SpamAssassin - Bayes_99
I've been messing around with SpamAssassin and MailScanner. Have been running it live on my network for couple months now and it works great. But I've had a few messages that were tagged as spam, and when I looked at it, the rule that was giving it the most score was BAYES_99. I've tried searching the Internet, reading SpamAssassin's site and other sites trying to find the ACTUAL ruleset for BAYES_99 but have had no luck. I'm trying to figure out what exactly in BAYES_99 is causing certain email to be tagged as SPAM. From what I can find out, BAYES_99 is part of the Bayesian system that is supposed to auto-learn SPAM from SpamAssassin. But I don't think I have the auto-learn capability turned on...so I'm very curious as to where its getting its list from and where it actually resides. Can someone please point me to the right direction? Thank you.
Basically, you need to return to the SpamAssassin website and search through their documentation and help files. The prodecure for teaching Bayes what is and is not spam involves creating two folders: ham and spam. Put examples of acceptable mail (ham) in one folder, and examples of unacceptable mail (spam) in the other for Bayes and sa-learn to use as learning tools. As ham gets directed incorrectly, and them to the ham folder. As spam gets directed incorrectly, add them to the spam folder. Gradually, over time, SpamAssassin becomes more precise in sorting ham from spam.
Search for the keywords whitelist and blacklist for more info.