AN ENHANCED INTRUSION DETECTION SYSTEM USING HONEYPOT AND CAPTCHA TECHNIQUES

Abstract

Internet is no doubt inevitable as it has a tremendous impact in our lives. Despite its importance, internet comes with many challenges, among which is security. From the literature, several attempts have been made to develop secure and user-friendly spam detection technique. But these attempts linger between these two fundamental issues: the robustness and the usability in CAPTCHA system, passiveness of Intrusion Detection System (IDS), which failed to detect some forms of novel attacks, flexibility to attacks and not efficient to users. In this work, honey CAPTCHA, an enhanced intrusion detection framework is designed to solve the above problems as it is capable of detecting crawlers’ attacks, resilient and efficient to users. The system is mainly considered as an option to a CAPTCHA-BASED IDS model, which suffers the above problems. The system outperforms the existing system considering its performance measure based on the proposed metrics that includes detection rate (DR) of 76%, 1.7 times the detection rate of the existing system with false positive rate (FPR) of 10% against the existing system that have 36% FPR, which proved that the system is more robust compared to the existing system. The usability of the system measured using BDR and BNR is 1.5 times that of the existing system, which shows how efficient the system is to users when compared to the existing system. Both systems were compared based on standard IDS evaluation metrics CID which proves that the system is 2.26 times better than the existing system