A Novel Approach for Intrusion Detection System in IoT Using Correlation-Based Hybrid Feature Selection and Harris Hawk Optimization Algorithm

With the rapid growth of the IoT, the number of devices connected to various networks has significantly increased. These devices generate vast amounts of data and are often deployed in open and unsecured environments, making them vulnerable to cyber-attacks. Therefore, ensuring the security of IoT networks has become a primary concern for researchers. One of the most effective methods for maintaining network security is using IDS. Intrusion detection monitors and analyzes incoming data to detect suspicious activities and potential attacks. Given the resource constraints of IoT devices and the complexity of the networks, improving the accuracy and efficiency of IDS is crucial. The primary goal of this research is to present a novel and optimized IDS for IoT networks. A hybrid feature selection method has been employed to enhance accuracy and reduce computational complexity, combining correlation-based filtering and wrapper methods using the (HHO) algorithm. In this approach, unnecessary features are removed, and essential features for classification are selected. Simulation results indicate that this method has achieved a 96.46% accuracy, outperforming traditional methods such as DT and SVM while improving false positive and false negative rates.