A Real-Time Smart Agent for Network Traffic Profiling and Intrusion Detection Based on Combined Machine Learning Algorithms

Cyber-intrusions are constantly growing due to the ineffectiveness of the traditional cyber security tools and filtering systems-based attacks detection. In the last decade, significant techniques of machine and deep learning were employed to resolve the cyber security issues. Unfortunately, the results are still imprecise with a lot of shortcomings. In this paper, we present a real-time cyber security agent based on honeypots technology for real-time data collection and a combination of machine learning algorithms for data modeling that enhances modeling accuracy.