<scp>RF</scp> ‐ <scp>FLIDS</scp> : A Novel Hybrid Intrusion Detection Model for Enhanced Anomaly Detection in <scp>IoT</scp> Networks

ABSTRACT The rapid proliferation of Internet of Things (IoT) technology has transformed modern industries, introducing significant security vulnerabilities that necessitate practical intrusion detection systems (IDSs). While random forest (RF) has shown promise for anomaly detection in IoT networks, its inherent bias toward majority classes in imbalanced datasets limits its effectiveness. Existing solutions primarily focus on data‐level techniques, such as oversampling or undersampling, or algorithm‐level adjustments, including cost‐sensitive learning, which often increase detection time or risk overfitting. This paper proposes RF‐FLIDS, a hybrid random forest and Fuzzy Logic‐based IDS, to address this challenge, providing a model‐level correction for class imbalance. RF‐FLIDS mitigates the overfitting risks associated with oversampling and eliminates the need for reweighting or modifying the RF algorithm, making it well suited for real‐time anomaly detection. The model was evaluated using two benchmark datasets, CICIoT2023 and Aposemat IoT‐23, and compared against traditional machine learning models, including Balanced Random Forest (BRF), logistic regression (LR), k‐nearest neighbor (KNN), decision trees (DT), and Gaussian Naive Bayes (GNB). Experimental results demonstrated the superior performance of RF‐FLIDS, achieving Balanced Accuracy of 92.52% and 92.92%, with significant improvements in Precision, Recall, and F1‐score. These findings confirm the robustness and adaptability of the proposed model for enhancing IoT security.