Abstract
This paper investigates the use of ensemble learning techniques to enhance the security of Internet of Things (IoT) networks through advanced Intrusion Detection Systems (IDS). Utilizing the CICIoT2023 dataset, the study evaluates four ensemble methods: Ensemble Voting, Random Subspace, Bayesian Model Averaging, and Boosting. The results demonstrate that Random Subspace and Bayesian Model Averaging significantly improve detection accuracy and robustness against various attack types, highlighting their potential in real-world IoT security applications.