Overview
Knowledge graphs represent real-world entities and the relationships between them as a directed graph. Nodes represent entities (people, places, concepts), edges represent relationships, and both can carry typed attributes. Unlike tabular data, knowledge graphs capture the structure of knowledge itself.
Key Technologies
- RDF (Resource Description Framework): The W3C standard for expressing knowledge graph triples (subject, predicate, object)
- SPARQL: Query language for RDF knowledge graphs
- OWL: Web Ontology Language for defining class hierarchies and inference rules
- JSON-LD: JSON-based serialization for linked data
Applications
- Semantic search and recommendation systems
- Enterprise data integration across heterogeneous sources
- Scientific knowledge representation (biomedical ontologies)
- Personal knowledge management (Obsidian, Roam)
Related
- Semantic Web — The broader vision of machine-readable linked data
- Decentralized Web — Decentralized knowledge graph infrastructure