This work describes the process of design, implementation, and results of different Machine Learning techniques to understand and predict meteorological data from the metropolitan area of Quito, Ecuador. Hourly data has been collected from 2004 to 2018 by the Environment Secretary of Quito in nine stations around the district. It focuses mainly on linear regression algorithms, time series analysis, and interpolation spatial analysis. During the definition of graphs to be implemented in the AirQ2 application, an analysis process was carried out on the contribution of each type of graph to obtain a better result when constructing prediction models on air pollution in Quito. Each graphic described here provides specific information and provides a guide to learn more about the different pollutants and stations of the collected data, as well as their contribution to the construction of the models’ prediction.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados