| 摘要 |
Under the backdrop of rural revitalization, rural landscapes serve as the core attraction of rural tourism, with their quality directly influencing visitor experiences. However, current rural landscape planning often adheres to the traditional "top-down" material zoning model, and neglects the perceptual experiences from the visitor's perspective, leading to a misalignment between landscape offerings and public emotional value demands. To address this core issue, Yanshou Town in Changping District, Beijing is taken as the research object in this paper, and a classification and reconstruction method for rural landscape features from a visitor-oriented perspective is explored. By scraping relevant travelogue data from RedNote, the study constructs an unsupervised machine learning framework based on the LDA topic model, conducting in-depth clustering of the semantic space perceived by visitors. The rural landscapes of Yanshou Town from the perspective of tourists are categorized into five types: mountainous agricultural and forestry landscapes, historical and cultural landscapes, seasonal floral landscapes, artistic experiential landscapes, and linear recreational landscapes. Each category is analyzed individually, and corresponding improvement suggestions are provided, offering a reference pathway for the precise governance and differentiated enhancement of rural landscape spaces in Yanshou Town. |
