From Tables to Graphs: Agentic Graph Modeling
summary
HyGM (Hypothetical Graph Modeling) converts MySQL schemas and sampled rows into a proposed graph model, iterates via LLM-generated structured output plus validation, and migrates data into memaph in-memory graph in seconds after one or two retries (demo uses Sakila). The system supports automatic and interactive modes where NLP edits rename relationships and add constraints, reports validation errors and warnings, and plugs into a GraphRAG retrieval pipeline for downstream search and reasoning.
AI-written summary · about
chapters (9)
- 0:00Introduction to talk
- 3:36What is a graph database
- 5:27Bridging SQL to graph models
- 7:12Hypothetical graph model demo
- 10:24Automatic migration process
- 12:10Interactive graph modeling
- 14:39Why move to graph databases
- 16:42Graph RAG for LLMs
- 18:47Workshop and closing
related talks
23mAgentic GraphRAG with MAGMA Manifold Agency with Graph Models and Augmentation
08mHarness Graph Power With Agentic AI: An Insights Discovery Use Case
28mEnhancing LLMs with Graph Technology
15mTraining and Using Generative AI Models with GraphQL
01h 40mKnowledge Graphs & GraphRAG: Techniques for Building Effective GenAI Applications
12mAgentic GraphRAG: Autonomous Knowledge Graph Construction and Adaptive Retrieval
21mAgentic GraphRAG: Autonomous Knowledge Graph Construction and Adaptive Retrieval
26mPractical GraphRAG Making LLMs smarter with Knowledge Graphs