An Efficient Three-phase Email Spam Filtering Technique

Email spam is one of the major problems of the today’s Internet, bringing financial damage to companies and annoying individual users. Many spam filtering techniques based on supervised machine learning algorithms have been proposed to automatically classify messages as spam or legitimate (ham). Naive Bayes spam filtering is a popular mechanism used to distinguish spam email from ham email. In this paper, we propose an efficient three-phase email spam filtering technique: Naive Bayes, Clonal Selection and Negative Selection. The experimental results applied on 10,000 email messages taken from the TREC 2007 corpus shows that when we apply the Clonal selection and Negative selection algorithms with the naive Bayes spam filtering technique the accuracy rate is increased than applying each technique alone.