User Witr <[EMAIL PROTECTED]> writes: > The reason I wanted to do in in the middle of the TMDA pipeline is > that the (potentially expensive) classification operation could be > avoided in all but the relatively few cases where confirmation was > indicated (about 10% of the time for me).
The success of the Bayesian technique that Paul Graham describes depends on a rich history of both spam and non-spam messages being assembled. Thus, you'd want it to see all your mail, not just 10% of it. > Of course the filter-then-TMDA versus TMDA-then-filter issue depends > to a certain extent on the relative expense of filter versus TMDA, > and for all I know bayesian filtering might be faster than TMDA... This is a subjective statement since "faster" will vary widely depending on the particular TMDA configuration (particularly # of incoming filter rules, what those rules are composed of, the size and content of your lists, etc.). I personally don't see much value in TMDA-then-filter since the only messages that will make it past TMDA should be legitimate ones. _____________________________________________ tmda-users mailing list ([EMAIL PROTECTED]) http://tmda.net/lists/listinfo/tmda-users
