Questions about Query's Idf and NEW Similarity fomula. Thanks!

[oiunt <xo...@gmail.com>]

        Nowadays, I’m doing research using the lucene to build the prototype.
I have read the book “Lucene in Action”, but can’t find the solution, so I
try to beg for your help.

I have 2 questions:
1)	I found that the term in query doesn’t use the idf but only tf, however,
in my application I have many queries and it will be inaccurate regardless
idf. My prototype will establish the traces between two kinds of document,
so the idf of the query is also important, could you give some advices?

2)	I try to use the WordNet to associate words, however, lucene provides
synonymy approach which considers the similarity between word “seek” and
word “search” is 1. In my research, I will consider the similarity
coefficient between words ranges from 0 to 1, and I revise the similarity
formula. I don’t know where to insert my formula into lucene.


         Thanks for your help!
-- 
View this message in context: http://www.nabble.com/Questions-about-Query%27s-Idf-and-NEW-Similarity-fomula.-Thanks%21-tf4127297.html#a11737297
Sent from the Lucene - Java Users mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org