I just upgraded to v.2.1, which has the BayesIt plugin v.5. Since upgrading, all mailing lists (including TBUDL) and friends' e-mails that have a bunch of cc's are filtered to the Junk folder. None of the adjustments I've made has made a difference.
So far I've marked them as "not junk", and I've raised the average score to 80 (from 50). I also added the addresses of TUBDL and others to the the white list of kludges in the options section. Obviously I'm missing something. 1. What is the difference between an average, maximum, and minimum score? 2. I've enabled "use white list of kludges", but should I put more (or something other) than the mailing address? And what does "optimize list" do? 3. The default setting for the number of "judging tokens" is 15, and the "enable advanced judging" is checked. I hate to sound so ignorant, but what are "judging tokens"? To increase accuracy, should I raise or lower this number? 4. Is one of my settings causing it to not "learn" when I mark TUBDL as not SPAM? -- Thanks, Rick ________________________________________________ Current version is 2.10.01 | 'Using TBUDL' information: http://www.silverstones.com/thebat/TBUDLInfo.html
