An Efficient Hybrid Approach Based on Deep Learning and Stacking Ensemble Using the Whale Optimization Algorithm for Detecting Attacks in <scp>IoT</scp> Devices

ABSTRACT With the increasing proliferation of the Internet of Things (IoT) and the rise of cyber‐attacks targeting these devices, the need for efficient attack detection methods has become increasingly urgent. In this paper, a novel approach is proposed based on ensemble deep transfer learning optimized using the Whale Optimization Algorithm (WOA) to detect attacks on IoT devices. The method employs multiple ensemble deep transfer learning (DTL) models that transfer knowledge from data in other domains to IoT‐specific data. Subsequently, the WOA algorithm is utilized to optimize the model parameters and combine them into an ensemble classifier. The performance of the proposed method is evaluated using the Edge‐IIoTset dataset, and the results demonstrate a detection accuracy of 99.6%, with a significantly reduced false alarm rate. These findings highlight the high efficiency of the proposed method in enhancing the security of IoT devices.