Enhancing Malicious Activity Detection in IoT-Enabled Network and IoMT Systems Through Metaheuristic Optimization and Machine Learning

The increasing prevalence of malicious activities in IoT-enabled healthcare and Internet of Medical Things (IoMT) systems necessitates robust intrusion detection mechanisms. This article introduces a novel approach combining meta-heuristic optimization and machine learning techniques to analyze network traffic for enhanced detection accuracy. Our proposed method utilizes eleven chaotic maps and the K-Nearest Neighbor (KNN) algorithm to identify malicious activity in IoMT and IoT network systems. Recognizing the significance of feature selection in network traffic intrusion detection, we employ the Chaotic Grey Wolf Optimizer (CGWO) to select the most relevant and impactful features for learning strategically. Our approach demonstrates superior performance through comprehensive experiments compared to well-known meta-heuristic algorithms and prior art methods, as evidenced by various evaluation metrics. This research contributes to advancing intrusion detection systems in healthcare IoMT and IoT, offering a reliable and efficient solution to safeguard against evolving cyber threats.