An Improved Binary Spider Wasp Optimization Algorithm for Intrusion Detection for Industrial Internet of Things

Ensuring network security, particularly within the Industrial Internet of Things (IIoT), has become paramount with the escalating reliance on Internet applications across diverse sectors, emphasizing the critical need for robust feature selection techniques in IIoT Intrusion Detection Systems (IDS). This paper introduces the Improved Binary Spider Wasp Optimizer (IBSWO) algorithm to address this pressing need. By merging the Spider Wasp Optimizer (SWO) with Genetic Algorithms (GAs) and leveraging flat crossover, the algorithm aims to enhance feature selection efficacy. Validation of the methodological framework was conducted using publicly available real-world datasets, including UNSW-NB15, TONIoT, and NCTUKM-IIOT. The results demonstrate the superior classification accuracy, precision, recall, and F1-measure of IBSWO compared to established Metaheuristic (MH) algorithms and machine learning techniques. Furthermore, the incorporation of flat crossover and transfer functions presents promising advancements in feature selection methodologies for IIoT IDS, offering implications for enhancing network security, and effectively detecting and mitigating evolving cyber threats.