[Classifier4j-devel] Where to put the stemmer?

[Matt C. <MCo...@my...>]

Provisions have been made for a custom tokenizer and a custom stop list.  The 
tokenizer excutes prior to the stop list being applied.  I initially thought 
that the stemmer would be part of the tokenizer, however, we know that we 
cannot stem before we apply the stop list.  Do we need to expand 
BayesianClassifier.java to accept an addition argument IStemmer?

If so, at what point do we pass the code to the Stemmer?

Looking around, I found the transformWord() method in BayesianClassifier.java 
and I called the stemmer method from there.  It works fine, but this is not a 
long term solution.

Matt Collier
RemoteIT
mco...@my...
877-4-NEW-LAN