A large-scale village classification model for tailored rural revitalization: A case study of Hubei province, China

A comprehensive understanding of village development patterns and the identification of different village types is crucial for formulating tailored planning for rural revitalization. However, a model for large-scale village classification to support tailored rural revitalization planning is still lacking. This study aims to develop a large-scale village classification model using the Gaussian Mixture Models to support tailored rural revitalization efforts. Firstly, we propose a multi-dimensional index system to capture the diverse features of massive villages. Secondly, the GMM clustering algorithm is applied to identify distinct village types based on their unique features. The model was employed to classify the 25,409 villages in Hubei province in China into four classes. Villages in these classes exhibit discernible differences in spatial distribution, topography, location, economic development level, industrial structure, infrastructure, and resource endowment. In addition, the GMM-based village classification model demonstrates a high level of agreement with evaluations made by planning experts, confirming its accuracy and reliability. In the empirical study, our model achieves an overall accuracy of 95.29%, signifying substantial concordance between the classifications made by planning experts and the results generated by our model. Based on the identified features, tailored paths are proposed for each village class for rural revitalization efforts.