Re: [Evolution] spamassassin learning

I´m working with Evolution 1.2.3. I have an Filter with spamassassin.
But how can spamassassin learn from incomming mails that are spam?

Spamassassin filters 90% of spam. but i want more! (99,9%...)

SpamAssassin 2.60 with Bayes database trained, network tests enabled and
additional blacklists kicks ass. 99.9% are very hard to reach, cause
that means 1 of 1000 mails is identified wrong. However, I got it to
99.8%, eventually raising... ;-)

First of all, you should collect SPAM mails in a special folder. And you
really should collect HAM (aka non-SPAM) mails as well, to train the
Bayes filter. Bayes won't kick in, unless you have trained it with at
least 200 HAM and 200 SPAM mails.

Evolution by default uses mbox format for the mail folders. Local mail
is stored in your ~/evolution/local/ directory. Check this dir for your
mail files containing HAM and SPAM. The mbox file is named 'mbox'.

Let SA learn your HAM and SPAM mails. You will have to run the command
from a terminal or you can write a script, if you want to call it on a
regular basis. Be sure the mail folders *only* contain HAM or SPAM, or
you will seriously screw your Bayes database.

# learn HAM
sa-learn --ham --showdots --mbox ~/evolution/local/Inbox/mbox

# learn SPAM
sa-learn --spam --showdots --mbox ~/evolution/local/SPAM/mbox

Adjust the paths as needed and run it on multiple mail folders, if you
have more than one containing HAM or SPAM messages.

Once SA has learned enough HAM and SPAM, the Bayesian classifier will
start working and seriously kicking ass.

For more information, please see -- especially
the FAQ and documentation pages.

(mails in german are welcome)

Well, I can translate this, if you don't understand. But this is an
international mailing list...


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; }}}

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]