Hi,
I have to change clucene scoring in the way that the number of documents that
contain a certain term
does not influence the final score when querying documents using this term.
In other words I have to make the score of a document independent of content
of the index - it should
be influenced only by the document itself and the query.
I understood that to achieve this it is enough to overwrite the Similarity and
make the inverse document
frequency - idf constant. This seems to work fine. But I noticed that the base
class Similarity contains two other
idf() functions which are not virtual - especially the float_t
Similarity::idf(Term* term, Searcher* searcher)
which retrieves the document frequency and calls the virtual idf() function to
compute the result.
I was wondering whether we could make those two other functions also virtual so
that I could save retrieving
the document frequency at all. I could just return the constant value.
Jiri
------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
http://p.sf.net/sfu/nokia-dev2dev
_______________________________________________
CLucene-developers mailing list
CLucene-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/clucene-developers
