组态视角下我国旅游产业发展的类型与路径选择——基于机器学习方法的探索  

The Type and Path Selection of Tourism Industry Development from the Perspective of Configuration:An Exploration Based on Machine Learning Methods

作　　者：廖杨月 余传鹏[1] 林春培[2,3] LIAO Yangyue;YU Chuanpeng;LIN Chunpei(Department of Tourism Management,South China University of Technology,Guangzhou 510006,China;School of Business Administration,Huaqiao University,Quanzhou 362021,China;Business Management Research Center,Huaqiao University,Quanzhou 362021,China)

机构地区：[1]华南理工大学旅游管理系,广东广州510006 [2]华侨大学工商管理学院,福建泉州362021 [3]华侨大学商务管理研究中心,福建泉州362021

出　　处：《旅游学刊》2024年第9期31-46,共16页Tourism Tribune

基　　金：国家社会科学基金重大项目“中国式现代化进程中文化和旅游深度融合发展研究”(23ZDA091);广东省自然科学基金青年提升项目“数字化认知变革驱动制造企业数字化转型的过程机制研究”(2024A1515030110);广东省哲学社会科学“十四五”规划学科共建项目“全面推进城市数字化转型的机制与对策研究”(GD22XGL01)共同资助。

摘　　要：运用机器学习方法识别旅游产业发展的复杂前因和组态路径,以此赋能我国区域协调发展和共同富裕目标实现。文章基于生产函数理论,以我国298个地级及以上城市为研究对象,采用K均值聚类算法将样本城市划分为发展受阻型、稳中求进型和全面辐射型3种群组类型,运用分类与回归树算法挖掘不同类型城市资源、技术和制度层面多维特征变量与旅游产业发展之间的复杂关系结构。研究发现:1)旅游产业发展的驱动要素具有耦合协调效应,体现为不同类型城市多维特征变量的横向耦合一致性和纵向等级分层性;2)高度相似城市因要素差异化配置获得不同旅游产业发展水平,表明每类城市都有适宜自身发展的组态条件,为推动区域协调发展提供现实基础;3)不同类型城市旅游产业高水平发展的驱动要素具有组合差异性,整体呈现殊途同归的作用效果,发展受阻型城市由“科技筑基-区域开放-文化吸引”驱动,稳中求进型城市由“经济引领-科技创新-数字赋能”驱动,全面辐射型城市由“文化吸引-交通增质”驱动。研究结论为我国城市旅游产业如何依据自身要素禀赋条件获得高水平发展提供了新思路和新参考依据。At present,the contradiction of inadequate and unbalanced regional development is still prominent.Developing a regional economy in a new way is related to the well-being of the people;it is also a major theoretical and practical problem to be solved urgently in our economic and social development in the new era.The 20th National Congress of the Communist Party of China once again wrote the goal of common prosperity into the report and continued to promote the coordinated development of regions to better meet the broad masses of people’s yearning for a better life.As a strategic pillar industry of our country,the tourism industry is an important driving force that promotes high-quality regional economic development and optimization and upgrading of industrial structure.Moreover,it will also promote the shared prosperity of our country.Combined with production function theory and a literature review,eight feature variables are selected from resource,technology and institution dimensions based on the reviewed research to examine the tourism industry’s personalized and differentiated driving path in different types of cities.Next,we obtain the field index data of 298 cities at the prefecture level or above and filter them.The entropy weight method is used to measure variables.After the corresponding feature variables are standardized by min-max normalization,the Kmeans algorithm is used to classify cities with similar characteristics into clusters(i.e.,hindered,stable and fully radiant).The feature difference radar map is drawn according to the mean value of the overall features of different city clusters,which are named to identify the overall heterogeneity of different types of cities.Lastly,taking the tourism industry development as the decision attribute and eight feature variables as the conditional attribute,the classification and regression tree algorithm is used to explore the potential decision rules of the tourism industry development in three types of city clusters.The results show that 1)the driving f

关 键 词：旅游产业发展 多维特征变量 组态视角 路径选择 机器学习 

分 类 号：F59[经济管理—旅游管理]