Hybrid AI Learning Approaches for Intrusion Detection: A Review

Intrusion detection is a critical cyber security method that keeps track of the progress of the network's software or hardware. To keep up with the ever-increasing rate and diversity of cyber threats, researchers have turned to Artificial Intelligence (AI) techniques to build intrusion detection systems (IDS) in controlling malicious behaviors and cyber-attacks. Recently, machine learning (ML) and deep learning (DL), branches of AI techniques, have gained momentum in the intrusion detection domain to deal with privacy concerns and security threats. Consequently, hybrid AI learning blends different learning techniques, such as traditional machine learning and modern AI methodologies, to tackle advanced and complex problems more efficiently. This article reviews on how these hybrid AI learning techniques are employed and achieved in intrusion detection processes, by exploring various IDS schemes, with their model, dataset, simulator, and evaluated metrics from the literature and scrutinizing their contributions and gaps. Finally, it discusses the issues revealed in the literature, and suggests future research directions for intrusion detection systems using hybrid AI learning approach.