This research explores the application of machine learning algorithms to predict energy consumption in educational buildings. The study evaluates three models: random forest (RF), long short-term memory (LSTM), and gradient boosting regressor (GBR). The goal is to enhance energy efficiency and sustainability in smart buildings by accurately forecasting energy usage. The research uses real-world data from three educational buildings, analyzing factors such as occupancy, temperature, and energy use to develop and test the models. The data was preprocessed and normalized, and performance metrics like RMSE, MAE, and MAPE were used to compare the models. The results show that GBR outperformed the other models in terms of accuracy, with high R-squared values and low error rates. The study highlights the importance of tailoring predictive models to the specific characteristics of each building's energy consumption. The findings suggest that GBR is a strong candidate for energy consumption prediction in educational buildings, offering improved accuracy and efficiency compared to traditional and deep learning methods. The research also discusses the potential of transformer models for future energy consumption forecasting and emphasizes the need for further optimization and exploration of advanced algorithms to enhance energy management in smart buildings.This research explores the application of machine learning algorithms to predict energy consumption in educational buildings. The study evaluates three models: random forest (RF), long short-term memory (LSTM), and gradient boosting regressor (GBR). The goal is to enhance energy efficiency and sustainability in smart buildings by accurately forecasting energy usage. The research uses real-world data from three educational buildings, analyzing factors such as occupancy, temperature, and energy use to develop and test the models. The data was preprocessed and normalized, and performance metrics like RMSE, MAE, and MAPE were used to compare the models. The results show that GBR outperformed the other models in terms of accuracy, with high R-squared values and low error rates. The study highlights the importance of tailoring predictive models to the specific characteristics of each building's energy consumption. The findings suggest that GBR is a strong candidate for energy consumption prediction in educational buildings, offering improved accuracy and efficiency compared to traditional and deep learning methods. The research also discusses the potential of transformer models for future energy consumption forecasting and emphasizes the need for further optimization and exploration of advanced algorithms to enhance energy management in smart buildings.