A Time-Robust Spam Classifier Based on Back-Propagation Neural Networks and Behavior-Based Features

Earlier works on detecting spam emails usually compare the contents of emails against specific keywords, which are not robust as the spammers frequently change the terms used in emails. In this paper, an back-propagation neural network is designed and implemented, which builds classification model by considering the behavior-based features revealed from emails´ headers and syslogs. Since spamming behaviors are infrequently changed, compared with the change frequency of keywords used in spams, behavior-based features are more robust with respect to the change of time; so that the behavior-based filtering mechanism outperform keyword-based filtering. The experimental results indicate that our methods are more useful in distinguishing spam emails than that of keyword-based comparison.