Taking southern Xinjiang as the research unit, this study constructs the evaluation index system of traditional rural disaster resilience from four aspects: rural industry, ecology, organization, and space. Then, the study comprehensively uses the analytic hierarchy process, entropy method, and K-means clustering algorithm to quantitatively measure and evaluate the target villages. The results show that the traditional villages in the Turpan and Kuqa areas in the northern part of the Taklimakan Desert have relatively high resilience, and the villages in the Hotan area and Bayinguoleng Mongolian Autonomous Prefecture in the southern part of the desert have relatively poor resilience. The overall resilience level tends to be strong in the east and north directions and weak in the west and south directions. From the contribution of the rural disaster resilience score, spatial disaster resilience is the main driving factor that affects the level of village resilience. Based on the score results of the factor layer, the income of the village, the distance between the village and its water source, and the number and size of public spaces in the village are the primary factors that affect the resilience. Most villages with strong resilience have obvious scoring advantages in industry and environment.
Zhang et al. (Fri,) studied this question.