▶ Watch ↗AI Engineer World's Fair 20241:39:52
Knowledge Graphs & GraphRAG: Techniques for Building Effective GenAI Applications
Read the full talk →Key ideas
Scroll to read ↓Build a retail graph, search product descriptions, personalize results through purchase paths, and use graph embeddings to supply recommendations for a seasonal email.
- Prepare the notebook and graph database0:26 ↗
- Find an item, personalize it, then suggest what goes with it5:59 ↗
- Build the customer–article–product graph11:56 ↗
- Generate embeddings before creating the index26:57 ↗
- Wrap retrieval without coupling it to one storage layout35:47 ↗
- Derive a peer group from shared purchases45:18 ↗
- Rerank semantic matches without losing cold-start results57:39 ↗
- Turn co-purchase structure into reusable recommendation links1:09:33 ↗
- Refresh embeddings and choose their size1:20:07 ↗
- Use graph queries for questions similarity cannot answer1:23:44 ↗
- Feed two retrieved product lists into one email prompt1:29:30 ↗
- Change the customer, then change the season1:37:04 ↗