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

Re: Spamassassin

On Wed, Oct 08, 2003 at 08:50:04AM +0200, Stephan Seitz wrote:
> Hi!
> On Wed, Oct 08, 2003 at 12:40:33AM +0200, Nicos Gollan wrote:
> >You need to bootstrap the bayesian filter with at least 250 spam and 
> Is there a download possibility to get so much spam mails? Since I 
> delete my spam, I don't have enough mails to train spamassassin.

As mentioned, there are places you can download a corpus of spam, but
it's not recommended because your filters will work better if they're
trained on the spam _you_ receive.

Keep in mind that it's very easy and very safe to train SA as you go. If
you just create a spam folder now and start dumping all your spam into
it, then run sa-learn over it periodically then they'll get added to
your database.

While your database is below 250 spam and 250 spam, SA will just use its
'standard' filters.  As soon as you hit that minimum, the Bayesian is
enabled (assuming it's enabled in your config of course). Then as long
as you keep feeding it ham and spam it'll keep getting more accurate.

>   -ScruLoose-   |     Everyone is entitled to an *informed* opinion.    <
>  Please do not  |                    - Harlan Ellison                   <
> reply off-list. |                                                       <

Attachment: pgpPY6_pSjUzM.pgp
Description: PGP signature

Reply to: