| 基于空间聚类分析的福建省各县市经济发展水平研究 |
| |
| 引用本文： | 陈小瑜,余明.基于空间聚类分析的福建省各县市经济发展水平研究[J].热带地理,2007,27(4):343-347. |
| |
| 作者姓名： | 陈小瑜 余明 |
| |
| 作者单位： | 福建师范大学,地理科学学院,福州,350007 |
| |
| 摘 要： | 将空间坐标和属性特征有机地结合起来,给出空间距离的定义,提出位置与属性一体化的空间聚类方法——基于空间距离的K-均值聚类法,利用M atlab编写其算法程序。同时以福建省各县市经济发展水平的类型划分作为研究实例,首先选取反映社会经济发展水平的8项指标,并利用G IS技术提取各县市的重心位置坐标,再利用该空间聚类算法将福建省各县市的经济发展水平聚类为5类,结果表明此算法能够综合考虑空间位置邻近性和属性特征相似性的聚类要求,对福建区域可持续发展有一定的指导意义。
|
| 关 键 词： | 空间距离 空间聚类 福建 经济发展水平 |
| 文章编号： | 1001-5221（2007）04-0343-05 |
| 修稿时间： | 2006-10-082007-01-04 |
A Study on Economic Development Level of the Counties and Cities in Fujian Based on Spatial Clustering Analysis |
| |
| GEN Xao-yu,YU Ming.A Study on Economic Development Level of the Counties and Cities in Fujian Based on Spatial Clustering Analysis[J].Tropical Geography,2007,27(4):343-347. |
| |
| Authors: | GEN Xao-yu YU Ming |
| |
| Institution: | College of Geographical Science, Fujian Normal University, Fuzhou 350007, China; 2. School of Resources and Environmental Science, Wuhan University , Wuhan 430079, China |
| |
| Abstract: | Currently,there are two deflections in applying the method of spatial clustering analysis,one is that those engaged in a study of GIS theories and technical tools make clustering according to the location of the spatial object,and take only the positional neighborhood into account,while ignoring the attributive similarity between the objects;the other is that those engaged in application of GIS and geographical research apply mechanically the traditional clustering methods,namely they simply consider the attributive similarity but ignore the positional neighborhood.In fact,the spatial object essentially has double signification with both positional and attributive characteristics.In the paper,three types of spatial distance measurement are defined,which combine positional and attributive characteristics.K-mean clustering algorithm based on spatial distance is programmed by Matlab.With the method,economic development level of the cities and counties in Fujian is taken as a case to carry out clustering analysis and classification.Eight indexes are chosen to reflect their economic development level,and the central position coordinates of all cities and counties are picked up by using technique of GIS,then the cities and counties in Fujian are divided into 5 types according to economic development level.The results show that the spatial clustering with both positional neighborhood and attributive similarity can reflect the fact with better objectivity. |
| |
| Keywords: | Spatial distance Spatial clustering analysis Fujian Economic development level |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
