Forecasting PM10 Concentrations in the Caribbean Area Using Machine Learning Models

In the Caribbean basin, particulate matter lower or equal to 10 ?m in diameter (PM10) has a huge impact on human mortality and morbidity due African dust. For first time this geographical area, theoretical framework of artificial intelligence is applied forecast PM10 concentrations. The aim study concentrations using six machine learning (ML) models: support vector regression (SVR), k-nearest neighbor (kNN), random forest (RFR), gradient boosting (GBR), Tweedie (TR), Bayesian ridge (BRR). Overall, with MBEmax = ?2.8139, results showed that all models tend slightly underestimate empirical data. GBR model gives best performance (r 0.7831, R2 0.6132, MAE 6.8479, RMSE 10.4400, IOA 0.7368). By comparing our other ML studies megacities, we found similar only three input variables, whereas previous use many variables Artificial Neural Network (ANN) models. All these features area.