[Spambayes] filter settings

[Tony Meyer]

> I've just updated my spambayes.  I enjoyed the previous
> version but have a question.  When I change the setting
> for filtering, is the higher number narrow the "acceptability"
> of incoming e-mails?

Basically, it works like this:

SpamBayes gives each message a score.  All good email should score really
close to 0, and all bad email should score really close to 100.  Messages
that SpamBayes is unsure about should score close to 50.  There's an example
graph showing how messages are typically distributed here:

<http://spambayes.org/images/chi2_graph.png>

By default, the plug-in assumes that anything over 90 is bad (spam).  If you
find that you get a lot of spam in your unsure folder that scores over (for
example) 80, and there is never any good mail over 80 (which there shouldn't
be), then you could move the score down to 80.  Similarly, anything under 15
is assumed to be good.  If all your good mail scores (for example) under 10,
you could move the threshold down to 10, and those other messages would end
up in the unsure folder (so a lower number could be said to narrow the
'acceptability' of incoming mail).  You may find that once you've done some
training (maybe now, depending on how long you had the old version) you
would do well with a tighter 'unsure' range - say 20 and 80.

Does this answer your question?

=Tony Meyer

---
Please always include the list (spambayes at python.org) in your replies
(reply-all), and please don't send me personal mail about SpamBayes. This
way, you get everyone's help, and avoid a lack of replies when I'm busy.