Forecasting Air Quality Dynamics: Employing Machine Learning Models for Enhanced Environmental Health Predictions

Increasing emissions of air pollutants and rapid urbanization have raised global concerns about air quality and public health. Therefore, predicting air quality is important to reduce its negative impacts. This research identifies the variables that most influence air quality and develops a Machine Learning-based prediction model to estimate the air quality index (AQI). Machine Learning models used in this research include Random Forest, Support Vector Machine (SVM), and Neural Network. In developing the model, historical air quality data from several large cities in the world were used to train and test the accuracy of the model predictions. The aim of this research is to improve the accuracy of air quality predictions and identify the main patterns that influence it.This research uses quantitative methods with secondary data collection from global air quality databases. Prediction models are built and validated using cross-validation techniques and evaluation metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The research results show that the Random Forest model has the best performance with an MAE of 2.5 and an RMSE of 3.7, compared to the SVM and Neural Network models. The Random Forest model is also able to identify the main variables that influence AQI, such as PM2.5, PM10 and NO2 concentrations. The prediction accuracy reached 92%, indicating that the model is reliable for real applications. These results provide important insights for policy makers and researchers to take mitigation actions against air pollution. In conclusion, the Machine Learning model developed in this research is effective for predicting air quality and can be used as a useful tool in environmental management. This research suggests using real-time data for more accurate predictions in the future.