[prev in list] [next in list] [prev in thread] [next in thread] 

List:       solr-user
Subject:    Re: score
From:       Sergio_GarcĂ­a_Maroto <marotosg () gmail ! com>
Date:       2023-02-25 8:15:35
Message-ID: CAD+Y64tYAz7Y0ZEKSh+FXUqQPfeHSBe-+eoUaxK8hD4k1uqsuw () mail ! gmail ! com
[Download RAW message or body]


Hi Mikhail.
That ´s right. Deleting that similarity class  and addding omitNorms=true
returned the expected behaviour,

Regards,
Sergio

On Fri, 24 Feb 2023 at 20:06, Mikhail Khludnev <mkhl@apache.org> wrote:

> Hello Sergio.
> I'm noticing a custom similarity class in the schema excerpt. I'm not sure
> how it's related to the problem.
> Can you check debugQuery=true and check explanations for matches, which are
> scored against expectation?
>
> On Fri, Feb 24, 2023 at 1:31 PM Sergio GarcĂ­a Maroto <marotosg@gmail.com>
> wrote:
>
> > Hi all,
> >
> > I am dealing with an scoring challenge I am not fully sure how to solve.
> > I have the following field which contains a list of job records in a
> > multivalue field. All data is concatenated as below
> > Problem is I am getting a smaller record with less jobs first due to
> > scoring penalizing the biggest records even they have more times the
> tokens
> > to be matched.
> > I used omitnorms=true and then I am getting same score.
> >
> > Question is?
> > Can i get more score for the record with more tokens changing something?
> >
> > Query:
> > ((SmartSearchS:"main director [$CU] [$PRJ] "~100)^5 OR
> > ((SmartSearchS:(main) AND (SmartSearchS:(director*))) OR
> > ((SmartSearchS:("main director")))^3))
> >
> > <doc>
> > <str name="PersonIDDoc">-730007</str> <arr name="JobSearch"> <str>[$CU]
> > [$PRJ] main director Ucona corporation Illinois Midwest(USA) United
> States
> > of America North America</str> </arr> <doc> <str
> name="PersonIDDoc">-730008
> > </str> <arr name="JobSearch"> <str>[$CU] [$PRJ] main director Ucona
> > corporation Illinois Midwest(USA) United States of America North
> America</
> > str> <str>[$CU] [$PRJ] main director Ucona corporation Illinois
> > Midwest(USA) United States of America North America</str> </arr> </doc>
> >
> > This the field and type
> > <field name="JobSearch" type="JobSearchField" indexed="true"
> stored="true"
> > multiValued="true" />
> >
> > <fieldType name="JobSearchField" class="solr.TextField"
> > positionIncrementGap
> > ="500"> <analyzer type="index"> <tokenizer class=
> > "solr.WhitespaceTokenizerFactory"/> <filter class=
> > "solr.WordDelimiterGraphFilterFactory" generateWordParts="0"
> > generateNumberParts="1" catenateWords="1" catenateNumbers="1"
> catenateAll=
> > "1" splitOnCaseChange="0" preserveOriginal="0"
> protected="protwords.txt"/>
> > <
> > filter class="solr.ASCIIFoldingFilterFactory"/> <filter class=
> > "solr.LowerCaseFilterFactory"/> <filter class=
> > "solr.SynonymGraphFilterFactory" synonyms="positionsynonyms.txt"
> > ignoreCase=
> > "true" expand="true"/> <filter class="solr.SynonymGraphFilterFactory"
> > synonyms="locationsynonyms.txt" ignoreCase="true" expand="true"/> <filter
> > class="solr.FlattenGraphFilterFactory"/> </analyzer> <analyzer
> > type="query">
> > <tokenizer class="solr.WhitespaceTokenizerFactory"/> <filter class=
> > "solr.WordDelimiterGraphFilterFactory" generateWordParts="0"
> > generateNumberParts="1" catenateWords="1" catenateNumbers="1"
> catenateAll=
> > "1" splitOnCaseChange="0" preserveOriginal="0"
> protected="protwords.txt"/>
> > <
> > filter class="solr.ASCIIFoldingFilterFactory"/> <filter class=
> > "solr.LowerCaseFilterFactory"/> </analyzer> <similarity class=
> > "com.spencerstuart.similarities.SpencerStuartNoSimilarity"></similarity>
> </
> > fieldType>
> >
> >
> > Thanks a lot
> > Sergio Maroto
> >
>
>
> --
> Sincerely yours
> Mikhail Khludnev
> https://t.me/MUST_SEARCH
> A caveat: Cyrillic!
>


[prev in list] [next in list] [prev in thread] [next in thread]