On Sat, 16 Oct 2010, Jostein Berntsen wrote:
On 15.10.10,14:32, Matthew G. Saroff wrote:
Just wondering. The implimentation of SpamAssassin on Panix does
not implement sa-learn, so I was wondering what you folks might use.
--
Matthew Saroff
I use spamassassin which works fine. On this site the author suggests
using Bogofilter together with bmf and qsf:
http://www.acme.com/mail_filtering/bayesian_frameset.html
Jostein
Well, with some searching around, it turns out that even though
Panix does not support the Bayesian filtering on SpamAssassin, it does
have Bogofilter installed.
So, I'm training the filter with bogofilter -n < filename and
bogofilter -s <filename right now, and I'm looking to impliment Procmail
recipes.
After a quick Google, it looks like the recipe to get Bogofilter
to tag a message would be
# Invoke bogofilter.
:0 fw
| $BOGOFILTER -p -e
Though some use the -u flag, which would filter and add the results to the
database?
:0fw
| bogofilter -u -e -p
Then check on the exbogosity flag, and filter spam, and unsure reaults
into their respective folder:
# If all filters agree it's maxi-spam, file it.
:0:
* ^X-Bogosity: Spam
$SPAMTRASH
:0:
* ^X-Bogosity: Unsure
unsure-bogofilter
Is this what other people use?
--
Matthew G. Saroff, E.I.T.
Owings Mills, MD
E-mail: msaroff(_at_)pobox(_dot_)com
Yadda, yadda, yadda, I gotta blog.
http://40yrs.blogspot.com
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)de
http://mailman.rwth-aachen.de/mailman/listinfo/procmail