I'm using spamassassin with bayes filtering, nightly I use sa-learn to go though the folders I have, learning spam and learning ham. This has gone on for about 6 weeks now with me moving the uncaught spam messages that are in my inbox into the spam folder. I know it has well over 200 messages as that was surpassed after the first week of using sa-learn. Tonight I did a count (this count is much in line with other counts I have done over the past two week). Out of 355 messages in the spam folder, 175 were labeled as spam. My question is, why is it only catching 49% of spam messages? I have the required # set to 4.0 Are there things I can put in place, other rules that are preformed or something to catch more spam? Autolearn always seems to be off when I look at the headers of spam messages, caught and uncaught.
Thanks, Brian
