[Snowball-discuss] German Stemmer

Richard Boulton richard at tartarus.org
Wed Nov 4 14:45:49 GMT 2009


2009/11/4 Tobias N. Sasse <tobi at byte23.de>:
> Why don't you find stopword removal useful in your scenario?

The main reason is that usually, I'm working with a probabilistic
search engine (eg, http://xapian.org/) in which the importance of
words is calculated partially based on their frequency.  For
stopwords, the frequency is very high, so the ranking tends to
automatically reduce their importance appropriately.

-- 
Richard



More information about the Snowball-discuss mailing list