ELECTRONIC MAIL SPAM ZOMBIES PURIFY IN EMAIL CONNECTION
Sathish Raja, Srikanth Venkatesan · International Journal of Advanced Research in Computer Science · 2013
Email Spammer are constantly pioneering the techniques to bypass anti-spam filters forcing many organizations to invest in spam email prevention mechanisms and solution. Traditional email spam filters aim at analyzing the email content to characterize the best features that are commonly included in email spams. However, this is observed that a crafty trick which is designed to avoid content-based filters will be endless owing to social and economic benefits of sending email spam. In view of this particular situation, there has been many research efforts towards doing email spam detection based on the reputation of the senders rather than what is contained in the email. Motivated by the fact that email spammer are prone to have unusual patterns/behavior and specific patterns of email communication, exploring the email social networks to detect email spams which has received much attention. Existing email spam detecting system aims at analyzing the IP address of e-mail to categorize the features that are commonly found e-mail spam. To resolve this problem an effective research has been made to create a solution that detects how the system generates that spam e-mail. In this paper a method is proposed that identifies the problem of spam e-mail. We present a procedure to generate the e-mail abstraction using HTML content in e-mail, and our newly devise abstraction can more effectively capture the near duplicate phenomenon of spam. This can be identified by tracking the performance of machines sending e-mail, trace the e-mail contents. Moreover, we design a complete spam detection system (A Multi-level Collaborative Spam Detection System), which possesse an efficient near-duplicates matching scheme and a progressive update scheme.