On Thu, 12 Feb 2004, The awesome and feared Collins Richey commented thusly,
> On Fri, 13 Feb 2004 09:41:37 +0600 (LKT)
> Grendel <[EMAIL PROTECTED]> wrote:
>
> > On Thu, 12 Feb 2004, The awesome and feared Glenn Johnson commented
> > thusly,
> >
>
> >
> > So the killfilter which i use is a bayesian filter, ie it has been
> > trained to recognise people in email lists who have a annoying
> > potential,
>
> Unfortunately, Grendel, many of your recent responses (not your
> technical queries) are prime candidates for your own killfilter,
> namely supremely annoying.
Fortunately the owners email addreses are excempt from the killfilter,
otherwise as you quite correctly pointed out I would be in trouble too :-)
On a side note if anyone is interested in trying bayesian mail filtering
it is very easy to setup. I would recommend getting bmf (bayesian mail
filter) from sourceforge.
extract the sources and compile and install it.
Now you have to train it, first of all feed a folder containing a group of
your normal email to it so that it recognises what is your normail email.
bmf -n -i normal_email_folder
Now download some spam from spamarchive.org and feed it to bmf
bmf -s -i 315.r2
Now edit your ~/.procmailrc and have the following in it,
# Invoke bmf as a filter
:0 fw
| bmf -p
# Filter spam
:0:
*^X-Spam-Status: Yes
mail/spam
Now as your mail gets delivered through procmail it will get scanned and
if it is spam it will go to the mail/spam folder.
If you receive a mail which is not spam but recognised as spam, then
export it as say message.txt and invoke
bmf -N -i message.txt
If you receive a spam which does nto get caught, then export it as say
message.txt
then
bmf -S -i message.txt
It gets better with time.
Grendel
--
Hi, I'm a signature virus. plz set me as your signature and help me spread
:)
--
[EMAIL PROTECTED] mailing list
