基于多源数据与机器学习的乡村旅游竞争力评价研究——以杭州市临安区为例

Evaluation of rural tourism competitiveness based on multi-source data and machine learning:A case study of Lin’an District in Hangzhou,China

旅游竞争力评价对推动乡村旅游产业可持续发展具有重要意义。在数字时代背景下,多源数据和机器学习方法能够从地理空间角度高效揭示相关要素特征,为乡村旅游竞争力的科学评价提供新视角与新手段。论文以杭州市临安区为例,基于多源遥感与互联网数据,构建乡村旅游竞争力评价指标体系,对比分析逻辑回归、支持向量机、随机森林、极限梯度提升树共4种机器学习模型在乡村旅游竞争力评价上的精度,并选取最优模型以揭示临安区村域尺度的乡村旅游竞争力格局。结果表明:①运用随机森林模型进行乡村旅游竞争力评价,其预测精度优于其他3种机器学习模型。②旅游资源、服务设施、交通可达性、政策条件是影响乡村旅游竞争力评价的主要指标。③高竞争力类村庄以条带状分布在临安区北部、西部地区,发展条件优越;中竞争力类以团块状分布在临安东部、中西部地区,在旅游资源品质、服务设施等方面存在欠缺;低竞争力村庄以斑块状分布在临安区中西部地区,生态环境优越、土地禀赋较好,但缺乏资源开发与政策支持。研究可为推动乡村旅游的可持续发展提供政策参考与方法借鉴。

Evaluating tourism competitiveness is important for ensuring the sustainable development of rural tourism.In the digital information era,multi-source data and machine learning methods can efficiently reveal the characteristics of relevant elements from a geospatial perspective,providing a new method for scientific evaluation of rural tourism competitiveness.Based on multi-source remote sensing and Internet data at the village level from 2019 to 2022,this study identified the rural tourism competitiveness in Lin'an District of Hangzhou City using four machine learning models,including logistic regression(LR),support vector machine(SVM),random forest(RF),and extreme gradient boosting tree(XGB),and the optimal model was selected to reveal the spatial pattern of competitiveness and analyze the critical indicators of identification.The results show that:1)The accuracy of the rural tourism competitiveness evaluation using the random forest(RF)model is better than the other three machine learning models.2)Tourism resources,service facilities,accessibility,and policy conditions are the main factors affecting the rural tourism competitiveness.3)Villages in the high tourism competitiveness category are distributed in strips in the northern and western areas of Lin'an District,with superior development conditions.The medium competitiveness villages are distributed in clumps in the eastern and central-western areas of the district,which have lower quality of tourism resources and service facilities.Low-competitiveness villages are distributed in patches in the central and western areas of the district,with superior ecological environment and land endowment,but lacking resource development and policy support.The study results may provide some policy references and technical supports for promoting the sustainable development of rural tourism.