Re[3]: BayesIt doesn't work as it should (was: S*P*A*M)

2005-10-04 Thread Stuart Cuddy
Hello Alexey, Monday, October 3, 2005, 8:34:56 PM, you wrote: No. Believe me, the problem exists. since both filters - my BayesIt and the Bayes Filter use so-called naive bayesian method, they are under the problem of overtraining. AFAIK the K9 creators use some solution to fight against it,

Re[3]: BayesIt doesn't work as it should (was: S*P*A*M)

2005-10-04 Thread Vili
Hello Alexey, No. Believe me, the problem exists. since both filters - my BayesIt and the Bayes Filter use so-called naive bayesian method, they are under the problem of overtraining. AFAIK the K9 creators use some solution to fight against it, and it is only the matter of time when such

Re[3]: BayesIt doesn't work as it should (was: S*P*A*M)

2005-10-03 Thread MikeD (2)
Hello Stuart, Monday, October 3, 2005, 2:34:52 PM, you wrote: SC I get too many false positives as it is I could not lower the SC threshold. No, if you are getting false positives you need to make the threshold number larger not smaller. SC I followed a lead someone else left to check to see