Tom Meunier <[EMAIL PROTECTED]> wrote: > If it's already 100% sure that it's spam, how is it helpful to train > it that it's spam? It's not like it's going to be 110% sure that it's > spam. It's already trained! > > Not trying to be a wise-ass, I've just seen this question come up > fairly often, and can't wrap my head around it.
There are 2 issues here: 1) *THIS* message scored 100%, but another message might not and teaching SA from this message could help that other message's score. 2) But real goal is that I have a line in my .procmailrc that looks for "autolearn=yes" and if it finds it, sends the message to /dev/null. I want to increase the number of spam messages I never have to deal with. Now your response is going to be "just use another line the .procmailrc script to look for "BAYES_99*[score: 1.0000]" and send that to /dev/null as well, right? Well, yeah, I could do that. But it: a) seems harder b) misses (1) above c) allows a definitive Bayes scores to be used in SAlearn -- + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Chris Barnes AOL IM: CNBarnes [EMAIL PROTECTED] Yahoo IM: chrisnbarnes Computer Systems Manager ph: 979-845-7801 Department of Physics fax: 979-845-2590 Texas A&M University ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
