Hello what are the best resources to learn about GraphRAG and put it into production? Also is Neo4j the best DB to deploy a graph rag solution? If so how does adding and removing documents work exactly? I can’t seem to find much info on document/chunk insertion and deletion.
Yes! So I run an edtech platform and we currently have SOTA RAG for individual files using an improved RAPTOR implementation. We want to scale this approach to the entire course as opposed to just one file. However Raptor seems to not scale well beyond a few files. This is where GraphRAG comes in handy. The issue with graphRAG though is that you have to recreate communities with every new file uploaded. This is extremely expensive of course. So I want to figure out a way where we can do entity detection once and cheaply cluster the new files with the old communities.
@biswaroop I know there is currently a PR open in Microsoft GraphRAG for this exact case. But would love to see if llamaindex has some ideas too. This is critical in bringing GraphRAG to production when files are dynamic.
