基于信息熵的空间对象群聚类算法

摘要

针对利用空间关系建立空间对象群聚类的问题,提出一种基于信息熵的空间对象群聚类算法ESOGC.该算法考虑空间数据的复杂性和数据之间的联系,根据邻域范围内信息熵的变化情况,捡起或放下当前空间对象群,从而实现对空间对象群的聚类.实验结果表明,该算法能解决空间对象群中对象类型、对象属性值和对象数量不一致性的问题.%For the clustering of spatial object group constructed based on spatial relationship, this paper presents a clustering algorithm for spatial object group based on information entropy, named ESOGC. ESOGC is different from the other clustering algorithms, and it takes variety data types and the number of objects into full account in spatial object group. Through the change of information entropy within a same region, ants determine whether to pick up or drop the current spatial object group to realize the clustering of spatial object group. Experimental results show it can solve the problems of different data types, attribute value, and number.