[Mimedefang] using sa-learn in conjunction with mimedefang
Kelson Vibber
kelson at speed.net
Wed Dec 24 00:26:41 EST 2003
On Tuesday 23 December 2003 5:35 pm, Lucas Albers wrote:
> Don't learn at such a low score. You are setting yourself for a bad bayes
> database.
Just to clarify, you can manually learn spam at any score. That's half of
what the bayes feature is for: catching spam that doesn't trigger existing
rules.
But Lucas' suggested auto-learn thresholds are good ones. You *really* don't
want it learning anything incorrectly!
--
Kelson Vibber
SpeedGate Communications, www.speed.net
More information about the MIMEDefang
mailing list