [ "Cognitive Architecture", "AI Research Area" ], "wikilinks": [] }, { "id": "workspace-analysis", "title": "Workspace Analysis", "domain": "robotics", "domain_name": "Robotics", "definition": "Workspace analysis determines the set of poses a robot's end-effector can reach, characterising the reachable and dexterous workspaces of a manipulator. It uses forward kinematics and joint-limit constraints to map the volume and shape of attainable positions and orientations. Workspace analysis guides robot selection, cell layout, task placement and reachability assessment for a given manipulator.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:workspace-analysis", "labels": [ "Workspace Analysis" ], "is_subclass_of": [ "Kinematics" ], "wikilinks": [] }, { "id": "workspace-templates", "title": "Workspace Templates", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Workspace Templates are pre-configured layouts, page structures, or project scaffolds that teams can instantiate to quickly start a new collaboration session with consistent conventions. They encode best practices for recurring activities such as sprint planning, meeting notes, or design reviews, reducing setup time for distributed teams. Templates are a standard feature in platforms such as Notion, Confluence, and Miro.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:workspace-templates", "labels": [ "Workspace Templates" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "world-building", "title": "World Building", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The creative and technical process of designing, constructing, and populating coherent virtual environments \u2014 including geography, architecture, atmosphere, lore, and rules \u2014 for use in games, virtual reality, film, and metaverse platforms. World building integrates procedural generation, 3D modelling, level design, environmental storytelling, and real-time rendering to produce navigable, persistent virtual worlds.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:world-building", "labels": [ "World Building" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [] }, { "id": "world-economic-forum", "title": "World Economic Forum", "domain": "governance", "domain_name": "Governance", "definition": "The World Economic Forum is an international organisation based in Switzerland that convenes leaders from business, government, and civil society to discuss global issues. It is best known for its annual meeting in Davos.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:world-economic-forum", "labels": [ "World Economic Forum" ], "is_subclass_of": [ "Legal and Regulatory", "owl:Thing" ], "wikilinks": [ "Governance", "Climate Finance", "owl:Thing", "https://www.weforum.org", "https://www.weforum.org/about/world-economic-forum/" ] }, { "id": "world-inequality-database", "title": "World Inequality Database", "domain": "data", "domain_name": "Data", "definition": "The World Inequality Database (WID.world) is an open, collaboratively maintained data resource that compiles harmonised series on the distribution of income and wealth within and between countries over long historical periods. It combines national accounts, tax records, surveys, and estimation methods to produce comparable inequality statistics, and underpins the World Inequality Report. It is a primary reference for empirical research on global inequality.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:world-inequality-database", "labels": [ "World Inequality Database" ], "is_subclass_of": [ "Data Management" ], "wikilinks": [] }, { "id": "world-instance", "title": "World Instance", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A runtime instantiation of a virtual world template that maintains isolated state, physics simulation, and user interactions for a bounded set of concurrent participants. World instances enable scalable multi-user virtual environments through dynamic spawning, load balancing, and state checkpointing, as seen in MMO dungeons, battle royale matches, and social VR rooms.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:world-instance", "labels": [ "World Instance" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "world-model", "title": "World Model", "domain": "ai", "domain_name": "Ai", "definition": "A World Model is an internal representation maintained by an intelligent agent\u2014biological or artificial\u2014that encodes beliefs about the structure, dynamics, and state of its environment, enabling prediction, planning, and counterfactual reasoning without requiring direct sensory input for every decision. In model-based reinforcement learning, a learned world model allows an agent to simulate future trajectories in latent space, dramatically improving sample efficiency compared to model-free approaches. World models compress high-dimensional sensory observations into compact representations that capture causally relevant structure, supporting long-horizon planning and generalisation to novel situations. They are central to current research on embodied AI, autonomous driving, and general-purpose robot manipulation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:world-model", "labels": [ "World Model" ], "is_subclass_of": [ "Knowledge Representation" ], "wikilinks": [] }, { "id": "world-models", "title": "World Models", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "World models are learned internal representations of an environment that predict how it evolves in response to actions. They let an agent plan and reason by simulating outcomes rather than acting directly in the world.", "entityType": "Class", "qualityScore": 0.6, "maturity": "emerging", "iri": "urn:ngm:class:world-models", "labels": [ "World Models" ], "is_subclass_of": [ "World Model" ], "wikilinks": [ "Neural Network", "Reinforcement Learning", "Planning", "Generative Models", "World Model" ] }, { "id": "world-trade-organization", "title": "World Trade Organization", "domain": "governance", "domain_name": "Governance", "definition": "The World Trade Organization (WTO) is the intergovernmental organisation, established in 1995 as successor to the GATT, that administers the multilateral rules governing trade between its 166 members. It provides the negotiated agreements covering goods, services, and intellectual property, a forum for further trade negotiations, a binding dispute-settlement mechanism, and monitoring of members' trade policies, anchored in the principles of non-discrimination (most-favoured-nation and national treatment), tariff bindings, and transparency.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:world-trade-organization", "labels": [ "World Trade Organization" ], "is_subclass_of": [ "Governance" ], "wikilinks": [ "Governance", "International Trade", "IMF", "OECD" ] }, { "id": "world-wide-web-consortium", "title": "World Wide Web Consortium", "domain": "security", "domain_name": "Security", "definition": "The World Wide Web Consortium (W3C) is the principal international standards organisation for the World Wide Web, founded in 1994 by Tim Berners-Lee to develop interoperable technical specifications and guidelines. Through a consensus process its member organisations and working groups produce Recommendations covering core web technologies such as HTML, CSS, the Document Object Model, and accessibility, as well as Semantic Web standards including RDF, OWL, and SPARQL. The W3C promotes a single, open, royalty-free Web that is accessible, secure, and consistent across browsers and devices.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:world-wide-web-consortium", "labels": [ "World Wide Web Consortium", "W3C" ], "is_subclass_of": [ "Standards Body" ], "wikilinks": [] }, { "id": "worldcoin", "title": "Worldcoin", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A digital identity and cryptocurrency project that uses iris biometrics to produce a cryptographic proof of unique human personhood, enabling sybil-resistant participation in online systems and distributing an associated cryptocurrency token.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:worldcoin", "labels": [ "Worldcoin" ], "is_subclass_of": [ "Digital Identity" ], "wikilinks": [ "Biometric Authentication", "Facial Recognition", "Sybil Resistance", "Digital Identity", "Identity Verification" ] }, { "id": "wormhole", "title": "Wormhole", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Wormhole is a decentralised, generic cross-chain messaging and asset-transfer protocol that connects heterogeneous blockchain networks by relying on a permissioned guardian network whose supermajority attestations produce signed Verified Action Approval (VAA) messages. On-chain core contracts on each supported network verify VAAs to lock, mint, or burn assets and to route arbitrary data payloads, making Wormhole a general-purpose interoperability layer rather than a simple token bridge. Originally launched in 2020 as a Solana\u2013Ethereum wrapped-asset bridge by Certus One, it has expanded to support over 30 layer-1 and layer-2 ecosystems under stewardship of the independent Wormhole Foundation.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wormhole", "labels": [ "Wormhole" ], "is_subclass_of": [ "Cross-Chain Bridge" ], "wikilinks": [] }, { "id": "wrapped-token", "title": "Wrapped Token", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A Wrapped Token is a digital asset on one blockchain that represents, at a 1:1 peg, an asset from a different blockchain or currency system, created by locking the original asset in a custody mechanism (custodian, bridge smart contract, or MPC wallet) and minting an equivalent representation on the target chain. Wrapped tokens enable cross-chain liquidity by making assets from non-smart-contract blockchains (such as Bitcoin) or from other chains available within DeFi ecosystems. WBTC (Wrapped Bitcoin on Ethereum, ERC-20) is the canonical example, allowing Bitcoin to participate in Ethereum-based lending, trading, and yield protocols. Wrapped tokens introduce custodial or bridge risk as the peg relies on the integrity of the locking and minting mechanism.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:wrapped-token", "labels": [ "Wrapped Token" ], "is_subclass_of": [ "Cryptocurrency Token" ], "wikilinks": [] }, { "id": "write-blocker", "title": "Write Blocker", "domain": "security", "domain_name": "Security", "definition": "A write blocker is a hardware or software device that intercepts write commands sent to a storage medium, allowing an investigator to read data from it without altering its contents. It is a standard tool in digital forensics, used to create forensically sound images of hard drives, USB media, and other storage before analysis, preserving evidentiary integrity and chain of custody. Hardware write blockers sit physically between the storage device and the acquisition workstation, intercepting commands at the interface level, while software write blockers achieve the same effect by filtering operating system calls. Their use is typically documented as part of standard evidence collection procedure to withstand challenge in court.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:write-blocker", "labels": [ "Write Blocker" ], "is_subclass_of": [ "Digital Forensics" ], "wikilinks": [] }, { "id": "write-ahead-logging", "title": "Write-Ahead Logging", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Write-Ahead Logging (WAL) is a durability technique in which changes are recorded to a sequential append-only log before they are applied to the main data store. By guaranteeing that the log is flushed to stable storage before the corresponding pages are modified, the system can recover a consistent state after a crash by replaying or undoing logged operations. WAL is foundational to transactional databases, providing atomicity and durability without expensive synchronous writes to scattered data pages.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:write-ahead-logging", "labels": [ "Write-Ahead Logging", "Write Ahead Log", "Write-Ahead Log" ], "is_subclass_of": [ "Data Persistence" ], "wikilinks": [] }, { "id": "wyoming-protocol", "title": "Wyoming Protocol", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The Wyoming Protocol is an open, peer-to-peer communication protocol used in the Home Assistant ecosystem to connect voice and audio services such as wake-word detection, speech-to-text, and text-to-speech. It defines a simple JSON-over-socket event framing that lets independent voice components run as networked services and interoperate locally without cloud dependencies. It matters as the backbone of Home Assistant's privacy-preserving local voice assistant pipeline.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:wyoming-protocol", "labels": [ "Wyoming Protocol" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "x-ray-astronomy", "title": "X-ray Astronomy", "domain": "space-science-and-systems", "domain_name": "Space Science And Systems", "definition": "", "entityType": "Class", "qualityScore": 0.0, "maturity": "draft", "iri": "urn:ngm:class:x-ray-astronomy", "labels": [ "X-ray Astronomy" ], "is_subclass_of": [], "wikilinks": [] }, { "id": "x-509-certificate", "title": "X.509 Certificate", "domain": "security", "domain_name": "Security", "definition": "A standardised digital certificate format that binds a public key to an entity identity, together with validity period and usage constraints, and is signed by a certificate authority within a public key infrastructure so that relying parties can cryptographically verify its authenticity.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:x-509-certificate", "labels": [ "X.509 Certificate", "Certificate Validation", "PKI X.509", "X.509", "X.509 Certificate Standard" ], "is_subclass_of": [ "Digital Certificate" ], "wikilinks": [ "Public Key Infrastructure", "Certificate Authority", "Authentication", "Transport Layer Security", "Digital Signature", "Digital Certificate" ] }, { "id": "x-402", "title": "X402", "domain": "blockchain", "domain_name": "Blockchain", "definition": "An open protocol that uses the HTTP 402 Payment Required status code to enable machine-to-machine payments for accessing web resources and services. It allows clients to pay programmatically before a request is served.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:x-402", "labels": [ "X402" ], "is_subclass_of": [ "Digital Currency" ], "wikilinks": [ "Stablecoin", "Decentralized Application", "Web3", "Digital Currency" ] }, { "id": "x509-standard", "title": "X509 Standard", "domain": "security", "domain_name": "Security", "definition": "The X.509 standard defines the format of public-key certificates that bind a public key to an identity, together with the structures for certificate paths and revocation lists used in public-key infrastructure. An X.509 certificate carries fields such as subject, issuer, validity period, public key and extensions, and is signed by a certificate authority so that relying parties can verify it by tracing a chain to a trusted root. It is the foundational certificate format underlying TLS, secure email and many authentication systems.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:x509-standard", "labels": [ "X509 Standard", "X.509 Standard" ], "is_subclass_of": [ "Digital Certificate" ], "wikilinks": [] }, { "id": "xacml", "title": "XACML", "domain": "security", "domain_name": "Security", "definition": "XACML (eXtensible Access Control Markup Language) is an OASIS standard that defines a declarative, XML-based language for expressing access-control policies and the requests and responses used to evaluate them. It specifies a reference architecture separating policy decision, enforcement, administration and information points, enabling fine-grained, attribute-based authorisation across heterogeneous systems. XACML lets organisations externalise authorisation logic from applications into centrally managed policies.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:xacml", "labels": [ "XACML" ], "is_subclass_of": [ "Attribute-Based Access Control" ], "wikilinks": [] }, { "id": "xbrl", "title": "XBRL", "domain": "governance", "domain_name": "Governance", "definition": "XBRL (eXtensible Business Reporting Language) is an open XML-based standard for tagging and exchanging business and financial reporting data in a machine-readable form. It uses taxonomies of defined concepts to attach semantic meaning, units, and context to each reported fact, enabling automated validation, comparison, and analysis. It matters because regulators worldwide mandate it for filings, making financial disclosures consistently structured and computable.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:xbrl", "labels": [ "XBRL", "XBRL Digital Tagging", "XBRL Inline Reporting", "XBRL Taxonomy" ], "is_subclass_of": [ "Technical Standard" ], "wikilinks": [] }, { "id": "xlnet", "title": "XLNet", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A generalised autoregressive pre-training model that learns bidirectional contexts by maximising the expected likelihood over all permutations of the token factorisation order, using a two-stream self-attention mechanism to avoid information leakage. XLNet integrates the Transformer-XL segment recurrence mechanism for long-range dependency modelling and outperforms BERT on 20 NLU benchmarks.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:xlnet", "labels": [ "XLNet" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "xml-encryption", "title": "XML Encryption", "domain": "security", "domain_name": "Security", "definition": "XML Encryption is a W3C specification defining how to encrypt all or part of an XML document, including nested and structured data, while preserving the surrounding XML structure. It supports encrypting arbitrary data, XML elements or element content, and can combine multiple encrypted parts within a single document using standard key-transport and key-wrapping conventions. It is used alongside XML Signature in SAML assertions to protect sensitive attribute values exchanged between identity providers and service providers.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:xml-encryption", "labels": [ "XML Encryption" ], "is_subclass_of": [ "Encryption" ], "wikilinks": [] }, { "id": "xml", "title": "XML", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "XML (Extensible Markup Language) is a W3C-standardised, text-based markup language for encoding documents and structured data in a format that is both human-readable and machine-processable. It defines a strict syntax of nested, attributed elements and supports schema languages (DTD, XML Schema, RELAX NG) for validation, plus a family of related standards for querying, transforming and namespacing. Once dominant for data interchange and configuration, it remains foundational to many enterprise, financial and document standards even as JSON has overtaken it for lightweight web interchange.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:xml", "labels": [ "XML" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [] }, { "id": "xr-accessibility-guideline", "title": "XR Accessibility Guideline", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Design recommendations and best practices ensuring XR applications and immersive experiences are usable by people with diverse abilities and disabilities.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:xr-accessibility-guideline", "labels": [ "XR Accessibility Guideline", "Accessibility Guideline" ], "is_subclass_of": [ "Governance and Safety", "Accessibility Standard" ], "wikilinks": [ "Accessibility Testing", "Assistive Technology Integration", "Best Practice", "Design Recommendation", "ETSI GR ARF 010", "Implementation Example", "ISO 9241-112", "Universal Access", "W3C XR Accessibility UR", "Accessibility Standard", "Inclusive XR Design", "InteractionDomain", "Middleware Layer", "User Research" ] }, { "id": "xr-accessibility-standards", "title": "XR Accessibility Standards", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Guidelines, best practices, and technical specifications for making extended reality experiences accessible to users with disabilities, addressing visual, auditory, motor, cognitive, and vestibular impairments through inclusive design principles and assistive technology compatibility.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:xr-accessibility-standards", "labels": [ "XR Accessibility Standards" ], "is_subclass_of": [ "Governance and Safety", "Accessibility" ], "wikilinks": [ "Accessibility", "metaverse" ] }, { "id": "xr-applications", "title": "XR Applications", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "XR Applications are software systems designed to run on extended reality platforms \u2014 encompassing augmented reality (AR), virtual reality (VR), and mixed reality (MR) \u2014 that blend real and virtual environments to deliver spatially aware, immersive user experiences. They span consumer entertainment, enterprise training, industrial maintenance, medical simulation, and remote collaboration domains.", "entityType": "Class", "qualityScore": 0.68, "maturity": "emerging", "iri": "urn:ngm:class:xr-applications", "labels": [ "XR Applications", "XR Application" ], "is_subclass_of": [ "Extended Reality" ], "wikilinks": [] }, { "id": "xr-content-delivery", "title": "XR Content Delivery", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "XR Content Delivery is the distribution of extended-reality assets, including 3D models, textures, spatial audio, and streamed scenes, to headsets and mobile devices with the low latency and high bandwidth that immersive rendering requires. It builds on content-delivery-network infrastructure but adds XR-specific concerns such as adaptive level-of-detail streaming and foveated compression. Reliable delivery is a precondition for responsive, motion-sickness-free XR experiences.