Reimagining heritage villages’ sustainability: machine learning-driven human settlement suitability in Hunan

Traditional villages play a crucial role in China’s cultural, ecological, and social fabric, as they preserve indigenous knowledge and foster sustainable practices. However, rapid urbanization poses significant threats to their population stability and cultural integrity, creating challenges for both preservation and development. While existing studies often concentrate on architectural restoration or tourism, they tend to overlook a comprehensive approach to human settlement suitability (HSS). In this study, we assess the HSS of 688 traditional villages in Hunan Province by employing the entropy weight method, principal component analysis (PCA), and K-means clustering. To investigate the factors influencing HSS, we utilize Geodetector and Random Forest models. Our results indicate that 39.1% of the villages fall into the Critical Suitability category, while 29.8% belong to the General Suitability category, with overall suitability decreasing from the south to the northwest. Based on their characteristics, the villages are categorized into three types: environmentally stressed, moderately developed, and economically advantaged. Among the various influencing factors, economic variables—particularly Per Capita Disposable Income—emerge as the primary drivers of HSS variation. This research proposes an innovative framework for evaluating rural sustainability and offers strategic insights for transforming traditional villages in a sustainable manner.