jpountz opened a new issue, #15768:
URL: https://github.com/apache/lucene/issues/15768

   ### Description
   
   Currently, duplicate terms in query strings see their scores boosted by 
their number of occurrences. It would likely be better to saturate the 
contribution of such repeated terms, as suggested in the original BM25 paper 
through the k3 tuning parameter.
   
   From a quick survey, k3 seems to often default to 7 or 8, see e.g. Terrier: 
https://github.com/terrier-org/terrier-core/blob/54e6faf802bb6275b53ad6de14a93a53389719b8/modules/core/src/main/java/org/terrier/matching/models/BM25.java#L48.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to