I have a dismax query where I check for values in 3 fields against documents in the index - a title, a list of keyword tags and then full-text of the document.
I usually get lots of results and I can see that the first results are OK - it's giving precedence to titles and tag matches, as my dismax boosts on title and keywords (normal boost and phrase boost). After say 20/30 good results I start to get matches based upon just the full-text, so these are less relevant. I am also facet.couting on my keyword tags (and presenting in the results as a way of filtering) and as you can imagine the counts are high because of the number of overall results. I want to somehow make the facet counts more associated with the higher relevancy results. My options as I see it are - 1) exclude full-text from the dismax altogether 2) configure the dismax normal boost on full-text to zero, but phrase boost to something higher (the aim here is to only really get a hit on the full-text if my search term is foound as a phrase in the full-text) 3) limit my results by relevancy or number of results If I do (3) above will the facet.counts respect the lower number of results - this is the overall aim really. Thank You Jason. -----Original Message----- From: Erick Erickson [mailto:[email protected]] Sent: Wed 03/11/2010 23:15 To: [email protected] Subject: Re: Filter by relevance Be aware, though, that relevance isn't absolute, it's only interesting #within# a query. And it's then normed between 0 and 1. So picking "a certain value" is rarely doing what you think it will. Limiting to the top N docs is usually more reasonable.... But this may be an XY problem. What is it you're trying to accomplish? Perhaps if you state the problem, some other suggestions may be in the offing.... Best Erick On Wed, Nov 3, 2010 at 4:48 PM, Jason Brown <[email protected]> wrote: > Is it possible to filter my search results by relevance? For example, > anything below a certain value shouldn't be returned? > > I also retrieve facet counts in my search queries, so it would be useful if > the facet counts also respected the filter on the relevance. > > Thank You. > > Jason. > > If you wish to view the St. James's Place email disclaimer, please use the > link below > > http://www.sjp.co.uk/portal/internet/SJPemaildisclaimer > If you wish to view the St. James's Place email disclaimer, please use the link below http://www.sjp.co.uk/portal/internet/SJPemaildisclaimer
