Quantifing the Imbalance of Spatial Distribution of Elderly Service with Muti-source Data

Abstract

With the growing challenge of aging populations around the world, the study of the elderly service is an essential initiative to accommodate the particular needs of the disadvantaged communities and promote social equity. Previous research frameworks are very casespecific with limited evaluation indicators that cannot be extended to other scenarios and fields. Based on multi-source data and Geographic Information System (GIS), this paper quantifies and visualises the imbalance in the spatial distribution of elderly services in 218 neighbourhoods in Shijingshan District, Beijing, China. Mortality data were obtained, and the most contributing indicators to mortality were investigated by correlation analysis. Finally, mapping between other facility indicators to mortality rates was constructed using machine learning to further investigate the factors influencing the quality of elderly services at the community level. The conclusion shows that the functional density of transportation facilities, medical facilities, living services facilities, and the accessibility of elderly care facilities are most negatively correlated with mortality. The correlation conclusion is combined with a machine learning prediction model to provide future recommendations for the construction of unbalanced elderly neighbourhoods. This research offers a novel systematic method to study urban access to elderly services as well as a new perspective on improving social fairness.