lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joshua O'Madadhain <jmad...@ics.uci.edu>
Subject RE: LIFO or FIFO??
Date Tue, 09 Jul 2002 03:49:12 GMT
Samir:

This is a FAQ: 

http://lucene.sourceforge.net/cgi-bin/faq/faqmanager.cgi?file=chapter.search&toc=faq#q31

which you can get to by going to the main Lucene page, following the
official FAQ link, and looking under "Searching".

Joshua

 jmadden@ics.uci.edu...Obscurium Per Obscurius...www.ics.uci.edu/~jmadden
  Joshua O'Madadhain: Information Scientist, Musician, Philosopher-At-Tall
 It's that moment of dawning comprehension that I live for--Bill Watterson
My opinions are too rational and insightful to be those of any organization.

On Mon, 8 Jul 2002, Samir Satam wrote:

> Thank you Brandon, That surely is what seems to be happening... Does
> anybody know the algorithm lucene uses to calculate hit score?
> 
> thanks again,
> samir
> 
> -----Original Message-----
> From: Brandon Jockman [mailto:brandonj@isogen.com]
> Sent: Monday, July 08, 2002 12:18 PM
> To: Lucene Users List
> Subject: Re: LIFO or FIFO??
> 
> 
> Samir,
> 
> I think its more likely that the more recently indexed articles had a higher
> 'score' and thus were displayed first in results. Hits are ranked based on
> an algorithm (see FAQ). You should printout the 'score' for each document
> hit and see that the first results displayed actually had a higher score
> than the older ones. I don't think FIFO/LIFO have anything to do with it...
> Look at the scoring algorithm; then try searching for things that should
> have a higher score in the older documents and see if they come up higher in
> your results.
> 
> Hope that helps,
> 
> -Brandon
> - brandonj@isogen.com
> 
> ----- Original Message -----
> From: "Samir Satam" <ssatam@copyright.com>
> To: "Lucene Users List" <lucene-user@jakarta.apache.org>
> Sent: Monday, July 08, 2002 11:03 AM
> Subject: LIFO or FIFO??
> 
> 
> > Hello,
> > I am new to Lucene, and during my development, I observed that...
> >
> > the articles indexed last were the articles that were returned first, in
> the results. As if the index was a LIFO. Now this is perfectly acceptable as
> long as we need to get the latest documents. (articles on WSJ published
> today were indexed today, nad hence are more recent and more relevant). But
> I also have to index the older articles.  So, I have 2 questions...
> > 1. Is my observation correct?
> > 2. I was wondering if anybody had similar problems, and how did they fix
> it?
> >
> >
> > thanks
> > Samir
> >
> > --
> > To unsubscribe, e-mail:
> <mailto:lucene-user-unsubscribe@jakarta.apache.org>
> > For additional commands, e-mail:
> <mailto:lucene-user-help@jakarta.apache.org>
> >
> >
> 
> 
> --
> To unsubscribe, e-mail:   <mailto:lucene-user-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>
> 
> 
> --
> To unsubscribe, e-mail:   <mailto:lucene-user-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>
> 
> 


--
To unsubscribe, e-mail:   <mailto:lucene-user-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:lucene-user-help@jakarta.apache.org>


Mime
View raw message