Using Applied Machine Learning to Detect Cyber-Security Threats in Industrial IoT Devices

Digital ubiquity has aided an industrial rise in Internet of Things (IoT) device integrability, machine-to-machine communication in manufacturing operations, and cybersecurity needs. Cyber-attacks from threats such as botnets exploit new forms of digital connectivity and threaten substantial financial losses for manufacturing enterprises. There is a need for an AI-powered network intrusion detection system adaptable enough to keep pace with growing digital landscapes. In this research, different Machine Learning (ML) and Deep Learning (DL) algorithms were deployed to detect botnet attacks on seven IoT devices. The primary objective was to develop secure and accurate models for the successful identification of security threats through botnet attacks. All models demonstrated high detection accuracy, precision, and recall rates above 90%. The decision tree (DT) model performed best, and exhibited 100% accuracy, precision, and recall for threat detection.