广义回归神经网络在空间数据聚类中的应用  

Application of generalized regression neural network in spatial data clustering

作　　者：卢建青[1] 陈银珠[1] 刘玉珠[1] 张锦 LU Jianqin;CHEN Yinzhu;LIU Yuzhu;ZHANG Jin(The Institution of Geological Surveying and Mapping,Hangzhou 310030,China;School of Geography and Information Engineering,China University of Geosciences,Wuhan 430074,China)

机构地区：[1]浙江省测绘大队,杭州310030 [2]中国地质大学地理与信息工程学院,武汉430074

出　　处：《导航定位学报》2020年第2期31-35,共5页Journal of Navigation and Positioning

摘　　要：针对空间数据聚类中由于空间数据本身的特点造成模糊C均值聚类算法无法满足使用要求的问题,提出1种改进的空间数据聚类算法:将模糊C均值聚类算法与广义回归神经网络相结合,得到结合广义神经网络的模糊C均值聚类算法;并将结合广义神经网络的模糊C均值聚类算法应用到空间数据的聚类中。实验结果表明,结合广义神经网络的模糊C均值算法在空间聚类方面比模糊C均值有着更好的效果,可以满足实际空间数据聚类的要求。Aiming at the problem that it is difficult to meet the requirement of spatial data clustering for the fuzzy C-means clustering algorithm due to the characteristics of spatial data itself,the paper proposed an improved spatial data clustering algorithm:the fuzzy C-means clustering algorithm was combined with the generalized regression neural network to get the generalized regression neural network-fuzzy C-means(GRNN-FCM)clustering algorithm,and GRNN-FCM was applied in spatial data clustering.Experiments showed that the proposed method would have better effect than the fuzzy C-means algorithm in spatial clustering,which could meet the requirements of actual spatial data clustering.

关 键 词：空间数据 空间聚类 模糊C均值聚类算法 结合广义神经网络的模糊C均值聚类算法 聚类效果 

分 类 号：P228[天文地球—大地测量学与测量工程]