Knowledge Graph

A knowledge base that uses a graph-structured data model or topology to integrate data.

How it works

A knowledge graph represents information as a network of entities (nodes) and relationships (edges). For example: (Barack Obama) -[born in]-> (Honolulu). Entities can have attributes, and relationships can have properties. Queries are answered by traversing the graph. Graph databases like Neo4j or RDF triplestores implement knowledge graphs. Large-scale examples include Google's Knowledge Graph (used to populate search result panels) and Wikidata.

Why it matters

Knowledge graphs provide structured, explicit facts that can be queried with precision — unlike LLMs which may confuse or conflate facts. Combining knowledge graphs with LLMs (Graph RAG) is an emerging approach to grounding language model responses in verifiable, structured facts, reducing hallucination while preserving the fluency and flexibility of LLM outputs. They are also fundamental to recommendation systems and entity resolution at scale.