Network Security in Cyberspace Using Machine Learning Techniques

Cyberspace has grown due to the widespread uses of the Internets and mobiles apps. Attacks on the Internet are more likely to be automated and to last longer. Cyber security techniques make it possible to improve security measures against cyber-attacks. Because modern fraudsters are so sophisticated, traditional security measures are no longer effective. Conventional security solutions must be revised when identifying new and evolving security threats. Many cyber security applications would only be possible using machine learning (ML) methods. Despite this continued improvement, major obstacles remain to be overcome before ML systems can be trusted. Cyberspace is rife with financially motivated, hostile actors eager to play games and exploit such ML flaws. The objective is to analyze ML methods for cyber security focusing on areas such as intrusion detections, spam detections, and malware detections on computer network and mobile network, to provide comprehensive overviews of the complexities ML methods faces in protecting cyberspaces from attack.