Semnan University International Journal of Nonlinear Analysis and Applications 2008-6822 2026 01 28 Prediction of air quality index using machine learning algorithms: A case study of Tehran 10451 10.22075/ijnaa.2025.34633.5179 EN Arshia Azizi Department of Mathematics and Computer Science, Damghan University, Damghan, Iran Sajjad Rahmany Department of Mathematics and Computer Science, Damghan University, Damghan, Iran Journal Article 2024 07 01 Air quality index (AQI) forecasting is a useful tool for increasing the general public 's awareness of the state of the air in the next days. This is one of the most significant problems facing any country. In this study, machine learning algorithms are used to predict the AQI in Tehran. The six important regression models are applied to forecast AQI on a daily basis. Models were compared and evaluated using statistical measures such as Mean Absolute Error (MAE), coefficient of determination, and root mean square error (RMSE). Based on these evaluations, the best model was selected. ExtraTreesRegressor is thought to be the best model for forecasting AQI in all seasons based on its outcomes. The results demonstrate that the ExtraTreesRegressor 's determination coefficient is nearly 1, and that the values of MAE and RMSE are respectively 0.002 and 0.004. https://ijnaa.semnan.ac.ir/article_10451_4b730bb369f87ad369fde0ffbf74a3c2.pdf