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:xr-content-delivery", "labels": [ "XR Content Delivery" ], "is_subclass_of": [ "Content Delivery Network" ], "wikilinks": [] }, { "id": "xr-device", "title": "XR Device", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Hardware equipment designed to create or enhance extended reality experiences, encompassing virtual reality headsets, augmented reality glasses, mixed reality displays, haptic feedback systems, and motion tracking sensors that enable immersive interaction with digital content.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:xr-device", "labels": [ "XR Device" ], "is_subclass_of": [ "Display and Rendering", "XR Hardware" ], "wikilinks": [ "metaverse", "XR Hardware" ] }, { "id": "xr-framework", "title": "XR Framework", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An XR Framework is a software architecture or SDK that provides unified runtime support, device abstraction, and application programming interfaces for extended reality experiences spanning augmented, virtual, and mixed reality. Frameworks such as OpenXR standardise access to headsets and input devices, whilst game engine integrations enable developers to target multiple XR platforms from a single codebase.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:xr-framework", "labels": [ "XR Framework" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "owl:Thing" ] }, { "id": "xr-hardware", "title": "XR Hardware", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "XR Hardware comprises the physical computing and sensing devices \u2014 including head-mounted displays, spatial computing headsets, hand-tracking controllers, haptic peripherals, and body-worn sensors \u2014 that enable augmented, mixed, and virtual reality experiences across the extended reality spectrum. These devices integrate high-resolution micro-display optics, inertial measurement units, inside-out positional tracking cameras, and wireless connectivity stacks to deliver low-latency immersive spatial content. Modern XR hardware increasingly embeds dedicated neural processing units for on-device spatial AI inference, eye-tracking modules for foveated rendering, and environmental depth sensors for real-time world reconstruction. The category spans standalone untethered headsets, tethered PC-class systems, and lightweight optical see-through spectacles optimised for always-on wearability.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:xr-hardware", "labels": [ "XR Hardware", "XR Hardware Device" ], "is_subclass_of": [ "Spatial Computing", "Display and Rendering" ], "wikilinks": [ "owl:Thing" ] }, { "id": "xr-headset", "title": "XR Headset", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A head-mounted display device that delivers immersive visual and audio experiences for virtual reality, augmented reality, or mixed reality applications, featuring integrated displays, tracking sensors, audio systems, and processing capabilities for rendering digital content in the user's field of view.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:xr-headset", "labels": [ "XR Headset" ], "is_subclass_of": [ "Display and Rendering", "XR Hardware" ], "wikilinks": [ "metaverse", "XR Hardware" ] }, { "id": "xr-meeting-space", "title": "XR Meeting Space", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Virtual collaboration environments accessed through extended reality devices that enable remote participants to interact as avatars or volumetric representations in shared 3D spaces, supporting meetings, presentations, design reviews, and team collaboration with spatial audio and gesture-based communication. XR meeting spaces extend traditional video conferencing by providing persistent, spatially aware environments where presence, proximity, and embodied interaction improve engagement and coordination.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:xr-meeting-space", "labels": [ "XR Meeting Space" ], "is_subclass_of": [ "Platform and Environment", "Virtual Collaboration" ], "wikilinks": [ "metaverse", "Virtual Collaboration" ] }, { "id": "xr-runtime-environment", "title": "XR Runtime Environment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "The software infrastructure layer that manages extended reality hardware and provides standardized APIs for XR applications, handling device abstraction, tracking systems, rendering pipelines, input processing, and compositor services to enable cross-platform XR development.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:xr-runtime-environment", "labels": [ "XR Runtime Environment", "OpenXR Runtime", "XR Runtime" ], "is_subclass_of": [ "Platform and Environment", "Runtime Environment" ], "wikilinks": [ "metaverse", "Runtime Environment" ] }, { "id": "xr-technical-standard", "title": "XR Technical Standard", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "An XR technical standard is a formal specification defining interfaces, data formats, or performance requirements for extended reality hardware and software \u2014 for example OpenXR for runtime APIs, WebXR for browser-based immersive experiences, and glTF for 3D asset interchange. Such standards reduce fragmentation and let content run across headsets from different vendors.", "entityType": "Class", "qualityScore": 0.35, "maturity": "draft", "iri": "urn:ngm:class:xr-technical-standard", "labels": [ "XR Technical Standard" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "owl:Thing" ] }, { "id": "xr-testing-infrastructure", "title": "XR Testing Infrastructure", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The hardware, software, and modological frameworks for validating extended reality applications, including automated testing tools, motion capture systems, user testing labs, performance profiling equipment, and simulation environments for quality assurance of VR, AR, and MR experiences.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:xr-testing-infrastructure", "labels": [ "XR Testing Infrastructure" ], "is_subclass_of": [ "Software Engineering", "Testing Infrastructure" ], "wikilinks": [ "metaverse", "Testing Infrastructure" ] }, { "id": "xapi", "title": "Xapi", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "xAPI (the Experience API, also known as Tin Can API) is an e-learning interoperability specification that captures learning experiences as actor-verb-object statements sent to a Learning Record Store. Unlike SCORM, it records learning that occurs outside a single course or platform, including informal, mobile and simulation-based activity. xAPI underpins rich learning analytics by providing a flexible, standardised data model.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:xapi", "labels": [ "Xapi", "xAPI" ], "is_subclass_of": [ "Education Technology" ], "wikilinks": [] }, { "id": "xml-schema-definition", "title": "Xml Schema Definition", "domain": "data", "domain_name": "Data", "definition": "XML Schema Definition (XSD) is a W3C-recommended language for describing and constraining the structure, content and data types of XML documents. An XSD declares permitted elements, attributes, ordering, cardinality and strongly typed values, enabling validators to confirm that an instance document conforms to an agreed contract. It supersedes the older Document Type Definition (DTD) by adding namespace awareness, a rich built-in type system and support for derivation, making it foundational to many enterprise, document and messaging standards.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:xml-schema-definition", "labels": [ "Xml Schema Definition", "XML Schema Definition" ], "is_subclass_of": [ "Data Schema" ], "wikilinks": [] }, { "id": "yaml", "title": "YAML", "domain": "standards", "domain_name": "Standards", "definition": "A human-readable data serialisation language (YAML Ain't Markup Language) that expresses mappings, sequences, and scalars through indentation-based structure rather than delimiters, and is a strict superset of JSON. Designed for legibility and hand-editing, it adds comments, anchors and aliases for reuse, and multi-document streams, which has made it the dominant format for configuration in the DevOps ecosystem \u2014 Kubernetes manifests, CI pipelines, Ansible playbooks, and OpenAPI definitions \u2014 despite well-known pitfalls around implicit typing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:yaml", "labels": [ "YAML" ], "is_subclass_of": [ "Data Serialization" ], "wikilinks": [ "Data Serialization", "JSON", "OpenAPI" ] }, { "id": "yearn", "title": "Yearn", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Yearn is a decentralised finance protocol that automates the allocation of deposited assets across yield-generating strategies, abstracting away manual position management for users.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:yearn", "labels": [ "Yearn", "Yearn Finance" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [ "Smart Contracts", "Yield Farming", "Liquidity Pool", "Treasury Management", "DeFi" ] }, { "id": "yield-aggregator", "title": "Yield Aggregator", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A yield aggregator is a decentralised-finance protocol that automatically routes deposited assets across multiple yield-bearing strategies \u2014 such as lending markets, liquidity pools, and staking \u2014 to maximise returns while compounding rewards on the user's behalf. By pooling capital and automating strategy selection, reward harvesting, and reinvestment, aggregators reduce gas costs and operational complexity for individual depositors. Modern aggregators frequently expose tokenised vault shares conforming to standards such as ERC-4626, making positions composable with the wider DeFi ecosystem.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:yield-aggregator", "labels": [ "Yield Aggregator" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "yield-farming", "title": "yield farming", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Yield farming is the practice of actively deploying digital assets into decentralised finance protocols\u2014including liquidity pools, lending markets, and staking contracts\u2014to earn token rewards, trading fees, or interest income, and optimising risk-adjusted returns by dynamically reallocating capital across multiple protocols. Participants typically receive liquidity provider (LP) tokens representing their proportional pool share, which can themselves be staked or deposited into yield aggregator contracts to compound rewards, creating layered risk and return exposure. The practice emerged prominently during the 2020 DeFi Summer as protocols adopted token-emission incentive models to bootstrap liquidity and governance participation. Yield farming strategies are sensitive to smart contract risk, impermanent loss, reward token price volatility, oracle manipulation, and evolving regulatory classification of token distributions.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:yield-farming", "labels": [ "Yield Farming" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "yield-generation", "title": "Yield Generation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Yield generation in the context of decentralised finance (DeFi) refers to the set of mechanisms by which cryptocurrency holders earn returns on their assets by deploying them into productive on-chain activities \u2014 including liquidity provision to automated market makers, lending on money market protocols, staking in proof-of-stake consensus, and participating in yield optimisation vaults \u2014 thereby converting idle digital assets into income-generating positions. Yield is typically expressed as an annualised percentage rate (APR) or annual percentage yield (APY) and derives from trading fees, interest payments from borrowers, block rewards, and protocol token emissions. Yield generation is a core economic primitive of DeFi, enabling capital efficiency but also introducing smart contract, liquidation, and impermanent loss risks.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:yield-generation", "labels": [ "Yield Generation", "DeFi Yield", "Yield Strategy", "Yield-Bearing Vault" ], "is_subclass_of": [ "DeFi" ], "wikilinks": [] }, { "id": "yield-optimisation", "title": "Yield Optimisation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Yield optimisation is the automated process of allocating deposited capital across decentralised finance protocols and strategies to maximise risk-adjusted returns, typically by continuously monitoring yield rates, harvesting rewards and rebalancing positions between lending markets, liquidity pools and staking contracts. It is the strategic objective that yield aggregators and vaults implement on a user's behalf, reducing the manual effort and gas cost of chasing the best available rate. Yield optimisation strategies must balance expected return against smart-contract risk, impermanent loss and reward-token volatility.", "entityType": "Class", "qualityScore": 0.55, "maturity": "established", "iri": "urn:ngm:class:yield-optimisation", "labels": [ "Yield Optimisation" ], "is_subclass_of": [ "Decentralised Finance" ], "wikilinks": [] }, { "id": "yjs-framework", "title": "Yjs Framework", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Yjs is an open-source, high-performance CRDT (Conflict-free Replicated Data Type) framework for building collaborative applications in JavaScript. It provides shared data types such as maps, arrays, and rich text that automatically merge concurrent edits without conflicts, and integrates with popular editors like ProseMirror, CodeMirror, and Quill. Yjs supports multiple transport backends including WebSocket, WebRTC, and IndexedDB for offline persistence.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:yjs-framework", "labels": [ "Yjs Framework" ], "is_subclass_of": [ "Protocol and Infrastructure" ], "wikilinks": [] }, { "id": "yoti", "title": "Yoti", "domain": "security", "domain_name": "Security", "definition": "A digital identity company that provides identity verification, age estimation and authentication services for individuals and organisations.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:yoti", "labels": [ "Yoti" ], "is_subclass_of": [ "Identity Verification" ], "wikilinks": [ "Biometric Authentication", "Identity Verification", "Age Verification", "Digital Identity", "Facial Recognition" ] }, { "id": "z-wave", "title": "Z Wave", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Z-Wave is a low-power wireless communication protocol designed specifically for home automation and smart home devices, operating in the sub-GHz frequency band (868 MHz in Europe, 908 MHz in North America). It uses a mesh networking topology where each device can relay signals, extending range and improving reliability throughout a building. Z-Wave supports up to 232 nodes per network and is governed by the Z-Wave Alliance, which maintains interoperability standards across manufacturers.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:z-wave", "labels": [ "Z Wave", "Z-Wave" ], "is_subclass_of": [ "Infrastructure", "Network and Communication" ], "wikilinks": [] }, { "id": "z-wave-js-ui", "title": "Z-Wave JS UI", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Z-Wave JS UI is an open-source application that wraps the Z-Wave JS driver to manage a Z-Wave mesh network of smart-home devices through a web interface and to bridge it to home-automation platforms via MQTT. It handles device inclusion, configuration, firmware updates, and network healing, exposing Z-Wave nodes as controllable entities. It matters as a primary integration path between Z-Wave hardware and Home Assistant.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:z-wave-js-ui", "labels": [ "Z-Wave JS UI", "Z-Wave JS" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "zk-rollup", "title": "ZK Rollup", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A ZK rollup is a Layer 2 scaling construction that executes transactions off-chain and posts a succinct validity proof, typically a zk-SNARK or zk-STARK, to a Layer 1 chain so the base layer can verify correctness without re-executing the batch. Compressed transaction data and the proof are published on-chain, inheriting the security of the underlying settlement layer while drastically reducing per-transaction cost. Unlike optimistic rollups, finality does not require a fraud-proof challenge window because validity is proven cryptographically.", "entityType": "Class", "qualityScore": 0.62, "maturity": "emerging", "iri": "urn:ngm:class:zk-rollup", "labels": [ "ZK Rollup", "ZK-Rollup", "zk-Rollup" ], "is_subclass_of": [ "Rollup" ], "wikilinks": [] }, { "id": "zk-snark", "title": "ZK-SNARK", "domain": "security", "domain_name": "Security", "definition": "A ZK-SNARK (Zero-Knowledge Succinct Non-Interactive Argument of Knowledge) is a cryptographic proof system that allows a prover to convince a verifier that a computational statement is true without revealing any information beyond the validity of the statement itself. The proof is succinct \u2014 its size and verification time are sub-linear (often constant or logarithmic) relative to the underlying computation \u2014 and non-interactive, requiring no back-and-forth messages between prover and verifier. Most constructions rely on bilinear pairings over elliptic curves and require a one-time trusted setup ceremony to generate a structured reference string.", "entityType": "Class", "qualityScore": 0.76, "maturity": "established", "iri": "urn:ngm:class:zk-snark", "labels": [ "ZK-SNARK", "Groth16 zk-SNARK", "Recursive SNARK", "SNARK", "ZK SNARK", "zk-SNARK", "zkSNARK" ], "is_subclass_of": [ "Zero-Knowledge Proof" ], "wikilinks": [ "Elliptic Curve Cryptography", "Groth16", "Rollup", "Zero-Knowledge Proof" ] }, { "id": "zk-snarks", "title": "ZK-SNARKs", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge - a cryptographic proof system enabling one party to prove possession of information without revealing the information itself, characterized by small proof sizes, fast verification, and no interaction required between prover and verif...", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:zk-snarks", "labels": [ "ZK-SNARKs" ], "is_subclass_of": [ "Governance and Safety", "Zero Knowledge Proof" ], "wikilinks": [ "metaverse", "Zero Knowledge Proof" ] }, { "id": "zk-starks", "title": "ZK-STARKs", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Zero-Knowledge Scalable Transparent Arguments of Knowledge: a cryptographic proof system enabling verification of computations without revealing underlying data. Unlike ZK-SNARKs, STARKs require no trusted setup, are conjectured to be quantum-resistant, and scale efficiently for large witness sizes, making them foundational for Layer 2 blockchain rollups and privacy-preserving computation.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:zk-starks", "labels": [ "ZK-STARKs", "ZK-STARK", "zk-STARK", "zkSTARK" ], "is_subclass_of": [ "Governance and Safety", "Zero Knowledge Proof" ], "wikilinks": [ "metaverse", "Zero Knowledge Proof" ] }, { "id": "zcash", "title": "Zcash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Zcash is a privacy-oriented cryptocurrency launched in 2016 that uses zero-knowledge proofs to allow transactions to be verified without revealing the sender, recipient or amount. It was the first widespread deployment of zk-SNARKs, succinct non-interactive arguments of knowledge, in a public blockchain. Zcash supports both transparent addresses, similar to Bitcoin, and shielded addresses that conceal transaction details, giving users a choice of privacy level. The protocol derives from the Zerocash academic proposal and is developed by the Electric Coin Company.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:zcash", "labels": [ "Zcash" ], "is_subclass_of": [ "Token and Asset", "Blockchain Domain" ], "wikilinks": [ "Zero Knowledge Proof", "Proof of Work", "Monero", "zkSync", "Blockchain Domain", "Ben-Sasson et al. 2014, Zerocash: Decentralized Anonymous Payments from Bitcoin" ] }, { "id": "zero-convolution", "title": "Zero Convolution", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A zero convolution is a 1x1 convolutional layer whose weights and bias are initialised to zero before training. Used in ControlNet, it connects a trainable copy of a diffusion model's encoder blocks to the frozen base network so that, at the start of training, the added conditioning branch contributes nothing and the combined model reproduces the original pretrained behaviour exactly. As training proceeds, the zero convolution's weights grow away from zero, progressively and stably introducing the effect of the new conditioning signal.", "entityType": "Class", "qualityScore": 0.55, "maturity": "emerging", "iri": "urn:ngm:class:zero-convolution", "labels": [ "Zero Convolution" ], "is_subclass_of": [ "ControlNet" ], "wikilinks": [] }, { "id": "zero-downtime-deployment", "title": "Zero Downtime Deployment", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Zero Downtime Deployment is a release strategy that updates a running service without interrupting user-facing availability. It relies on patterns such as rolling updates, blue-green switchovers, or canary releases combined with health checks and load-balancer draining so that traffic only reaches instances ready to serve it. Achieving it requires backward-compatible changes, graceful connection shutdown, and idempotent operations across the transition.