lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From EJLeVin1 <ejlev...@gmail.com>
Subject Lucene Per-User Relevancy
Date Wed, 13 May 2009 16:03:08 GMT

Hi,
     I am kind of new to Lucene so please bear with me if what I'm asking
sounds ridiculous. I am trying to get lucene to have higher amounts of
relevancy based on keywords that the user has marked in their profile as
important, and also based on keywords that the user dislikes.  Right now I
am doing my preferences through pulling the TermVector of the Document the
user rates up or down, then subsequently recording the terms for that as
either positive or negative on the users profile.  

The problem I am running into is this: 

1) How can I get those terms to affect the ranking that Lucene uses?  I
realize I could use term boosting, but the number of terms that will be
voted up or down will be very large, and I'm guessing that would be highly
inefficient... Which brings me to my second question.

2) Is there a way to get the number of keywords down for documents?  I
already am doing stemming, and removing the stop words.  I was wondering if
there was a way to tell Lucene to only use terms that have a higher
frequency then X.  Is that a bad idea?  Thoughts?

Thanks so much for your input, I really appreciate it.

-Eric
-- 
View this message in context: http://www.nabble.com/Lucene-Per-User-Relevancy-tp23524822p23524822.html
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


Mime
View raw message