Internetworkers: http://www.ibiblio.org/internetworkers/

[Joey O'Doherty <joey AT odoherty.net>]

Woah, check this out:  Spamassassin now does bayesian filtering!

From: http://spamassassin.org/dist/README

Since SpamAssassin now includes a Bayesian learning filter (in version
2.50 on), it is worthwhile training SpamAssassin with your collection of
non-spam and spam, if possible.  This will make it more accurate for your
incoming mail.  Do this using the "sa-learn-spam" and "sa-learn-nonspam"
commands, like so:

        sa-learn-spam ~/Mail/saved-spam-folder
        sa-learn-nonspam ~/Mail/inbox
        sa-learn-nonspam ~/Mail/other-nonspam-folder

Use as many mailboxes as you like.  Note that SpamAssassin will remember
what mails it's learnt from, so you can re-run this as often as you like.


Thomas C. Meggs wrote:
> If any of you guys happen to have your email going through a place where 
> you can use something like procmail, check out a program called 
> spamassassin. [http://spamassassin.org/]
>
> Basically it's a perl script that is used to filter your incoming 
> messages. It applies a set of rules which apply different points to a 
> message, and then if a message reaches a certain threshhold of points, 
> it is dubbed as spam with a special header. Then with procmail you can 
> have it relocate said spam messages into another folder, or to /dev/null.
>
> I've been using it for the last 6 months or so with a very close to 100% 
> success rate. The only messages that it has trapped have been highly 
> suspect to begin with. There have, however, been 2 occasions where a 
> piece of spam for a specific site/product came through, but I just added 
> the domains to the configurable blackhole list SpamAssassin provides.
>
> Here is an example message which has been moved into my spam folder, it 
> was the only message in the entire folder which had the word 'Banana' in 
> it, so I thought it was appropriate...
>
> --- Begin Message ---
> Return-Path: <perf-errors.6409.138103.4703545.501.0.4 AT b.list-email.net>
> Received: from out017.list-email.net (out017.list-email.net 
> [66.180.247.67])
>     by hoser.sphear.org (8.12.3/8.12.3/Debian -4) with SMTP id 
> gBR7FgRp001324
>     for <tom AT plik.net>; Fri, 27 Dec 2002 02:15:44 -0500
> To: tom AT plik.net
> From: BananaFritters-R4L <newsletter AT tfmb.net>
> Subject: *****SPAM***** THOMAS, thank you!
> Date: Thu, 26 Dec 2002 23:06:46 -0800
> Message-ID: <1271446623-1463792382-1040972806 AT topica.com>
> Errors-To: <perf-errors.6409.138103.4703545.501.0.4 AT b.list-email.net>
> Reply-To: <perf-remove.6409.138103.4703545.0.0.4 AT b.list-email.net>
> Mime-Version: 1.0
> Content-Type: text/plain
> X-Spam-Status: Yes, hits=14.0 required=5.0 
> tests=PLING,REMOVAL_INSTRUCTIONS,EXCUSE_14,NO_OBLIGATION,EXCUSE_3,CLICK_BELOW,CLICK_HERE_LINK,DIFFERENT_REPLY_TO 
> version=2.20
> X-Spam-Flag: YES
> X-Spam-Level: **************
> X-Spam-Checker-Version: SpamAssassin 2.20 (devel $Id: SpamAssassin.pm,v 
> 1.77 2002/04/06 19:28:30 hughescr Exp $)
> X-Spam-Prev-Content-Type: multipart/alternative;
>     boundary="1561637555.1463793406.1040972411"
>
> SPAM: -------------------- Start SpamAssassin results 
> ----------------------
> SPAM: This mail is probably spam.  The original message has been altered
> SPAM: so you can recognise or block similar unwanted mail in future.
> SPAM: See http://spamassassin.org/tag/ for more details.
> SPAM:
> SPAM: Content analysis details:   (14 hits, 5 required)
> SPAM: Hit! (0.5 points)  Subject has an exclamation mark
> SPAM: Hit! (3.8 points)  BODY: Gives instructions for removal from list
> SPAM: Hit! (1.1 points)  BODY: Tells you how to stop further SPAM
> SPAM: Hit! (2.6 points)  BODY: There is no obligation.
> SPAM: Hit! (2.7 points)  BODY: Claims you can be removed from the list
> SPAM: Hit! (1.5 points)  BODY: Asks you to click below
> SPAM: Hit! (1.8 points)  BODY: Tells you to click on a URL
> SPAM: Hit! (0.0 points)  Reply-To: massively different from From: or To:
> SPAM:
> SPAM: -------------------- End of SpamAssassin results 
> ---------------------
>
> --1561637555.1463793406.1040972411
> Content-Type: text/plain
>
> Recipe4Living delivers the tastiest offers on the web. We
> respect your privacy. If you no longer wish to receive
> these yummy offers, see the removal instructions at the
> bottom of the email.
>
> ___________________________________________________________
>
> Recipes4Living    Recipes and advice for a healthly life
> ___________________________________________________________
>
> This recipe for Banana Fritters has so many possibilities I
> don't know where to begin. Serve these warm for dessert
> with the topping of your choice. I can think of several that
> sound great. Chocolate sauce, peanut butter sauce,
> whipped cream, marshmallow sauce. Add a sprinkle of
> chopped nuts and don't forget the cherry on top.
>
> I just thought of another…caramel sauce. Wow!
>
> Click on the link below for Banana Fritters from heaven
> recipe.
>
> http://c.tfmb.net/maaaJ5DaaU2WGaatTLNe/
>
> ** SNIP ** SNIP ** (there is more, but its just a bunch of drivel)
> --- End Message ---
>
> There are a few other programs out there which utilize Bayesian 
> statistical analysis on Spam, which after a lengthy training process 
> have a slightly higher success rate, but SpamAssasin has them far beat 
> with its ease of use and setup.
>
> At any rate, I highly reccomend that any of you with a growing Spam 
> problem check it out! I've been in a Spam Free heaven for enough time 
> that now I rarely hesitate when it comes to giving out my email address 
> to questionable companies.
>
> Regards,
> Tom
>
> Joey Carr wrote:
>
> > Most of all the spam I receive at my work email is WHOIS related.  I
> > registered a domain name and within two months I was getting more spam on
> > that account than any other I've ever had.  Registering a domain is, 
> > IMHO,
> > the worst thing you could do to avoid spam.
> >
> > On Sun, 19 Jan 2003, Sil Greene wrote:
> >
> >
> > > Hmmm... This must be the first WHOIS-related spam I've gotten.  I'm 
> > > sure no one here would pay first and ask questions later, but just in 
> > > case, do verify with WHOIS that the domain is owned before clicking 
> > > the link in the original email.  In this case, my registrar's regular 
> > > prices were *slightly* cheaper.
> > >
> > > --Sil
>
>
> ---
> Come and play at the InterNetWorkers Web site!  
> http://www.ibiblio.org/internetworkers/
> You are currently subscribed to InterNetWorkers mailing list
> To unsubscribe visit 
> http://lists.ibiblio.org/mailman/listinfo/internetworkers


--
pub  1024D/B663781B 2001-11-13 Joey O'Doherty <joey(at)odoherty(dot)net>
Key fingerprint = F76B 9ACA 4197 C707 6E4D  2B78 E430 101A B663 781B