Intrusion detection using machine learning-hardened domain generation algorithms

Abstract

Machine learning has recently been applied in a variety of areas in information technology due to its superiority over the typical computer algorithms. The machine learning approaches are being integrated into cybersecurity detection approaches with the primary aim of supporting or providing an alternative to the first line of defense in networks. Although the automation of these detection and analysis systems is potent in today’s changing technological environment, the usefulness of machine learning in cybersecurity requires evaluation. In this research, we present an analysis and address cybersecurity concerns of machine learning techniques used in the detection of intrusion, spam, and malware. The analysis will entail the evaluation of the current maturity of the machine learning solutions when identifying their primary limitations, which has prevented the immediate adoption of machine learning in cybersecurity.