Using Machine Learning Techniques to Address IoT Forensics Challenges
Boitumelo Nkwe, Michael E. Kyobe · 2025
While the Internet of Things$\text{IoT}$provides many useful benefits to organisations, they present cybersecurity challenges and difficulties in conducting$\text{IoT}$digital forensic investigations. Machine Learning (ML) technologies promise to address these challenges; however, little is still known about the challenges and ML techniques and methodologies for$\text{IoT}$digital forensics investigations. This study aims to provide an understanding of the challenges involving$\text{IoT}$forensics investigations in recent years and the contribution of ML techniques and frameworks in addressing them. A systematic literature review, guided by Khan's et al approach (2003) was conducted. The findings indicate that$\text{IoT}$digital forensic investigations are impeded by the heterogeneous nature of$\text{IoT}$, the lack of privacy, and lack of standardization of the investigation process, among other factors. ML supported digital forensic technologies and methodologies provide several advantages in$\text{IoT}$investigations compared to traditional forensic methods. However, more research is still needed to leverage the capabilities of AI and ML in digital quantum forensics, virtual reality evidence analysis, mobile-based investigations, and data collection in cloud environment.