• v1.38b bayesian filtering

    By Keith Strickland 2 decades ago

    I'm kind of new to bayesian filtering so this may be a stupid question. But, it's my understanding that by using the bayesian filter and populating mailspam.nsf with spam mail that I should see kSpam start blocking spam based upon things other than the rules I have defined in kspamcon.nsf. If this is correct, and I may be totally wrong, when should I expect to see a difference in the amount of spam that kSpam is catching? Now, I do see kSpam catching spam that is defined by deny rules in kspamcon.nsf and also allowing mail defined by allow rules. Maybe my understanding of a bayesian filter is wrong, in which case I'll go read more :)



    Thanks,

    Keith

    • you need...

      By Tom Lyne 2 decades ago

      to have spam mail in mailspam.nsf and good mail, "ham" in mailgood.nsf . 200 in each is enough to start, then enable the Bayesian filter and see what happens,



      -tom

      • I've got

        By Keith Strickland 2 decades ago

        754 spam mails in mailspam.nsf but only about 90 in mailgood.nsf. I'll add more to mailgood.nsf and see what happens. Thanks for the quick response.



        Keith