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:zero-downtime-deployment", "labels": [ "Zero Downtime Deployment", "Zero-Downtime Deployment" ], "is_subclass_of": [ "Continuous Deployment" ], "wikilinks": [] }, { "id": "zero-knowledge", "title": "Zero Knowledge", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Cryptographic proof systems enabling a prover to convince a verifier that a statement is true without disclosing any information beyond the validity of the statement itself. zk-SNARKs and zk-STARKs are the dominant proof systems, with applications spanning blockchain scalability through ZK-rollups, privacy-preserving identity verification, confidential DeFi transactions, and verifiable AI model outputs. Foundational to privacy-preserving computation and Layer 2 scaling architectures.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:zero-knowledge", "labels": [ "Zero Knowledge", "Zero Knowledge Architecture", "ZeroKnowledge" ], "is_subclass_of": [ "Governance and Safety", "Privacy-Enhancing Computation (PEC)" ], "wikilinks": [ "AIModel", "CryptographicProof", "dt:enables", "dt:protects", "dt:secures", "dt:validates", "dt:verifies", "InteractiveProtocol", "NIZKProof", "preservesPrivacy", "PrivacyCoin", "PrivacyPreserving", "PrivacyProtocol", "PrivateTransaction", "provesStatement", "ScalableRollup", "SelectiveDisclosure", "usedIn", "UserData", "VerifiableComputation" ] }, { "id": "zero-shot-learning", "title": "Zero Shot Learning", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "Zero-shot learning is a machine-learning setting in which a model performs a task on classes or instances it has never seen during training, generalising from auxiliary information such as semantic attributes, natural-language descriptions or a shared embedding space. Modern instances exploit large pre-trained language and vision-language models that align inputs and labels in a common representation, enabling prediction by similarity rather than by fitting task-specific examples. It contrasts with few-shot and supervised learning by requiring no labelled examples of the target classes.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:zero-shot-learning", "labels": [ "Zero Shot Learning", "Zero-Shot Learning" ], "is_subclass_of": [ "Transfer Learning", "Meta-Learning" ], "wikilinks": [] }, { "id": "zero-trust-architecture", "title": "zero trust architecture", "domain": "security", "domain_name": "Security", "definition": "Zero Trust Architecture (ZTA) is a cybersecurity paradigm that abandons the notion of a trusted network perimeter and instead requires explicit, continuous verification of every principal \u2014 user, device, or service \u2014 for every request regardless of network origin. Grounded in least-privilege access, ZTA evaluates contextual signals including identity assertions, device health posture, behavioural risk, and data sensitivity at each access decision point, enforced by a Policy Decision Point and Policy Enforcement Point pair operating across micro-segmented environments. Formalised in NIST SP 800-207, ZTA has become the dominant enterprise security reference model in response to the dissolution of traditional perimeter boundaries by cloud-native, remote-work, and supply-chain threat vectors.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:zero-trust-architecture", "labels": [ "Zero Trust Architecture", "Zero Trust", "Zero Trust Network Access", "Zero Trust Security", "Zero Trust Security Framework", "Zero-Trust Access Control", "Zero-Trust Architecture", "Zero-Trust Security", "ZeroTrustArchitecture" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "zero-trust", "title": "Zero Trust", "domain": "security", "domain_name": "Security", "definition": "A security architecture built on the principle of 'never trust, always verify': no user, device, or workload is trusted by virtue of its network location, and every access request must be continuously authenticated, authorised, and encrypted based on identity, device posture, and context, replacing the traditional perimeter model in which anything inside the corporate network was implicitly trusted.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:zero-trust", "labels": [ "Zero Trust" ], "is_subclass_of": [ "Security Architecture" ], "wikilinks": [ "Security Architecture", "Authentication", "Least Privilege", "Perimeter Security" ] }, { "id": "zero-day-exploits", "title": "Zero-Day Exploits", "domain": "security", "domain_name": "Security", "definition": "A zero-day exploit is an attack technique that targets a software or hardware vulnerability unknown to the vendor and for which no patch yet exists, giving defenders zero days to prepare. Because the flaw is undisclosed, such exploits bypass signature-based defenses and are highly valuable to attackers, state actors, and exploit brokers. They matter as a central concern in cybersecurity, surveillance, and offensive cyber operations.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:zero-day-exploits", "labels": [ "Zero-Day Exploits", "Zero-Day", "Zero-Day Vulnerability" ], "is_subclass_of": [ "Cybersecurity" ], "wikilinks": [] }, { "id": "zero-knowledge-machine-learning", "title": "zero-knowledge machine learning", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Cryptographic techniques enabling verification that a machine learning model executed correctly on specific inputs without revealing model weights, training data, or private inputs. ZKML bridges zero-knowledge proofs and machine learning, allowing on-chain verification of off-chain AI inference with privacy guarantees.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:zero-knowledge-machine-learning", "labels": [ "Zero-Knowledge Machine Learning" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "zero-knowledge-proof-zkp", "title": "Zero-Knowledge Proof (ZKP)", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A cryptographic protocol that enables one party (the prover) to prove to another party (the verifier) that a statement is true without revealing any information beyond the validity of the statement itself.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:zero-knowledge-proof-zkp", "labels": [ "Zero-Knowledge Proof (ZKP)" ], "is_subclass_of": [ "Security and Identity", "Cryptography" ], "wikilinks": [ "Algebraic Circuits", "Bulletproofs", "Computational Complexity Theory", "Confidential Transactions", "Cryptographic Hash Function", "Cryptographic Verification System", "Interactive ZKP", "Non-Interactive ZKP", "Number Theory", "OMA3 + Reed Smith", "PLONK", "Polynomial Commitment Scheme", "Privacy-Preserving Identity", "Privacy-Preserving Protocol", "Private Authentication", "Verifiable Computation", "Elliptic Curve Cryptography", "Middleware Layer", "TrustAndGovernanceDomain", "zk-SNARKs" ] }, { "id": "zero-knowledge-proof", "title": "Zero-Knowledge Proof", "domain": "security", "domain_name": "Security", "definition": "A cryptographic protocol allowing one party (prover) to convince another party (verifier) that a statement is true without revealing any information beyond the validity of the statement itself. Zero-knowledge proofs provide privacy-preserving verification in blockchain systems, enabling private transactions, identity attestation, and scalable computation via ZK-rollups.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:zero-knowledge-proof", "labels": [ "Zero-Knowledge Proof", "BC-0023-zero-knowledge-proofs", "BC-0202-zero-knowledge-proofs", "BC-0315-zero-knowledge-proof", "Non-Interactive Zero-Knowledge", "Non-Interactive Zero-Knowledge Proof", "Zero Knowledge Proof", "Zero Knowledge Proofs", "Zero-Knowledge Predicate Proof", "Zero-Knowledge Proofs", "ZeroKnowledgeProof" ], "is_subclass_of": [ "Cryptographic Primitive", "Blockchain Entity", "CryptographicPrimitive" ], "wikilinks": [ "IEEE 2418.1", "ISO/IEC 23257:2021", "NIST NISTIR", "Blockchain Entity", "CryptographicDomain", "CryptographicPrimitive", "SecurityLayer" ] }, { "id": "zero-knowledge-rollup", "title": "Zero-Knowledge Rollup", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A zero-knowledge rollup (ZK-rollup) is a layer-2 scaling construction that executes transactions off-chain in batches and posts a succinct validity proof to a base chain attesting that the new state was computed correctly. Because the proof cryptographically guarantees correctness, the base chain need not re-execute the transactions, achieving high throughput while inheriting the security of the underlying ledger. ZK-rollups offer near-instant finality once a proof is verified, distinguishing them from optimistic designs that rely on challenge periods.", "entityType": "Class", "qualityScore": 0.62, "maturity": "established", "iri": "urn:ngm:class:zero-knowledge-rollup", "labels": [ "Zero-Knowledge Rollup", "Zero Knowledge Rollup" ], "is_subclass_of": [ "Rollup" ], "wikilinks": [] }, { "id": "zero-shot-prompting", "title": "Zero-Shot Prompting", "domain": "ai", "domain_name": "Ai", "definition": "Zero-shot prompting is the technique of instructing a language model to perform a task using only a description, without providing worked examples. It relies on knowledge the model acquired during pretraining.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:zero-shot-prompting", "labels": [ "Zero-Shot Prompting" ], "is_subclass_of": [ "Prompt Engineering" ], "wikilinks": [ "Language Model", "Prompt Engineering", "In-Context Learning", "Few-Shot Learning", "Large Language Models" ] }, { "id": "zero-trust-architecture-zta", "title": "Zero-Trust Architecture (ZTA)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Security model requiring continuous verification of all entities and transactions with least-privilege access enforcement, eliminating implicit trust within metaverse network boundaries.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:zero-trust-architecture-zta", "labels": [ "Zero-Trust Architecture (ZTA)" ], "is_subclass_of": [ "Governance and Safety", "Security Architecture" ], "wikilinks": [ "Authentication Protocol", "Authorization Framework", "Breach Containment", "Continuous Verification", "Cybersecurity Framework", "Device Authentication", "Dynamic Access Control", "Encryption", "ENISA 2024", "Insider Threat Mitigation", "ISO 27001", "Least-Privilege Access Control", "Logging and Monitoring", "Microsegmentation", "Network Segmentation", "NIST SP 800-207", "Policy Decision Point", "Policy Enforcement Point", "Security Information and Event Management (SIEM)", "Threat Detection" ] }, { "id": "zigbee", "title": "Zigbee", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Zigbee is a low-power, low-data-rate wireless mesh networking protocol stack built atop the IEEE 802.15.4 physical and MAC layer standard, designed for battery-operated IoT sensors, actuators, and control devices in home automation, industrial monitoring, and building management. Maintained by the Connectivity Standards Alliance (formerly the Zigbee Alliance), it defines application-layer profiles, mesh routing, and AES-128 security on top of IEEE 802.15.4, supporting coordinator, router, and sleepy end-device roles within self-healing mesh topologies. Operating predominantly at 2.4 GHz (globally) as well as 868 MHz and 915 MHz regional bands, Zigbee targets data rates up to 250 kbps and enables coin-cell battery lifetimes of years, making it a foundational IoT wireless technology that coexists and increasingly converges with Thread and the Matter smart-home standard.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:zigbee", "labels": [ "Zigbee", "Zigbee Home Automation", "Zigbee Protocol", "Zigbee Radio Dongle" ], "is_subclass_of": [ "IEEE 802.15.4" ], "wikilinks": [] }, { "id": "zigbee2-mqtt", "title": "Zigbee2MQTT", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Zigbee2MQTT is an open-source bridge that connects Zigbee smart-home devices to MQTT, allowing them to be controlled independently of any proprietary vendor hub or cloud. Using an inexpensive Zigbee coordinator radio, it translates device messages into MQTT topics with a large database of supported devices. It matters as a vendor-neutral, locally controlled integration path for Zigbee hardware into platforms like Home Assistant.", "entityType": "Class", "qualityScore": 0.72, "maturity": "emerging", "iri": "urn:ngm:class:zigbee2-mqtt", "labels": [ "Zigbee2MQTT" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "zookeeper", "title": "ZooKeeper", "domain": "distributed-systems", "domain_name": "Distributed Systems", "definition": "An Apache open-source coordination service for distributed systems that exposes a replicated, hierarchical key-value namespace of 'znodes' with strict ordering guarantees, ephemeral nodes, and watches, over which applications build leader election, distributed locks, configuration management, and group membership; consistency across the ensemble is maintained by the ZAB atomic-broadcast protocol, a Paxos-influenced consensus design.", "entityType": "Class", "qualityScore": 0.8, "maturity": "mature", "iri": "urn:ngm:class:zookeeper", "labels": [ "ZooKeeper" ], "is_subclass_of": [ "Distributed Systems" ], "wikilinks": [ "Etcd", "Leader Election", "Paxos" ] }, { "id": "zoom-meetings", "title": "Zoom Meetings", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Zoom Meetings is a cloud-based video conferencing platform that enables synchronous audio, video, and screen-sharing sessions for distributed teams. It supports features such as breakout rooms, meeting recording, polling, and waiting-room controls, making it suitable for everything from one-on-ones to large-scale webinars. Its widespread adoption has made it a de-facto standard term for video calls in remote-work contexts.", "entityType": "Class", "qualityScore": 0.7, "maturity": "established", "iri": "urn:ngm:class:zoom-meetings", "labels": [ "Zoom Meetings" ], "is_subclass_of": [ "Workspace Tools" ], "wikilinks": [] }, { "id": "ai-core-concepts", "title": "ai core concepts", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Foundational artificial intelligence technologies \u2014 including machine learning paradigms (supervised, unsupervised, reinforcement), neural architectures (neural networks, transformers, CNNs), generative models, and classical symbolic reasoning \u2014 that collectively enable intelligent system behaviour. In metaverse contexts these concepts drive NPC dialogue, environment perception, procedural content generation, and federated privacy-preserving learning.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:ai-core-concepts", "labels": [ "ai core concepts" ], "is_subclass_of": [ "AI Research Area" ], "wikilinks": [ "AIBias", "AIGovernance", "ContentGeneration", "EnvironmentPerception", "LLM", "NPCInteraction", "TransformerModel", "TransparencyRequirements", "AIEthics", "AISystem", "ComputerVision", "ConvolutionalNeuralNetwork", "DataPipeline", "Explainability", "ExplainableAI", "FederatedLearning", "GenerativeAI", "GenerativeModel", "MachineLearning", "MachineLearning" ] }, { "id": "ai-application", "title": "AI Application", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "AI Application is the taxonomy hub for deployed uses of artificial intelligence across domains \u2014 encompassing conversational AI, computer vision, generative AI, healthcare AI, autonomous systems, fraud detection, and multimodal AI. It bridges AI research and real-world deployment, grouping systems by the function they perform rather than the techniques they use.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ai-application", "labels": [ "AI Application", "AI Application Domain" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-governance-and-ethics", "title": "AI Governance and Ethics", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "A taxonomy hub encompassing the policies, frameworks, principles, and regulatory mechanisms that govern the responsible development, deployment, and oversight of artificial intelligence systems. This category covers ethical guidelines, accountability structures, safety requirements, bias mitigation, transparency mandates, and legal compliance obligations that collectively shape how AI systems are built and governed.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-governance-and-ethics", "labels": [ "AI Governance and Ethics" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-model-architecture", "title": "AI Model Architecture", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Taxonomy hub encompassing the structural designs, layer configurations, and computational graphs used to construct machine-learning models, including transformer architectures, convolutional networks, recurrent networks, diffusion models, and mixture-of-experts designs. Architecture choices determine model capacity, training efficiency, and deployment characteristics.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-model-architecture", "labels": [ "AI Model Architecture" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-research-area", "title": "AI Research Area", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Taxonomy hub classifying the principal sub-disciplines and research programmes within artificial intelligence, including machine learning, natural language processing, computer vision, and AI safety. Each constituent area represents an active community with distinct methodologies, benchmarks, and applications that collectively advance the broader AI field.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:ai-research-area", "labels": [ "AI Research Area" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "ai-technique", "title": "AI Technique", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Taxonomy hub encompassing the principal methods and algorithmic approaches used in artificial intelligence, spanning symbolic, statistical, and neural paradigms. AI Technique organises the broad spectrum of techniques \u2014 from classical reasoning and search to contemporary deep learning and generative models \u2014 that implement intelligent behaviour in systems.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:ai-technique", "labels": [ "AI Technique" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "based-on", "title": "based on", "domain": "robotics", "domain_name": "Robotics", "definition": "\"Based on\" is a provenance and derivation relation used in robotics standards and quality management frameworks to indicate that a system, design, or specification is derived from or conforms to an authoritative reference standard such as ISO 9001 or functional-safety norms. Within the robotics domain it records the normative lineage of robot software architectures, safety certifications, and performance benchmarks relative to their governing standards.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:based-on", "labels": [ "based on" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [ "Safety Standard", "Robotics" ] }, { "id": "bc-cryptographic-primitive", "title": "Cryptographic Primitive", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy hub for the foundational cryptographic building blocks underpinning blockchain systems, including hash functions, digital signatures, asymmetric encryption, zero-knowledge proofs, and cryptographic commitments. These primitives collectively provide the security properties of integrity, authenticity, and privacy that make distributed ledger technology trustworthy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-cryptographic-primitive", "labels": [ "Cryptographic Primitive (Blockchain)", "Cryptographic Primitives" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "bc-defi-and-economics", "title": "DeFi and Economics", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy category grouping decentralised finance protocols and blockchain-native economic mechanisms, including automated market makers, liquidity pools, yield farming, stablecoins, tokenisation, DAOs, and the incentive structures that govern on-chain economies.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-defi-and-economics", "labels": [ "DeFi and Economics" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "bc-governance-and-regulation", "title": "Governance and Regulation", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy hub covering the legal, policy, and organisational frameworks that govern blockchain and distributed ledger deployments. This category spans compliance obligations, decentralised governance mechanisms, regulatory technology, and the evolving global ruleset for digital assets and smart contracts.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:bc-governance-and-regulation", "labels": [ "Governance and Regulation" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "bc-network-component", "title": "Network Component", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy hub for the node and network-layer components of a blockchain system. Network components are the participant entities and communication infrastructure through which blocks and transactions are propagated, validated, and stored across a distributed ledger.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-network-component", "labels": [ "Network Component (Blockchain)" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "bc-protocol-and-consensus", "title": "Protocol and Consensus", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy category grouping all protocol-level and consensus-layer concepts within the blockchain domain, including proof-of-work and proof-of-stake variants, Byzantine fault-tolerant algorithms, finality models, and the rules governing block production and chain selection.