Explainable Boosting Ensemble Methods for Intrusion Detection in Internet of Medical Things (IoMT) Applications

This research focuses on developing an Intrusion Detection System (IDS) for Internet of Medical Things (IoMT) applications, utilizing the CICIoMT-2024 dataset. It approaches the problem as a classification challenge, classifying traffic as either one of eighteen types of attacks or benign. The study analyzes the effectiveness of three boosting algorithms—XGBoost, AdaBoost, and CatBoost—in determining their accuracy for identifying various types of attack and benign traffic. The experimental results reveal that the XGBoost classifier achieves an accuracy of 95.01%, AdaBoost records an accuracy of 92.89%, and CatBoost attains an accuracy of 88.37%. Additionally, eXplainable Artificial Intelligence (XAI) is employed to enhance the interpretability of the machine learning models.