Spatial heterogeneity of built environment's impact on urban vitality using multi-source big data and MGWR.

This study proposes a novel interpretative framework combining multi-source big data with Multiscale Geographically Weighted Regression (MGWR) to examine spatial variations in built environment-vitality relationships. Using multi-source data including LBS big data, Weibo check-in data, Dianping data, Baidu heat map data, POI, Street view image, and so on, we analyzed how four dimensions of environment influence urban vitality across different urban contexts. The analysis reveals significant spatial heterogeneity in built environment-vitality relationships, with varying effects across urban locations. Road density is generally associated with overall urban vitality. Diversity stands out as a prerequisite for stimulating urban vitality. Most indicators have specific conditions for enhancing vitality: life-service facilities should focus on promoting their diversity; population density and intensive urban development do not necessarily lead to increased vitality; the impact of long-distance transportation facilities is more prominent than that of conventional transportation; areas with a natural environment and high sidewalk coverage exhibit higher vitality. The findings suggest the need for context-sensitive approaches to urban design and planning interventions.