Indian Journal of Science and Technology

Abstract

Objective: This paper aims to analyze the top ten vulnerabilities present in the Linux kernel and to classify them into different classes based on security metrics. Methods/Statistical Analysis: Comparative evaluation of different tree based, Bayesian and instance based classifiers w.r.t. three security metrics integrity impact, confidentiality impact and availability impact are presented. The performance of these classifiers is evaluated on the basis of accuracy. Findings: All classifiers are showing highest accuracy in terms of integrity. It is very critical security metrics as it is directly proportional to vulnerability score. Application: Accurately classifying integrity will help in making the system more secure and vulnerable proof.
Keywords: Analysis, Classifiers, Linux Kernel Vulnerabilities, Security Metrics, Vulnerabilities, Vulnerability Score