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-protocol-and-consensus", "labels": [ "Protocol and Consensus" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "bc-token-and-asset", "title": "Token and Asset", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Taxonomy hub for blockchain-based digital tokens and on-chain asset representations. Covers fungible tokens, non-fungible tokens, stablecoins, security tokens, governance tokens, and the standards and mechanisms by which digital value is issued, transferred, and managed on distributed ledgers.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:bc-token-and-asset", "labels": [ "Token and Asset" ], "is_subclass_of": [ "Blockchain" ], "wikilinks": [] }, { "id": "blockchain-core-concepts", "title": "blockchain core concepts", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The foundational technical concepts of blockchain and distributed ledger technology, encompassing consensus mechanisms (PoW, PoS), cryptographic primitives (public-key encryption, digital signatures, hash functions, Merkle trees), smart contracts, tokens, and governance frameworks. These concepts collectively enable trustless transaction validation, DeFi, NFTs, and metaverse digital economies.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:blockchain-core-concepts", "labels": [ "blockchain core concepts" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [ "Consensus", "CryptographyFundamentals", "DecentralisedFinance", "LayerTwo", "MetaverseEconomy", "NFT", "NFT", "PrivacyEnhancements", "PublicKeyEncryption", "Scalability", "BlockchainGovernance", "Blockchain Oracle", "ConsensusAlgorithm", "CrossChainBridge", "Cryptocurrency", "DAO", "DataStructure", "DecentralizedStorage", "DigitalSignature", "HashFunction" ] }, { "id": "cat-ai-infrastructure", "title": "AI Infrastructure", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Intermediate category of the Artificial Intelligence domain covering the compute, data, tooling, and platform infrastructure that supports training, fine-tuning, serving, and orchestration of machine-learning and foundation-model systems \u2014 including accelerators, distributed-training frameworks, model registries, and inference platforms.", "entityType": "Class", "qualityScore": 0.85, "maturity": "established", "iri": "urn:ngm:class:cat-ai-infrastructure", "labels": [ "AI Infrastructure (Artificial Intelligence)", "AI Infrastructure Providers" ], "is_subclass_of": [ "Artificial Intelligence" ], "wikilinks": [] }, { "id": "community", "title": "community", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "A community is a group of people who interact around a shared interest, project, or platform, contributing to its development, governance, or use.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:community", "labels": [ "community", "Community", "Community Moderation" ], "is_subclass_of": [ "Communication Technology", "owl:Thing" ], "wikilinks": [ "open source", "Governance", "owl:Thing" ] }, { "id": "continuous-improvement", "title": "continuous improvement", "domain": "ai", "domain_name": "Ai", "definition": "Continuous improvement is an iterative organisational and engineering philosophy in which processes, products, and systems are systematically and incrementally refined over time through structured cycles of measurement, analysis, experimentation, and implementation, rather than through large-scale periodic overhauls. Rooted in Japanese manufacturing philosophy (kaizen), it was formalised in quality management frameworks including the Plan-Do-Check-Act cycle, ISO 9001, and Lean methodologies, and has since been adopted throughout software engineering via DevOps, agile retrospectives, and MLOps pipelines. The practice relies on feedback loops that surface inefficiencies, defects, or opportunities close to their point of origin, enabling rapid corrective action and cumulative quality gains.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:continuous-improvement", "labels": [ "continuous improvement", "Continuous Improvement" ], "is_subclass_of": [ "Software Development Process" ], "wikilinks": [] }, { "id": "d-yd-x", "title": "dYdX", "domain": "blockchain", "domain_name": "Blockchain", "definition": "dYdX is a decentralised exchange focused on perpetual-futures trading, allowing users to take leveraged long and short positions on cryptocurrencies. Earlier versions operated on Ethereum using a layer-two scaling solution with an off-chain order book and on-chain settlement, while a later version migrated to a purpose-built application-specific blockchain in the Cosmos ecosystem. It combines the order-book trading experience of centralised venues with non-custodial settlement.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:d-yd-x", "labels": [ "dYdX" ], "is_subclass_of": [ "Decentralised Finance", "Decentralised Finance Domain" ], "wikilinks": [ "Order Book", "Smart Contract", "Perpetual Futures", "Leveraged Trading", "GMX", "Cosmos", "Decentralised Finance Domain" ] }, { "id": "dc-communication", "title": "Communication Technology", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Taxonomy category covering the network protocols, media channels, and software systems that enable distributed teams to communicate across space and time, spanning real-time video conferencing, asynchronous messaging, WebRTC transport, MQTT messaging, and immersive communication modalities.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dc-communication", "labels": [ "Communication Technology" ], "is_subclass_of": [ "Distributed Collaboration" ], "wikilinks": [] }, { "id": "dc-protocol-and-infra", "title": "Protocol and Infrastructure", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Taxonomy hub for the communication protocols, data-interchange formats, and underlying infrastructure that enable distributed collaboration. Encompasses decentralised messaging protocols, identity systems, content-addressed storage, version control, and the networking substrate on which remote collaborative workflows operate.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dc-protocol-and-infra", "labels": [ "Protocol and Infrastructure" ], "is_subclass_of": [ "Distributed Collaboration" ], "wikilinks": [] }, { "id": "dc-telepresence", "title": "Telepresence", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Telepresence is the distributed-collaboration taxonomy hub covering technologies that create a convincing sense of physical co-presence across remote locations \u2014 spanning robotic telepresence, haptic-feedback telepresence, virtual-reality telepresence, and social presence theory. It is a peer category to Workspace Tools and Communication Technology within the distributed collaboration domain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dc-telepresence", "labels": [ "Telepresence (Distributed Collaboration)", "5G Telepresence", "Immersive Telepresence", "Presence and Telepresence" ], "is_subclass_of": [ "Distributed Collaboration" ], "wikilinks": [] }, { "id": "dc-workspace-tools", "title": "Workspace Tools", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Workspace Tools is the distributed-collaboration taxonomy hub for software and platform technologies that support remote and hybrid team productivity \u2014 including collaboration platforms, shared whiteboards, asynchronous video, meeting AI assistants, and immersive workspaces. It is a peer category to Telepresence and Communication Technology within the distributed collaboration domain.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:dc-workspace-tools", "labels": [ "Workspace Tools" ], "is_subclass_of": [ "Distributed Collaboration" ], "wikilinks": [] }, { "id": "did-nostr", "title": "did:nostr", "domain": "security", "domain_name": "Security", "definition": "did:nostr is a W3C DID Core-conformant decentralised identifier method that represents a Nostr public key as a universally resolvable DID of the form did:nostr:{pubkey}, where {pubkey} is the 64-character hex secp256k1 key already used by the Nostr protocol. Because the identifier is the key, no registration, server, or fee is required: anyone with a Nostr keypair already has a DID. Resolution degrades gracefully across three layers \u2014 HTTP fetch of a .well-known/did/nostr/{pubkey}.json document, relay-enhanced resolution from Nostr kind 0 (metadata) and kind 10002 (relay list) events, and an offline fallback that derives a minimal valid DID document from the key alone. Keys are expressed in W3C Multikey/Multibase form (secp256k1 multicodec), letting Nostr identities authenticate to any DID-aware system, including Solid login, SSO, and verifiable-credential flows.", "entityType": "Class", "qualityScore": 0.74, "maturity": "emerging", "iri": "urn:ngm:class:did-nostr", "labels": [ "did:nostr", "DID Nostr", "DID-Nostr", "nostr-did" ], "is_subclass_of": [ "DID Method" ], "wikilinks": [] }, { "id": "e-idas-2-0", "title": "eIDAS 2.0", "domain": "governance", "domain_name": "Governance", "definition": "eIDAS 2.0 (Regulation EU 2024/1183) is the amended EU framework for electronic identification, authentication, and trust services, entered into force in May 2024 as a successor to the original eIDAS Regulation (EU 910/2014). Its cornerstone is the European Digital Identity Wallet (EUDIW), a mandatory public-infrastructure obligation requiring each EU member state to offer citizens, residents, and businesses a certified wallet for storing qualified electronic attestations of attributes, verifiable credentials, and qualified electronic signatures. The regulation obliges online platforms with more than 45 million EU users to accept the wallet as an authentication mechanism, dismantling private-sector identity lock-in and creating a public-interest counterweight to data concentration. eIDAS 2.0 aligns with W3C Verifiable Credentials, Decentralised Identifiers, OpenID for Verifiable Credential Issuance (OID4VCI), and OpenID for Verifiable Presentations (OID4VP), embedding open interoperability requirements and selective-disclosure privacy protections into binding EU law.", "entityType": "Class", "qualityScore": 0.74, "maturity": "established", "iri": "urn:ngm:class:e-idas-2-0", "labels": [ "eIDAS 2.0", "eIDAS 2" ], "is_subclass_of": [ "Digital Identity Framework" ], "wikilinks": [] }, { "id": "eidas-regulation", "title": "eIDAS Regulation", "domain": "security", "domain_name": "Security", "definition": "The eIDAS Regulation is the European Union framework on electronic identification and trust services for electronic transactions in the internal market. It establishes legal recognition and cross-border interoperability for electronic identification schemes and for trust services such as electronic signatures, seals, timestamps and website authentication. Its revision introduces the European Digital Identity Wallet, extending qualified trust services to verifiable digital credentials.", "entityType": "Class", "qualityScore": 0.62, "maturity": "mature", "iri": "urn:ngm:class:eidas-regulation", "labels": [ "eIDAS Regulation" ], "is_subclass_of": [ "Regulatory Compliance" ], "wikilinks": [] }, { "id": "e-idas", "title": "eIDAS", "domain": "governance", "domain_name": "Governance", "definition": "A European Union regulation establishing a framework for electronic identification and trust services for electronic transactions across the internal market, defining assurance levels, legal effects for electronic signatures, seals, timestamps and certificates, and mandating mutual recognition of notified national identity schemes.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:e-idas", "labels": [ "eIDAS", "European Commission eIDAS", "eIDAS Compliance", "eIDAS Regulation" ], "is_subclass_of": [ "Regulatory Framework" ], "wikilinks": [ "Digital Signature", "Public Key Infrastructure", "Digital Identity", "Identity Verification", "Certificate Authority", "Regulatory Framework" ] }, { "id": "ecash", "title": "ecash", "domain": "blockchain", "domain_name": "Blockchain", "definition": "A privacy-preserving digital cash system based on David Chaum's blind-signature cryptography, enabling bearer-token payments in which the issuing mint cannot link individual transactions to users. Modern implementations such as Cashu and Fedimint build Chaumian eCash on top of Bitcoin and the Lightning Network, enabling censorship-resistant micropayments for human and autonomous-agent use.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:ecash", "labels": [ "ecash", "Ecash Token" ], "is_subclass_of": [ "Token and Asset" ], "wikilinks": [ "Agent Coordination", "AI Agents", "Anonymity Protocols", "Anonymous Transactions", "Ark Protocol", "Autonomous Systems", "Bitcoin Privacy", "Blind Signatures", "BOLT", "Byzantine Tolerance", "Chaumian Mint", "CoinJoin", "Confidential Transactions", "Consensus Mechanisms", "Cryptographic Protocols", "David Chaum", "Decentralized Commerce", "Digital Cash", "Economic Networks", "Fedimint" ] }, { "id": "fast-ai", "title": "fast.ai", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "fast.ai is a research lab, online course series, and open-source Python deep learning library founded by Jeremy Howard (former president and chief scientist of Kaggle, founder of FastMail and Optimal Decisions Group) and Rachel Thomas (UCSF / University of San Francisco data ics researcher), buil...", "entityType": "Class", "qualityScore": 0.52, "maturity": "established", "iri": "urn:ngm:class:fast-ai", "labels": [ "fast.ai" ], "is_subclass_of": [ "AI Application", "Deep Learning Library", "PyTorch Ecosystem Project", "Open Source ML Framework", "Educational Platform", "Applied AI Research Lab" ], "wikilinks": [ "1cycle Learning Rate Schedule", "ACL", "Adam Optimiser", "Answer.AI", "Answer.AI Website", "Applied AI Research Lab", "Audio Classification", "Beginner-Friendly Deep Learning", "Brown et al. 2020 GPT-3", "Callback System", "Claudette", "Code-First Education", "Collaborative Filtering", "Convolutional Neural Networks", "Cosine Annealing", "CUDA GPU", "DataBlock API", "Deep Learning Library", "Democratised Deep Learning", "Dettmers Belkada Howard 2024 FSDP-QLoRA" ] }, { "id": "g-rpc", "title": "gRPC", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "gRPC (gRPC Remote Procedure Call) is an open-source, high-performance remote procedure call framework developed by Google and released in 2015, built on HTTP/2 transport and Protocol Buffers as the interface definition language and serialisation format. It supports four communication patterns\u2014unary, server-streaming, client-streaming, and bidirectional streaming\u2014enabling efficient, strongly typed, low-latency communication between services in polyglot distributed systems. gRPC generates client and server stubs in over a dozen programming languages from a single .proto service definition, making it the dominant choice for internal microservices communication in cloud-native architectures. Its binary encoding and multiplexed HTTP/2 connections deliver significantly lower overhead than REST/JSON at high throughput.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:g-rpc", "labels": [ "gRPC", "gRPC Specification", "gRPC Transport" ], "is_subclass_of": [ "Communication Protocol" ], "wikilinks": [] }, { "id": "gaming", "title": "gaming", "domain": "metaverse", "domain_name": "Metaverse", "definition": "Gaming is the activity of playing electronic games and the industry that produces them. It spans hardware, software, online services, and competitive play.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:gaming", "labels": [ "gaming", "Dynamic Gaming", "Gaming", "Multiplayer Gaming", "Traditional Gaming" ], "is_subclass_of": [ "Game Development" ], "wikilinks": [ "Game Engine", "GameFi", "Virtual World", "Game Development", "https://en.wikipedia.org/wiki/Video_game", "https://en.wikipedia.org/wiki/Video_game_industry" ] }, { "id": "gl-tf-3-d-file-format", "title": "glTF (3D File Format)", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A royalty-free, open-standard 3D asset transmission format developed by Khronos Group that efficiently specifies scene structure, geometry, materials, animations, and other properties for real-time rendering.", "entityType": "Class", "qualityScore": 0.35, "maturity": "established", "iri": "urn:ngm:class:gl-tf-3-d-file-format", "labels": [ "glTF (3D File Format)" ], "is_subclass_of": [ "Standards and Interoperability" ], "wikilinks": [ "3D Asset Exchange", "3D Content Pipeline", "Animation Channels", "Asset Interchange System", "Base64 Encoding", "Content Interoperability", "Cross-Platform Compatibility", "EWG/MSF taxonomy", "JSON Schema", "Khronos Group", "Material Definition", "Mesh Data", "MIME Types", "Runtime Rendering", "Texture References", "URI Specification", "Binary Buffer", "Binary Encoding", "CreativeMediaDomain", "DataLayer" ] }, { "id": "gl-tf", "title": "glTF", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "glTF (GL Transmission Format) is an open royalty-free standard from the Khronos Group for the efficient transmission and loading of 3D scenes and models, defining geometry, materials, animation and scene structure in a compact runtime format optimised for direct upload to graphics APIs.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:gl-tf", "labels": [ "glTF", "glTF Asset Specification", "glTF Format" ], "is_subclass_of": [ "3D File Format" ], "wikilinks": [ "3D Model", "Material Definition", "Asset Interoperability", "Real-Time Rendering", "WebXR", "3D File Format" ] }, { "id": "i-proov", "title": "iProov", "domain": "security", "domain_name": "Security", "definition": "iProov is a British biometric technology company founded in 2011 that specialises in remote face verification and liveness detection for digital identity assurance. Its core technology, Genuine Presence Assurance, uses controlled illumination sequences to verify that a real, live person is present during an online authentication event, distinguishing genuine users from spoofing attacks using photographs, videos, or synthetic deepfakes. iProov is deployed by governments, financial institutions, and healthcare providers to underpin Know Your Customer (KYC) onboarding, border control systems, and high-assurance remote authentication. The platform integrates with broader identity document verification and identity orchestration frameworks, providing a biometric binding layer within electronic identity (eID) and digital wallet schemes.", "entityType": "Class", "qualityScore": 0.75, "maturity": "established", "iri": "urn:ngm:class:i-proov", "labels": [ "iProov" ], "is_subclass_of": [ "Biometric Authentication" ], "wikilinks": [ "Facial Recognition", "Identity Verification System", "Onfido", "Biometric Authentication" ] }, { "id": "implementation-examples", "title": "implementation examples", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Concrete, production-oriented code samples and reference designs demonstrating how theoretical concepts translate into functioning systems. In metaverse and blockchain contexts these span smart contract patterns (DeFi AMMs, NFT contracts, DAO governance), gas optimisation techniques, security patterns such as checks-effects-interactions, and upgradeability proxies. They serve as educational resources and accelerators for developers, reducing deployment risk through audited, annotated reference implementations.", "entityType": "Class", "qualityScore": 0.4, "maturity": "emerging", "iri": "urn:ngm:class:implementation-examples", "labels": [ "implementation examples" ], "is_subclass_of": [ "Content and Assets" ], "wikilinks": [ "BestPractices", "CrossChainBridging", "DAOGovernance", "DeFiProtocols", "MetaversePlatformAPIs", "NFTContracts", "PrivacyPreservingProtocols", "SmartContractExamples", "MetaverseDomain" ] }, { "id": "infra-computing-and-cloud", "title": "Computing and Cloud", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Taxonomy hub covering computing paradigms and cloud service models within the infrastructure domain, including distributed systems, edge computing, serverless architectures, and specialised hardware accelerators. It organises concepts ranging from physical compute infrastructure to platform and software services delivered over networks.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:infra-computing-and-cloud", "labels": [ "Computing and Cloud" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "infra-data-management", "title": "Data Management", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Taxonomy hub covering the storage, movement, governance, and quality of data assets within the infrastructure domain, spanning databases, data lakes, ETL pipelines, metadata registries, and data governance frameworks. It provides the structural substrate on which analytics, machine learning, and application workloads depend.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:infra-data-management", "labels": [ "Data Management (Infrastructure)", "Health Data Management" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "infra-legal-and-regulatory", "title": "Legal and Regulatory", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Taxonomy hub for legal and regulatory concepts within the infrastructure domain, covering compliance frameworks, data protection law, intellectual property, AI regulation, and governance structures that constrain and guide technology deployment.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:infra-legal-and-regulatory", "labels": [ "Legal and Regulatory" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "infra-network-and-comms", "title": "Network and Communication", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Taxonomy hub covering the protocols, architectures, and physical/virtual infrastructure that enable data exchange between systems and devices. This category spans networking layers, communication protocols, edge computing, distributed systems, and the transport fabric underpinning cloud and IoT deployments.