i thought bayes knew if a message was whitelisted or blacklisted and used that knowledge to prevent impartial bayes learning? Am i wrong in thinking this was ever the case?
If bayes doesnt use whitelisting/blacklisting to determine auto_learn, then every whitelisted mail gets learned as ham and every blacklisted gets learned as spam, which is a very bad thing.
*COUGH* *clears throat*
I'll repeat myself...Please read carefully this time.
No, bayes does not use the score contributions of whitelisting in
determining wether or not to auto-learn, but it can still autolearn if the
non-whitelisted score is over/under a threshold.
In other words, bayes autolearning behaves as if whitelisting and blacklisting does not exist. The score contributions of white/black listing are removed entirely calculating the score for the autolearner.
Thus, whitelisted messages DO NOT get ANY special treatment. They will NOT all be learned as ham, because the -100 score bias is IGNORED by the bayes autolearner. ie: a whitelisted message with a score of -98.0 is judged by the autolearner as if it had a score of +2.0, because the -100 for the whitelist is removed from the calculations entirely.
------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
