Enhancing IIoT Security: A Stacked Intelligent Ensemble Model for Intrusion Detection

The embracing of Industrial Internet of Things (IIoT) technology revolutionized industrial operations through smart automation and data-driven decision making. However, expansion of huge connected devices spreading across operational network has substantially increased cybersecurity threats. The enhanced operational decision-making benefits of IIoT face increasing security challenges because of its expanding complex operational structure that leads to various attacks and data breaches alongside illegal access and other destructive intrusions. The research proposes a stacking-based intelligent intrusion detection system (IDS) that utilizing ensemble learning to enhance threat detection capabilities in IIoT networks. The proposed IDS employes a meta-classifier to combine base classifier decisions from multiple base classifiers of machine learning and deep learning techniques such as Naïve Bayes, k-Nearest Neighbours, Logistic Regression, and Multi-Layer Perceptron for predicting cyber threats. The Edge-IIoTset dataset is utilized for extensive experiments to measure the performance of proposed IDS on multiple classes in IIoT security situations. The proposed IDS demonstrates successful and efficient IIoT attack prediction by achieving accuracy at 99.88% combined with precision at 99.88%, recall at 99.9%, and F1-score at 99.88%.