lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: Using HitCollector to Collect First N Hits
Date Sat, 22 Aug 2009 11:45:35 GMT
> I’m using Lucene 2.4.1 and I’m trying to use a custom
> HitCollector to collect only the first N hits (not the best hits) for
> performance. 

You mean that you do not need score calculation therefore you do not want results sorted by
relevancy. Just you need is a Boolean Retrieval Model, right? All results will have ConstantScore
(0 or 1).

If yes, I think omitting norms and tfs of your searchable fields can improve your search performance
without writing custom HitCollector.

void org.apache.lucene.document.Field.setOmitNorms(true) 
Using Field.Index.ANALYZED_NO_NORMS in constructor.
void org.apache.lucene.document.Field.setOmitTf(true)  

But i am not sure those changes are enough to convert Vector Space Model to Boolean Retrieval
Model. I think there should be another trick to not to calculate score/similarity at all.

Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message