Citizens and urban settings have benefited significantly from major advances in new life and services provided by smart homes (SHs), which are fully capable of controlling physical objects in real time, collecting very sensitive information, and delivering intelligent data to citizens in many fields, such as transportation, healthcare, smart buildings and parking, public safety, traffic systems, and smart agriculture. The architecture, on the other hand, may encounter a variety of security and privacy challenges at multiple levels.It is in the applied experimental session testing the performance of various machine learning (ML) models for threat detection that one may get a thorough grasp of where and how Data Science can offer value to IoT network security. The findings serve as a foundation for illustrating the advantages of integrating new technology for the purpose of forecasting risks and concerns. Implementing machine learning into intelligent security systems, in addition, increases the requirement for a multi-disciplinary strategy and data infrastructure to manage the whole lifespan of a security product (Software Engineering end-to-end, including ML and Data DevOps). In this paper, we give an intelligent security algorithm for home data privacy and security in IoT. Experiments using the publicly available IoT-23 dataset, which contains labeled information on malicious and benign IoT network traffic, are used to complete the case study. The benign situations were received directly from the actual hardware and were not faked in any way or form. This enabled real-time network activity to be observed and evaluated. Therefore, models provide accurate outputs that may be used to forecast and identify vulnerabilities on Internet of Things-based systems. Furthermore, the lab might be expanded to accommodate the development of commercial and industry demos to demonstrate the benefits of building intrusion detection systems that use machine learning algorithms.