MailScanner and SA auto-learning

Kai Schaetzl maillists at conactive.com
Wed Apr 5 12:31:22 IST 2006


Martin Hepworth wrote on Wed, 5 Apr 2006 09:02:09 +0100:

> Has your bayes DB got the required 200 ham AND spam messages? I'm not sure 
> the bayes functions work at all without the required 400 seed emails..

Bayes works, autolearning doesn't. See my last posting. All the high scoring 
spam has too few header hits.

Kai

-- 
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www.conactive.com





More information about the MailScanner mailing list