Power Amazon Bedrock applications with Neo4j knowledge graph
summary
Shows a working pipeline that uses Amazon Bedrock (via SageMaker Studio) to parse SEC EDGAR Form 13 XML, extract holdings and CUSIPs, and auto-build a Neo4j knowledge graph with vector embeddings for semantic + graph traversal. A LangChain-driven RAG flow converts natural questions into Cypher, runs graph queries, and returns concise answers (examples: top‑10 holdings, which managers own FAANG). Code and demo artifacts are public on GitHub and the chatbot demo succeeds end‑to‑end.
AI-written summary · about
chapters (8)
- 0:00Introduction to GenAI
- 1:33Graph Database Benefits
- 2:54Architecture Overview
- 3:55Demo Setup and Workflow
- 5:13Graph-based RAG approach
- 7:13Data Processing Demo
- 13:48Chatbot Functionality
- 18:22Resources and Next Steps
related talks
28mBuilding database queries with Amazon Bedrock
24mBuilding GenAI applications with Amazon Bedrock
15mGraph feature engineering with Neo4j and Amazon SageMaker
31mLamRAG: From Data to Constructive Insights using Amazon Bedrock
38mGenerative AI - Architectures and applications in depth
35mEmpower your Bedrock agent with GraphRAG
43mNavigating a Full Stack Knowledge Architecture from Database to LLM
28mImproving responses in generative AI applications