PERFORMANCE ANALYSIS OF CAPTCHA BASED BLOCKING AND RESPONSE OF AN INTRUSION DETECTION MODEL USING SIGNATURE

Abstract

Intrusion Detection System is the process of intelligently monitoring the events occurring in a computer system or network, analyzing them for signs of violations of a security policy. Its primary aim is to protect the availability, confidentiality and integrity of critical networked information systems. This paper considered and reviewed a CAPTCHA based intrusion detection model. A method of incorporating signature was used along with the CAPTCHA in the intrusion detection model to clear the controversy identified in the existing model. The signature provides a means of identifying intruders that are able to by-pass the system as legitimate users. The model was implemented using a website hosted online. Dataset obtained from the site was analyzed based on Naïve Bayes classification model using confusion matrices. Implementation of the data analysis was carried out using RStudio software package. Analyzed results shows a better Detection Rate (DR), Accuracy (CR) and False Positive Rate (FPR). This shows that the developed system has significant capability of identifying intelligent spywares targeted at breaking CAPTCHA