You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by "EXTERNAL Taminidi Ravi (ETI, AA-AS/PAS-PTS)" <ex...@us.bosch.com> on 2015/07/20 16:16:34 UTC

Rule for Score

Hi All, Can you someone explain or refer me right place to know more on how the Score is calculated, I am seeing it has few attribute like, termfrequency, document frequeuncy, weight, boost also it is say sum of , product of..

Is there any example of understanding the basics of these attributes and the logic for calculating the score.

Thanks

Ravi

RE: Rule for Score

Posted by "EXTERNAL Taminidi Ravi (ETI, AA-AS/PAS-PTS)" <ex...@us.bosch.com>.
Thanks Shawn, I will Check that.


Ravi

-----Original Message-----
From: Shawn Heisey [mailto:apache@elyograg.org] 
Sent: Monday, July 20, 2015 10:23 AM
To: solr-user@lucene.apache.org
Subject: Re: Rule for Score

On 7/20/2015 8:16 AM, EXTERNAL Taminidi Ravi (ETI, AA-AS/PAS-PTS) wrote:
> Hi All, Can you someone explain or refer me right place to know more on how the Score is calculated, I am seeing it has few attribute like, termfrequency, document frequeuncy, weight, boost also it is say sum of , product of..
> 
> Is there any example of understanding the basics of these attributes and the logic for calculating the score.

If you add "debugQuery=true" to your URL request, then you will get a full accounting of how the score was calculated for each document in the results.  I think "debug=query" also works.

Looking at the source code for DefaultSimilarity is probably also helpful.  Here's the javadoc for that class:

http://lucene.apache.org/core/5_2_0/core/org/apache/lucene/search/similarities/DefaultSimilarity.html

Much more detailed info is available by clicking on the parent class link (TFIDFSimilarity) on this page.

Thanks,
Shawn

Re: Rule for Score

Posted by Shawn Heisey <ap...@elyograg.org>.
On 7/20/2015 8:16 AM, EXTERNAL Taminidi Ravi (ETI, AA-AS/PAS-PTS) wrote:
> Hi All, Can you someone explain or refer me right place to know more on how the Score is calculated, I am seeing it has few attribute like, termfrequency, document frequeuncy, weight, boost also it is say sum of , product of..
> 
> Is there any example of understanding the basics of these attributes and the logic for calculating the score.

If you add "debugQuery=true" to your URL request, then you will get a
full accounting of how the score was calculated for each document in the
results.  I think "debug=query" also works.

Looking at the source code for DefaultSimilarity is probably also
helpful.  Here's the javadoc for that class:

http://lucene.apache.org/core/5_2_0/core/org/apache/lucene/search/similarities/DefaultSimilarity.html

Much more detailed info is available by clicking on the parent class
link (TFIDFSimilarity) on this page.

Thanks,
Shawn