Entity Resolved Knowledge Graphs
summary
Builds a Neo4j knowledge graph by running entity resolution over multiple open datasets (beneficial ownership, sanctions lists, GLEIF, Panama Papers, leaked bank records and PPP loan data) to produce an entity-resolved overlay that preserves original records and shows why merges occur. Connects that KG to a vector layer and uses GraphRAG via LlamaIndex for chat queries and visualization, surfacing surprising cross-dataset links used for sanctions and UBO linking and practical PPP-fraud detection in production.
AI-written summary · about
chapters (7)
- 0:00Introduction to investigations
- 5:27Investigative data modeling
- 12:52Unbundling KG construction
- 14:57Entity resolution fundamentals
- 21:04Architecture and workflows
- 24:09NLP for lexical graphs
- 27:01Demos and resources
related talks
44mBuilding Knowledge Graphs- 44mBuilding Knowledge Graphs
46mBuilding Knowledge Graphs
42mMaking OSS AI Real with Knowledge Graphs
27mAdaptive Knowledge Graphs for Dynamic RAG Systems
25mKnowledge graphs for financial institutions
15mMultiple Knowledge Graphs for Sophisticated RAG
22mBuilding Contextual Knowledge Graphs to Adapt, Evaluate, and Leverage LLMs