Michael 'Moose' Dinn wrote: > > Has anyone taken a huge spam database and sent it through some sort of > genetic learning program to see if spam can be identified that way? > > More of a curiosity thing than anything else.
We use quite a bit of machine learning stuff plugged into SpamAssassin's core engine here. Some bayesian probability, some neural nets, and some decision trees. The coolest thing about it is it tracks stuff we don't have rules for yet (though it's not the type of thing where we could extract a rule, sadly), the worst thing about it is that it becomes quite specific to the training set, so the training set has to be quite diverse. It's pretty accurate though, and I'm happy enough with the results that I'm keeping it. Though I'm not about to replace the other rules. Matt. ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Oh, it's good to be a geek. http://thinkgeek.com/sf _______________________________________________ Spamassassin-talk mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
