sentence-transformers

/

all-MiniLM-L6-v2

like 1.7k

Hey !

I have an issue with my project : I have a bunch of medical drug file and I'm trying to find specific file for specific type of drugs

What I did is search for keywords in every pdf with a retriever retriever = db.as_retriever(search_kwargs={'k': 2}) and embedding, it returns me paragraph, using sentence transformers.

Then I compare with the specificity of the drug I am looking for by comparing embedding => so I'm comparing paragraphs with keywords,

The result are not the good files, do you have any ideas how to improve it, I was thinking to summarize the paragraphe before doing the comparison.

Thanks fo the help :))))