Against the backdrop of widening intra-urban disparities in well-being, a core challenge lies in identifying and regulating environment-related well-being indicators across multiple urban settings in ways that respond to heterogeneous resident needs and enhance perceived fairness. This study innovatively incorporates inequality in subjective well-being (SWB) as a feedback mechanism into the regulation framework of objective urban well-being (OUW), and empirically examines this approach across multiple urban settings—namely the citywide, urban core, and suburban areas of Beijing. Community-level SWB indicators were derived from residents’ expressions on social media, while spatial inequality was quantified using local Gini coefficients. An interpretable machine learning model (XGBoost + SHAP) was then employed to identify key environmental determinants and their threshold effects. The results reveal that: (1) SWB inequality is higher in suburban areas, while the urban core faces a greater risk of extreme inequality. (2) Inequality “hotspots” are concentrated along the urban periphery and urban–rural transition zones, while the historic core and new residential areas form relatively balanced “coldspots.” (3) Experience-oriented factors dominate SWB inequality in the urban core (e.g. green environments), whereas opportunity-oriented factors play a stronger role in suburban areas (e.g. daily services), indicating systematically distinct inequality-generating mechanisms. (4) The effects of the objective environment on SWB inequality are characterised by pronounced non-linearity and context dependence. Based on these findings, this study advocates a perception-oriented governance pathway for well-being equity and formulates differentiated strategies for different urban areas, providing data-driven and human-centred guidance for building happier and more equitable cities. • Measure intra-urban SWB inequality using social media data and local Gini indices. • Integrate SWB inequality into objective well-being models for equity evaluation. • Identify threshold ranges and marginal effects of key objective well-being factors. • Differentiate core–suburban models to reveal spatial heterogeneity of determinants. • Classify six community types representing distinct inequality patterns and drivers.
Zhu et al. (Fri,) studied this question.