Re: [Evolution] evolution junk filtering

[guenther <guenther rudersport de>]

> hmm...I thought it would work, but it still does not...I have been
> marking mail junk for a while, and I know it takes a while to learn, but
> it does not get sorted into the junk folder even if the exact same
> message is received again

OK, some more insight is necessary:

Marking mails as Junk (SPAM) or not Junk (HAM) calls SpamAssassin and
lets it learn this message. Anyway, learning doesn't necessarily mean,
that SA knows "this mail and everything very similar is SPAM".

Learning means, SA learns from the words in the mail and adjusts the
level of SPAMmyness of these words, using a Bayesian database. Which
effectively means special words can be a sign of either SPAM or HAM.

The SA Bayes classifier won't work out of the box. To kick in, it needs
to be trained properly. IIRC Bayes should kick in at about 200 SPAMs and
200 HAMs learned. (If it would work before being trained, this would
lead to a lot of false positives and false negatives.)

In fact, SA does auto-learn mails -- which means, even without manually
marking mails as HAM or SPAM, SA trains the Bayes database from that
mail. However, this would only happen for mails which do score high
either as SPAM or HAM. This isn't the case for most SPAM mails.

...guenther


-- 
char *t="\10pse\0r\0dtu\0  ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4";
main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1:
(c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}