摘要
以数据挖掘为工具,结合县域经济发展理论,探究湖南省县域经济发展聚集态势.构建了湖南省县域经济发展评价指标体系,对数据集应用基于原型的二分K均值聚类、基于图的AGNES聚类、加权主成分距离聚类与基于密度的DBSCAN聚类,根据邻近度矩阵的相关度选择最优聚类方法 .结果显示,湖南省县域经济发展按特征可归为四类,各类县市特征显著,类别间发展差距较大.根据模型的理论结果对现有的湖南省政府乡村振兴三大发展梯队的划分进行调整,为地方政府实施乡村振兴战略提供一定的理论基础,具有较好的参考价值.
Taking data mining as a tool while combining the theory of county economy,this paper dedicates to detecting the potential patterns of county-level economic development in Hunan province.After the assessment index system of county economic development in Hunan province was constructed,four clustering algorithms of different type are applied to detect the development patterns of counties in Hunan province,the optimal method is selected by the principle of maximum correlation coefficient of the proximity matrix.The result shows that all county villages in Hunan province can be divided into four types according to their economic growth patterns.Each division has a distinctive feature,and development disparities exist between them.Adjustments are made to develop the three echelons partition proposed by Hunan Provincial Government,and may help local governments to carry out rural revitalization strategies.
作者
丁玲
屈晶晶
王文强
DING Ling;QU Jing-jing;WANG Wen-qiang(Hunan Key Laboratory for Computation and Simulation in Science and Engineering,Xiangtan University,Xiangtan 411105 China)
出处
《湘潭大学学报（自然科学版）》
CAS
2019年第1期110-120,共11页
Journal of Xiangtan University(Natural Science Edition)
基金
国家自然科学基金项目(11571373
11671343)
湖南省教育厅重点项目(18A049)
关键词
聚类算法
乡村分类
乡村振兴战略
clustering algorithm
county classification
rural revitalization strategy
作者简介
通信作者:王文强(1971—),男,湖南邵阳人,教授.E-mail:wwq@xtu.edu.cn
