You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by Daniel Quinlan <qu...@pathname.com> on 2004/08/19 11:43:32 UTC

Re: Why HAM learn on BAYES_99?

Larry Rosenman <le...@lerctr.org> writes:

> Why did the below message get learned as HAM on a bayes_99?

http://wiki.apache.org/spamassassin/AutolearningNotWorking

-- 
Daniel Quinlan
http://www.pathname.com/~quinlan/

Re: Why HAM learn on BAYES_99?

Posted by Daniel Quinlan <qu...@pathname.com>.
Matt Kettler <mk...@comcast.net> writes:

> Dan.. AFAIK that's an outright bug in spamassassin. One of the rules of 
> autolearning introduced in the 2.61 series is to never autolearn anything 
> that strongly contradicts existing learning (ie: never learn a message as 
> ham that would be bayes_99)...

I'm not sure we really fixed it to begin with.

> Did that intentional feature somehow get reverted out of 3.x?
>
> http://bugzilla.spamassassin.org/show_bug.cgi?id=2437

*shrug*

Daniel

-- 
Daniel Quinlan
http://www.pathname.com/~quinlan/

RE: Why HAM learn on BAYES_99?

Posted by Larry Rosenman <le...@lerctr.org>.
Matt Kettler wrote:
> At 02:43 AM 8/19/2004 -0700, Daniel Quinlan wrote:
>>> Why did the below message get learned as HAM on a bayes_99?
>> 
>> http://wiki.apache.org/spamassassin/AutolearningNotWorking
> 
> Dan.. AFAIK that's an outright bug in spamassassin. One of the rules
> of autolearning introduced in the 2.61 series is to never autolearn
> anything that strongly contradicts existing learning (ie: never learn
> a message as ham that would be bayes_99)...   
> 
> Did that intentional feature somehow get reverted out of 3.x?
> 
> http://bugzilla.spamassassin.org/show_bug.cgi?id=2437

That was my bug back then too. 

I really think that bayes_99 needs to be higher score, and that the 
Fix to not auto-learn as ham when bayes_99 is present needs to be put back
in. 

This one slipped through with JUST the bayes hit. :(

LER


-- 
Larry Rosenman                     http://www.lerctr.org/~ler
Phone: +1 972-414-9812                 E-Mail: ler@lerctr.org
US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749



Re: Why HAM learn on BAYES_99?

Posted by Matt Kettler <mk...@comcast.net>.
At 02:43 AM 8/19/2004 -0700, Daniel Quinlan wrote:
> > Why did the below message get learned as HAM on a bayes_99?
>
>http://wiki.apache.org/spamassassin/AutolearningNotWorking

Dan.. AFAIK that's an outright bug in spamassassin. One of the rules of 
autolearning introduced in the 2.61 series is to never autolearn anything 
that strongly contradicts existing learning (ie: never learn a message as 
ham that would be bayes_99)...

Did that intentional feature somehow get reverted out of 3.x?

http://bugzilla.spamassassin.org/show_bug.cgi?id=2437