Knowledge Graphs : Adding a New Dimension to Vector Search and Generative AI
summary
Claims that grounding LLMs with knowledge graphs reduces hallucination and adds explainability while improving relevance by re-ranking vector results. Uses Neo4j vector index plus LangChain modules, Aura for hosting, and a Graph Data Science workflow (FastRP) to build a co‑purchase graph and generate “customers also like” edges. Demonstrates on an H&M dataset: vector-only hit at rank 7 moves to rank 3 after a simple GQL traversal boost (multiply vector score by graph signal). Code, Collab notebooks and repo are available.
AI-written summary · about
chapters (9)
- 0:00Introduction to graph AI
- 2:33Bridging silos with graphs
- 11:21Industry standards for graphs
- 12:20Why RAG needs knowledge graphs
- 14:51Building graph applications
- 16:12Tooling and acceleration
- 17:36Semantic search journey
- 19:50Practical implementation demo
- 34:15Learning resources
related talks
27mWhen Vectors Aren’t Enough: Unlocking AI’s Hidden Connections with Graphs
41mEnhancing LLMs with Graph Technology
28mData-Empowered AI Assistants
23mKnowledge Graphs and LLMs in Action
39mEnhancing RAG with Neo4j Knowledge Graph
46mBuilding Knowledge Graphs
38mLeveraging Knowledge Graphs for RAG: A Smarter Approach to Contextual AI Applications
01h 00mGraph Mastery to AI