Innovation Owner
Mr. THANADOL PINTASIRI
Student
Details
This study compares the performance of five ensemble machine learning methods for predicting the Air Quality Index (AQI) using a dataset from India's Central Pollution Control Board (CPCB) covering 2021 to 2023.
This study aims to evaluate and compare the performance of predicting the air quality index (AQI) using five ensemble machine learning methods:
- Random Forest
- XGBoost
- CatBoost
- Stacking ensemble of Random Forest and XGBoost
- Stacking ensemble of Random Forest, SVR, and MLP
The research utilizes a dataset from the Central Pollution Control Board of India (CPCB), comprising 15 pollutant variables and 9 meteorological variables collected between January 2021 and December 2023, totaling 1,024,920 records. Performance was measured using RMSE, MAE, and the coefficient of determination. The results indicate that the stacking ensemble of Random Forest and XGBoost performed best, achieving a minimum RMSE of 0.1040, a minimum MAE of 0.0675, and a maximum coefficient of determination of 0.8128. SHAP-based interpretation confirmed that PM2.5 and PM10 were the two most significant variables impacting global predictions.
Objective
To study and compare the performance of air quality index prediction using five ensemble machine learning methods: Random Forest, XGBoost, CatBoost, stacking ensemble of Random Forest and XGBoost, and stacking ensemble of Random Forest, SVR, and MLP.
To study and compare the performance of air quality index prediction using five ensemble machine learning methods:
- Random Forest
- XGBoost
- CatBoost
- Stacking ensemble of Random Forest and XGBoost
- Stacking ensemble of Random Forest, SVR, and MLP