knowledge graph
A knowledge graph is a data structure that represents knowledge as entities connected by labeled relationships, with entities as nodes and relationships as edges.
A knowledge graph organizes knowledge as a network in which entities such as people, places, products or concepts are nodes and the named relationships between them are edges. Where a relational database stores data in separate tables, a knowledge graph puts relationships at the center, which suits queries that follow connections from one entity to another.
The term gained wide recognition after Google introduced its Knowledge Graph to search in 2012. Knowledge graphs are often paired with ontologies that define the concepts and relationship types allowed. In AI, they are used to give large language models structured information and to add relational context to retrieval-augmented generation (RAG).
This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.