Abstract: As China enters the era of mass tourism, tourism has emerged as a pivotal force in driving rural revitalization. Traditional villages boast abundant resources for the development of rural tourism, yet these resources remain inadequately explored and rationally utilized. This study takes South Jiangxi as a case in point, meticulously analyzing the evaluation big data from five key traditional villages in this region on pertinent tourism websites, complemented by field research. The objective is to establish a feedback mechanism from tourism evaluations to the advancement of tourism construction in traditional villages, thereby further tapping into the tourism development potential of these villages on the premise of preservation. Firstly, based on evaluation time data, this paper dissects the travel preferences of people on a monthly and annual basis. Secondly, drawing upon the types of evaluation content data, it primarily analyzes text data through lexical frequency, semantic networks, and emotional attitudes. This analysis is integrated with settlement spatial identification, organization, and construction orientation to forge an evaluation feedback pathway. Furthermore, by combining the photographic data from the evaluation content with field research, it delves into the perceived imagery of the tourism space in traditional villages. Lastly, grounded in evaluation analysis and feedback, with spatial optimization as the thread, the study endeavors to propose enhancement suggestions for the construction of traditional villages as tourist destinations, encompassing overall strategy, route planning, and node imagery.