Research and implementation of spam filtering algorithm based on single table document
Zhang Yong-hua · Jisuanji gongcheng yu sheji · 2009
The limitation of Naive Bayes algorithm used for filtering mails is analyzed. The conditional Naive Bayes method doesn’t consider the different features between the legitimate mail and the junk in the classification process. For the individual user’s personal needs, a revised E-mail filter using only spammy words on the basis of Naive Bayes algorithm and the risk minimum is researched, and through experiments to prove its feasibility. Finally, a comprehensive model of E-mail filtering based on a list of black-and-white, rule and content is built.