Support vector machines (SVM) for IoT device classification in cloud environments

To manage the resources effectively and secure their usage, particularly when using the IoT in smart home automation and other devices, there is a need to classify IoT devices in the cloud. This paper explores the possibilities of Support Vector Machines (SVM) to provide improved scalability and classification of cloud-based systems in IoT devices. Current techniques frequently fail to perform well and introduce additional computation costs when they are required to handle high dimensional data, skewed data and real time processing requirementsThe Smart Home Automation-based Machine Learning (SHA-ML) framework leverages SVM and distributed computing for scalable, efficient IoT device classification. It handles high-dimensional data with minimal training, improving accuracy, latency, and resource use, demonstrating SVM’s effectiveness for real-time cloud-based smart home automation. This approach does not only address the issues that are occurring at this point, but also enables the hassle-free and secure implementation of IoT solutions. The recommended approach has an accuracy of 98.74% when it comes to classification, a latency rate of 43.52%, a resource utilization rate of 97.68%, a processing demands rate of 98.34%, and a scalability rate of 96.57%.