Understanding Accessibility of Health and Fitness with Big Data Techniques: Facility Visualization in Shanghai with Multi-Source Data

Abstract

Accessibility of health and fitness services is important for the rehabilitation in the post-epidemic era. The emergence of multi-source data offers new and timely approaches to understanding health and fitness facilities (HFFs) in cities. Taking Shanghai as an example, this paper obtained the HFFs dataset through platforms such as GaoDe, and crawled data were accurately classified into Street level, District level and Municipal level. Based on spatial data mining, the spatial differentiation of current HFFs was understand. The analysis through Kernel Density Estimation (KDE) and the Multi-mode Two-step Floating Catchment Area Method (M2SFCA) indicates that HFFs basically meet the needs of most residents in the city within their service scope. The Lorentz curve results shows that top 20% community residents enjoy up to 73.66% of HFFs. This shows that there is a large difference in HFF accessibility among community residents. This article presents a visualization example of spatial computing in understanding health accessibility and can provide a timely reference for policy making.