期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2020
卷号：98
期号：3
页码：429-440
出版社：Journal of Theoretical and Applied
摘要：This study deals with an Improved Completely Automated Turning test to tell Computers and Humans Apart (CAPTCHA)-Based Intrusion Detection System Based on Redirector Model. In advanced technological world today, internet usage is inevitable but there are certain Network security challenges such as intelligent spywares that are capable of breaking and bypassing the CAPTCHA. This study aims at designing CAPTCHA-based IDS so as to identify the intelligent spywares that are capable of breaking CAPTCHAs in Intrusion Prevention System (IPS). Therefore, the study tries to design IDS with intelligent redirector using CAPTCHA- Trap of detecting Intrusion by intelligent spywares. The work is set to design an email website which was integrated with the new CAPTCHA based IDS system and host it online. A dummy honeypot system was designed in which the IP addresses of the software intruders that intruded into the system were captured. The work also implemented and evaluated the performance metrics which includes Detection Rate (DR), Precision (PR), False Positive Rate (FPR), correctly and incorrectly classified instances of the system against the Existing system. The researchers use captured IP addresses as the datasets for the study. The work employed Waikato Environment for Knowledge Analysis (WEKA) and Python to analyze the experimental data. The findings of the study indicate a high percentage of Detection Rate and a very low False Positive rate as compared to the existing system. It is recommended that website users and webmasters should incorporate the use of the Improved CAPTCHA-based IDS, IPS and Honeypot so as to block intruders and collect information about them for necessary measures.
