I. Introduction
The Internet of Things (IoT) has been involved in various services, including smart cities, smart healthcare, and smart manufacturer, to enhance the quality of life, the efficiency of urban services, operation, and competitiveness [1]. A distributed network made up of IoT devices connected via wired or wireless networks continuously interacts with the outside world to provide a variety of data sources, including images, text, video, and other sorts of data. The distributed network also enables efficient and effective sharing of IoT data resources and information. However, due to the substantial amount of data generated by IoT sensors, an intelligent system is necessary to operate the system autonomously, as IoT devices are typically resource-limited and cannot independently execute machine learning algorithms. With the aid of edge computing, IoT clusters can form intelligent networks when combined with machine learning [2]. However, having an effective machine learning model necessitates extensive data from many IoT clusters, which is often difficult to collect and utilize due to privacy concerns, security risks, and other associated challenges [3].