Security of IoT-Cloud Systems Based Machine Learning

Cloud Storage plays a vital role as an intermediary layer that connects objects and applications, simplifying complexities and enabling seamless functionality. The Internet of Things represents a network of interconnected artifacts that serves diverse purposes. Ensuring security and staying updated with emerging techniques are essential for meeting the requirements of IoT-Cloud systems. The integration of Cloud and IoT introduces concerns regarding the trustworthiness of cloud providers and the transparency of data storage. Multitenancy Cloud Services Storage Systems raise worries about maintaining confidentiality and integrity, potentially resulting in unintended data exposure. The prevailing skepticism surrounding cloud service providers classifies this vulnerability as an internal threat within the IT industry. This paper extensively investigates the security of Machine Learning-based IoT-Cloud systems, highlighting the encountered challenges and the employed Machine Learning techniques. The prevailing evaluation metrics are examined, and previous researches are analyzed and compared. In the concluding section, the paper proposes an intelligent model for detecting existing and emerging attacks in IoT-Cloud systems.