Assessment of Fine Particulate Matter for Port City of Eastern Peninsular India Using Gradient Boosting Machine Learning Model

Atmosphere

Seema Mehandia¹  Manoj Sharma²  Naresh Kumar³  Vikas Jangra⁴  Sumit Kumar⁵  Pawan Kumar⁶  Shallu Sharma⁷ 

[1] Department of Biotechnology, University Institute of Engineering and Technology (UIET), Panjab University, Chandigarh 160014, India;Department of Electronics and Communication Engineering, Giani Zail Singh Campus College of Engineering & Technology, Maharaja Ranjit Singh Punjab Technical University, Bathinda 151001, India;Department of Electronics and Communication Engineering, University Institute of Engineering and Technology (UIET), Panjab University, Chandigarh 160014, India;Department of Statistics, Panjab University, Chandigarh 160014, India;Division of Research and Development, Centre for Space Research, School of Electronics and Electrical Engineering (SEEE), Lovely Professional University, Phagwara 144411, Punjab, India;Materials Research Application Lab (MARL), Department of Nano Sciences & Materials, Central University of Jammu, Jammu 181143, India;Neuroimaging & Neuro Spectroscopy Lab, National Brain Research Centre, Manesar, Gurugram 122051, Haryana, India;

关键词: air pollution;    PM_2.5;    gradient boosting;    machine learning;    CatBoost regression model;   

DOI  :  10.3390/atmos13050743

来源: DOAJ

【 摘 要 】

An assessment and prediction of PM_2.5 for a port city of eastern peninsular India is presented. Fifteen machine learning (ML) regression models were trained, tested and implemented to predict the PM_2.5 concentration. The predicting ability of regression models was validated using air pollutants and meteorological parameters as input variables collected from sites located at Visakhapatnam, a port city on the eastern side of peninsular India, for the assessment period 2018–2019. Highly correlated air pollutants and meteorological parameters with PM_2.5 concentration were evaluated and presented during the period under study. It was found that the CatBoost regression model outperformed all other employed regression models in predicting PM_2.5 concentration with an R² score (coefficient of determination) of 0.81, median absolute error (MedAE) of 6.95 µg/m³, mean absolute percentage error (MAPE) of 0.29, root mean square error (RMSE) of 11.42 µg/m³ and mean absolute error (MAE) of 9.07 µg/m³. High PM_2.5 concentration prediction results in contrast to Indian standards were also presented. In depth seasonal assessments of PM_2.5 concentration were presented, to show variance in PM_2.5 concentration during dominant seasons.

【 授权许可】

Unknown