Andrew Sykes wrote:
Stefano,
Would it be possible to create some seed data, or does some exist for
the Bayesian filter corpus, it seems to me that certain tokens are
always going to be considered as spam (v1agra/ciali$ etc...)
No, bayesian technique works fine when trained by the user.
It should be easy enough to start sending a few spam messages to the feeder.
E.g: with a basic "standard" corpus that consider v1agra spam and has no
informations about James being ham your message would have been deleted
by my bayesian.
Stefano
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
