Real Time Air Quality Evaluation Model using Machine Learning Approach

Abstract

In recent years, the world is being industrialized day-by-day which ultimately compels our concentration towards air quality. A gradual increase in population along with the raise in usage of vehicles and consumption of conventional energy leads to air pollution which subsequently accelerates the deterioration of air quality. And air pollution has its severe impact on human health. Many researchers have proposed various methodologies for predicting and forecasting the air quality. But it is rather important to predict the future air quality in order to reduce its impact. Therefore, this paper proposes an air quality evaluation system for future prediction. The current experiment includes three modules namely Preparation of Data, Forecasting AQI and Evaluating Air Quality. Data preparation is collecting real time data and formatting it as an input to next module. Sparse Spectrum GPR (SSGPR) is used in this study to forecast, whereas cloud model to evaluate air quality. The proposed model is capable of modelling the fuzziness and randomness. Finally, the entire model is evaluated using performance metrics like MAE, RSME and MAPE.