An Algorithm of Filtering Junk E-mail based on Cognition-Learning Bayes

A new algorithm of filtering junk-mail based on the Cognition-Learning Bayes is proposed in this paper, which can improve the accuracy of mail filtering conspicuously by using Cognition-Learning Bayes method to assure that the category’s feature vector will always stand for it's real features. The experimental result shows that the new algorithm gains obviously high precision and recall, and is a promising method for filtering junk mail.