The field reference applies only to the single term, quoted phrase, or
parenthesized sub-query that follows it, so maybe you meant:
(*:* -field1:"ABC XYZ")^1000
or
(*:* -field1:(ABC XYZ))^1000
Your pick as to whether you require the terms to be adjacent or merely
somewhere in the field. You could also use a slop factor on the quoted
phrase
-- Jack Krupansky
-----Original Message-----
From: mechravi25
Sent: Monday, October 01, 2012 8:00 AM
To: solr-user@lucene.apache.org
Subject: How does negative query boosting works in solr 3.6.1
I am using a negative query boosting for a particular field as given in the
solr wiki
(*:* -field1:ABC XYZ)^1000
When I do a search now, the top results does not contain value "ABC XYZ"
from field1.
When I give a search value as ABC DEF and accordingly change the negative
query boosting in bq parameter,
I notice that the 1st result fetched has the value "ABC DEF" in field1.
My question is will the negative query boosting have to be changed for each
value searched or is
the results are returned based on the scoring for each fields?
I have not given any index time boosts and how does solr does scoring
internally for other fields?
I tested this search in two solr instances (solr 1.4 and solr 3.6.1) and the
scoring was higher in solr 3.6.1 compared to
solr 1.4 for the same string. I am using the same set of config files but I
am not sure why the scoring is higher in 3.6.1.
Please guide me.
--
View this message in context:
http://lucene.472066.n3.nabble.com/How-does-negative-query-boosting-works-in-solr-3-6-1-tp4011166.html
Sent from the Solr - User mailing list archive at Nabble.com.