FLeSMA: a firewall level spam mitigation approach through a genetic classifier model

Abstract

Spamming is a serious issue for the Internet community. As well as incurring financial losses to many organisations, spamming causes mental frustration, and the spam content (much of which includes indecent material) causes embarrassment and presents moral issues. Many techniques and solutions have been presented over the past few years, but most of them cannot effectively deter the spamming issue totally. The major reason for the failure of these solutions is that spammers use the very protocols used for email transmission and forwarding to flood the network. One reason for the failure of new solutions is the high drop rate of legitimate emails by the receiver entity. Various solutions have been tried to deter spamming, but in vain. Although in theoretical tests these techniques provide an effective defence against the email message flood, the practical implementation of 100% spam protection is still far from reality. We describe a technical solution to prevent end users from receiving spam messages. Although our approach is simple and intuitive, it provides significant resistance against spam attack generation. Rather than countering the spamming at the receiver side,our solution counters the spamming attack at the point of origination, making it a more deterrent solution against spam based attacks.