search

microsoft / graphrag

A modular graph-based Retrieval-Augmented Generation (RAG) system
https://microsoft.github.io/graphrag/
MIT License
16.86k stars 1.58k forks source link

[Issue]: <title> #768

Closed 1linS closed 3 weeks ago

1linS commented 1 month ago

Is there an existing issue for this?

Describe the issue

When using local search to obtain entities based on a query, the map_query_to_entities function utilizes the text_embedding_vectorstore.similarity_search_by_text to calculate similarity. Is this function calculating the similarity between the query and the description_embedding? Can it be replaced with the similarity calculation of the entities' embeddings?

Steps to reproduce

No response

GraphRAG Config Used

# Paste your config here

Logs and screenshots

No response

Additional Information

natoverse commented 1 month ago

The similarity search is comparing the query against the embeddings for the entity.description field. That is currently the only field that we compute embeddings for by default, and we have found good success with that as a way of matching relevant entities to a query

github-actions[bot] commented 1 month ago

This issue has been marked stale due to inactivity after repo maintainer or community member responses that request more information or suggest a solution. It will be closed after five additional days.

github-actions[bot] commented 3 weeks ago

This issue has been closed after being marked as stale for five days. Please reopen if needed.