A Content-Based Approach for Tracking Concept Drift in Email Spam Filtering

The continued growth of Email usage, which is naturally followed by an increase in unsolicited emails so called spams, motivates research in spam filtering area. In the context of spam filtering systems, addressing the evolving nature of spams, which leads to obsolete the related models, has been always a challenge. In this paper an adaptive spam filtering system based on language model is proposed which can detect concept drift based on computing the deviation in email contents distribution. The proposed method can be used along with any existing classifier; particularly in this paper we use Naïve Bayes method as classifier. The proposed method has been evaluated with Enron data set. The results indicate the efficiency of the method in detecting concept drift and its superiority over Naïve Bayes classifier in terms of accuracy.