Real-Time Data Transformation for Knowledge Graph with LLMs
summary
Shows a concrete pipeline that continuously turns markdown into a live graph: CocoIndex ingests files, calls GPT-4o to extract titles, summaries, and subject–predicate–object triples, and streams nodes and edges into Neo4j with declared labels and primary keys. Claims real‑time export and automatic upserts so entities deduplicate, document→term citations persist, and incremental processing keeps the graph in sync with source changes. Includes code snippets for data classes, extraction calls, and the declarative export mapping.
AI-written summary · about
chapters (7)
- 0:00Introduction to CocoIndex
- 1:44Knowledge graph terminology
- 3:40Document data extraction
- 5:30Entity and triple extraction
- 7:33Mapping data to Neo4j
- 9:33Exporting relationships
- 10:47Simplifying graph maintenance
related talks
30mBuilding Knowledge Graphs With LLMs
28mGrounded Intelligence Boosting LLMs with Real-Time Graph Context
20mPractical GraphRAG: Making LLMs smarter with Knowledge Graphs
28mThe Present and Future of AI-Enabled Knowledge Graph Generation
53mPractical GraphRAG - Making LLMs smarter with Knowledge Graphs
42mMaking OSS AI Real with Knowledge Graphs
25mEnhance LLM Applications' Trustworthiness and Explainability
28mBuilding Knowledge Graph-Based Agents with Structured Text Generation