The area of the Internet of Things (IoT) is rapidly growing, raising severe security concerns for the entire network. Due to its high traffic volume and real-time operation, a security framework is essential. The paradigm of the Internet of Things (IoT) requires pervasive connectivity to billions of heterogeneous devices. Devices in the Internet of Things (IoT) are very limited in processing capacity, memory, and storage. Therefore, a lightweight security process is to be provided for IoT devices in other to create the required development. This paper introduces several developments in lightweight versions of various known security processes and analyzes each relevant process. With a machine-learning-based model process that provides a lightweight solution that works by selecting the best features leveraging well-known filter and wrapper methods for feature selection. The strategy can be evaluated over different datasets collected from varying network scenarios.