山地康养旅游空间适宜性评估——基于随机森林算法的云南实证  

Spatial Suitability Assessment on Health-Wellness Tourism in Mountainous Regions:An Empirical Study in Yunnan Province of China Based on Random Forest Algorithm

作　　者：杨友宝 杜诗雨 YANG Youbao;DU Shiyu(College of Tourism,Hunan Normal University,Changsha 410081,China)

机构地区：[1]湖南师范大学旅游学院,长沙410081

出　　处：《山地学报》2025年第1期46-58,共13页Mountain Research

基　　金：国家自然科学基金(42201224);湖南省教育厅重点项目(22A0047)。

摘　　要：中国山地康养旅游普遍存在空间配置受限、资源利用低效等问题,科学识别与解析山地康养旅游空间适配格局及其驱动机制,是实现山地康养旅游高质量发展的基础。既有研究受制于山区可达性差而导致数据采集困难,同时基于传统统计模型的空间适宜性评估存在结果精度不足与结论可靠性问题。本研究通过整合地理空间数据、网络平台数据与社会经济统计数据等多源信息,创新性采用随机森林算法(RF)模型,精准评估云南省山地康养旅游适宜性,并系统探讨了其影响因素。结果表明:(1)RF模型在十折交叉验证中表现出良好的模型匹配性(平均准确率84.4%,召回率74.0%,F 1-Score 77.5%,AUC 0.87),表明该模型对于云南省山地康养旅游适宜性评估结果具有较为准确的解释力。(2)云南省山地康养旅游形成“高度适宜-适度适宜-边际适宜-不适宜”四级开发格局,其中前两类占全省国土面积的85.92%,主要分布在云岭-哀牢山以西区域;边际适宜区(2.73%)主要分布在云南省三大水源区;不适宜区(10.81%)呈“斑块状”镶嵌分布。(3)云南省山地康养旅游适宜性开发格局的形成是自然本底、大气环境、资源特质和社会经济等多因素综合作用的结果,特别是归一化植被指数(NDVI)、距水域距离及气候条件等因素,对山地康养旅游适宜性具有显著影响,应作为开发规划与决策中的关键考量因素。研究成果不仅为云南省康养旅游空间优化提供量化依据,更为高原山区“防-治-养”一体化康养模式构建提供理论支撑。Health-wellness tourism in mountainous regions in China generally faces issues such as limited spatial allocation of resource with inefficient utilization.Scientific determination of a spatial suitability pattern and its driving mechanisms for health-wellness tourism in a mountainous region is fundamental to achieving high-quality development.Unfortunately,previous studies were constrained by difficulties in data collection due to poor accessibility in mountainous areas,and spatial suitability assessments based on traditional statistical models suffer from insufficient accuracy and reliability of results.In this study,it integrated multi-source information including geospatial data,network platform data,and socio-economic statistics,and innovatively adopted a model of Random Forest Algorithm(RF)to accurately assess the suitability of health-wellness tourism in mountainous regions of Yunnan Province,China,and systematically discussed its influencing factors.(1)The RF model exhibited good model fit in ten-fold cross-validation(average accuracy rate of 84.4%,recall rate of 74.0%,F 1-Score of 77.5%,and AUC of 0.87),indicating that the model had accurate explanatory power for assessing the suitability of health-wellness tourism in mountainous regions of Yunnan Province.(2)Health-wellness tourism in Yunnan formed a four-level development pattern of suitability“high-moderate-marginal-unsuitable”,of which the first two categories account for 85.92%of the province's land area and were mainly distributed in the area west of the Yunling Mountains-the Ailao Mountains;the areas in marginal suitability(2.73%)were mainly distributed in the three major water source regions of Yunnan;unsuitable areas(10.81%)were scatteredly distributed in patches.(3)The formation of spatial suitability development pattern for health-wellness tourism in mountainous regions of Yunnan is the result of a combination of multiple factors,including natural settings,atmospheric environment,resource characteristics,and socio-economic factors.In partic

关 键 词：山地康养旅游 适宜性 随机森林算法 山区经济 云南 

分 类 号：F592[经济管理—旅游管理]