> >to the end of my user_prefs, and I still get lots of high Bayes scores. Is > >that normal? I more than a little surprised at some of the messages it > >flags as spam. > > Have you used check_bayes to take a look at what tokens are in your bayes db? > > you shouldn't get high bayes scores unless your training had a lot of > mis-placed email, or you have a lot of bayes poision.. > > What's your spam/ham training ratio like?
Initially it was pretty close, but going forward there tends to be much more spam than ham, probably around 4:1. I did the initial training with my collection of "caught" spam and with my current month's "sent-mail" plus one or two other folders of saved ham. -- Jack Gostl [EMAIL PROTECTED] ------------------------------------------------------- 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