Evaluation of Machine Learning Algorithms Used on Attacks Detection in Industrial Control Systems

The Industrial Internet of Things corresponds to several industrial devices that are equipped with sensors connected to networks gathering and sharing data. These devices are being used by the industry, providing a new global industrial system on a scale never seen before, called Industry 4.0. The conjunction of industrial IoT and intelligent automation has been an asset for many enterprises, allowing the machines to take on tasks that previous generations of automation could not handle. On the other hand, the number of cyber attacks is increasing since the industrial devices have become connected to the Internet. In this paper, separate machine learning (ML) algorithms for instance Random Forest, Support Vector Machine (SVM), Decision Tree, Artificial Neural Networks (ANN), K-Nearest Neighbors (KNN) and Naive Bayes are evaluated for attacks detection against ICS and their performance metrics are recorded. The outcome shows great execution of machine learning algorithms in identifying assaults, furthermore, shows a meager erroneous alarm rate thus implies, it identifies ordinary traffic very well.