SecuIIoT: Hyper-Tuned Adaptive Learning for Cognitive Detection of Industrial IoT (IIoT) Cyber Threats

Industrial automation has been advanced by the explosive growth of the Industrial Internet of Things (IIoT), but also has become markedly more susceptible to cyber attacks. This paper introduces intelligent and adaptive security framework called SecuIIoT which discovered the attacks in IIoT. SecuIIoT uses Hyper-tuned Ensemble Learning models that include algorithms, like ResNet50 as pre-trained model, XGBoost for feature extraction and Logistic Regression (LR) for final classification, with automatic hyper parameter optimization in order to increase the detection accuracy and reduce the rate of false positive (FP). Utilizing a cognitive approach to learning, the cognitive learning system dynamically learns from new threat patterns and changing network behaviors. The model is trained and tested from China IIoT cyber-attacks benchmark datasets and enables to achieve better accuracy of 98.78%, precision of 98.78%, recall of 97.34%, and F1-score of 98.41% compare to the traditional classifiers.