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:infra-network-and-comms", "labels": [ "Network and Communication", "Network and Communications" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "infra-security-and-identity", "title": "Security and Identity", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "A taxonomy hub encompassing the cryptographic primitives, authentication mechanisms, access control systems, and identity management frameworks that protect infrastructure components and verify principals. This category includes digital signatures, certificate authorities, encryption protocols, zero-trust architectures, and decentralised identity systems that collectively secure networked systems and digital interactions.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:infra-security-and-identity", "labels": [ "Security and Identity" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "infra-software-engineering", "title": "Software Engineering", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Taxonomy hub for software engineering concepts within the infrastructure domain, encompassing the practices, processes, tools, and architectures used to design, build, test, and maintain software systems at scale.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:infra-software-engineering", "labels": [ "Software Engineering (Infrastructure)" ], "is_subclass_of": [ "Infrastructure" ], "wikilinks": [] }, { "id": "k-anonymity-in-datasets", "title": "k-Anonymity in Datasets", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "k-Anonymity is a dataset privacy property that guarantees each record is indistinguishable from at least k-1 other records with respect to quasi-identifier attributes\u2014fields such as age, gender, and postal code that can be combined to re-identify individuals. Anonymization is achieved through generalisation (replacing specific values with broader categories) and suppression (removing highly identifying records), producing equivalence classes where all members share identical quasi-identifier values.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:k-anonymity-in-datasets", "labels": [ "k-Anonymity in Datasets", "K-Anonymity", "k-Anonymity" ], "is_subclass_of": [ "AI Technique", "Privacy Preserving Analytics" ], "wikilinks": [ "Li et al. (2007)", "Machanavajjhala et al. (2007)", "Sweeney (2002)", "AIEthicsDomain", "ConceptualLayer" ] }, { "id": "libp2p", "title": "libp2p", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "libp2p is a modular network stack and peer-to-peer networking framework originally extracted from the IPFS project that enables developers to build decentralised applications with configurable transport, security, and protocol multiplexing layers. It abstracts over TCP, QUIC, WebSocket, and WebRTC transports, applies encryption via Noise Protocol or TLS 1.3, multiplexes streams with Yamux or mplex, and provides peer discovery, routing, and publish-subscribe messaging as composable modules. libp2p powers the peer layer of Ethereum 2.0, Filecoin, Polkadot, and hundreds of other decentralised systems, providing a battle-tested foundation for production-grade peer-to-peer networking.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:libp2p", "labels": [ "libp2p" ], "is_subclass_of": [ "Peer-to-Peer Network" ], "wikilinks": [] }, { "id": "m-bert", "title": "mBERT", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "Multilingual BERT (mBERT) is a variant of the BERT encoder pre-trained jointly on Wikipedia text in 104 languages using masked language modelling and next-sentence prediction objectives. By sharing a single vocabulary and model weights across all languages, mBERT learns cross-lingually aligned representations that support zero-shot cross-lingual transfer: a model fine-tuned for a task in one language can be applied directly to another language without that language's labelled data. mBERT is the foundational benchmark for evaluating multilingual language understanding and cross-lingual transfer learning.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:m-bert", "labels": [ "mBERT" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "m-t5", "title": "mT5", "domain": "artificial-intelligence", "domain_name": "Artificial Intelligence", "definition": "mT5 (Multilingual T5) is a massively multilingual pre-trained text-to-text transformer model developed by Google Research, covering 101 languages through pre-training on the mC4 multilingual Common Crawl corpus. It extends the T5 architecture's unified text-to-text framework to multilingual settings, treating all NLP tasks as sequence-to-sequence problems. mT5 enables strong cross-lingual transfer and zero-shot performance on low-resource languages, making it a foundational model for multilingual NLP applications including translation, question answering, and information retrieval.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:m-t5", "labels": [ "mT5" ], "is_subclass_of": [ "AI Model Architecture" ], "wikilinks": [ "MetaverseDomain" ] }, { "id": "metaverse-core-concepts", "title": "metaverse core concepts", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Metaverse core concepts is an index class grouping the foundational building blocks of persistent shared virtual spaces: immersive rendering technologies (Augmented Reality, Spatial Computing, Haptic Feedback), virtual environments (Virtual World, Social VR, Digital Twin), digital identity and economic primitives (Digital Avatar, Virtual Asset, Virtual Economy), and cross-platform integration enablers (Interoperability). Together these concepts define the design space for metaverse platforms and standards such as those developed by the Metaverse Standards Forum.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:metaverse-core-concepts", "labels": [ "metaverse core concepts" ], "is_subclass_of": [ "Platform and Environment" ], "wikilinks": [ "Decentraland", "The Sandbox", "Augmented Reality", "AugmentedReality", "Digital Avatar", "DigitalAvatar", "Digital Twin", "DigitalTwin", "Haptic Feedback", "HapticFeedback", "Interoperability", "MetaverseDomain", "Social VR", "SocialVR", "Spatial Computing", "SpatialComputing", "Virtual Asset", "VirtualAsset", "Virtual Economy", "VirtualEconomy" ] }, { "id": "micro-ros", "title": "micro-ROS", "domain": "robotics", "domain_name": "Robotics", "definition": "micro-ROS is a framework that brings the ROS 2 programming model and communication to resource-constrained microcontrollers, allowing embedded devices to participate as first-class nodes in a ROS graph.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:micro-ros", "labels": [ "micro-ROS" ], "is_subclass_of": [ "Robot Operating System" ], "wikilinks": [ "Embedded Systems", "Real-Time Operating System", "Real-Time Control", "ROS", "DDS Middleware", "Robot Operating System" ] }, { "id": "performance", "title": "performance", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "Performance is a measure of how efficiently a system completes its work, commonly expressed through throughput, latency, and resource usage.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:performance", "labels": [ "performance", "Performance", "Performance Analytics", "Performance Engineering", "System Performance" ], "is_subclass_of": [ "Computing and Cloud", "owl:Thing" ], "wikilinks": [ "Observability", "owl:Thing" ] }, { "id": "rb-0007-collaborative-robot", "title": "rb 0007 collaborative robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A collaborative robot (cobot) is an industrial robot designed and certified to work in direct physical proximity with human operators within a shared workspace, without requiring conventional hard safety guarding. Defined under ISO/TS 15066, cobots implement one or more of four collaborative operation modes \u2014 safety-rated monitored stop, hand guiding, speed-and-separation monitoring, and power-and-force limiting \u2014 each constraining robot behaviour so that human contact does not result in injury. Cobots typically feature lightweight, rounded structures, intrinsic force/torque sensing, and redundant safety-rated control architectures compliant with ISO 10218-1/-2.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0007-collaborative-robot", "labels": [ "rb 0007 collaborative robot" ], "is_subclass_of": [ "Robot Type", "Industrial Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0008-autonomous-robot", "title": "rb 0008 autonomous robot", "domain": "robotics", "domain_name": "Robotics", "definition": "An autonomous robot is a robotic system capable of performing tasks in unstructured environments without continuous human intervention, relying on onboard sensing, perception, decision-making, and actuation. Autonomy spans a spectrum from simple programmed responses to full cognitive agency; key enabling technologies include SLAM, path planning, machine learning-based perception, and safety-certified control architectures.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0008-autonomous-robot", "labels": [ "rb 0008 autonomous robot" ], "is_subclass_of": [ "Robot Type" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0012-wheeled-mobile-robot", "title": "rb 0012 wheeled mobile robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A wheeled mobile robot (WMR) is a ground-based autonomous or semi-autonomous robot that uses wheels for locomotion. WMRs range from differential-drive platforms to omnidirectional holonomic designs; they rely on odometry for dead-reckoning, SLAM for map building, and path-planning algorithms for autonomous navigation in structured or semi-structured environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0012-wheeled-mobile-robot", "labels": [ "rb 0012 wheeled mobile robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0017-rescue-robot", "title": "rb 0017 rescue robot", "domain": "robotics", "domain_name": "Robotics", "definition": "A rescue robot is a mobile robotic system specifically designed or adapted for deployment in disaster, emergency, or hazardous environments where direct human access is dangerous or impossible. Rescue robots perform tasks such as victim search-and-detection, structural reconnaissance, rubble traversal, gas sensing, and teleoperated manipulation of debris. They typically integrate rugged locomotion systems (tracked, legged, or serpentine), multiple sensor modalities (thermal, depth camera, LIDAR, gas detectors), and semi-autonomous navigation with human-on-the-loop teleoperation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0017-rescue-robot", "labels": [ "rb 0017 rescue robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0018-inspection-robot", "title": "rb 0018 inspection robot", "domain": "robotics", "domain_name": "Robotics", "definition": "An inspection robot is a mobile robotic system designed to autonomously or semi-autonomously navigate environments that are hazardous, inaccessible, or inefficient for human workers, in order to assess the structural, mechanical, or operational condition of infrastructure, equipment, or facilities. Equipped with sensor arrays (cameras, LiDAR, ultrasonic, thermal IR), inspection robots collect data for predictive maintenance, safety auditing, and non-destructive testing across industries such as oil and gas pipelines, power-grid infrastructure, bridges, and confined industrial spaces.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0018-inspection-robot", "labels": [ "rb 0018 inspection robot" ], "is_subclass_of": [ "Robot Type", "Mobile Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0019-exoskeleton-robot", "title": "rb 0019 exoskeleton robot", "domain": "robotics", "domain_name": "Robotics", "definition": "An exoskeleton robot is a wearable robotic device that attaches to the human body and augments, supports, or replaces the wearer's physical capabilities. Exoskeletons may be powered (actuated) or passive (spring-and-damper based), and are used for physical rehabilitation after stroke or spinal cord injury, industrial worker fatigue reduction, and military load-carrying augmentation. Safety is paramount because the robot shares the mechanical structure with a human; ISO 13482 governs safety requirements for personal care robots including exoskeletons.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0019-exoskeleton-robot", "labels": [ "rb 0019 exoskeleton robot" ], "is_subclass_of": [ "Robot Type", "Exoskeleton Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0021-robot-kinematics", "title": "rb 0021 robot kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot kinematics is the study of the geometry and motion of robot mechanisms \u2014 particularly manipulator arms \u2014 without regard to the forces or torques that cause motion. It encompasses forward kinematics (computing end-effector pose from joint configurations), inverse kinematics (computing joint configurations for a desired end-effector pose), Jacobian analysis relating joint velocities to Cartesian velocities, and singularity analysis. Kinematic models are foundational inputs to motion planning, trajectory generation, and control law design.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0021-robot-kinematics", "labels": [ "rb 0021 robot kinematics" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0022-robot-dynamics", "title": "rb 0022 robot dynamics", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot dynamics is the study of the relationship between the forces and torques applied to a robot's joints and links and the resulting motion of the robot. It encompasses forward dynamics (computing accelerations from applied torques), inverse dynamics (computing required torques to achieve a desired motion), and the derivation of equations of motion via Newton-Euler or Lagrangian formulations. Dynamic models are essential for model-based controllers such as computed-torque control, optimal control, and trajectory optimisation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0022-robot-dynamics", "labels": [ "rb 0022 robot dynamics" ], "is_subclass_of": [ "Actuation and Control", "Robot Dynamics" ], "wikilinks": [ "Death of the Internet", "Google", "RoboticsDomain" ] }, { "id": "rb-0023-degrees-of-freedom", "title": "rb 0023 degrees of freedom", "domain": "robotics", "domain_name": "Robotics", "definition": "Degrees of Freedom (DOF) in robotics denotes the number of independent parameters required to fully specify the configuration of a robot mechanism. Each revolute or prismatic joint contributes one DOF; a 6-DOF manipulator can achieve arbitrary position and orientation in three-dimensional space. The number of DOF constrains the robot's reachable workspace, its dexterity, and the complexity of its kinematic and dynamic models.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0023-degrees-of-freedom", "labels": [ "rb 0023 degrees of freedom", "Degrees of Freedom" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0024-workspace", "title": "rb 0024 workspace", "domain": "robotics", "domain_name": "Robotics", "definition": "The workspace of a robot manipulator is the total volume of space that the end-effector can reach, given the joint range limits of all links. The reachable workspace encompasses every point the tool-centre-point (TCP) can attain in at least one orientation, while the dexterous workspace is the subset reachable in all possible orientations. Workspace volume, shape, and dexterity distribution are primary design criteria for robot selection and cell layout, and must account for self-collisions, payload, and safety exclusion zones.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0024-workspace", "labels": [ "rb 0024 workspace" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0026-robot-joint", "title": "rb 0026 robot joint", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot joint is a mechanical articulation connecting two adjacent robot links that permits one or more degrees of relative motion\u2014translational (prismatic) or rotational (revolute). Robot joints are the fundamental kinematic elements that determine a manipulator's workspace and degrees of freedom; their dynamic properties, including inertia, backlash, and compliance, critically influence both control precision and safe interaction with the environment.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0026-robot-joint", "labels": [ "rb 0026 robot joint" ], "is_subclass_of": [ "Actuation and Control", "Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0027-robot-link", "title": "rb 0027 robot link", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot link is a rigid (or semi-rigid) structural body that forms one segment of a robot's kinematic chain, connecting adjacent joints and transmitting forces and torques between them. Links define the geometry of the robot's workspace by their length, mass, and inertial properties, and together with joints they constitute the Denavit-Hartenberg representation used in forward and inverse kinematics. Link stiffness, material choice, and mass distribution directly affect dynamic performance, vibration characteristics, and safety in human-robot interaction.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0027-robot-link", "labels": [ "rb 0027 robot link" ], "is_subclass_of": [ "Actuation and Control", "Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0028-forward-kinematics", "title": "rb 0028 forward kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Forward kinematics (FK) is the mathematical process of computing the position and orientation of a robot's end effector in Cartesian space given a known set of joint angles or displacements. Using a chain of homogeneous transformation matrices (typically expressed using Denavit-Hartenberg parameters), FK provides a unique, computationally deterministic mapping from joint space to task space. It is used in motion planning, visualisation, and safety monitoring to determine where the robot tool tip is at any instant, and underpins workspace analysis, collision checking, and trajectory verification.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0028-forward-kinematics", "labels": [ "rb 0028 forward kinematics" ], "is_subclass_of": [ "Actuation and Control", "Kinematics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0029-inverse-kinematics", "title": "rb 0029 inverse kinematics", "domain": "robotics", "domain_name": "Robotics", "definition": "Inverse kinematics (IK) is the computational process of determining the joint angles or actuator lengths required to place a robot's end-effector at a desired position and orientation in task space. Unlike forward kinematics, IK is typically under-determined or over-determined and requires iterative numerical solvers, analytical solutions, or Jacobian-based methods; it is a foundational component of motion planning, manipulation, and trajectory execution in robotic systems.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0029-inverse-kinematics", "labels": [ "rb 0029 inverse kinematics" ], "is_subclass_of": [ "Actuation and Control", "Kinematics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0030-jacobian-matrix", "title": "rb 0030 jacobian matrix", "domain": "robotics", "domain_name": "Robotics", "definition": "The Jacobian matrix is a mathematical mapping that relates joint-space velocities to Cartesian end-effector velocities for a robotic manipulator. It is central to differential kinematics, velocity control, force-torque transformation, and singularity analysis; its pseudo-inverse enables computation of joint velocities from desired Cartesian motions in inverse kinematics.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0030-jacobian-matrix", "labels": [ "rb 0030 jacobian matrix" ], "is_subclass_of": [ "Actuation and Control", "Kinematics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0031-singularity", "title": "rb 0031 singularity", "domain": "robotics", "domain_name": "Robotics", "definition": "A kinematic singularity is a robot configuration in which the Jacobian matrix loses rank, causing the manipulator to lose one or more degrees of freedom in Cartesian space. At a singularity, certain end-effector motions become unachievable regardless of joint velocity magnitudes, and inverse kinematics solutions either vanish or require unbounded joint speeds. Singularity avoidance and singularity-robust inverse kinematics (via damped least-squares) are critical for safe, continuous robot motion, particularly near workspace boundaries and for wrist configurations of six-axis arms.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0031-singularity", "labels": [ "rb 0031 singularity" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain", "Singularity" ] }, { "id": "rb-0032-manipulability", "title": "rb 0032 manipulability", "domain": "robotics", "domain_name": "Robotics", "definition": "Manipulability is a scalar or ellipsoid measure of a robotic manipulator's capacity to move and exert forces in arbitrary directions from a given joint configuration. Introduced by Tsuneo Yoshikawa, the manipulability measure w = sqrt(det(J\u00b7J\u1d40)) quantifies how far a configuration is from kinematic singularity: higher values indicate greater dexterity, while w = 0 indicates a singular configuration where motion in at least one direction is lost. Manipulability is used in motion planning, redundancy resolution, and task-space control to avoid singular regions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0032-manipulability", "labels": [ "rb 0032 manipulability", "Manipulability", "Manipulability Ellipsoid" ], "is_subclass_of": [ "Actuation and Control", "Robot Kinematics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0033-payload", "title": "rb 0033 payload", "domain": "robotics", "domain_name": "Robotics", "definition": "Payload in robotics refers to the maximum mass that a robot can carry or manipulate while maintaining its specified performance characteristics, including accuracy, repeatability, and speed. Payload capacity is a fundamental design parameter for robot arms, mobile robots, and drones, directly constraining the tools, parts, or cargo the robot can handle. It is typically quoted at the wrist or end-effector mounting face and may vary with configuration and speed; dynamic payload accounts for inertial forces during acceleration.