Hi folks, I'm having an issue where I want MLT to operate on multiple fields, one of which contains a large number of terms (that is, each document in the index has many terms for this field) and the others only a few terms per document. In my situation, the fields with the fewer terms I am boosting, because they are what I'm particularly interested in, and the field with many terms is to be the "fallback" (so a lower boost).
The problem is that because there are so many common terms in the one field across the documents in the index, MLT is returning only interesting terms from this field. If I increase the mlt.maxqt I can find the terms from the other (more important) fields, but then I have included many more terms from the already over-influencing field. Hopefully I have explained this well enough. My question is if there is a mechanism in place (I haven't been able to find one) to make those fields I have boosted the ones that come up first in the list of interesting terms. Thanks, Nick Spacek
