http://bugzilla.spamassassin.org/show_bug.cgi?id=3847





------- Additional Comments From [EMAIL PROTECTED]  2004-09-30 19:25 -------
Responding to the last paragraph of Loren's comment:

The only reason for a "reliability" specifier is to give the perceptron a hint
when the actual data is in the noise level -- E.g., if there are no ham hits in
the corpus. If rules such as RFCI get a high score because they hit spam that
other rules do not and that offsets a high FP rate, that is a differnet issue.
The perceptron has all the information it needs to score it.

A philosophical question: If RFCI was perfect at hitting otherwise missed spam
with no FPs except roaringpenguin.com, and the mail used to score the perceptron
had much less than 1 in 2500 mails from roaringpenguin.com, is it correct to let
the rule get a very high score?




------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

Reply via email to