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0033-payload", "labels": [ "rb 0033 payload" ], "is_subclass_of": [ "Actuation and Control", "Robot" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0034-repeatability", "title": "rb 0034 repeatability", "domain": "robotics", "domain_name": "Robotics", "definition": "Repeatability is a key performance metric of robot manipulators that quantifies the closeness of agreement between successive commanded returns to the same target position or pose, measured under identical conditions. It is formally defined in ISO 9283 as the radius of a sphere enclosing a specified percentage of attained positions from repeated attempts. Repeatability is distinct from accuracy \u2014 a robot may be highly repeatable but systematically offset from the commanded target \u2014 and is critical for applications such as assembly, welding, and precision manufacturing.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0034-repeatability", "labels": [ "rb 0034 repeatability" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0035-accuracy", "title": "rb 0035 accuracy", "domain": "robotics", "domain_name": "Robotics", "definition": "Robot accuracy is the closeness of agreement between a robot's commanded pose and its actual achieved pose, measured as the mean positional or orientational error across multiple repeated commanded positions. Defined in ISO 9283, it encompasses pose accuracy, path accuracy, and static compliance accuracy, and is distinct from repeatability, which measures the spread of repeated attempts at the same pose rather than deviation from the commanded value.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0035-accuracy", "labels": [ "rb 0035 accuracy" ], "is_subclass_of": [ "Actuation and Control", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "rb-0036-resolution", "title": "rb 0036 resolution", "domain": "robotics", "domain_name": "Robotics", "definition": "Resolution, as defined by ISO 8373, is the smallest increment of motion or measurement that a robot system can distinguish or command. For manipulators it typically refers to the minimum step size achievable in Cartesian or joint space; for sensors it denotes the smallest detectable change in the measured quantity. Resolution is distinct from accuracy and repeatability, and it places a fundamental lower bound on the precision of both positioning and perception tasks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0036-resolution", "labels": [ "rb 0036 resolution" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0037-dexterity", "title": "rb 0037 dexterity", "domain": "robotics", "domain_name": "Robotics", "definition": "Dexterity, in robotics, is the ability of a manipulator to achieve a broad range of end-effector orientations and positions within its workspace, particularly in the vicinity of a given point, without encountering kinematic singularities. It is quantified by measures such as the Jacobian condition number, manipulability ellipsoid volume, or isotropy index. High dexterity enables a robot arm to approach objects from many angles, reconfigure during a task, and avoid joint limits \u2014 properties critical for assembly, surgery, and unstructured manipulation tasks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0037-dexterity", "labels": [ "rb 0037 dexterity" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0038-compliance", "title": "rb 0038 compliance", "domain": "robotics", "domain_name": "Robotics", "definition": "In robotics, compliance is the mechanical property of a robot joint or end-effector describing its tendency to yield under applied forces or torques, quantified as the inverse of stiffness. Compliant behaviour is essential for safe human-robot interaction, allowing robots to absorb contact forces without rigid collision, and is actively exploited in impedance and admittance control strategies to achieve gentle, force-sensitive manipulation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0038-compliance", "labels": [ "rb 0038 compliance" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "Open Source", "RoboticsDomain" ] }, { "id": "rb-0039-stiffness", "title": "rb 0039 stiffness", "domain": "robotics", "domain_name": "Robotics", "definition": "Stiffness, in robotics, is the ratio of applied force (or torque) to the resulting displacement (or angular deflection) of a robot link, joint, or end-effector. High stiffness yields precise positioning at the cost of storing large elastic energy that can be hazardous in contact; low stiffness (compliance) absorbs impact and is preferred in human-robot collaboration. Variable-stiffness actuation and impedance control allow robots to modulate stiffness dynamically, trading accuracy against safety depending on task context.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0039-stiffness", "labels": [ "rb 0039 stiffness" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0040-backlash", "title": "rb 0040 backlash", "domain": "robotics", "domain_name": "Robotics", "definition": "Backlash in robotics is the lost motion in a mechanical transmission \u2014 the angular or linear displacement of the output element when the input reverses direction without producing corresponding output movement, caused by clearance gaps between mating gear teeth or other drive components. Backlash degrades positional accuracy and repeatability, introduces nonlinearity into the control loop, and can cause oscillation or chattering in feedback control systems. Minimising backlash is critical for precision robot joints, and it is typically reduced through the use of preloaded gear pairs, harmonic drives, or cycloidal transmissions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0040-backlash", "labels": [ "rb 0040 backlash" ], "is_subclass_of": [ "Actuation and Control", "Backlash" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0041-inertia", "title": "rb 0041 inertia", "domain": "robotics", "domain_name": "Robotics", "definition": "In robotics, inertia refers to the resistance of a robot's links and payload to changes in motion, quantified by the inertia tensor for rotational dynamics and mass for translational dynamics. Accurate inertia modelling is essential for dynamic control, trajectory planning, and compliance with power-and-force-limiting safety requirements, because high inertia directly increases the impact forces during unintended contact.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0041-inertia", "labels": [ "rb 0041 inertia" ], "is_subclass_of": [ "Actuation and Control", "Newton-Euler Dynamics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0043-torque", "title": "rb 0043 torque", "domain": "robotics", "domain_name": "Robotics", "definition": "Torque is the rotational force applied about a joint axis, expressed in Newton-metres (N\u00b7m). In robot dynamics, joint torques are the primary control inputs that drive links through desired trajectories; torque limits constrain the feasible workspace and influence payload capacity, and torque sensing enables compliant and force-controlled interaction.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0043-torque", "labels": [ "rb 0043 torque" ], "is_subclass_of": [ "Actuation and Control", "Robot Dynamics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0044-velocity", "title": "rb 0044 velocity", "domain": "robotics", "domain_name": "Robotics", "definition": "In robotics, velocity refers to the time-derivative of position, encompassing both linear velocity (metres per second) at the end-effector or a body-frame point, and angular velocity (radians per second) describing rotational rate. Velocity is the central quantity in differential kinematics: the Jacobian matrix maps joint-space velocity vectors to Cartesian task-space velocities. Velocity limits are safety-critical parameters in collaborative robot standards (ISO/TS 15066) where end-effector speed directly determines permissible human contact force.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0044-velocity", "labels": [ "rb 0044 velocity" ], "is_subclass_of": [ "Actuation and Control", "Robot Kinematics" ], "wikilinks": [ "Computer Vision", "RoboticsDomain" ] }, { "id": "rb-0045-acceleration", "title": "rb 0045 acceleration", "domain": "robotics", "domain_name": "Robotics", "definition": "In robotics, acceleration is the rate of change of velocity with respect to time, expressed for each joint (joint-space acceleration) or for the robot's end-effector (task-space acceleration), measured in rad/s\u00b2 or m/s\u00b2 respectively. Acceleration profiles govern the dynamic forces and torques that a manipulator must generate, coupling directly into Newton-Euler equations of motion. Limiting acceleration is central to safety (reducing impact forces) and to trajectory smoothness in collaborative applications.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0045-acceleration", "labels": [ "rb 0045 acceleration" ], "is_subclass_of": [ "Actuation and Control", "Newton-Euler Dynamics" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0047-feedback-control", "title": "rb 0047 feedback control", "domain": "robotics", "domain_name": "Robotics", "definition": "A closed-loop control strategy in which sensor measurements of a system's output are continuously compared to a reference setpoint, and the resulting error signal drives corrective actuator commands. Feedback control is the foundational mechanism for stable, accurate robotic motion, enabling autonomous robots and telerobotic systems to compensate for disturbances, model uncertainty, and environmental variation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0047-feedback-control", "labels": [ "rb 0047 feedback control" ], "is_subclass_of": [ "Actuation and Control", "Closed-Loop Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0048-pid-controller", "title": "rb 0048 pid controller", "domain": "robotics", "domain_name": "Robotics", "definition": "A PID (Proportional-Integral-Derivative) controller is the most widely deployed feedback control algorithm in robotics and industrial automation. It computes a control output as the weighted sum of three terms: the proportional term (reacts to the current error magnitude), the integral term (eliminates steady-state error by accumulating past errors), and the derivative term (anticipates future error by responding to the rate of change). PID controllers are used in robot joint position and velocity loops, temperature regulation, and process control, often augmented with feed-forward terms to improve performance under known dynamics.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0048-pid-controller", "labels": [ "rb 0048 pid controller" ], "is_subclass_of": [ "Actuation and Control", "Control Theory" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0049-motion-planning", "title": "rb 0049 motion planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Motion planning is the computational process of determining a sequence of valid robot configurations or control inputs that moves a robot from an initial state to a goal state while satisfying constraints such as obstacle avoidance, joint limits, and dynamic feasibility. It bridges high-level task specification and low-level actuation, encompassing path planning, trajectory optimisation, and task-and-motion planning (TAMP). Sampling-based methods (RRT, PRM) and optimisation-based approaches are the dominant paradigms, increasingly augmented by learning-based techniques for dynamic and uncertain environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0049-motion-planning", "labels": [ "rb 0049 motion planning" ], "is_subclass_of": [ "Navigation and Planning", "Robotics" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "rb-0051-trajectory-planning", "title": "rb 0051 trajectory planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Trajectory planning is the process of computing a time-parametrised path \u2014 specifying position, velocity, and acceleration profiles \u2014 that moves a robot from a start configuration to a goal while satisfying kinematic and dynamic constraints, joint limits, and task requirements. It extends path planning by assigning timing to waypoints, enabling smooth, jerk-limited motions suitable for real-time execution by a robot controller.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0051-trajectory-planning", "labels": [ "rb 0051 trajectory planning" ], "is_subclass_of": [ "Navigation and Planning", "Motion Planning" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0052-collision-avoidance", "title": "rb 0052 collision avoidance", "domain": "robotics", "domain_name": "Robotics", "definition": "Collision avoidance in robotics is the capability of a robot system to detect and react to obstacles \u2014 including other robots, infrastructure, and human operators \u2014 so as to prevent physical contact that could cause damage or injury. It encompasses both reactive techniques (e.g. potential field methods, velocity obstacles) and proactive planning approaches that embed free-space constraints into the trajectory from the outset. It is a prerequisite for safe autonomous navigation and collaborative operation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0052-collision-avoidance", "labels": [ "rb 0052 collision avoidance" ], "is_subclass_of": [ "Navigation and Planning", "Motion Planning" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0053-force-control", "title": "rb 0053 force control", "domain": "robotics", "domain_name": "Robotics", "definition": "Force control is a robot control paradigm in which the controller regulates the interaction force between the end-effector and its environment, rather than purely tracking a desired position trajectory. By closing the loop on measured contact forces from a force-torque sensor, the robot can perform compliant tasks such as surface grinding, peg-in-hole insertion, and human-robot handover without requiring precise environmental models. The main architectures are impedance control, admittance control, and hybrid position/force control, each suited to different environment stiffness regimes.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0053-force-control", "labels": [ "rb 0053 force control" ], "is_subclass_of": [ "Actuation and Control", "Control System" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0054-position-control", "title": "rb 0054 position control", "domain": "robotics", "domain_name": "Robotics", "definition": "Position control is a robot control strategy in which the primary objective is to drive each joint or end-effector to a desired spatial configuration, using closed-loop feedback from encoders or resolvers to minimise position error. It forms the foundation for precise manipulation tasks and is typically implemented via PID controllers, and may be combined with force or impedance control for compliant operation in contact-rich environments.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0054-position-control", "labels": [ "rb 0054 position control" ], "is_subclass_of": [ "Actuation and Control", "Motion Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0055-velocity-control", "title": "rb 0055 velocity control", "domain": "robotics", "domain_name": "Robotics", "definition": "Velocity control is a robot control strategy in which joint or Cartesian velocities are the primary commanded quantities, with a feedback controller (typically PID-based) continuously correcting deviations between desired and measured velocities. It is used for smooth trajectory following, compliant interaction with soft contacts, and speed-and-separation monitoring safety functions that require real-time speed capping. Velocity control is distinct from position control \u2014 it does not inherently resist positional drift \u2014 and from torque control, which acts at the force level.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0055-velocity-control", "labels": [ "rb 0055 velocity control" ], "is_subclass_of": [ "Actuation and Control", "Motion Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0056-impedance-control", "title": "rb 0056 impedance control", "domain": "robotics", "domain_name": "Robotics", "definition": "Impedance control is a robot interaction-control strategy that regulates the dynamic relationship between end-effector force and motion by imposing a desired mechanical impedance (mass, damping, stiffness) on the robot's behaviour at the point of contact. Rather than commanding precise positions or forces independently, impedance control allows compliant, safe physical interaction with humans or uncertain environments by shaping the robot's apparent mechanical response. It is fundamental to collaborative robotics, enabling robots to yield to external forces in a controlled manner without requiring an explicit force setpoint.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0056-impedance-control", "labels": [ "rb 0056 impedance control" ], "is_subclass_of": [ "Actuation and Control", "Interaction Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0057-admittance-control", "title": "rb 0057 admittance control", "domain": "robotics", "domain_name": "Robotics", "definition": "Admittance control is an interaction-control strategy in which a robot measures contact forces and torques and converts them into corresponding desired motion (position or velocity) using a virtual mass-spring-damper model. It is the dual of impedance control: the robot senses force input and renders motion output, enabling compliant, safe physical human-robot interaction in collaborative tasks.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0057-admittance-control", "labels": [ "rb 0057 admittance control" ], "is_subclass_of": [ "Actuation and Control", "Interaction Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0060-optimal-control", "title": "rb 0060 optimal control", "domain": "robotics", "domain_name": "Robotics", "definition": "Optimal control is a branch of control theory that computes control inputs minimising (or maximising) a performance criterion \u2014 such as energy consumption, time, or tracking error \u2014 subject to system dynamics and constraints. In robotics, optimal control underpins trajectory optimisation, model predictive control, and reinforcement learning-based policy synthesis.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0060-optimal-control", "labels": [ "rb 0060 optimal control" ], "is_subclass_of": [ "Actuation and Control", "Control Theory" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0061-nonlinear-control", "title": "rb 0061 nonlinear control", "domain": "robotics", "domain_name": "Robotics", "definition": "Nonlinear control encompasses control strategies designed for systems whose dynamics cannot be adequately described by linear differential equations. In robotics, virtually all manipulators and mobile platforms exhibit nonlinear behaviour due to inertia coupling, Coriolis terms, gravity loading, and joint friction. Nonlinear control techniques\u2014including computed-torque control, sliding-mode control, feedback linearisation, Lyapunov-based methods, and model predictive control\u2014explicitly account for these nonlinearities to achieve stable, high-performance trajectory tracking across the full operating envelope.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0061-nonlinear-control", "labels": [ "rb 0061 nonlinear control" ], "is_subclass_of": [ "Actuation and Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0062-model-predictive-control", "title": "rb 0062 model predictive control", "domain": "robotics", "domain_name": "Robotics", "definition": "Model Predictive Control (MPC) is an advanced optimal control strategy that uses an explicit mathematical model of the plant to predict future system behaviour over a finite receding horizon, then solves an optimisation problem at each control step to determine the input sequence that minimises a cost function subject to state and input constraints. Only the first element of the computed sequence is applied before the optimisation is repeated. In robotics, MPC enables constraint-aware trajectory tracking, force regulation, and whole-body motion planning that classical PID controllers cannot achieve.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0062-model-predictive-control", "labels": [ "rb 0062 model predictive control" ], "is_subclass_of": [ "Actuation and Control", "Optimal Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0063-sliding-mode-control", "title": "rb 0063 sliding mode control", "domain": "robotics", "domain_name": "Robotics", "definition": "Sliding Mode Control (SMC) is a nonlinear robust control technique that drives system states onto a predefined sliding surface in state space and then maintains them on that surface using discontinuous (switching) control actions. Once on the sliding manifold, the system dynamics become insensitive to matched disturbances and parameter uncertainties, making SMC highly robust for robot manipulators with uncertain dynamics, friction, and external loads. A key challenge is chattering \u2014 high-frequency oscillation caused by the switching law \u2014 which is addressed through boundary layer methods and higher-order SMC variants.