摘要
区域划分是依据人口和社会经济指标将行政统计单元或其他地理实体划分成若干个不同水平或类别的集合。由于大多数的人口和社会经济指标来源于面状数据-行政统计单元,常用的区域划分的空间聚类方法是基于面状数据的,本文通过分析现有面状数据的聚类算法特点和不足,进而提出一种新的算法,该方法提出将面状统计单元进行网格划分,引入基于网格密度聚类算法的思想,克服现有面状聚类的诸多缺点,打破行政区划的限制,更好地发现潜在信息。
Spatial clustering is a process of grouping a set of spatial objects into clusters so that objects within a cluster have high similarity to one another,,and are nearby in spatial location.Popula- tion and economic data are always gained from administration cells, which is presented as area data in GIS,so regional division can be seen as one kind of spatial clustering based on area data.In order to overcome disadvantages of administration cells,this paper applies the method of grid information system,which transforms spatial data from administration cells to grid cells,and then distributes popula- tion and economic data on the grid cells,which are used as basic u- nits to store and analyze data,and have better accuracy.Then a new Grid-based clustering algorithm based on grid information system is introduced in order to overcome difficulties resulting from subjectivity of density threshold.
出处
《测绘科学》
CSCD
北大核心
2007年第z1期66-69,共4页
Science of Surveying and Mapping
关键词
空间聚类
区域划分
网格聚类
Spatial Clustering
Regional Division
Gridbased Clustering
