Mapping Population Distribution Based on XGBoost Using Multisource Data

Mapping fine-scale distribution of the population is essential to the study of human activities, where more reliable open-access big data could be excavated with the help of machine learning models. However, the combination of multisource datasets and multidimensional features for population estimat...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xin Zhao, Nan Xia, Yunyun Xu, Xuefeng Huang, Manchun Li
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	CatBoost gradient boosted decision tree (GBDT) light gradient boosting machine (lightGBM) multisource data population mapping random forest (RF) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/8d4589b2cbf74680a2742075702c11bb
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:8d4589b2cbf74680a2742075702c11bb
record_format	dspace
spelling	oai:doaj.org-article:8d4589b2cbf74680a2742075702c11bb2021-11-24T00:00:19ZMapping Population Distribution Based on XGBoost Using Multisource Data2151-153510.1109/JSTARS.2021.3125197https://doaj.org/article/8d4589b2cbf74680a2742075702c11bb2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9601310/https://doaj.org/toc/2151-1535Mapping fine-scale distribution of the population is essential to the study of human activities, where more reliable open-access big data could be excavated with the help of machine learning models. However, the combination of multisource datasets and multidimensional features for population estimation was still unclear, and different models also needed comparison. Thus, in this study, related features from multisource data were first extracted, including building data, geographic big data, remote sensing data, and basic geographic data. Then, the effective indicators with higher contribution weight were selected from multisource data, which can reduce the noise and unstable model fitting. Finally, the population distribution map for 100-m grid was obtained in Shenzhen in 2019, and estimation results for five tree-based ensemble learning models were also compared at community scale, including random forest (RF), gradient boosted decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost). Our results showed that: 1) building data and geographic big data could better reflect the spatial heterogeneity of the population; 2) indicators selection could effectively improve the estimation accuracy of the population mapping; and 3) compared with other models, XGBoost had the largest <italic>R</italic><sup>2</sup> (80%), the smallest RMSE, and MAE, the most percentage of accurate estimation communities (−0.3<RE<0.3, 65%), and a shorter train time. Therefore, XGBoost was chosen for mapping population distribution instead of GBDT, LightGBM, CatBoost, and RF. Our proposed method for population mapping can help to optimize the allocation of resources and guide a more scientific path for urban development.Xin ZhaoNan XiaYunyun XuXuefeng HuangManchun LiIEEEarticleCatBoostgradient boosted decision tree (GBDT)light gradient boosting machine (lightGBM)multisource datapopulation mappingrandom forest (RF)Ocean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11567-11580 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	CatBoost gradient boosted decision tree (GBDT) light gradient boosting machine (lightGBM) multisource data population mapping random forest (RF) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	CatBoost gradient boosted decision tree (GBDT) light gradient boosting machine (lightGBM) multisource data population mapping random forest (RF) Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Xin Zhao Nan Xia Yunyun Xu Xuefeng Huang Manchun Li Mapping Population Distribution Based on XGBoost Using Multisource Data
description	Mapping fine-scale distribution of the population is essential to the study of human activities, where more reliable open-access big data could be excavated with the help of machine learning models. However, the combination of multisource datasets and multidimensional features for population estimation was still unclear, and different models also needed comparison. Thus, in this study, related features from multisource data were first extracted, including building data, geographic big data, remote sensing data, and basic geographic data. Then, the effective indicators with higher contribution weight were selected from multisource data, which can reduce the noise and unstable model fitting. Finally, the population distribution map for 100-m grid was obtained in Shenzhen in 2019, and estimation results for five tree-based ensemble learning models were also compared at community scale, including random forest (RF), gradient boosted decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost). Our results showed that: 1) building data and geographic big data could better reflect the spatial heterogeneity of the population; 2) indicators selection could effectively improve the estimation accuracy of the population mapping; and 3) compared with other models, XGBoost had the largest <italic>R</italic><sup>2</sup> (80%), the smallest RMSE, and MAE, the most percentage of accurate estimation communities (−0.3<RE<0.3, 65%), and a shorter train time. Therefore, XGBoost was chosen for mapping population distribution instead of GBDT, LightGBM, CatBoost, and RF. Our proposed method for population mapping can help to optimize the allocation of resources and guide a more scientific path for urban development.
format	article
author	Xin Zhao Nan Xia Yunyun Xu Xuefeng Huang Manchun Li
author_facet	Xin Zhao Nan Xia Yunyun Xu Xuefeng Huang Manchun Li
author_sort	Xin Zhao
title	Mapping Population Distribution Based on XGBoost Using Multisource Data
title_short	Mapping Population Distribution Based on XGBoost Using Multisource Data
title_full	Mapping Population Distribution Based on XGBoost Using Multisource Data
title_fullStr	Mapping Population Distribution Based on XGBoost Using Multisource Data
title_full_unstemmed	Mapping Population Distribution Based on XGBoost Using Multisource Data
title_sort	mapping population distribution based on xgboost using multisource data
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/8d4589b2cbf74680a2742075702c11bb
work_keys_str_mv	AT xinzhao mappingpopulationdistributionbasedonxgboostusingmultisourcedata AT nanxia mappingpopulationdistributionbasedonxgboostusingmultisourcedata AT yunyunxu mappingpopulationdistributionbasedonxgboostusingmultisourcedata AT xuefenghuang mappingpopulationdistributionbasedonxgboostusingmultisourcedata AT manchunli mappingpopulationdistributionbasedonxgboostusingmultisourcedata
_version_	1718416107718574080

Mapping Population Distribution Based on XGBoost Using Multisource Data

Ejemplares similares