Evaluating trends of airborne contaminants by using support vector regression techniques

Abstract

Monitoring, modeling and forecasting of air quality parameters are important topics in environmental and health research due to their impact caused by exposing to air pollutants in urban environments. The aim of this article is to show that forecast of daily airborne pollution using support vector machines (SVM) is feasible in regression mode. Results are presented using data measurements of Particulate Matter of aerodynamical size on the order of 10 and 2.5 micrograms (PMx) in London-Bloomsbury at south England.