A Hybrid Ensemble Framework for Intrusion Detection in Internet of Things Networks

The spreading of Internet of Things (IoT) tools has brought about a revolution in multiple sectors by facilitating smooth connectivity and data transmission. But the rapid expansion has also made IoT net- works vulnerable to serious security risks, triggering the need for reliable Intrusion Detection Systems (IDS). This paper offers an innovative ensemble-based IDS in this research that is tailored to IoT scenarios. By utilizing the Max Voting technique, the approach combines the best features of three machine learning algorithms—Decision Tree (DT), Gaussian Naive Bayes (GNB), and Extreme Gradient Boosting (XGBoost). Through the integration of multiple models, the ensemble method outperforms individual classifiers in detection, hence alleviating their limitations. The empirical results achieve an accuracy of 99.85%, indicating their effectiveness. The findings show that the ensemble approach offers a strong and well-balanced security mechanism against a variety of cyber threats, especially when Max Voting is used.