RAG vs GraphRAG
A practical comparison of chunk retrieval and graph-guided retrieval, including when to use RAG, GraphRAG, or a hybrid system.
ContextGraph Articles
Practical notes on GraphRAG systems.
A focused reading list on retrieval, entity extraction, and graph-shaped context.
Taxonomy and canonical nodes
Why ContextGraph needs taxonomy, aliases, and merge safeguards so names like CDAI and Content Discovery AI resolve to the same service or concept.
Choosing graph storage for GraphRAG
A comparison of Neo4j, Azure Cosmos DB for Apache Gremlin, Neptune, ArangoDB, PostgreSQL, and vector-store boundaries for ContextGraph storage.
Permission-aware GraphRAG
Why graph traversal must respect user, tenant, document, node, edge, and explanation-level access boundaries before context reaches the model.
GraphRAG evaluation
How to evaluate answer quality, graph paths, citations, permissions, freshness, and retrieval traces so GraphRAG can be trusted in production.
Entity extraction for GraphRAG
Why entity extraction is mandatory, how GLiNER-style models help, what realistic sample data should include, and how many entity types to start with.