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0063-sliding-mode-control", "labels": [ "rb 0063 sliding mode control" ], "is_subclass_of": [ "Actuation and Control", "Control Theory" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0064-computed-torque-control", "title": "rb 0064 computed torque control", "domain": "robotics", "domain_name": "Robotics", "definition": "Computed torque control (CTC) is a model-based robot control strategy that uses the full inverse dynamics model of a manipulator to compute the joint torques required to follow a desired trajectory. By cancelling the nonlinear dynamics \u2014 including Coriolis, centripetal, and gravitational terms \u2014 CTC transforms the closed-loop system into a set of independent linear double-integrators, enabling simple PD outer-loop controllers to achieve high-accuracy tracking. Its effectiveness depends on the fidelity of the dynamic model and is sensitive to parameter uncertainty.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0064-computed-torque-control", "labels": [ "rb 0064 computed torque control" ], "is_subclass_of": [ "Actuation and Control", "Model Based Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0065-visual-servoing", "title": "rb 0065 visual servoing", "domain": "robotics", "domain_name": "Robotics", "definition": "Visual servoing is a robot control technique that uses real-time visual feedback from a camera to regulate the motion of a robot toward a goal configuration. Image-based visual servoing (IBVS) minimises an image-feature error directly, while position-based visual servoing (PBVS) reconstructs 3-D pose from vision before computing Cartesian control signals. The approach is widely used in manipulation, assembly, and tracking tasks where precise end-effector placement relative to a visually perceived target is required.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0065-visual-servoing", "labels": [ "rb 0065 visual servoing" ], "is_subclass_of": [ "Perception and Sensing", "Feedback Control" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0066-robot-sensor", "title": "rb 0066 robot sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A robot sensor is a transducer or measurement device integrated into or used by a robot system to acquire information about the robot's own state or its surrounding environment. Robot sensors span proprioceptive types \u2014 such as encoders, IMUs, and force-torque sensors \u2014 and exteroceptive types such as vision systems, LiDAR, radar, and proximity sensors. The data they provide forms the perceptual basis for control, motion planning, and safety functions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0066-robot-sensor", "labels": [ "rb 0066 robot sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0067-force-torque-sensor", "title": "rb 0067 force torque sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "A force-torque sensor (FT sensor) is a transducer mounted at a robot's wrist or tool centre point that simultaneously measures all six components of mechanical load: three orthogonal forces (Fx, Fy, Fz) and three orthogonal torques (Tx, Ty, Tz). These measurements enable the robot controller to monitor contact forces in real time, supporting force-controlled assembly, surface-following tasks, human-robot contact detection, and safety-critical power-and-force limiting under ISO/TS 15066. FT sensors are the primary feedback device for impedance and admittance control strategies.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0067-force-torque-sensor", "labels": [ "rb 0067 force torque sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0068-vision-system", "title": "rb 0068 vision system", "domain": "robotics", "domain_name": "Robotics", "definition": "A robotic vision system is an integrated sensor subsystem that captures and processes visual data\u2014using cameras, depth sensors, or LiDAR\u2014to provide a robot with spatial awareness, object recognition, and scene understanding capabilities. Vision systems underpin tasks including visual servoing, part inspection, SLAM-based localisation, and human-robot interaction, and conform to ISO 8373:2021 perception-system definitions.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0068-vision-system", "labels": [ "rb 0068 vision system" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "Aoki2003", "Hyper personalisation", "RoboticsDomain" ] }, { "id": "rb-0069-lidar", "title": "rb 0069 lidar", "domain": "robotics", "domain_name": "Robotics", "definition": "A remote sensing technology that measures distances by emitting laser pulses and calculating time-of-flight to generate precise three-dimensional point clouds of the surrounding environment, enabling robots and autonomous vehicles to perceive and navigate physical space with centimetre-level accu...", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:rb-0069-lidar", "labels": [ "rb 0069 lidar" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "3D Mapping", "Camera Sensor", "Drone Navigation", "ICRA 2023", "IMU", "ISO 8373:2021", "Obstacle Detection", "Radar", "Robotics Ontology Working Group", "Simultaneous Localisation and Mapping", "Autonomous Navigation", "Autonomous Vehicle", "Mobile Robot", "Point Cloud", "RoboticsDomain", "Sensor Fusion", "SLAM" ] }, { "id": "rb-0070-tactile-sensing", "title": "rb 0070 tactile sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Tactile sensing is the robotic capability to detect and measure contact forces, pressures, textures, and slip at points of physical interaction between the robot and its environment or human operators. Realised through arrays of pressure sensors, capacitive skins, or piezoelectric films distributed across end-effectors and link surfaces, tactile sensing enables compliant grasping, contact-triggered safety stops, and rich feedback for teleoperation. It is a key enabling technology for human-robot collaboration under ISO/TS 15066.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0070-tactile-sensing", "labels": [ "rb 0070 tactile sensing" ], "is_subclass_of": [ "Perception and Sensing", "Tactile Sensor" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0072-encoder", "title": "rb 0072 encoder", "domain": "robotics", "domain_name": "Robotics", "definition": "An encoder in robotics is a proprioceptive transducer that converts the angular or linear position of a joint or actuator shaft into a digital electrical signal, providing the position and velocity feedback essential for closed-loop control. Encoders are categorised as incremental (providing relative position pulses) or absolute (outputting a unique code for each position), with absolute rotary encoders being preferred in safety-critical collaborative robot applications due to their power-loss resilience. High-resolution encoders directly determine robot accuracy, repeatability, and the fidelity of safety functions such as speed limitation.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0072-encoder", "labels": [ "rb 0072 encoder" ], "is_subclass_of": [ "Perception and Sensing", "Proprioceptive Sensor" ], "wikilinks": [ "Robotics", "RoboticsDomain" ] }, { "id": "rb-0073-imu", "title": "rb 0073 imu", "domain": "robotics", "domain_name": "Robotics", "definition": "An Inertial Measurement Unit (IMU) is an electronic device that measures and reports a body's specific force, angular rate, and sometimes magnetic field using a combination of accelerometers, gyroscopes, and optional magnetometers. In robotics, IMUs provide high-frequency proprioceptive feedback for state estimation, pose tracking, and stabilisation control, and are commonly fused with odometry or SLAM algorithms to reduce drift.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0073-imu", "labels": [ "rb 0073 imu" ], "is_subclass_of": [ "Perception and Sensing", "Inertial Measurement Unit" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0075-range-finder", "title": "rb 0075 range finder", "domain": "robotics", "domain_name": "Robotics", "definition": "A range finder is an exteroceptive sensor that measures the distance between a robot and surrounding objects or surfaces by emitting a signal (laser, ultrasonic, or infrared) and timing its return. Range finders are fundamental to obstacle detection, safety zone monitoring, and environment mapping in mobile and collaborative robots.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0075-range-finder", "labels": [ "rb 0075 range finder", "Range Sensor" ], "is_subclass_of": [ "Perception and Sensing", "Exteroceptive Sensor" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0077-depth-camera", "title": "rb 0077 depth camera", "domain": "robotics", "domain_name": "Robotics", "definition": "A depth camera is a sensor that captures per-pixel distance information alongside a conventional intensity image, producing a registered RGB-D data stream or raw point cloud. In robotics, depth cameras are used for 3D scene reconstruction, obstacle avoidance, object recognition, and SLAM. Common operating principles include structured light projection (e.g. Intel RealSense), time-of-flight measurement, and stereo triangulation; each involves different trade-offs in range, resolution, and outdoor usability.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0077-depth-camera", "labels": [ "rb 0077 depth camera", "Depth Cameras" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0078-infrared-sensor", "title": "rb 0078 infrared sensor", "domain": "robotics", "domain_name": "Robotics", "definition": "An infrared (IR) sensor is an exteroceptive sensor that detects and measures infrared radiation emitted or reflected by objects in the environment, enabling proximity detection, thermal profiling, and obstacle identification in robotic systems. IR sensors operate across near-infrared (NIR), short-wave, mid-wave, and long-wave bands; common robotics applications include proximity switches, line-following, and thermal imaging for human detection. They complement other ranging sensors such as LiDAR and ultrasonic devices, and are frequently fused in multi-modal perception pipelines.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0078-infrared-sensor", "labels": [ "rb 0078 infrared sensor" ], "is_subclass_of": [ "Perception and Sensing" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0080-radar", "title": "rb 0080 radar", "domain": "robotics", "domain_name": "Robotics", "definition": "Radar (Radio Detection and Ranging) is an active sensing technology that emits radio-frequency electromagnetic pulses and detects reflected returns to measure the range, velocity, and bearing of objects in the environment. In robotics, radar sensors provide reliable obstacle detection and velocity estimation across diverse weather and lighting conditions where cameras and LiDAR may degrade, making them particularly valuable for autonomous ground vehicles, drones, and safety-critical proximity monitoring in industrial settings.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0080-radar", "labels": [ "rb 0080 radar" ], "is_subclass_of": [ "Perception and Sensing", "rb 0066 robot sensor" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0087-safety-standard", "title": "rb 0087 safety standard", "domain": "robotics", "domain_name": "Robotics", "definition": "A safety standard in robotics is a normative document, published by a recognised standards body, that specifies requirements and guidelines to achieve acceptable levels of safety for robot systems, their components, and their operating environments. Key robotics safety standards include ISO 10218-1/2 for industrial robots, ISO/TS 15066 for collaborative operation, and ISO 13482 for personal care robots. Such standards define hazard categories, risk assessment methodology, required safety functions, performance levels, and verification procedures that manufacturers and system integrators must satisfy before deployment.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0087-safety-standard", "labels": [ "rb 0087 safety standard" ], "is_subclass_of": [ "Safety and Standards", "Robot Standard" ], "wikilinks": [ "NVIDIA Omniverse", "PEOPLE", "RoboticsDomain" ] }, { "id": "rb-0088-iso-13482-compliance", "title": "rb 0088 iso 13482 compliance", "domain": "robotics", "domain_name": "Robotics", "definition": "ISO 13482 compliance refers to conformance with ISO 13482:2014, the international safety standard for personal care robots that interact with humans in non-industrial settings. Achieving compliance requires demonstrating that hazards have been identified, risks reduced to acceptable levels via protective measures, and that the robot meets defined performance and safety integrity criteria throughout its operational life.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0088-iso-13482-compliance", "labels": [ "rb 0088 iso 13482 compliance" ], "is_subclass_of": [ "Safety and Standards", "Functional Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0089-risk-assessment", "title": "rb 0089 risk assessment", "domain": "robotics", "domain_name": "Robotics", "definition": "Risk Assessment in robotics is the systematic process of identifying hazards associated with a robot system, estimating the severity and probability of potential harm, and determining whether risks are acceptable or require mitigation. It is mandated by ISO 10218-1/-2 for industrial robots and ISO 13482 for personal care robots, and forms the foundation for selecting appropriate safeguarding measures and collaborative operation modes.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0089-risk-assessment", "labels": [ "rb 0089 risk assessment" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0090-emergency-stop", "title": "rb 0090 emergency stop", "domain": "robotics", "domain_name": "Robotics", "definition": "An Emergency Stop (E-stop) is a safety function that immediately removes power or halts the motion of a robot system upon activation, bringing the robot to a controlled or uncontrolled halt to prevent injury or damage. Defined under ISO 10218 and IEC 60204-1, it is a hardwired, independently monitored stop category (typically Stop Category 0 or 1) that takes priority over all other controls. E-stops must be clearly marked, accessible, and self-latching so that the robot cannot restart until the stop is deliberately reset.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0090-emergency-stop", "labels": [ "rb 0090 emergency stop", "Emergency Stop" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety", "rb 0087 safety standard" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0091-safety-rated-monitored-stop", "title": "rb 0091 safety rated monitored stop", "domain": "robotics", "domain_name": "Robotics", "definition": "A safety-rated monitored stop (SRMS) is a robot stopping function in which the robot halts motion while the control system continuously monitors joint positions to verify the robot remains stationary, without removing power to the actuators. Unlike an emergency stop, SRMS allows rapid resumption of operation when the hazardous condition clears, and is central to collaborative robot (cobot) safety architectures defined in ISO 10218 and ISO/TS 15066.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0091-safety-rated-monitored-stop", "labels": [ "rb 0091 safety rated monitored stop", "Safety-Rated Monitored Stop" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0092-protective-stop", "title": "rb 0092 protective stop", "domain": "robotics", "domain_name": "Robotics", "definition": "A protective stop is a safety-initiated cessation of robot motion that halts all hazardous movement when a safety function is triggered, without necessarily cutting power to the drive system. Unlike an emergency stop, a protective stop permits automatic restart once the triggering condition is resolved, making it a standard mechanism in collaborative robot cells operating under ISO 10218 and ISO/TS 15066. It is fundamental to speed-and-separation monitoring and power-and-force-limiting safety strategies.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0092-protective-stop", "labels": [ "rb 0092 protective stop" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0093-speed-limitation", "title": "rb 0093 speed limitation", "domain": "robotics", "domain_name": "Robotics", "definition": "Speed limitation is a collaborative robotics safety function that restricts the maximum velocity of robot joints or the tool centre point to a defined safe threshold, as specified by ISO/TS 15066 and ISO 10218. It is a key mechanism for enabling safe human-robot collaboration in shared workspaces by ensuring robot motion cannot exceed speeds that would cause unacceptable injury risk upon contact. Speed limitation operates continuously or is activated when a human is detected within a monitored zone.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0093-speed-limitation", "labels": [ "rb 0093 speed limitation" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0094-power-and-force-limiting", "title": "rb 0094 power and force limiting", "domain": "robotics", "domain_name": "Robotics", "definition": "Power and Force Limiting (PFL) is a collaborative robot safety mode defined in ISO/TS 15066 in which the robot's mechanical power, force, and momentum are continuously constrained so that any contact with a human remains below biomechanical injury thresholds. Unlike speed-and-separation monitoring, PFL allows direct physical contact by ensuring that contact forces can never exceed prescribed quasi-static and transient limits.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0094-power-and-force-limiting", "labels": [ "rb 0094 power and force limiting", "Contact Force Limit", "Force Limiting", "Force-Limited Operation", "Power and Force Limiting" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0095-safety-zone", "title": "rb 0095 safety zone", "domain": "robotics", "domain_name": "Robotics", "definition": "A safety zone is a defined spatial region around a robot or autonomous system within which human presence or other objects trigger protective actions such as speed reduction, monitored stops, or full emergency stops. Safety zones are configured and monitored via safeguarding devices and are a central mechanism for achieving collaborative and safe robot operation in accordance with ISO 10218 and ISO/TS 15066.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0095-safety-zone", "labels": [ "rb 0095 safety zone" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0096-safeguarding", "title": "rb 0096 safeguarding", "domain": "robotics", "domain_name": "Robotics", "definition": "Safeguarding, in the context of collaborative and industrial robotics, refers to the ensemble of physical barriers, electronic devices, and procedural controls deployed to prevent hazardous contact between robots and humans. Safeguarding measures include hard guards, light curtains, safety zones, and monitored-stop functions, and are mandated by standards such as ISO 10218 and ISO/TS 15066. Effective safeguarding is complementary to inherent safe design: it mitigates residual risks that cannot be eliminated through power-and-force limiting or speed reduction alone.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0096-safeguarding", "labels": [ "rb 0096 safeguarding" ], "is_subclass_of": [ "Safety and Standards", "Robot Safety" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0100-safety-integrity-level", "title": "rb 0100 safety integrity level", "domain": "robotics", "domain_name": "Robotics", "definition": "Safety Integrity Level (SIL) is a discrete measure of the reliability required for a safety function in a robotic or automated system, defined by IEC 61508 on a four-level scale (SIL 1\u20134). A higher SIL demands greater hardware fault tolerance, stricter software development processes, and more comprehensive validation to ensure the safety function reduces risk to a tolerable level. SIL is assigned during risk assessment and drives the entire safety lifecycle of a system.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0100-safety-integrity-level", "labels": [ "rb 0100 safety integrity level" ], "is_subclass_of": [ "Safety and Standards", "Cobot Safety Levels" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0101-performance-level", "title": "rb 0101 performance level", "domain": "robotics", "domain_name": "Robotics", "definition": "Performance Level (PL) is a discrete safety metric defined in ISO 13849-1 that quantifies the ability of a safety-related control system to perform a safety function under foreseeable conditions. Levels range from PLa (lowest) to PLe (highest), each corresponding to a target probability of dangerous failure per hour. PL is determined through risk assessment and verified against the required PL (PLr) derived from the hazard and risk analysis of the application.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0101-performance-level", "labels": [ "rb 0101 performance level", "Performance Level", "Performance Level Method" ], "is_subclass_of": [ "Safety and Standards" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0103-collaborative-operation", "title": "rb 0103 collaborative operation", "domain": "robotics", "domain_name": "Robotics", "definition": "Collaborative operation is a mode of robotic system use in which a robot and one or more human operators work together within a shared workspace, as defined in ISO/TS 15066 and ISO 10218. It encompasses the specific operating conditions, safety functions, and interaction patterns that govern the coexistence of humans and robots without a fixed separating safeguard. The concept underpins cobot deployment strategies including hand guiding, speed and separation monitoring, power and force limiting, and safety-rated monitored stop.