Dear Sam, > Also, if I want to test some type of technique, what levels of spam > filtering/fp/fn are people getting? What percentage points should I > shoot for?
One reason that improving SpamBayes is hard is that it already does such a good job. I have mine_received_headers turned on and I've added a patch that tokenizes the IP addresses that URLs point to. On my mail, I get a false-negative rate that's under 0.5%. The false-positive rate that I get is much lower than that. Regards, Matt _______________________________________________ spambayes-dev mailing list spambayes-dev@python.org http://mail.python.org/mailman/listinfo/spambayes-dev
