From the most recent event, global warming has now caused harm to humanity and the Earth’s ecosystem. This study investigated cooling systems in National Capital Region (NCR) commercial buildings and the factors dynamic to the growing demand for electrical energy within these systems. This would also augment Republic Act (RA) No. 11285 to improve the energy consumption in the Philippines. Various statistical methods evaluated the behavior data collected. Further, the data are analyzed using a machine learning algorithm Random Forest Regression, RBF Kernel Support Vector Regression, and Gradient Boosting Regression. However, gradient-boosting regression, with an R2 value of 0.2719, shows a better model performance in this study. In addition, this paper analyzed the variable importance of every feature by using a percent increase in MSE. As a result, building activity has the largest influence on the model, cities ranked second, and cooling degree days ranked third. Further, partial dependencies are used to understand how the high-importance features impact cooling energy consumption.