Comprehensive Comparative Analysis of Attack Detecting Using Different Machine Learning Algorithms in IIoT

Industrial Internet of Things (IIoT) is a subset of Internet of Things (IoT) which involves interconnected industrial devices to improve industrial system’s productivity and operational capability. However, these smart systems are suspectable to different types of attack and that raises a strong security concern as such attacks cause a huge damage to the industrial sector. In this project, we provide a comparative analysis on IIOT-related dataset using several supervised and unsupervised machine learning algorithms and evaluate their detection performance in terms of accuracy, recall, precision. Furthermore, we conduct the experiments using three different train/test split scenarios and observe how these scenarios impact the algorithms’ performance.