Data Engineering for Graph-based Retrieval Augmented Generation GraphRAG using Neo4j
summary
Shows a concrete pipeline that turns SEC Edgar PDFs and Form 13 tables into a Neo4j-backed GraphRAG: chunking 10‑K text, storing chunks and form rows as nodes, adding embeddings and summaries, linking chunks with Minimal Viable Graph sequences and hierarchical summaries, and querying via Cypher templates, dynamic text‑to‑Cypher (fine‑tuned models) or vector search. Demo uses NetApp data, Lucene vector store with HNSW and Lucene indexing, geospatial indexes, and reports improved factuality and explainability over plain vector RAG.
AI-written summary · about
chapters (7)
- 0:00Introduction and overview
- 3:12Basics of RAG and GraphRAG
- 8:56Property graph model explained
- 15:16Building knowledge graphs
- 24:52Retrieval patterns for GraphRAG
- 36:32Demo and practical examples
- 44:10Resources and closing
related talks
37mBuilding advanced RAGs with Elasticsearch OpenSearch
29mRAG for Data Engineers: Why Your Metadata Is the Missing Piece
44mEnhance Your Generative AI Applications with RAG
17mScaling Enterprise-Grade RAG: Lessons from Legal Frontier
32mRevolutionizing ETL for RAG with
50mBuild a ChatGPT RAG Data Pipeline
30mHow Graph Data Science Can Turbocharge Your Knowledge Graph
30mBeyond the Silos Achieving Data Harmony With Graph Based ER