Prediction of Air Pollutant PM10 using Various SVM Models

The air pollution is majorly caused by 3 harmful pollutants viz. SO
2
, NO
2
and PM (Particulate Matter). Inside particulate matter, the solid particles are present mixed up with liquid droplets. The PM is of two types viz. PM10 and PM2.5. The diameter of PM10 particles is 10 mm or less, and that of PM2.5 particles is 2.5 mm or less. These particles can cause very harmful consequences on mankind. The PM particles are formed as a result of chemical reaction taking place between SO
2
and NO
2
. So, the knowledge of SO
2
and NO
2
concentrations can be a base for the prediction of PM. In this article, an effort has been put to predict PM10 component for given SO
2
and NO
2
pollutant concentrations using SVM (Support Vector Machine) techniques with different kernel functions in MATLAB (R2021a). The prediction performance of all prediction models is evaluated and compared.