Question Regarding Japanese Tokenizer in Databricks Vector Search #28072
yamaguchiraf
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Does Databricks Vector Search support a Japanese tokenizer? Or does it accept the specification of a tokenizer? Currently, we are using hybrid search, but the search performance for Japanese proper nouns is low.
Beta Was this translation helpful? Give feedback.
All reactions