首页 | 本学科首页   官方微博 | 高级检索  
     

对似近分析的一点改进及其拓广
引用本文:王晓佳,谭远德. 对似近分析的一点改进及其拓广[J]. 生物数学学报, 1994, 0(4)
作者姓名:王晓佳  谭远德
作者单位:西南农业大学园艺系!重庆,630716(王晓佳),广西师范大学生物系!桂林,541004(谭远德)
摘    要:本文给出了似近分析一个重要的改进,并将改进的似近分析推广为m维似近分析,同时给出了两个数学定理及其证明。根据多维似近系数的数学性质,提出了一种新的聚类方法,即逐步多维聚类法。实例分析证明,这种聚类法比二维系数聚类优越,聚类功能强,所获得的结果更为客观。

关 键 词:似近系数  相似(相关)系数  聚类

AN IMPROVEMENT ON RESEMBLANCE-NEAR ANALYSIS AND ITS GENERALIZATION
Wang Xiaojia,Tan Yuande. AN IMPROVEMENT ON RESEMBLANCE-NEAR ANALYSIS AND ITS GENERALIZATION[J]. Journal of Biomathematics, 1994, 0(4)
Authors:Wang Xiaojia  Tan Yuande
Abstract:An improyement on resemblance=near analysis was made and then the improved resemblance-near analysis was generalized to become a resemblance-near analysis of m dimensions,at the same time,two theorems in the resemblance-near analysis were drawn from this improvement,and proofel in mathematics in this article.On the basis of the nature of resemblance-near coefficient of multiple dimensions,a new cluster method---the cluster used by parameters,(A) of step b/step muliple dimensions was put foreward.Example analysis showed that this cluster method has an advance over the one used by parameter,l(2),of two dimensions in cluster function and objective degree of cluster tree.
Keywords:Resemblance-near coefficient  Resemblane (correlaion) cofficient  Cluster
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号