基于贝叶斯过滤算法的反垃圾邮件策略

结合传统贝叶斯算法的数学定理．给出了朴素贝叶斯过滤算法以及该算法在反垃圾邮件中的六个处理步骤，算法通过渐进学习的方式分别建立三个哈希表，并预置相应的阀值来判定收到的邮件是否为垃圾邮件．以实例方式说明了此算法为基础的反垃圾邮件策略，最后指出了朴素贝叶斯算法存在的问题及可能改进的方向，对防范和处理垃圾邮件提供理论依据和实践参考．

Strategy of Anti-Spam Based on Bayesian Filtering Algorithm

Based on the traditional Bayesian theory, Naive Bayesian filtering algorithm and its six processing steps in anti-spam are presented. The algorithm creates three Hash Tables respectively by gradual study of mails, and presets threshold accordingly to determine whether a mail received is spam or not. Examples are given to prove this algorithm is a basic strategy to anti-spam. Finally, problems and future improvement of Naive Bayesian filtering algorithm are pointed out, which provides theoretical basis and practical reference to keep away and deal with spam.