Predicción de la Contaminación Atmosférica por Partículas (PM2.5 y SO2) mediante un Modelo de Regresión de Crestas.

Abstract

Atmospheric pollution is one of the biggest environmental problems today, due to the serious effects it generates not only for the environment but also for human health as a consequence of accelerated socio-economic growth, which mainly denotes industrial activity, which has caused a considerable increase in the concentration of particulate matter such as PM 2.5, O3, NO2, SO2 and CO, which are some of the most incident variables in air quality monitoring, according to the registry prepared by the RMCA (The Air Quality Monitoring Network) for the city of Bogotá. However, the possibility of detecting polluting agents in advance could contribute to the formulation of alternatives and management plans for control and mitigation that prevent the spread of these variables, especially in sectors with greater environmental threat. Therefore, the objective of this article is to analyze the efficiency of the “ridge regression” algorithm for the study and prediction regarding the average fluctuation of polluting gases such as PM 2.5 and SO2 mainly, taking as a reference area the town of Kennedy.