Knowledge graphs for financial institutions
summary
BMO builds a Knowledge Graph with Neo4j to join 12 million customers, 880 billion in assets and federated relational data with unstructured docs and social media; a Neo4j plugin uses Tika and OpenNLP to extract entities from PDFs/Word files, a vendor supplies article sentiment, CSVs load structured data, and a React/Node.js dashboard with Neo4j driver surfaces obscure multi‑hop links and sends alerts on negative sentiment — the proof‑of‑concept uncovers previously hidden guarantor/commodity/product relationships and proves the approach works.
AI-written summary · about
chapters (9)
- 0:00Introduction to talk
- 1:02What is Neo4j
- 6:58BMO’s challenge overview
- 9:04Data integration challenges
- 13:00Knowledge graph model
- 15:08Solution architecture
- 19:00Dashboard and insights
- 21:07Capabilities gained
- 23:35Q&A and conclusion
related talks
24mWhy You Need a Knowledge Graph And How to Build It
44mBuilding Knowledge Graphs- 44mBuilding Knowledge Graphs
30mEntity Resolved Knowledge Graphs
15mGraphRAG-Powered AI Agents: Transforming Corporate Metadata into Intelligent Knowledge
46mBuilding Knowledge Graphs
24mYour company, as a Knowledge Graph - the foundation of cybersecurity’s future
21mFrom Data to Knowledge to Action - The Graph Intelligence Platform