Resilience Security of Resource-Based Cities in China Based on a Machine Learning Approach: Driving Factors and Regional Heterogeneity

Abstract Enhancing the urban resilience security (URS) of resource-based cities (RBCs) is critical to overcoming the resource curse and responding to multiple external shocks. Nevertheless, existing studies still fall short in capturing the complexity and spatial heterogeneity of URS driving mechanisms. This study uses the Critic-Topsis method (a multi-attribute decision analysis approach) to quantitatively assess the URS of 115 RBCs in China from 2011 to 2022, combined with machine learning algorithms to identify the driving factors and their regional differences. The results show that the overall URS exhibits a W-shaped evolution trend, with a spatial distribution characteristic of stronger in the east, weaker in the west; higher in the central, lower in the south. Economic resilience and infrastructure resilience have a universal impact across RBCs. In the eastern and central regions, URS is driven by economic development and transportation, while in the western and northeastern regions, the focus is on ecological transformation. This study provides a scientific basis for the construction of URS in RBCs and the formulation of differentiated policies.