Classification methods in the detection of new malicious emails

Abstract

A serious security threat today is malicious emails, especially new, unseen Internet worms and viruses often arriving as email attachments. These new malicious emails are created at the rate of thousands every year. Current anti-virus systems attempt to detect these new malicious email viruses with signatures generated by hand but it is often times costly. In this paper, we present some classification methods that detect new, unseen malicious emails accurately and automatically. The classification method found discrepancy behaviors in data set and use these behaviors to detect new malicious email viruses. Comparison results show the accuracy in the detection of new malicious emails. In order to improve the detection accuracy, the prototype of the bagged classifier is utilized in the implementation of our malicious email detection system.