AN INTEGRATED SYSTEM FOR DETECTION AND IDENTIFICATION OF SPAMBOT WITH ACTION SESSION AND LENGTH FREQUENCY

Abstract

We propose an integrated system for detection and identification of spambot with action session and length frequency basedon the notion of control-flow graph, which models interactions of the end-user’s machine and browser with the Web site,and assists a lot in detecting possible anomalies. User’s interaction with the web is premised on Document Object Model(DOM) Events since the DOM forms a representation of the Web page which shows acceptance of asynchronous input fromthe user. The DOM is a platform-independent, event-driven interface which accepts input from the user and allows programsand scripts to access and update the content of the page. Proof of concept will be established by deploying the DOMantiSpambot as an add-on for Mozilla Firefox using JavaScript.

Keywords: Antispam, DOM, detection, frequency, Action Session, Browsers and Web.