Hallo! Du (Blars Blarson) hast geschrieben:
>In article <[EMAIL PROTECTED]> [EMAIL PROTECTED] writes:
>>on 2: we would need a whole spam-mail including all headers so we can
>> find charakteristika to filter on, so i now take all nominated
>> postings, and try to find patterns with some black
>> shell-magic.
>
>Is "sa-learn --spam" considered "black shell magic"? Of course you
>need to "sa-learn --ham" on some non-spam too. That's what I do with
>nominated messages to the BTS, and messages with a certain range of
>spamassassin scores.
no, it's not that easy, and my 'black-shell-magic' refers to getting
the whole nominated postings from the listarchive.
Also we (the listmasters) don't train the bayes-filter, we add also
rules to procmail and SpamAssassin if nessecary.
Cord
--
http://lists.debian.org
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
