RESUMEN
The aim of this work is to investigate possible relationships between air quality and the spread of the pandemic. We evaluate the performance of machine learning techniques in predicting new cases. Specifically, we describe a cross-correlation analysis on daily COVID-19 cases and environmental factors, such as temperature, relative humidity, and atmospheric pollutants. Our analysis confirms a significant association of some environmental parameters with the spread of the virus. This suggests that machine learning models trained using environmental parameters might provide accurate predictions about the number of infected cases. Our empirical evaluation shows that temperature and ozone are negatively correlated with confirmed cases (therefore, the higher the values of these parameters, the lower the number of infected cases), whereas atmospheric particulate matter and nitrogen dioxide are positively correlated. We developed and compared three different predictive models to test whether these technologies can be useful to estimate the evolution of the pandemic.