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0103-collaborative-operation", "labels": [ "rb 0103 collaborative operation" ], "is_subclass_of": [ "Human-Robot Interaction", "Collaborative Operation" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0104-hand-guiding", "title": "rb 0104 hand guiding", "domain": "robotics", "domain_name": "Robotics", "definition": "Hand guiding is a collaborative robot operation mode in which an operator physically moves the robot arm by direct physical contact, typically through a dedicated hand-guiding device, while the robot's safety systems monitor force, speed, and separation. It is one of four collaborative operation modes defined in ISO/TS 15066 and ISO 10218, enabling intuitive programming-by-demonstration and safe human-robot co-manipulation tasks.", "entityType": "Class", "qualityScore": 0.0, "maturity": "emerging", "iri": "urn:ngm:class:rb-0104-hand-guiding", "labels": [ "rb 0104 hand guiding", "Hand Guiding", "Hand Guiding Mode" ], "is_subclass_of": [ "Human-Robot Interaction", "Human Robot Interaction" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "rb-0105-speed-and-separation-monitoring", "title": "rb 0105 speed and separation monitoring", "domain": "robotics", "domain_name": "Robotics", "definition": "Speed and Separation Monitoring (SSM) is a collaborative robot safety function in which the speed of the robot is continuously regulated based on the measured distance between the robot and any human operator in the shared workspace. When the separation distance decreases below defined thresholds the robot slows or stops, and it resumes normal speed once sufficient separation is restored. SSM is standardised under ISO/TS 15066 as one of the four permitted collaborative operation modes.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:rb-0105-speed-and-separation-monitoring", "labels": [ "rb 0105 speed and separation monitoring", "Speed Monitoring", "Speed and Separation Monitoring" ], "is_subclass_of": [ "Safety and Standards", "Light Curtain" ], "wikilinks": [ "RoboticsDomain" ] }, { "id": "remarkable-e-ink-knowledge-tablet", "title": "reMarkable E-Ink Knowledge Tablet", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "The reMarkable is an e-ink tablet designed for distraction-free writing, note-taking, and document annotation, running a Linux-based operating system that exposes SSH access and supports a rich ecosystem of open-source tools. Its platform enables handwriting recognition, LaTeX generation, AI-assisted prompt workflows, and integration with knowledge management systems such as Obsidian. The device occupies a distinct niche as a digital-analogue bridge for researchers and knowledge workers seeking pen-on-paper fidelity with machine-readable output.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:remarkable-e-ink-knowledge-tablet", "labels": [ "reMarkable E-Ink Knowledge Tablet", "Remarkable" ], "is_subclass_of": [ "Software Engineering" ], "wikilinks": [] }, { "id": "robo-actuation-and-control", "title": "Actuation and Control", "domain": "robotics", "domain_name": "Robotics", "definition": "Taxonomy hub covering the mechanisms, algorithms, and feedback systems that translate commands into physical motion in robotic systems. Encompasses actuator hardware, control algorithms such as PID and adaptive control, motion planning, and trajectory execution that collectively govern a robot's dynamic behaviour.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robo-actuation-and-control", "labels": [ "Actuation and Control", "Robo Actuation and Control" ], "is_subclass_of": [ "Robotics" ], "wikilinks": [] }, { "id": "robo-navigation-and-planning", "title": "Navigation and Planning", "domain": "robotics", "domain_name": "Robotics", "definition": "Taxonomy hub for methods that enable a robot to determine its position, build a map of its environment, and compute collision-free paths to goals. It encompasses simultaneous localisation and mapping, path planning, motion planning, obstacle avoidance, sensor fusion, and autonomous navigation, forming the core competency for mobile robot autonomy.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robo-navigation-and-planning", "labels": [ "Navigation and Planning" ], "is_subclass_of": [ "Robotics" ], "wikilinks": [] }, { "id": "robo-perception", "title": "Perception and Sensing", "domain": "robotics", "domain_name": "Robotics", "definition": "Taxonomy hub for all robotic perception and environmental sensing technologies, covering sensor hardware (LiDAR, cameras, IMUs, tactile sensors), sensor-fusion algorithms, SLAM, object detection, and visual odometry. These capabilities provide robots with the situational awareness required for autonomous navigation and manipulation.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:robo-perception", "labels": [ "Perception and Sensing" ], "is_subclass_of": [ "Robotics" ], "wikilinks": [] }, { "id": "robo-robot-type", "title": "Robot Type", "domain": "robotics", "domain_name": "Robotics", "definition": "Taxonomy hub classifying robots by physical configuration, kinematics, and operational domain, including industrial manipulators, mobile platforms, humanoids, collaborative robots, aerial and marine systems, surgical robots, and service robots.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robo-robot-type", "labels": [ "Robot Type" ], "is_subclass_of": [ "Robotics" ], "wikilinks": [] }, { "id": "robo-safety-and-standards", "title": "Safety and Standards", "domain": "robotics", "domain_name": "Robotics", "definition": "Taxonomy hub for robotics safety requirements, certification standards, collision avoidance, force-limiting mechanisms, and regulatory compliance frameworks governing safe robot operation alongside humans in industrial and service environments.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:robo-safety-and-standards", "labels": [ "Safety and Standards" ], "is_subclass_of": [ "Robotics" ], "wikilinks": [] }, { "id": "ruv-fann-multi-agent-swarm-framework", "title": "ruv-FANN Multi-Agent Swarm Framework", "domain": "infrastructure", "domain_name": "Infrastructure", "definition": "RuV Agents refers to the ruv-swarm agent framework developed within the ruv-FANN project, providing a multi-agent orchestration layer for coordinating LLM-backed autonomous agents across tasks such as coding, research, and data processing. The framework integrates with inference APIs and supports rate-limit bypass strategies through unified LLM platform routing. It represents a practical implementation of multi-agent coordination patterns for software engineering and knowledge-work automation.", "entityType": "Class", "qualityScore": 0.35, "maturity": "emerging", "iri": "urn:ngm:class:ruv-fann-multi-agent-swarm-framework", "labels": [ "ruv-FANN Multi-Agent Swarm Framework", "RuV Agents" ], "is_subclass_of": [ "Computing and Cloud" ], "wikilinks": [] }, { "id": "s-btc", "title": "sBTC", "domain": "blockchain", "domain_name": "Blockchain", "definition": "sBTC is an asset on the Stacks blockchain designed to represent Bitcoin in a decentralised manner so that it can be used in smart contracts. It is backed by bitcoin held under the protocol.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:s-btc", "labels": [ "sBTC" ], "is_subclass_of": [ "Stacks" ], "wikilinks": [ "Bitcoin Network", "Smart Contract", "DeFi", "Layer 2 Scaling", "Stacks", "https://www.stacks.co/sbtc", "https://docs.stacks.co" ] }, { "id": "sc-content-and-assets", "title": "Content and Assets", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Taxonomy hub for all content types and digital assets within the spatial computing domain, encompassing 3D models, avatars, NFTs, textures, rendering pipelines, and the digital asset lifecycle from creation through distribution. This category unifies the asset-creation and asset-management concerns that underpin immersive experiences.", "entityType": "Class", "qualityScore": 0.0, "maturity": "established", "iri": "urn:ngm:class:sc-content-and-assets", "labels": [ "Content and Assets", "Spatial Computing Content and Assets" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "sc-display-and-rendering", "title": "Display and Rendering", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Taxonomy hub for display and rendering technologies within spatial computing, encompassing the pipelines, hardware, and algorithms that transform scene geometry into visual output for AR, VR, and mixed reality experiences. This category bridges hardware display devices with software rendering techniques, covering both real-time and offline methods optimised for immersive perceptual fidelity.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sc-display-and-rendering", "labels": [ "Display and Rendering" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "sc-governance-and-safety", "title": "Governance and Safety", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Taxonomy hub for governance, safety, regulation, and privacy concepts within the spatial computing domain, encompassing frameworks for responsible XR and metaverse deployment, data protection, content moderation, risk management, and user safety standards.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sc-governance-and-safety", "labels": [ "Governance and Safety" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "sc-interaction", "title": "Interaction Technology", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Interaction Technology is the spatial-computing taxonomy hub for all modalities through which users perceive and manipulate virtual and mixed-reality environments \u2014 including hand tracking, eye tracking, gaze control, haptics, voice interaction, and VR controllers. It sits alongside Display and Rendering and Platform and Environment as a peer category within spatial computing.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sc-interaction", "labels": [ "Interaction Technology", "Spatial Computing Interaction" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "sc-platform-and-environment", "title": "Platform and Environment", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Taxonomy hub covering the software runtimes, hardware platforms, operating systems, and execution environments that host spatial computing applications. This category encompasses the infrastructure layer on which XR experiences are built and deployed, spanning cloud, edge, and on-device compute contexts.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sc-platform-and-environment", "labels": [ "Platform and Environment" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "sc-standards-and-interop", "title": "Standards and Interoperability", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "Taxonomy hub for open standards, file formats, APIs, and cross-platform interoperability specifications that enable spatial computing content, devices, and services to work together. This category spans industry consortia output such as OpenXR, OpenUSD, glTF, and WebXR, along with avatar portability and cross-platform identity standards.", "entityType": "Class", "qualityScore": 0.8, "maturity": "established", "iri": "urn:ngm:class:sc-standards-and-interop", "labels": [ "Standards and Interoperability" ], "is_subclass_of": [ "Spatial Computing" ], "wikilinks": [] }, { "id": "secp256k1-elliptic-curve", "title": "secp256k1 Elliptic Curve", "domain": "blockchain", "domain_name": "Blockchain", "definition": "secp256k1 is a specific elliptic curve defined over a 256-bit prime field, standardised by the SEC and chosen for its efficient, verifiable parameters. It underpins ECDSA and Schnorr signatures used by Bitcoin, Nostr, and many other systems for key generation and digital signing. Its near-rigid, low-entropy parameters reduce concern about hidden weaknesses.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:secp256k1-elliptic-curve", "labels": [ "secp256k1 Elliptic Curve" ], "is_subclass_of": [ "Cryptographic Primitive" ], "wikilinks": [] }, { "id": "secp256k1", "title": "secp256k1", "domain": "blockchain", "domain_name": "Blockchain", "definition": "secp256k1 is a Koblitz elliptic curve defined by the short Weierstrass equation y\u00b2 = x\u00b3 + 7 over the 256-bit prime field \u1d3d\u209a (where p = 2\u00b2\u2075\u2076 \u2212 2\u00b3\u00b2 \u2212 977), standardised in SEC 2 by the Standards for Efficient Cryptography Group (SECG) and widely adopted as the asymmetric cryptographic backbone of Bitcoin, Ethereum, and dozens of subsequent blockchain protocols. Its specific Koblitz-form parameters permit the Frobenius endomorphism optimisation (GLV decomposition), yielding scalar multiplication roughly 30% faster than equivalent-security NIST curves without requiring a random-looking but opaque seed. The curve underpins both the Elliptic Curve Digital Signature Algorithm (ECDSA) used for transaction authorisation and, via BIP 340, the Schnorr signature scheme that enables key and signature aggregation in Taproot-based Bitcoin smart contracts.", "entityType": "Class", "qualityScore": 0.74, "maturity": "mature", "iri": "urn:ngm:class:secp256k1", "labels": [ "secp256k1", "Secp256k1" ], "is_subclass_of": [ "Elliptic Curve Cryptography" ], "wikilinks": [] }, { "id": "security-audit-guide", "title": "security audit guide", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "A structured procedural document establishing methodologies, checklists, and frameworks for systematically evaluating the security posture of blockchain systems, smart contracts, and cryptographic infrastructure. Security audit guides codify best practices across phases of automated analysis, manual code review, and formal verification, enabling auditors to identify vulnerabilities such as reentrancy, integer overflow, and access control weaknesses before deployment.", "entityType": "Class", "qualityScore": 0.7, "maturity": "emerging", "iri": "urn:ngm:class:security-audit-guide", "labels": [ "security audit guide" ], "is_subclass_of": [ "Governance and Safety" ], "wikilinks": [ "Blockchain", "Cryptography", "MetaverseDomain", "SmartContract" ] }, { "id": "supports", "title": "supports", "domain": "standards", "domain_name": "Standards", "definition": "Supports is a relational predicate used in the knowledge graph to indicate that one entity provides backing, compatibility, or enabling capability for another.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:supports", "labels": [ "supports" ], "is_subclass_of": [ "Software Engineering", "owl:Thing" ], "wikilinks": [ "Interoperability", "owl:Thing" ] }, { "id": "tc-0002-collaborative-document-editing", "title": "tc 0002 collaborative document editing", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Collaborative Document Editing - Real-time or asynchronous shared editing technology enabling multiple distributed users to simultaneously create, modify, and comment on digital documents with live synchronization and version control.", "entityType": "Class", "qualityScore": 0.5, "maturity": "established", "iri": "urn:ngm:class:tc-0002-collaborative-document-editing", "labels": [ "tc 0002 collaborative document editing", "Collaborative Document Editing", "Document Collaboration", "Shared Document Editing" ], "is_subclass_of": [ "Workspace Tools", "Telepresence", "Knowledge Co-Creation", "Remote Collaboration" ], "wikilinks": [ "Asynchronous Communication", "Asynchronous-Synchronous Communication", "Change Tracking", "Cloud Storage", "Collaborative Inquiry", "Collaborative Research", "Collaborative Thinking", "Collaborative Writing", "Comments", "Confluence", "Connectivism", "Constructivism", "CRDT (Conflict-free Replicated Data Type)", "Credential Verification", "Distributed Editing Networks", "Distributed Teamwork", "Etherpad", "Google Workspace", "HackMD", "High Contrast" ] }, { "id": "tc-0003-telepresence-robot", "title": "tc 0003 telepresence robot", "domain": "distributed-collaboration", "domain_name": "Distributed Collaboration", "definition": "Telepresence Robot - Mobile robotic platform with audio-visual and manipulation capabilities enabling a remote operator to have a physical embodied presence, interact with environments, and perform tasks at a distance while maintaining situational awareness through real-time sensory feedback.", "entityType": "Class", "qualityScore": 0.5, "maturity": "emerging", "iri": "urn:ngm:class:tc-0003-telepresence-robot", "labels": [ "tc 0003 telepresence robot" ], "is_subclass_of": [ "Telepresence", "Telepresence Technology", "Mobile Robot", "Remote Collaboration" ], "wikilinks": [ "5G/6G Communication", "5G Low-Latency Networks", "Agency and Control", "Alternative Control Methods", "Audio", "Audio Description", "Augmented Reality Overlay", "Autonomous Navigation Support", "Avatar Representation", "Blended Environments", "Brain-Computer Interface", "Camera System", "Computer Vision Understanding", "Control Joystick", "Deliberate Practice", "Depth", "Direct Teleoperation", "Electric Motors", "Embodied Cognition", "Embodied Collaboration" ] }, { "id": "transaction-fees", "title": "transaction fees", "domain": "blockchain", "domain_name": "Blockchain", "definition": "Transaction fees are amounts paid by users to have their transactions included in a blockchain block. They compensate miners or validators and help prioritise transactions when capacity is limited.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:transaction-fees", "labels": [ "transaction fees" ], "is_subclass_of": [ "Transaction" ], "wikilinks": [ "Bitcoin Network", "Transaction Validation", "Bitcoin Mining", "Transaction", "https://developer.bitcoin.org/devguide/transactions.html", "https://en.bitcoin.it/wiki/Transaction_fees" ] }, { "id": "v-llm", "title": "vLLM", "domain": "machine-learning", "domain_name": "Machine Learning", "definition": "vLLM is an open-source library for high-throughput serving of large language models. It introduced paged attention, a memory management technique that reduces waste in the key-value cache during generation by managing attention KV cache in fixed-size blocks analogous to virtual memory paging.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:v-llm", "labels": [ "vLLM" ], "is_subclass_of": [ "Inference Serving", "AI Infrastructure (Artificial Intelligence)" ], "wikilinks": [ "KV Cache", "GPU", "Model Serving", "Inference Serving", "Large Language Models", "Latency" ] }, { "id": "visionflow", "title": "visionflow", "domain": "spatial-computing", "domain_name": "Spatial Computing", "definition": "visionflow is a project for rendering and interacting with knowledge graphs in a 3D, GPU-accelerated environment, including extended reality interfaces. It combines graph layout computation with real-time visualisation.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:visionflow", "labels": [ "visionflow" ], "is_subclass_of": [ "Knowledge Graph" ], "wikilinks": [ "GPU", "Real-Time Rendering", "Data Aggregation", "Computer Graphics", "Knowledge Graph" ] }, { "id": "x402-and-l402-payment-protocols", "title": "x402 and l402 payment protocols", "domain": "blockchain", "domain_name": "Blockchain", "definition": "x402 and l402 payment protocols is a blockchain and distributed systems concept and a type of blockchain.", "entityType": "Class", "qualityScore": 0.5, "maturity": "draft", "iri": "urn:ngm:class:x402-and-l402-payment-protocols", "labels": [ "x402 and l402 payment protocols", "L402 Protocol", "X402 Protocol" ], "is_subclass_of": [ "DeFi and Economics" ], "wikilinks": [ "32 bytes", "402 Payment Required", "402 status code", "account registration", "ACINQ", "agent economy", "Agent Federations", "agent-to-agent", "Agriculture Automation", "AI", "AI agent", "AI agents", "AI-first", "AI inference", "AI/ML", "AI-powered", "AI systems", "AML compliance", "amount", "analytics" ] }, { "id": "x-dai", "title": "xDai", "domain": "blockchain", "domain_name": "Blockchain", "definition": "xDai was the original name for the stable-token chain that became Gnosis Chain, an Ethereum-compatible sidechain using a stablecoin for transaction fees. It now operates under the Gnosis Chain name.", "entityType": "Class", "qualityScore": 0.6, "maturity": "established", "iri": "urn:ngm:class:x-dai", "labels": [ "xDai" ], "is_subclass_of": [ "Gnosis Chain", "Network Component (Blockchain)" ], "wikilinks": [ "Ethereum", "DeFi", "Sidechain", "Gnosis Chain", "https://www.gnosischain.com", "https://docs.gnosischain.com" ] }, { "id": "zk-sync", "title": "zkSync", "domain": "blockchain", "domain_name": "Blockchain", "definition": "zkSync is an Ethereum layer-2 scaling network developed by Matter Labs that uses zero-knowledge rollup technology to settle transactions on Ethereum with validity proofs. Its main network, zkSync Era, is a zk-rollup with an EVM-compatible execution environment, allowing many Ethereum smart contracts and tools to be used with minimal changes. By posting succinct proofs that each batch of transactions is valid, the network achieves Ethereum-level settlement security without an optimistic challenge period. zkSync is part of a broader effort to scale Ethereum through zero-knowledge cryptography.", "entityType": "Class", "qualityScore": 0.72, "maturity": "established", "iri": "urn:ngm:class:zk-sync", "labels": [ "zkSync" ], "is_subclass_of": [ "Network Component", "Blockchain Domain" ], "wikilinks": [ "Ethereum", "Rollup", "Zero Knowledge Proof", "Decentralised Finance Domain", "Optimism", "Arbitrum", "Polygon", "Blockchain Domain" ] } ]