lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nitin Chauhan <>
Subject Help for Implementing Most relevance Search algorithm in lucene for my project
Date Tue, 23 Dec 2014 06:08:07 GMT

I wanted to implement "most relevant search" in Lucene for my project. I am currently using
the Lucene provided by Hybris 5.3 i.e. Lucene 4.6.1. The scenario is that I have type ahead
functionality (autosuggest) implemented already in the project so when user starts typing
in the input box, after typing three characters he/she starts to see a drop down with suitable
options. Now I want those options to be based on Most relevant search .How do I move ahead
around similarities, scorer, score-boosting etc. to get this done. How do I enable getting
score of the document from scoredoc now as

searcher.setDefaultFieldSortScoring(true, false);  doesn't seem to work in Lucene 4.6.1

Also which similarity is best to build this up . The default is good enough for me or BM25
or any other ?

Nitin Chauhan

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message