Hello Kevin, Friday, January 9, 2004, 4:56:54 PM, you wrote:
KR> Once I have run the sa-learn routine on a group of thousands of messages, is KR> it necessary to keep all of those messages to run the next time I run KR> sa-learn? It takes a while for sa-learn to process thousands of message so KR> I was wondering if I only ran sa-learn on the new hand sorted messages, KR> would it cause a problem? No. Once you've sa-learned an email correctly, whether by auto or manual activity, there's no need to sa-learn it again. KR> Naturally the new group would be from 50-300 messages where the whole group KR> might be 5000 as recommended. Once a day I collect all the spam from my spamtraps, and my ham from my hamtraps, and review them to verify 1) they contain only spam or ham as desired, and 2) they do not contain any technical anti-spam discussion (this list and similar personal email are the only things I've found that will poison my bayes database). I then sa-learn that collection of spam and ham, and then concatenate them onto a weekly spam and a weekly ham mbox. Each week I migrate that weekly mbox into my mass-check corpus. Each spam and each ham is learned only once. Bob Menschel ------------------------------------------------------- This SF.net email is sponsored by: Perforce Software. Perforce is the Fast Software Configuration Management System offering advanced branching capabilities and atomic changes on 50+ platforms. Free Eval! http://www.perforce.com/perforce/loadprog.html _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk