> I'm still fiddling around with these spams that have a bunch of > one-letter > words hiding drugs for sale: > > V k I p A m G i R u A v > V j A v L s I t U w M g > X g A f N a A f X q > C x I e A a L g I c S l > > followed by a url: > > http://www.prouceteir.com > > followed by some presumably benign text:
It took some training for me before my SpamBayes started to recognize those reliably, but it seems that my old hack to tokenize URL's IPs helps: Spambayes spam score 0.998 for message "Re: your VtAGRiA" Spambayes clues: 0.000 *H*, 0.997 *S* [. . .] 0.845 url-ip:222.52.1.11/32 0.845 url-ip:222.52.1/24 0.845 url-ip:222.52/16 [. . .] 0.976 url-ip:222/8 Excellent arguments about why doing that is a bad idea have been made, but it seems to work for me. Regards, Matt _______________________________________________ spambayes-dev mailing list spambayes-dev@python.org http://mail.python.org/mailman/listinfo/spambayes-dev
