should I change score for Bayes_90?

[Chris Campbell]

I have a similar issue... and I was wondering if anyone responded on this
thread... I would like to know if others are changing the bayes* scores...
I know I have wrote about 100 custom rules, and changed about 50
points.....but i haven't touched the bayes yet... mostly because I haven't
found
a good way to use sa-learn on my gateway mailrouter boxes yet... (internal
users are  lotus notes users, and it doesn't support forward as attachment
:/)


So, is anyone else doing this sort of stuff?

/chris





                    Chris Yuzik
                    <chris at fractalweb.       To:     MAILSCANNER at JISCMAIL.AC.UK
                    com>                     cc:
                    Sent by:                 Subject:     should I change score for Bayes_90?
                    MailScanner
                    mailing list
                    <MAILSCANNER at JISCM
                    AIL.AC.UK>


                    10/17/2003 12:25
                    PM
                    Please respond to
                    chris






I'm getting quite a number of spam messages that just sneak under the
threshold of 5.0. While I'm tempted to just decrease the threshold, I don't
want to increase the number of false-positives either.

I notice that in each of these messages, Bayes_90 is triggered, but that
only
adds 3.0 to the score. If Bayes is 90% certain that this message is spam,
should I "trust" it a little more by say, giving it a 4.0 weight instead of
3.0?

Chris