Find answers from the community

Updated 2 months ago

Hello,

Hello,
Suppose I have thousands of one-page documents (such as CVs) and I want to find the top 5 documents that meet a specific criteria. Do you have any recommendations for the architecture I should use? In light of the fact that these are unrelated one-page documents, I'm wondering if I should include embeddings at all (assuming I'm not trying to save money on LLM).
TIA!!

7 comments

TTeemu

Which criteria are you using? Semantic/vector search can work pretty well for those types of use-cases, just depends on exactly what type of information/criteria you're searching for

yyoelk

What do you mean by semantic search? Does it use embeddings?

yyoelk

I think embedding might not work well if my criteria is "5+ years experience in software development "

yyoelk

I guess the llm would need to scan the whole CV to decide whether it matches the criteria or not

TTeemu

Yeah in that case you might need to, semantic search doesn't do well with numbers

TTeemu

Yeah embeddings use the semantic meaning to find similar text snippets

yyoelk

I guess sometimes you just need to apply a simple solution without using the superpowers of additional libraries

Add a reply