#community-help

Questions on Hybrid Search Feature in Upcoming Release

TLDR Chetan inquired about the compatibility and parameter control of hybrid search feature. Jason clarified its functionality, indicating its compatibility with manually created vec field and parameters affecting keyword vs semantic search results.

Powered by Struct AI
3
2mo
Solved
Join the chat
Jul 24, 2023 (2 months ago)
Chetan
Photo of md5-f5bf390d5711f8a2f96bd93d1bc97d20
Chetan
05:32 PM
a few questions about the hybrid search feature in the upcoming release candidate:
• is it still compatible with documents indexed with a manually created vec field? or do i have to let typesense do the embedding for me from specific field
• is it possible to change the parameter controlling how keyword search results are weighed vs semantic search results?
Jason
Photo of md5-8813087cccc512313602b6d9f9ece19f
Jason
07:30 PM
> is it still compatible with documents indexed with a manually created vec field? or do i have to let typesense do the embedding for me from specific field
Yup, in 0.25.0.rc53 and later, you can send both a q parameter and also vector_query parameter to do a hybrid search and combine the results together, without having to use built-in embedding generation
07:31
Jason
07:31 PM
> is it possible to change the parameter controlling how keyword search results are weighed vs semantic search results?
This is how results are ranked in hybrid search (it’s called Rank Fusion): https://gist.github.com/jasonbosco/f4187f6b4f585d2dc8902af85408994a#hybrid-search-rank-fusion

So by default keyword search is given more priority over semantic search.

You can influence keyword search ranking and semantic search ranking independently, but not both of them together