Back to AI Glossary
Retrieval & DataKG

Knowledge Graph

A structured representation of information as a network of entities and their relationships, enabling machines to reason about connections between concepts.

Why it matters

Knowledge graphs give AI systems a structured understanding of how things relate to each other — something that pure vector search misses. They are essential for complex domains with rich entity relationships.

Structure

A knowledge graph stores data as triples: subject → predicate → object. For example: "Claude" → "developed_by" → "Anthropic". This graph structure captures relationships that are invisible in flat document stores, enabling multi-hop reasoning and relationship discovery.

Knowledge graphs + AI

The combination of knowledge graphs and LLMs is increasingly powerful:

  • GraphRAG — uses graph traversal alongside vector retrieval for more connected, contextual answers.
  • Entity extraction — LLMs can automatically build knowledge graphs from unstructured text.
  • Fact verification — graph structure provides a ground-truth reference for validating LLM outputs.

Popular technologies

Neo4j is the dominant graph database. Property graph models (Neo4j, Amazon Neptune) are more practical for most teams than RDF/SPARQL-based approaches. For AI-specific use cases, LangChain and LlamaIndex both offer graph-aware retrieval modules.

Related Terms

Retrieval-Augmented Generation(RAG)— A technique that grounds a language model's output in external data by retrieving relevant documents before generating a response.Embeddings— Dense numerical representations of text (or other data) in a high-dimensional vector space, where similar meanings are placed closer together.

On the AI Radar

Knowledge Graphs— Graph-based knowledge representation that enables AI systems to reason over structured relationships rather than relying solely on unstructured text retrieval. The convergence with LLMs through GraphRAG has made knowledge graphs practical for reducing hallucinations and enabling multi-hop reasoning.GraphRAG— Combining knowledge graphs with retrieval-augmented generation for structured, relationship-aware AI reasoning. Moving from research pattern to production standard as naive RAG reaches its limits.
Back to AI Glossary