FINRA's knowledge graphs improve regulatory investigations
summary
FINRA builds a nightly-updated investigative knowledge graph that turns stacks of documents into quads by extracting text with Amazon Textract, detecting entities and relations with Amazon Comprehend, and bulk-loading triples into Amazon Neptune; the web UI visualizes connections, links nodes back to original documents, and lets investigators enrich and export graphs. The pipeline scales to enterprise volumes, reproduces a months-long insider-trading trail automatically, uncovers additional evidence, and materially speeds and standardizes investigations.
AI-written summary · about
chapters (12)
- 0:00Intro
- 0:35Agenda
- 1:10What are graphs
- 1:45When to use graphs
- 3:07Neptune
- 5:55Data structure
- 7:10Integrations
- 8:31Document Extraction
- 12:30Regulatory Mission
- 23:24What we achieved
- 27:30Next steps
- 30:30Conclusion
related talks
32mFINRA's knowledge graphs improve regulatory investigations
15mGraphRAG-Powered AI Agents: Transforming Corporate Metadata into Intelligent Knowledge
25mKnowledge graphs for financial institutions
30mModernizing data collection at FINRA with Amazon DocumentDB
26mAdvancing Financial Fraud Detection with Graph Dbs
27mEnhancing Financial Analytics With Event-Driven Architectures and Graph Neural Networks
28mFraud Graph with Amazon Neptune and Machine Learning
30mEntity Resolved Knowledge Graphs