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

Re: A question about : [Fwd: JULY 6th Lead Training 3 tips for working leads]



On Thu, Jul 08, 2004 at 06:39:50AM +0200, Mezig wrote:
> Phillip Hofmeister wrote:
> >You should start by updating any Bayesian filters you have on your
> >machine and then deleting the message.  After you have done this you

I always keep my spam archived in a separate mailbox; it's good for
training new machines.

> By the way have you a good link about bayesian filters.., my 
> spamassassin is very cheap as is my english :( !
> i can read a little post, not all a documentation!

I don't know any french documentation. To train your spamassassin you
can sort all your spam to a single mailbox-file and the do a "sa-learn
--spam --mbox /path/to/file" which makes sa learn that all mails in
that file are to be recognized as spam.
For good trainig you may need lots of spam and lots of ham (ask your
friends for samples; "sa-learn --ham ..." makes sa learn 'nice' mails

configuration is up to you.

> To end, i thought, someone could made something special against such a 
> post. Sorry i mismake :(!

AFAIK there is already a working spam-filter installed for the ML.
That's why there are not really many spam-mails here. There's always a
chance spam slips through.


HTH
Horst


-- 
#debian.de
< stoffel_> was wurde aus sex & drugs & rock'n roll?
< Lam_al_Adie> stoffel_: dieter bohlen, Harald juhnke und peter kraus?



Reply to: