Nichols, William <[EMAIL PROTECTED]> wrote: > I attended a seminar yesterday about spam. One of the idea that was > pitched as being in a product was training the inbound spam detecting > on the patterns of outbound mail. > > They talked about their filters looking at outbound mail, building > patterns off it, then applying those patterns to inbound mail to help > with false positives / negatives.
I suppose this would work - IF you assume that all of your outbound email is non-spam. If you have someone sending spam out, you'd hose your Bayes db pretty quickly. -- + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + Chris Barnes AOL IM: CNBarnes [EMAIL PROTECTED] Yahoo IM: chrisnbarnes Computer Systems Manager ph: 979-845-7801 Department of Physics fax: 979-845-2590 Texas A&M University ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
