Developed to protect connected devices against complex cyber threats, the system, titled the Stacking Ensemble-based IoT Attack Detection Model, or SE-IADM, integrates multiple classification algorithms with hybrid feature selection and data balancing.

Conventional intrusion detection systems frequently struggle to handle the complexity, heterogeneity, and dynamic nature of modern IoT attacks.

When tested on the ToNIoT benchmark data, the ensemble achieved 99.41 percent accuracy for binary attack detection and 96.64 percent accuracy for multiclass classification, demonstrating consistent detection capabilities across distinct network environments.

Subject of Research: Intrusion detection in Internet of Things networks using stacked ensemble machine learning Article Title: SE-IADM: an optimized stacked ensemble model for smart IoT attack detection using hybrid feature selection Article References: Khan, M. Z., Bokhari, M. U., Alam, M., Husain, M. S., Khan, M. Z., Ashraf, Z., & Masoodi, F. S. (2026).

SE-IADM: an optimized stacked ensemble model for smart IoT attack detection using hybrid feature selection.