Byteplant Forum

Home » CleanMail Support » CleanMail SpamAssassin Filter » Bayesian filtering - false positives
Bayesian filtering - false positives [message #622] Thu, 09 September 2004 18:22
JamesGurney
Messages: 1
Registered: September 2004
Junior Member
I have had a few emails which have been falsely deemed spam due to Bayesian filtering. Can anyone advise on what is happening here please. Clearly I would like to have these not identified as spam inthe future. Can the rules be modified or the system somehow taught? I know about the whitelist and have used that successfully for other email sources. I was wondering if there was a way of getting the Bayesian analysis more correct in the first place.

Thanks

Best regards

James
Re: Bayesian filtering - false positives [message #623 is a reply to message #622] Fri, 10 September 2004 12:01 Go to previous message
support
Messages: 919
Registered: April 2004
Senior Member
> I have had a few emails which have been falsely deemed spam
> due to Bayesian filtering. Can anyone advise on what is
> happening here please. Clearly I would like to have these not
> identified as spam inthe future. Can the rules be modified or
> the system somehow taught? I know about the whitelist and have
> used that successfully for other email sources. I was wondering
> if there was a way of getting the Bayesian analysis more
> correct in the first place.

The Bayes filter can be teached using the sa-learn utility. Please see
http://www.byteplant.com/support/nospamtoday/howtolearnspam.html
and
http://www.byteplant.com/support/nospamtoday/howtolearnexchange.html
for detailed information and remember to specify --ham instead of --spam if you
want to teach false positives as ham.



Customer Support
Byteplant GmbH
Previous Topic: Forced delete of "Bad" SPAMs
Next Topic: How to LEARN.
Goto Forum:
  


Current Time: Sun Dec 04 23:22:34 CET 2016