The lifeblood of contemporary cultures is energy. Due to population growth and rising luxury demands, the world's energy consumption and related CO2 emissions have rapidly increased in recent decades. Predicting building energy use is crucial for energy management, planning, and conservation. The objective of the project is to create a machine learning model to lower building emissions and energy use by utilising smart sensors in either residential or commercial buildings. In this project, a smart sensor will be used to develop a machine learning model, method, design, or appliance to reduce energy consumption and emissions in residential or commercial buildings. The sensor uses real-time occupancy sensor networks, changing space schedules, weather forecasts, and other environmental conditions to accurately estimate room occupancy. Set points and schedules can be successfully adjusted by a skilled operator.The sensor detects various things, including CO2 levels, sound levels, ambient light, and door status. Using machine learning methods, these may be used to precisely estimate how many people are in each room. The amount of knowledge that a human being can process,however good it may be, limits their abilities. There are numerous chances to utilise external data sources, such as real-time occupancy sensors. This technology can be used to predict future occupancy, reduce energy management, plan energy management, and conserve energy by predicting future occupancy and consuming less energy.