Knowledge Graphs & GraphRAG: Techniques for Building Effective GenAI Applications
summary
Workshop shows a working GraphRAG pipeline that combines Neo4j (sandbox) + OpenAI embeddings + LangChain retrievers to reduce hallucination and personalize e‑commerce output. It loads H&M Kaggle data in Colab, builds text embeddings and an HNSW vector index, creates FastRP graph embeddings and KNN-derived customer_also_likes edges, and reranks vector results by co‑purchase score. Final product is a Gradio email generator that pulls 10 search hits + recommendations and returns seasonally tuned marketing emails — demo runs end‑to‑end live.
AI-written summary · about
chapters (8)
- 0:00Workshop introduction
- 10:20Graph data modeling
- 13:18Building the knowledge graph
- 18:13Vector search overview
- 30:06Implementing vector search
- 45:30Graph pattern search
- 1:09:35Graph data science
- 1:29:30Building the LLM app
related talks
53mPractical GraphRAG - Making LLMs smarter with Knowledge Graphs
27mAdaptive Knowledge Graphs for Dynamic RAG Systems
36mAgentic GraphRAG: AI’s Logical Edge
36mGraphRAG Explained 🔥 Building Smarter LLMs
01h 41mGraph Intelligence: Enhance Reasoning and Retrieval Using Graph Analytics
01h 17mGraph Database and Retrieval Augmented Generation
42mMaking OSS AI Real with Knowledge Graphs
26mPractical GraphRAG Making LLMs smarter with Knowledge Graphs