lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From parnab kumar <>
Subject Using new similarities in Lucene 4.0
Date Tue, 30 Oct 2012 14:20:28 GMT
Hi all,

      Lucene 4 has introduced several state of the art ranking functions. I
was wondering how could i make use of those similarities .These models
obviously uses some more term and collection statistics as opposed to
the traditional Vector Space Model . I guess these statistics
are  precomputed during indexing .Should there be any changes to the
indexing mechanism ,like specifying the indexing options to make use of
these models .


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