Securing The Internet of Vehicles (IoVs): Review of Machine Learning Techniques for Malware Detection

Interconnectivity in the Internet of Vehicles (IoVs), enabled by its development and implementation in real-world, facilitates seamless communication between vehicles, infrastructure, and backend systems but also introduces important cybersecurity risks, with malware posing a dangerous threat. Conventional security solutions that rely on signature-based detection, fail against new variants. This current study explores Machine Learning (ML) as a proactive solution for enhancing detection in IoVs systems. Through investigation of new studies, the current study, aims to explore various ML approaches for IoVs malware uncovering, including supervised, unsupervised, and deep learning techniques. It evaluates their effectiveness, strengths, and limitations in terms of accuracy, efficiency, and flexibility. These novel techniques analyze specific features, ensuring accurate detection of malware during the analysis process. The study meticulously divides both traditional-based approaches and ML methodologies, considering findings from previous and recent research papers. Research conclusions derived from rigorous experiments are evaluated, and algorithm performances are systematically compared based on accuracy detection. As the massive implementation of IoVs depends on the ability to establish trust and confidence against evolving cyber threats, it becomes imperative, making the integration of advanced security measures, particularly those rooted in ML to enhance security against malware threats in IoVs network.