Graph Intelligence: Enhance Reasoning and Retrieval Using Graph Analytics
summary
Shows a reproducible GraphRAG workflow that loads embeddings into Neo4j, builds a vector index on node property embeddings, runs KN&N k=25 to link similar chunks, applies community detection (Louvain/Leiden/label propagation) and computes PageRank/betweenness, and then collapses duplicates for efficiency. Dataset examples: ~17k nodes, ~774k relationships, 7 labels, 27 rel types; one cluster contains 49 near‑duplicate chunks (avg sim 0.98). Also creates co‑occurrence edges from retriever hits to trace conversation paths.
AI-written summary · about
chapters (19)
- 0:00Intro
- 5:47The Basics
- 7:57External Data
- 13:43How does it work
- 15:09What other elements are there
- 21:57Graph Analytics Workshop
- 25:02Power of Graphs
- 27:32Create Instance
- 29:17Database
- 30:42Account Creation
- 31:37Neo4j
- 32:17Neo Agent
- 33:32Documentation
- 34:07Conversations
- 37:47Chunking
- 43:39Dual Labels
- 47:33Backup and Restore
- 51:07Notebook Setup
- 57:14Connect Cluster Curate
related talks
01h 40mKnowledge Graphs & GraphRAG: Techniques for Building Effective GenAI Applications
27mAdaptive Knowledge Graphs for Dynamic RAG Systems
27mWhen Vectors Aren’t Enough: Unlocking AI’s Hidden Connections with Graphs
24mAgentic GraphRAG: AI’s Logical Edge
49mModern Graph Databases
30mHow Graph Data Science Can Turbocharge Your Knowledge Graph
30mGraph RAG: Bringing together graph and vector search to empower retrieval
24mEmpowering Lifelong Learning: Transforming Data into Insights With GraphRAG