on Tue, Oct 07, 2003 at 04:27:31PM -0500, Alan Shutko (ats@acm.org) wrote:
> Kjetil Kjernsmo <kjetil@kjernsmo.net> writes:
>
> > On Tuesday 07 October 2003 19:48, Alfredo Valles wrote:
>
> >> I was near suicide when some good guy in this list recommended me
> >> spamassissin. It's so easy to get to work and once that you train the
> >> bayesian filter bye bye to all the stupids swen mails.
> >
> > Yeah, that's one option. I considered it, but the problem is, if you
> > feed the learner with tons of similar viruses, how good will it be to
> > kill spam...?
>
> In my experience, still really, really good. I haven't been getting
> any additional false possitives, and I haven't been missing any more
> spam than normal.
>
> This is with bogofilter.
Dittos.
I've created a "spam-learn" folder, over which "sa-learn --spam --dir"
is run via cronjob every 30 minutes. False negatives get tossed there.
Filtering is now *very* good.
Peace.
--
Karsten M. Self <kmself@ix.netcom.com> http://kmself.home.netcom.com/
What Part of "Gestalt" don't you understand?
Bush/Cheney '04: Over a billion Whoppers served.
Attachment:
signature.asc
Description: Digital signature