Evaluation of Machine Learning Models for Ozone Concentration Forecasting in the Metropolitan Valley of Mexico

Abstract

In large and densely populated cities, the concentration of pollutants such as ozone and its dispersion is related to effects on people’s health; therefore, its forecast is of great importance to the government and the population. Given the increased computing capacity that allows for processing massive amounts of data, the use of machine learning (ML) as a tool for air quality analysis and forecasting has gotten a significant boost. This research focuses on evaluating different models, such as Random Forest (RF), Support Vector Regression (SVR), and Gradient Boosting (GB), to forecast ozone (O3) concentration 24 h in advance, using data from the Mexico City Atmospheric Monitoring System using meteorological variables that influence the phenomenon of ozone dispersion and formation.