Applying negative boost to some documents after rescoring.

o19s / elasticsearch-learning-to-rank

Plugin to integrate Learning to Rank (aka machine learning for better relevance) with Elasticsearch

http://opensourceconnections.com/blog/2017/02/14/elasticsearch-learning-to-rank/

Apache License 2.0

1.48k stars 370 forks source link

Applying negative boost to some documents after rescoring. #391

Open meg261995 opened 3 years ago

meg261995 commented 3 years ago

Hi Team,

So I have a set of documents which are passed to my lambdamart model for rescoring. However after the rescoring, I still want to reduce the relevance of some of the documents. Is this possible? If not is there any other method to do this?

nathancday commented 2 years ago

After the re-scoring phase you can no longer change the score of a document.

Could you use sorting on [condition, _score] to bury some of these unwanted docs? I guess you could also craft/insert features into your models if you wanted to handle it there.

Sorry for the very delayed response.