Machine learning for detection of cyberattacks on industrial control systems

Abstract

Senior executives for industrial systems are increasingly facing the need to reassess their cyber risk as cyberattacks are on a steep rise. This is because of the rapid digitalization of traditional industries, designed to work for decades at a time when security was not a priority. Simultaneously, the available tools to detect these attacks have also increased. This thesis aims to help researchers and industry leaders understand how to implement machine learning (ML) as an early detection tool for anomalies (cyberattacks being a subset of anomalies) in their processes. With learnings from an end-to-end implementation of some state-of-the-art machine learning models and a literature survey, this thesis highlights the critical focus areas for managers looking to implement ML tools. The thesis also helps managers to understand research metrics and converts them into business goals that would allow for better decision-making and resource allocation.