An investigation of information granulation techniques in cybersecurity

Abstract

Information Granulation in the context of Granular Computing provides a viable alternative for finding solutions to complex problems using granules. Issues such as intrusions, malware exigencies, spam and user unauthorized access still remain challenging in cybersecurity. Moreover, as the prevalence of undetected attacks due to system design flaws and system development flaws become rampant in the cybersecurity systems. Although, numerous techniques that have been applied have shown very good prospects, there are several difficulties in managing cyber-attacks from the angle of biometric recognition systems which are commonly used in cybersecurity. These challenges has positioned cybersecurity issues to be regarded as complex and uncertain which requires techniques such as information granulation to unravel a sustainable solution. This paper investigates how information granulation techniques are used in cybersecurity detection models with the aim of providing a holistic view of the current status of research in this area. In this paper, we proposed a framework that applied the principle of justifiable granularity (PJG) in the feature extraction module of a finger-vein recognition system using granular support vector machines as classifier to justify the effectiveness of information granulation in strengthening a verification system in a cybersecurity setting. We benchmark our result with state-of-the-art biometric verification systems, and our approach shows promising contribution in that direction.