Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning
Gridded gross domestic product (GDP) data are a crucial land surface parameter for many geoscience applications. Recently, machine learning approaches have become powerful tools in generating gridded GDP data. However, most machine learning approaches for gridded GDP estimation seldom consider the g...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/12/3/123 |
| _version_ | 1797611376271687680 |
|---|---|
| author | Zekun Xu Yu Wang Guihou Sun Yuehong Chen Qiang Ma Xiaoxiang Zhang |
| author_facet | Zekun Xu Yu Wang Guihou Sun Yuehong Chen Qiang Ma Xiaoxiang Zhang |
| author_sort | Zekun Xu |
| collection | DOAJ |
| description | Gridded gross domestic product (GDP) data are a crucial land surface parameter for many geoscience applications. Recently, machine learning approaches have become powerful tools in generating gridded GDP data. However, most machine learning approaches for gridded GDP estimation seldom consider the geographical properties of input variables. Therefore, in this study, a geographically weighted stacking ensemble learning approach was developed to generate gridded GDP data. Three algorithms—random forest, XGBoost, and LightGBM—were used as base models, and the linear regression in stacking ensemble learning was replaced by geographically weighted regression to locally fuse the three predictions. A case study was conducted in China to demonstrate the effectiveness of the proposed approach. The results showed that the proposed GDP downscaling approach outperformed the three base models and traditional stacking ensemble learning. Meanwhile, it had good predictive power on county-level GDP test data with R<sup>2</sup> of 0.894, 0.976, and 0.976 for the primary, secondary, and tertiary sectors, respectively. Moreover, the predicted 1 km gridded GDP data had a high accuracy (R<sup>2</sup> = 0.787) when evaluated by town-level GDP data. Hence, the proposed GDP downscaling approach provides a valuable option for generating gridded GDP data. The generated 1 km gridded GDP data of China from 2020 are of great significance for other applications. |
| first_indexed | 2024-03-11T06:26:59Z |
| format | Article |
| id | doaj.art-55de84502c9c48678be09254c80abcfd |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-11T06:26:59Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-55de84502c9c48678be09254c80abcfd2023-11-17T11:28:23ZengMDPI AGISPRS International Journal of Geo-Information2220-99642023-03-0112312310.3390/ijgi12030123Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble LearningZekun Xu0Yu Wang1Guihou Sun2Yuehong Chen3Qiang Ma4Xiaoxiang Zhang5College of Hydrology and Water Resources, Hohai University, Nanjing 210098, ChinaFlood and Drought Disaster Prevention and Protection Center of Heilongjiang Province, Haerbin 150006, ChinaCollege of Hydrology and Water Resources, Hohai University, Nanjing 210098, ChinaCollege of Hydrology and Water Resources, Hohai University, Nanjing 210098, ChinaResearch Center on Flood and Drought Disaster Reduction, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaCollege of Hydrology and Water Resources, Hohai University, Nanjing 210098, ChinaGridded gross domestic product (GDP) data are a crucial land surface parameter for many geoscience applications. Recently, machine learning approaches have become powerful tools in generating gridded GDP data. However, most machine learning approaches for gridded GDP estimation seldom consider the geographical properties of input variables. Therefore, in this study, a geographically weighted stacking ensemble learning approach was developed to generate gridded GDP data. Three algorithms—random forest, XGBoost, and LightGBM—were used as base models, and the linear regression in stacking ensemble learning was replaced by geographically weighted regression to locally fuse the three predictions. A case study was conducted in China to demonstrate the effectiveness of the proposed approach. The results showed that the proposed GDP downscaling approach outperformed the three base models and traditional stacking ensemble learning. Meanwhile, it had good predictive power on county-level GDP test data with R<sup>2</sup> of 0.894, 0.976, and 0.976 for the primary, secondary, and tertiary sectors, respectively. Moreover, the predicted 1 km gridded GDP data had a high accuracy (R<sup>2</sup> = 0.787) when evaluated by town-level GDP data. Hence, the proposed GDP downscaling approach provides a valuable option for generating gridded GDP data. The generated 1 km gridded GDP data of China from 2020 are of great significance for other applications.https://www.mdpi.com/2220-9964/12/3/123gross domestic product (GDP)gridded GDPensemble learninggeographically weighted regressionChina |
| spellingShingle | Zekun Xu Yu Wang Guihou Sun Yuehong Chen Qiang Ma Xiaoxiang Zhang Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning ISPRS International Journal of Geo-Information gross domestic product (GDP) gridded GDP ensemble learning geographically weighted regression China |
| title | Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning |
| title_full | Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning |
| title_fullStr | Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning |
| title_full_unstemmed | Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning |
| title_short | Generating Gridded Gross Domestic Product Data for China Using Geographically Weighted Ensemble Learning |
| title_sort | generating gridded gross domestic product data for china using geographically weighted ensemble learning |
| topic | gross domestic product (GDP) gridded GDP ensemble learning geographically weighted regression China |
| url | https://www.mdpi.com/2220-9964/12/3/123 |
| work_keys_str_mv | AT zekunxu generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning AT yuwang generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning AT guihousun generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning AT yuehongchen generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning AT qiangma generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning AT xiaoxiangzhang generatinggriddedgrossdomesticproductdataforchinausinggeographicallyweightedensemblelearning |