Bm25 tune parameters using elasticsearch

I indexed the data with some b and k1 values of the default bm25 algorithm. The thing is implementing relevance feedback on the search engine to tune those b and k1 values. How can I not use reindexing for each time and supply those values at runtime for querying, so that this entire process would be a lot less painful?

Thanks in advance.

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.