CODEWITHSUNDEEP
solrfilterschematokenize
Andrew
Andrew

Reputation: 58

Solr Tokeniser/Filter not scoring catenated/tokenised words

I'm having trouble with solr field types, basically I can't seem to get it to return a decent score for stupidly similar terms.

Heres an example: A store called "Bagelkingdom"

I have two records for this - "Bagelkingdom" and "Bagel Kingdom"

When i search on "Bagelkingdom" I get a score of 14, it returns "Bagel Kingdom" too, but only with a score of 4!

When i do the reverse search on "Bagel Kingdom" I get a score of 13 and it again returns "Bagelkingdom", but this time with a score of 0.16.

I won't paste all my schema file here - I've tried so many different filters the current one doesn't do much.

At the moment my filter is;

     <fieldType name="text_catenate" class="solr.TextField"  positionIncrementGap="2"  autoGeneratePhraseQueries="true">
  <analyzer>
            <filter class="solr.LowerCaseFilterFactory"/>
            <filter class="solr.WordDelimiterFilterFactory" generateWordParts="1"
                    generateNumberParts="1"
                    catenateWords="1"
                    catenateNumbers="1"
                    catenateAll="1"
                    splitOnCaseChange="0"
                    splitOnNumerics="0"
                    preserveOriginal="1"/>
                    <tokenizer class="solr.KeywordTokenizerFactory"/>
  </analyzer>
</fieldType>

Upvotes: 0

Views: 31

Answers (1)

alexf
alexf

Reputation: 1301

Maybe you can try to analyze your score to fix it. You can use for example this tool to observe how score is build.

Upvotes: 1

Related Questions