Subject: Re: spam filtering
bryan@ak.net
Date: Fri May 09 2003 - 15:00:30 AKDT
On Fri, May 09, 2003 at 07:11:57AM -0800, David J. Weller-Fahy <lists@weller-fahy.com> wrote:
>
> I've got the same thing running here. The only thing I've noticed is
> that (since the update to .11.1 and .12.2) I'm getting a few more false
> negatives every week. I think I may have to retrain the good/bad
> wordlists with the new version.
I started filtering yesterday, and I'm getting false negatives too.
I only trained for a few days, but I thought that would be enough
with as much spam as I get. Still, the filtering is a big improvement,
and when I get non-filtered spam, I write them to a 'spam' file, and
run 'bogofilter -Ns <spam' to reclassify it.
> > It wasn't that hard to set up, and I'm happy with the situation.
> > I would have expected Bayesian filtering to be complicated, but
> > bogofilter takes care of the details, and is easy to use from a
> > user perspective.
>
> Nope, easy as pie. Need to figure out how to setup 3 state filtering,
> though: SPAM, UNSURE, and NOT SPAM would be better, then I could just
> train on the UNSURE and completely misfiled.
>
> Have you tried anything with the 3 states instead of two yet?
No, I haven't. It would be easy enough to tell bogofilter to do that,
but what should be done with the results? Would I have to have a
third, "unsure" mailbox? Two is enough for me. I'll continue to
reclassify the mistakes, and things should shake out sooner or later.
-- Bryan Medsker bryan@ak.net--------- To unsubscribe, send email to <aklug-request@aklug.org> with 'unsubscribe' in the message body.
This archive was generated by hypermail 2a23 : Fri May 09 2003 - 15:00:55 AKDT