Secure Net: Leveraging Machine Learning to Curb Cyber Attacks

Cyber security enhanced by machine learning is advancing across social networks, cloud technologies, online banking, mobile devices, and more. Machine learning applications in cyber security are categorized by system (network-based or not), learning phase (supervised or unsupervised), and challenge. This paper provides an overview of machine learning techniques and explores developments in detection methods for cyber security risks. It focuses on using machine learning to detect cache spoofing attacks in Named Data Networks. A machine learning approach is developed using ensemble models arranged in a Bayesian Network Classifier (BNC). Detecting a Content Poisoning Attack (CPA) requires monitoring node behavior and analyzing metric changes across the network.