Powering Legal Research with Knowledge Graphs and Large Language Models
summary
Ingests EU Commission cases across five domains (mergers, state aid, antitrust/cartels, DMA, foreign subsidies), filters to nine fields and legal bases Articles 101/102/105, then models an ontology and loads a Neo4j knowledge graph. An LLM (OpenAI) translates English queries to Cypher and NeoDash/Bloom renders subgraph visualizations; live demo returns totals, counts by legal base and sector, and a Microsoft-linked case subgraph. Result cuts manual case-linking and speeds case selection for investigations.
AI-written summary · about
chapters (9)
- 0:00Introduction and team
- 1:50Competition law primer
- 6:08Example of a competition case
- 12:30Pain points in legal research
- 15:02Proposed solution overview
- 16:27Building the knowledge graph
- 19:21Demo with NeoDash
- 24:48Impact and future applications
- 25:41Conclusion and Q&A
related talks
44mGive Your LLMs a Left Brain
25mEnhance LLM Applications' Trustworthiness and Explainability
30mBuilding Knowledge Graphs With LLMs
20mPractical GraphRAG: Making LLMs smarter with Knowledge Graphs
14mLLMs and Knowledge Graphs: A Case Study in Blue(y)
28mGrounded Intelligence Boosting LLMs with Real-Time Graph Context
30mMaking Virtuous Cycles: Lessons From the Frontier of LLM-Knowledge Graph Interaction
52mGive Your LLMs a Left Brain