Multi Value field

[John Seer]

Hello,

I  have 100k index with documents with one searchable field in it. That field has multiple values for example

doc( search: X search : X Y search: X Y Z id:1)
doc( search: X Y K id:2)

I am using Standard Analyzer for building and searching, and having problem with scores

if the term is "X Y" the document 2 is getting higher score then document 1.


Please can someone suggest way that doc 1 will score higher?
Maybe I need to use some specific query type?

Thanks

[markharw00d]

> if the term is "X Y" the document 2 is getting higher score then  
> document 1.


That may be length normalisation at play. Doc 2 is shorter so may be  
seen as a better match for that reason.
Using the "explain" function helps illustrate the break down of scores  
in matches.

You could try indexing that field without "norms" or use a custom  
Similarity at query time to ignore the length normalisation factor.

Cheers
Mark



On 7 Jul 2009, at 19:31, John Seer wrote:

>
> Hello,
>
> I  have 100k index with documents with one searchable field in it.  
> That
> field has multiple values for example
>
> doc( search: X search : X Y search: X Y Z id:1)
> doc( search: X Y K id:2)
>
> I am using Standard Analyzer for building and searching, and having  
> problem
> with scores
>
> if the term is "X Y" the document 2 is getting higher score then  
> document 1.
>
>
> Please can someone suggest way that doc 1 will score higher?
> Maybe I need to use some specific query type?
>
> Thanks
>
>
>
> --
> View this message in context: http://www.nabble.com/Multi-Value-field-tp24378975p24378975.html
> Sent from the Lucene - Java Users mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@...
> For additional commands, e-mail: java-user-help@...
>

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

[John Seer]

I already tried to use custom similarity (I set all methods to return 1f)-  doesn't work.
I just try norms idea as well no change

markharw00d wrote:

> if the term is "X Y" the document 2 is getting higher score then  
> document 1.


That may be length normalisation at play. Doc 2 is shorter so may be  
seen as a better match for that reason.
Using the "explain" function helps illustrate the break down of scores  
in matches.

You could try indexing that field without "norms" or use a custom  
Similarity at query time to ignore the length normalisation factor.

Cheers
Mark



On 7 Jul 2009, at 19:31, John Seer wrote:

>
> Hello,
>
> I  have 100k index with documents with one searchable field in it.  
> That
> field has multiple values for example
>
> doc( search: X search : X Y search: X Y Z id:1)
> doc( search: X Y K id:2)
>
> I am using Standard Analyzer for building and searching, and having  
> problem
> with scores
>
> if the term is "X Y" the document 2 is getting higher score then  
> document 1.
>
>
> Please can someone suggest way that doc 1 will score higher?
> Maybe I need to use some specific query type?
>
> Thanks
>
>
>
> --
> View this message in context: http://www.nabble.com/Multi-Value-field-tp24378975p24378975.html
> Sent from the Lucene - Java Users mailing list archive at Nabble.com.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [e-mail]
> For additional commands, e-mail: [e-mail]
>


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

[markharw00d]

> I just try norms idea as well no change

You'll need to look at searcher.explain() for the two docs or post a  
Junit or code example that can be executed which shows the issue


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