Matthew Cline <[EMAIL PROTECTED]> writes: > Why not do like Bayes, and promote all of them? Then the GA can > assign lower scores to the rules with more FPs.
I don't trust the GA/perceptron to assign good scores for range rules with very few hits. Also, the additional rules don't really cover much territory unless Bayes. Daniel -- Daniel Quinlan anti-spam (SpamAssassin), Linux, http://www.pathname.com/~quinlan/ and open source consulting
