Cyber Attack Detection for Industrial IoT using ML Techniques

The industrial sector is seeing a rapid expansion of the Industrial Internet of Things (IIoT), and protecting IIoT systems against cyber threats is essential. This study suggests a machine-learning-based intrusion detection solution for the Industrial Internet of Things. The suggested system classifies incoming network data as malicious or benign using two separate machine classifier algorithms: Decision Tree and Support Vector Machine. The Edge-IIoTsetdataset, which comprises 157,800 samples—133499 attack samples and 24301 normal samples was utilized in this study. To train and evaluate the suggested intrusion detection system, the dataset underwent preprocessing and features were extracted. The accuracy, recall, precision, and F1-score were used to assess the suggested system's performance. The experiments' findings demonstrate that the suggested system successfully identified malicious traffic in the IIoT network with high recall, accuracy, and precision. With 96.27% accuracy, the Decision Tree algorithm was the most accurate, followed by the support vector machine with 74.88% accuracy. It is anticipated that the suggested cyber-attack detection system will improve IIoT system security and lessen the possibility of cyberattacks in the industrial sector. To increase their efficacy and offer a security solution, it is advised to integrate the suggested system with the current IIoT security systems.