Nonlinear and Threshold Effects of the Green Environment, Built Environment, and Socioeconomic Data on the Urban Vitality Index within the Fourth Ring Road in Beijing

This study employs Shapley Additive Explanations and an interpretable XGBoost model to quantify the influence of various factors on the urban vitality index (UVI), revealing the spatial distribution patterns and underlying mechanisms of key drivers. By integrating multisource data—including the green environment, built environment, and socioeconomic indicators—the study not only achieves accurate UVI prediction but also visualizes both global and local variable impacts, providing deeper insights into complex urban systems. The results show significant spatial clustering and heterogeneity of the UVI within Beijing’s Fourth Ring Road. Average housing prices, park density, and the normalized difference vegetation index are positively associated with the UVI, while building density and population density exhibit negative correlations. Additionally, road network density and functional diversity positively contribute to the UVI, highlighting the importance of transportation accessibility and multifunctional spatial structures. The introduction of kernel-weighted spatial features further enhances the model’s capacity to explain spatial heterogeneity. By quantifying the relative contributions and nonlinear interactions of influencing factors, this study offers robust analytical tools and scientific support for urban planning and resource allocation, ultimately contributing to the optimization of urban spatial layout and the promotion of sustainable urban vitality.