[Mimedefang] Re: SPAM/HAM Trap

[Jonas Eckerman]

Daniel Aquino wrote:

> Well does auto learning learn more spam than ham or does it even out
> both of them ?

SpamAssassin simply learns mail that scores above 
bayes_auto_learn_threshold_spam as spam, and mail that scores 
below bayes_auto_learn_threshold_nonspam as ham. The resulting 
proportion between learned ham/spam therefore is a function of 
your mailflow and those two variables.

Please read the documentation for SpamAssassin. This is mentioned 
in the documentation for 
Mail::SpamAssassin::Plugin::AutoLearnThreshold, pointed to by
the documentation for Mail::SpamAssassin::Conf. In more detail.

Regards
/Jonas
-- 
Jonas Eckerman, FSDB & Fruktträdet
http://whatever.frukt.org/
http://www.fsdb.org/
http://www.frukt.org/