The Empirical Analysis of Machine Learning Approaches for Enhancing the Cyber security for better Quality

Abstract

In recent years, there have been significant advances in both technologies & tactics so in area of cyber security, with (ML) machine learning at the forefront of the transformation. It is the ability to obtain security event characteristics or findings from cyber security information and then develop a matching information model that will allow a security system to become autonomous and smart. The widespread proliferation and the usage of Web and Smartphone applications has increased the size of cyber world as a consequence. When a computerized assault takes too long to complete, the internet becomes vulnerable. Security measures may be improved by recognizing and reacting to cyber-attacks, thanks to cyber security techniques. Security measures that were previously used aren't any longer appropriate because scammers have learned how to evade them. It is getting more difficult to detect formerly unknown and unpredictable security breaches, which are growing more widespread. Cyber security is becoming more dependent on machine learning (ML) techniques. Machine learning algorithms' dependability remains a major challenge, given its continual advancement. It is possible to find malicious hackers in internet that are ready to exploit ML defects that have been made public. A thorough review of machine learning techniques safeguarding cyberspace against attacks is provided in this paper, which presents a literature review on Cyber security using machine learning methods, such as vulnerability scanning, spam filtering, or threat detection on desktop networks as well as smart phone networks. Among other things, this paper provides brief descriptions of each machine-learning technique and security info, essential machine-learning technology, and evaluation parameters for a classification method.