[CLUE-Tech] Spam filters.

[Sean LeBlanc]

On 10-01 12:17, David Anselmi wrote:
> This article
> 
> http://catless.ncl.ac.uk/Risks/22.22.html#subj15
> 
> references a spam tool based on statistics rather than keyword filters 
> and block lists.  Might be interesting to those running spam filters.

This does look interesting. I always thought a neural net approach would be
pretty cool: have hooks to the program from MUAs like mutt and such, and
send mail to it via hooks - this IS spam, this one is NOT spam, until you
can get enough training for the neural net. Repeat as necessary for any spam
that slips through, or for any false positives. 

Is the program that is based on the technique in the article (bogofilter)
actually adaptive, though? I'd try it out, except that SpamAssassin is
actually working really well for me right now. I like the ability to report
to Vipul's Razor, too.

-- 
Sean LeBlanc:seanleblanc at americanisp.net  
http://users.americanisp.net/~seanleblanc/
Get MLAC at: http://sourceforge.net/projects/mlac/
He is no lawyer who cannot take two sides. 
(contributed by Chris Johnston)