At Tue Sep 23 21:45:13 2003, Renato G. Troitino wrote: > > Just curious... when a message get enought points to be a spam, it should be > set autolearn=yes right? So why all the messages gets autolearn=no even > getting 15 hits???
Learning spam as ham (or ham as spam) is really bad. If you start learning messages incorrectly, you will seriously screw up your bayes database. If you screw it up enough, it'll start mis-classifying messages. Auto-learning is dangerous if it's not done very carefully. For that reason, only the most spammy-looking messages get learned as spam, and only the most hammy-looking messages get learned as ham. There's a wide range where SpamAssassin plays it safe and doesn't learn, rather than possibly making a mistake. Martin -- Martin Radford | "Only wimps use tape backup: _real_ [EMAIL PROTECTED] | men just upload their important stuff -o) Registered Linux user #9257 | on ftp and let the rest of the world /\\ - see http://counter.li.org | mirror it ;)" - Linus Torvalds _\_V ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
