Machine-Based Effects of Data-Driven Governance on Urban Environmental Resilience

To unleash the role of Data-Driven Governance (DDG) in urban environmental resilience, this study addresses the effects of DDG, a new technology model, on urban environmental resilience. Employing a CatBoost-SHAP machine learning model, we investigate the following questions: (1) can DDG truly enhance urban environmental resilience? (2) Through what mechanisms might such improvement occur? (3) And what theoretical foundations support these mechanisms? The findings reveal that the impact of DDG on urban environmental resilience is multidimensional and positive. It significantly enhances urban environmental resilience primarily through the core mediating mechanisms of industrial upgrading and market integration. Furthermore, a hybrid institutional environment where market-driven forces and policy guidance collaborate can best unlock governance dividends. This paper contributes to a novel integrated framework by theoretically bridging the gap between DDG and urban environmental resilience. Methodologically, we advance non-linear mechanism detection via the CatBoost-SHAP. Additionally, this study offers context-specific strategies for diverse institutional settings.