Exploring the Landscape of IoT Ransomware Prediction Through Machine Learning Techniques: A Comprehensive Survey
Résumé
The rapid expansion of Internet of Things (IoT) devices has revolutionized various sectors. It enhances automation, facilitates data collection, and enables real-time monitoring. However, it has also exposed these interconnected systems to significant security risks, particularly ransomware attacks, an increasingly common threat capable of causing severe damage to individuals and organizations. To deal with this issue, it is necessary to leverage machine learning techniques to come up with a robust early detection solution to protect IoT infrastructures against ransomware effectively. This survey reviews state-of-the-art solutions for IoT ransomware prediction using machine learning techniques by mainly focusing on their analysis tasks, including detection, classification, and early detection. The survey also introduces a multi-criteria taxonomy to categorize existing solutions systematically for different aspects. This taxonomy allows us to compare the solutions and highlight the gaps in the research literature. The findings of this survey show that there is still significant potential for advancing the state-of-the-art and addressing existing research gaps. Furthermore, we analyze the advantages and limitations of the proposed solutions, highlight unresolved challenges, and suggest future research directions