Hey Doug, Thanks for reply. Details regarding my use case:
I need to query on two fields in my indexes, and will consider score only for the one which has highest score. So i create two boolean queries [these queries are made up of other differet tokens, so chose boolean query] on those two fields, and added as disjuncts on one dismax query. Now when i call setboost on the original dismax query it sets the boost on all its disjunct queries, so then this whole dismax returns boost*boost as boost gets applied twice. So then what i have to do is apply boost on dismax and then after that remove boost from its child to retain sanctity. I hope i made it clear. Thanks and regards, Gajendra Dadheech On Sat, Sep 26, 2015 at 9:02 PM, Doug Turnbull < dturnb...@opensourceconnections.com> wrote: > Hi Gajendra > > What do you mean by "behaving odd"? What odd behavior do you see? > > My colleague John Berryman and I have seen many ways dismax has "behaved > odd" and we've tried to write about them (it quite often surprises me too) > > > http://opensourceconnections.com/blog/2013/07/02/getting-dissed-by-dismax-why-your-incorrect-assumptions-about-dismax-are-hurting-search-relevancy/ > > http://opensourceconnections.com/blog/2013/03/07/the-anatomy-of-a-dismax-query/ > > http://opensourceconnections.com/blog/2013/04/15/querying-more-fields-more-results-stop-wording-and-solrs-mm-min-should-match-argument/ > Chapter 6 in Relevant Search <http://manning.com/turnbull>, where I > discuss > query-string methods & term centric search (dismax is a rudimentary form of > term centric search) > > Hope that helps, > -Doug > > > On Sat, Sep 26, 2015 at 11:05 AM, Gajendra Dadheech <gajju3...@gmail.com> > wrote: > > > Hi, > > > > As i can see in solr code, DisJunctionMaxQuery extends Query, so by > default > > is inherits the functionaltiy. So Query class has one method called > > SetBoost, this is behaving very odd when applied on a DisMax Query. Can > > someone guide me regarding how calling SetBoost on a dismax query would > > impact its dijuncts [which could be dismax or boolean] > > > > Thanks and regards, > > Gajendra Dadheech > > > > > > -- > *Doug Turnbull **| *Search Relevance Consultant | OpenSource Connections > <http://opensourceconnections.com>, LLC | 240.476.9983 > Author: Relevant Search <http://manning.com/turnbull> > This e-mail and all contents, including attachments, is considered to be > Company Confidential unless explicitly stated otherwise, regardless > of whether attachments are marked as such. >
