Solr: Accelerating phonetic variations?

Question

I am trying to search in two fields, each of which has its own boost coefficient and includes phonetic variations, but the results with these variations should always be ranked lower in the results. The current problem is that results with phonetic changes in a field with a higher increase are preferable to results with an exact match in a field with a lower gain.

In schema.xml, I have a field called "text" containing two other searchable fields ("name" and "description" implemented via copyField), each with its boost factor (defined in the loss of SearchHandler in the solrconfig.xml file ) This field has a solr.PhoneticFilterFactory filter with DoubleMetaphone on the "index" and "query" analyzers. As far as I understand, phonetic variations of each word are added to the query and index.

My question is, how can I tell solr to give a separate boost factor (e.g. 0.3) for phonetic variations?

Answer 1

One possible solution is to create two more fields: Suppose your source fields are called "title" and "description". Create "title_phonetic" and "description_phonetic" copy_fields and add only phonetic variations to these fields. Then, use the analyzer parser to give different enhancements for these fields.