Best way to index Confluence pages?

Hi Weaviators,

I’m new to the Weaviate and the RAG scene. I’ve been playing around with LlamaIndex and essentially followed this guide but using the Confluence Reader for data ingestion. - https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/WeaviateIndexDemo.html

For my use-case, I am trying to build a chat over my organization’s Confluence pages, but I’m having trouble getting the retrieval to retrieve the right pages without having to use ExactMatchFilter. This seems like a really manual process that I think would dampen user experience. I was wondering if anybody knows the best way to index the pages in such a way that users can just provide a general query to the RAG app like “What was discussed in the 5/11/2023 meeting notes?” and have it return an LLM generated analysis of the page titled “5/11/2023 Meeting Notes”.

Any help would be appreciated. Thanks!

Hi @Ben6529 !! Welcome to our commnunity :hugs:

Have you tried using the hybrid search? if you want to do exact search together with semantic, you can leverage this option and change the alpha parameter to your needs.

Let me know if that helps.

Thanks

Not yet. I’ll give that a shot! Thanks!