摘要
基于欧氏距离的聚类方法往往会得到球形簇,直接计算球形簇的球心及半径有很大困难,提出了一种算法以逐渐逼近的方式确定这两个参数。理论分析和实验结果都证明该算法能够达到精确度要求。
The clustering method based on Euclidean distance usually leads to spherical clus- ter. It is very difficult to compute the center and radius of the spherical cluster. Proposes a algorithm to determine the two parameters with gradually approaching method. The theoretical analysis and ex- perimental results both show that the algorithm reaches the demand of precision.
出处
《江汉大学学报（自然科学版）》
2013年第5期62-64,共3页
Journal of Jianghan University：Natural Science Edition
基金
武汉市科技局基金资助项目(201250499145-21)
关键词
球形簇
算法
逼近
精确度
spherical cluster
algorithm
approach
precision
