The urban shrinkage phenomenon is increasingly common. However, at the same time, it is accompanied by local growth within the region, and its complex impact mechanism still lacks in-depth research. This paper takes Daqing, a resource-based city, as an example and constructs a multi-dimensional urban growth/shrinkage type definition matrix. The random forest classification model is used to quantitatively analyze the relevant factors affecting the classification matrix by comparing different models. The results show that: (1) The growth/shrinkage of Daqing City leads to a concentric structure of intensive growth—expansion growth—expansion shrinkage—intensive shrinkage that gradually changes from the center to the periphery in space. (2) The random forest classification model can better explain the cause mechanism of urban growth/shrinkage. (3) The growth/shrinkage of resource-based cities in the transformation stage is mainly affected by economic factors, such as traffic accessibility and urban morphology related to the resource economy. This study provides an evaluation framework from qualitative to quantitative, providing useful references for research and planning policy formulation in related fields.