Boosting Boolean Should's & Best Approach for Address


Couldn't find anything in the documentation about how do you go about boosting an individual should match? I'd hoped it would be a simple case of adding the boost: n into the object as with multi_search but it's not effecting my query score.

A sort of related problem, I originally split an address into multiple indexes but if I do a full address against one of these individual indexes it may come out with a very high match because rather than having '9 Cherrytree Grove' the index may just contain '9'.

Is it best just to concatenate the whole string and let elastic handle it even if the ordering of words in the address being used within the source isn't the same as that stored?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.