The Research In Anti-spam With Bayesian Data-mining Algorithm

Today, many means can be applied to resolve the problem of spam. Content-based spam filtering is one of the mainstream technologies used so far. It includes two-asides: rule-based approach and bayes-based approach.The article introduced NBC (Navie Bayesian Classifier), and based on this algorithm a new algorithm ---Bayes Parameter Estimation is proposed in this paper. The E-mail filtering experimental results between the NBC and Bayes Parameter Estimation show that the algorithm of Bayes Parameter Estimation have fairly satisfactory performance, the newer algorithm has much higher rate to converge than the naive Bayes one.