ENHANCING ML-BASED ANOMALY DETECTION IN DATA MANAGEMENT FOR SECURITY THROUGH INTEGRATION OF IOT AND CLOUD COMPUTING

Abstract

Cloud computing (CC) and the Internet of Things (IoT) are widely used technologies across a variety of fields including science, medicine, security, and economics, providing efficient data management and automation capabilities. However, data quality and security challenges remain significant concerns for both technologies. Low-cost sensors and controllers that make up IoT devices are often vulnerable to security breaches and generate diverse and difficult-to-manage data. Although CC provides services for diverse user requirements, its data security and management transparency are still areas of concern. To mitigate vulnerabilities, IoT data must be stored in cloud or edge services using SQL or NoSQL databases, and secure connections between CC and IoT must be established. CC provides high-performance hardware and software to users on a rental basis, reducing the need for significant infrastructure investments. This study investigates the interaction between CC and IoT, focusing on protocols and security measures for microcontrollers that generate and transmit data to the cloud, enable real-time data retrieval after local server storage, and facilitate anomaly detection through analysis. The primary objective of this thesis is to enhance MLbased anomaly detection in data management through the integration of IoT and cloud computing, while also identifying benefits and challenges associated with integrating IoT and cloud technologies, assessing the security measures implemented in database management for IoT devices and cloud computing systems, and providing insights into optimal methods for implementing database management and security measures in the interaction between IoT and cloud technologies. Experimental tests and evaluations demonstrate the effectiveness of our proposed solution