Research On Email Filtering Mechanism Based On Cloud-Computing Techniques

Since the Internet came out, its main application Email has developed very fast. But spam emails subsequently not only occupied network resource seriously, but threaten network security and brought lots of troubles to our daily life. Today’s techniques all can’t clear the spam completely, and the spam maker always tries a new way to make spam, so it’s necessary to improve modern anti-spam techniques.The traditional email filtering system such as the Naive Bayesian-based email system has a strong classifying capacity, but the prophase mail-training and-learning consume mass system and network resources and affect system efficiency. At the later classify decision period, the Bayesian anti-spam filtering seldom thinks about whether the threshold is suitable or not, so the Recalling or the Precision is reduced.Cloud computing is a new research point. It’s mainly for concurrent processing of mass dataset. Compared to traditional compute techniques, cloud computing can improve the executive efficiency largely. The main contribution of this article is it proposed a cloud computing-based email filtering mechanism. Specifically, the main contribution of this article is as follows:First, it imports cloud computing and map/reduce techniques, so its concurrent property is improved, especially for dealing with a large number of emails;Second, the filtering algorithm has improved, so it has a better performance on efficiency;Third, two improved related decision algorithms are proposed, they have a better performance on mail filter accuracy both in the abstract and the experiment.The experiment data indicates that the cloud computing-based email filtering mechanism this article proposed has better expression on recalling, precision, and execute efficiency.