China’s Wealth Capital Stock Mapping via Machine Learning Methods

The frequent occurrence of extreme weather and the development urbanization have led to continuously worsening climate-related disaster losses. Socioeconomic exposure is crucial in risk assessment. Social assets at mainly include buildings, machinery equipment, infrastructure. In this study, wealth capital stock (WKS) was selected as an indicator for measuring social wealth. However, existing WKS estimates not been gridded accurately, thereby limiting further Hence, multisource remote sensing POI data were used disaggregate 2012 prefecture-level into 1000 m × grids. Subsequently, ensemble models built via stacking method. performance verified by evaluating comparing three base with model. model attained more robust prediction results (RMSE = 0.34, R2 0.9025), its spatially presented a realistic asset distribution. produced research offer reasonable accurate socioeconomic map compared ones, providing important bibliography This study may also be adopted learning refining spatialization data.