Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications
Air pollution poses a significant global challenge, impacting human health and environmental sustainability worldwide. Accurate air quality forecasting is essential for effective mitigation strategies, particularly in rapidly urbanizing regions. This study focuses on Shenyang, China, as a representa...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/16/7/776 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839616515196846080 |
|---|---|
| author | Zhaoxin Xu Huajian Zhang Andong Zhai Chunyu Kong Jinping Zhang |
| author_facet | Zhaoxin Xu Huajian Zhang Andong Zhai Chunyu Kong Jinping Zhang |
| author_sort | Zhaoxin Xu |
| collection | DOAJ |
| description | Air pollution poses a significant global challenge, impacting human health and environmental sustainability worldwide. Accurate air quality forecasting is essential for effective mitigation strategies, particularly in rapidly urbanizing regions. This study focuses on Shenyang, China, as a representative case to analyze air quality dynamics and develop a high-precision forecasting tool. Using a comprehensive six-year dataset (2020–2025) of daily air quality and meteorological measurements, a rigorous preprocessing pipeline was applied to ensure data integrity. Five gradient-boosted decision-tree models were trained and combined through a ridge-regularized stacking ensemble to enhance the predictive accuracy. The ensemble achieved an R<sup>2</sup> of 94.17% and a mean absolute percentage error of 7.79%, outperforming individual models. The feature importance analysis revealed that ozone, PM<sub>10</sub>, and PM<sub>2.5</sub> concentrations are the dominant drivers of daily air quality fluctuations. The resulting forecasting system delivers robust, interpretable predictions across seasonal variations, offering a valuable decision support tool for urban air quality management. This framework demonstrates how advanced machine learning techniques can be applied in a Chinese urban context to inform global air pollution mitigation efforts. |
| format | Article |
| id | doaj-art-0f6c1b12a85b45a8b7eeba62fca33a86 |
| institution | Matheson Library |
| issn | 2073-4433 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj-art-0f6c1b12a85b45a8b7eeba62fca33a862025-07-25T13:13:21ZengMDPI AGAtmosphere2073-44332025-06-0116777610.3390/atmos16070776Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global ImplicationsZhaoxin Xu0Huajian Zhang1Andong Zhai2Chunyu Kong3Jinping Zhang4School of Mechanical and Power Engineering, Shenyang University of Chemical Technology, Shenyang 110142, ChinaSchool of Mechanical and Power Engineering, Shenyang University of Chemical Technology, Shenyang 110142, ChinaDepartment of Electrical Engineering and Information Technology, Shandong University of Science and Technology, Qingdao 266042, ChinaSchool of Mechanical and Power Engineering, Shenyang University of Chemical Technology, Shenyang 110142, ChinaSchool of Mechanical and Electrical Engineering, Yunnan Open University, Kunming 650223, ChinaAir pollution poses a significant global challenge, impacting human health and environmental sustainability worldwide. Accurate air quality forecasting is essential for effective mitigation strategies, particularly in rapidly urbanizing regions. This study focuses on Shenyang, China, as a representative case to analyze air quality dynamics and develop a high-precision forecasting tool. Using a comprehensive six-year dataset (2020–2025) of daily air quality and meteorological measurements, a rigorous preprocessing pipeline was applied to ensure data integrity. Five gradient-boosted decision-tree models were trained and combined through a ridge-regularized stacking ensemble to enhance the predictive accuracy. The ensemble achieved an R<sup>2</sup> of 94.17% and a mean absolute percentage error of 7.79%, outperforming individual models. The feature importance analysis revealed that ozone, PM<sub>10</sub>, and PM<sub>2.5</sub> concentrations are the dominant drivers of daily air quality fluctuations. The resulting forecasting system delivers robust, interpretable predictions across seasonal variations, offering a valuable decision support tool for urban air quality management. This framework demonstrates how advanced machine learning techniques can be applied in a Chinese urban context to inform global air pollution mitigation efforts.https://www.mdpi.com/2073-4433/16/7/776AQI prediction modelKalman filteringspecial engineeringK-fold cross validationmulti model stacking fusion |
| spellingShingle | Zhaoxin Xu Huajian Zhang Andong Zhai Chunyu Kong Jinping Zhang Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications Atmosphere AQI prediction model Kalman filtering special engineering K-fold cross validation multi model stacking fusion |
| title | Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications |
| title_full | Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications |
| title_fullStr | Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications |
| title_full_unstemmed | Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications |
| title_short | Stacking Ensemble Learning and SHAP-Based Insights for Urban Air Quality Forecasting: Evidence from Shenyang and Global Implications |
| title_sort | stacking ensemble learning and shap based insights for urban air quality forecasting evidence from shenyang and global implications |
| topic | AQI prediction model Kalman filtering special engineering K-fold cross validation multi model stacking fusion |
| url | https://www.mdpi.com/2073-4433/16/7/776 |
| work_keys_str_mv | AT zhaoxinxu stackingensemblelearningandshapbasedinsightsforurbanairqualityforecastingevidencefromshenyangandglobalimplications AT huajianzhang stackingensemblelearningandshapbasedinsightsforurbanairqualityforecastingevidencefromshenyangandglobalimplications AT andongzhai stackingensemblelearningandshapbasedinsightsforurbanairqualityforecastingevidencefromshenyangandglobalimplications AT chunyukong stackingensemblelearningandshapbasedinsightsforurbanairqualityforecastingevidencefromshenyangandglobalimplications AT jinpingzhang stackingensemblelearningandshapbasedinsightsforurbanairqualityforecastingevidencefromshenyangandglobalimplications |