Spam detection system: a new approach based on interval type-2 fuzzy sets

Today, most Internet users use email to communicate electronically. They depend on the Internet to deliver their important emails safely and to the right recipients. However, the fast growth of Internet users and their use of email together with the exponential increase of unsolicited users sending spam have made the email system less reliable. An email can falsely be markedly a spam filter on its way to the recipient  or even get buried among junk mailing the recipient’s inbox. There are several intelligent anti-spam filters which use different artificial intelligence methods to detect spam including neural networks and fuzzy logic systems. This paper presents an interval based type-2 fuzzy spam detection system. Our results show that interval type-2 fuzzy set is an effective technique for spam detection and email classification. The proposed system enables the user to have more control over the various categories of spam and allows for filter personalization.