Detection of Web Spambot in the Presence of Decoy Actions

摘要

Based on the recent research and statistics by Symantec, significant amount of all global web traffic and email traffic is marked as spam. Spambot is basically a robot that maliciously traverses the World Wide Web (WWW), and gathers information, email addresses, etc. For the spammer. The increasing growth of spam bot sophistication advances in the introduction of Spam 2.0, which infiltrate legitimate Web 2.0 unsolicited. This leads to various unwanted outcomes, such as the appearance of spam pages as the top search engines results due to excessive usage of popular terms, unreal web-pages visit rate, spam emails, and wastes of resources. Here we present an efficient method to detect web spam bot in the presence of decoy actions, by applying efficient approximate string-matching techniques. Our preliminary experimental results show that the proposed method is successful for the classification of web spam bot in the presence of decoy actions, hence eliminating spam in Web 2.0 applications.