The length normalization "factor" is a very coarse value, so it may not be
fine-grained enough to distinguish these particular field lengths. Normally,
it is a short vs. long distinction rather than actual length.
In any case, add &debugQuery=true to your query and look at the "explain"
section see how "norm" is either different or the same for these three
documents in the results. The norm may in fact be fine, but maybe some other
factors overwhelm the overall score.
-- Jack Krupansky
-----Original Message-----
From: Timothy Potter
Sent: Monday, March 11, 2013 10:43 AM
To: solr-user@lucene.apache.org
Subject: Re: Boost maximum match in a field
I'm curious if the default ranking doesn't already return these in 3,2,1
order. Doc 3 should get an implicit boost with norms enabled for your title
field, so make sure the title field has omitNorms="false", ie. in
schema.xml:
<field name="title" ... omitNorms="false"/>
Tim
On Mon, Mar 11, 2013 at 8:02 AM, Nicholas Ding <nicholas...@gmail.com>wrote:
Hello,
I was wondering how to boost a maximum match in a field. For example, you
have few documents has different length of title.
Doc 1:
Title: Ford Car Body Parts
Doc 2:
Title: 2012 Ford Car
Doc 3:
Title: Ford Car
If user searching for "Ford Car", how to make the Doc 3 has the highest
score?
Thanks
Nicholas
