: I think coord works at the document level, I was thinking of having : something that worked at a field level, against a 'principle/primary' : field.
I'm not sure what you mean by "works at hte document level" ... coord is used y the BooleanQuery scoring mechanism to define how scores should be affected when a document doens't match all terms in the query. Mikhail's suggestion was that with an appropriately definied coord method in a custom similarity, you could probably get close to what you are asking about using the "query()" function on a BooleanQuery containing hte terms you are interested in. It may be easier then that though ... you might want to take a look at the termfreq() and norm() functions .. combined with map() (to ensure you get a "1" for docs that match a term, no matter what the termfreq() is) you could probably get proportionate values to what you are looking for -- but the denominators won't be the exact number of terms in the field unless you customize the norm function in your similarity. in general though this smells like an XY problem, because the kind of boosting you seem to be trying to achieve sounds like exactly what the normal TF/IDF scoring algorithm will give you. so perhaps you should tell us more about some real world specifics of hte types of data/query you are using, what types of results you are seeing, and the types of results you want... https://people.apache.org/~hossman/#xyproblem XY Problem Your question appears to be an "XY Problem" ... that is: you are dealing with "X", you are assuming "Y" will help you, and you are asking about "Y" without giving more details about the "X" so that we can understand the full issue. Perhaps the best solution doesn't involve "Y" at all? See Also: http://www.perlmonks.org/index.pl?node_id=542341 -Hoss
