Provides 2 custom similarity plugins for SOLR:
- Constant similarity To be used in places where you can't use a constant scorer. It simply assigns a constant score to a result.
- Bounded similarity A similarity that produces bounded, more predictable values. It is aimed to score small fields like names, keywords, categories, etc. It is possible to switch-off the tf-componentor the idf-component.
SOLR nowadays supports plugins per index. Install by doing following steps:
- goto the solr_home of you index
- create <solr_home>/lib directory
- copy the jar from the release into this directory
- stop / start SOLR.
Edit your managed-schema file and put following code in there:
<similarity class="nl.bitmanager.solr.similarities.BoundedSimilarityFactory">
<float name="max_idf">0</float>
<float name="max_tf">0.2</float>
<float name="force_tf">-1</float>
<float name="bias_tf">0.6</float>
<bool name="discount_overlaps">true</bool>
</similarity>
The above settings are the default, so you can omit them if you wish.
The meaning of the parameters is:
| Parameter | Default | Meaning |
|---|---|---|
| max_idf | 0 | Limits the influence of the idf. |
| max_tf | 0.2 | Limits the influence of the tf. |
| force_tf | -1 | Forces a tf (if >= 0). |
| bias_tf | 0.6 | Correction value in the division. |
| discount_overlaps | true | Determines whether overlap tokens (Tokens with 0 position increment) are ignored when computing norm. |
The total score is calculated as idf * tf, where 1<=idf<=(1+max_idf) and 0<=tf<=(1+max_tf).
As you can see, the idf component is switched off by default!
It is also possible to assign a similarity per fieldType. In that case, you have to put the above <similarity> node as a child of your <fieldType> node.