lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Markus Jelsma <markus.jel...@openindex.io>
Subject Re: omitNorms and very short text fields
Date Tue, 01 Apr 2014 19:27:44 GMT
Yes, that will work. And combined with your other question scores will always be equal even
if cinderella or chuck occur more than once in one document.



Walter Underwood <wunder@wunderwood.org> schreef:Just double-checking my understanding
of omitNorms.

For very short text fields like personal names or titles, length normalization can give odd
results. For example, we might want these two to score the same for the query "Cinderella".

* Cinderella
* Cinderella (Diamond Edition) (Blu-ray + DVD + Digital Copy) (Widescreen)

And these two for the query "chuck":

* Chuck House
* Check E. Cheese

I think that omitNorm=true on those fields will give that behavior. Is that the right approach?

wunder
--
Walter Underwood
wunder@wunderwood.org



Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message