Hey,
I haven't seen much traffic on this list in a while. Here's an
announcement, though: I just got my AI::Categorize package in a state
where it's ready for other people to look at, so I uploaded it to CPAN:
The uploaded file
AI-Categorize-0.03.tar.gz
has entered CPAN as
file: $CPAN/authors/id/K/KW/KWILLIAMS/AI-Categorize-0.03.tar.gz
size: 15741 bytes
md5: 4de80054597876615caef87a6f6b8a0f
The package provides several algorithms for looking at a bunch of
pre-categorized training documents, learning categorization rules, then
categorizing new documents based on those rules.
I'm going to give a talk on the module at YAPC::America and again at
TPC, so if anybody has any feedback I'd love to hear it. So far there
are two categorization algorithms implemented, Naive Bayes and k-Nearest
Neighbor. There's a generalized API for categorization, so it's
relatively easy to add new categorization schemes. In this respect it's
kind of like DBI and its various DBD drivers (though the organization is
different).
There's also a simple tool, AI::Categorize::Evaluate, that helps compare
the results of categorization under various conditions.
------------------- -------------------
Ken Williams Last Bastion of Euclidity
[EMAIL PROTECTED] The Math Forum
