无序样品聚类分析——基于第四组距离法和质心法  

Cluster analysis of disordered samples:based on the methods of the fourth set of distance and the centroid method

作　　者：胡纯严 胡良平 Hu Chunyan;Hu Liangping(Graduate School,Academy of Military Sciences PLA China,Beijing 100850,China;Specialty Committee of Clinical Scientific Research Statistics of World Federation of Chinese Medicine Societies,Beijing 100029,China)

机构地区：[1]军事科学院研究生院,北京100850 [2]世界中医药学会联合会临床科研统计学专业委员会,北京100029

出　　处：《四川精神卫生》2024年第S1期90-94,共5页Sichuan Mental Health

摘　　要：本文目的是介绍与无序样品聚类分析有关的基本概念、计算方法、两个实例以及使用SAS实现计算的方法。基本概念包括不对称名义变量、不对称比率变量、Dice系数和Kulcynski 1系数;计算方法涉及度量两个样品之间距离大小的第四组距离法(共8个)和度量两类样品之间距离大小的质心法(centroid法);两个实例分别为“九个离婚理由是否适用于美国各州”以及“缺血性中风病患者首发证候情况”;借助SAS软件,对第一个实例的数据进行了无序样品聚类分析,并对SAS输出结果做出了解释。The purpose of the paper was to introduce the basic concepts,calculation methods,two examples and the implementation of SAS calculation methods related to the cluster analysis of disordered samples.The basic concepts included asymmetric nominal variables,asymmetric ratio variables,Dice coefficient and Kulcynski 1 coefficient.The calculation methods involved measuring the distance between two samples using the fourth set of distance methods(8 in total)and measuring the distance between two classes of samples using the centroid method.The data in the two examples were"whether the nine divorce reasons apply to each state in the United States"and"the first symptom situation of ischemic stroke patients".Using SAS software,Cluster analysis of disordered samples was performed on the data in the first instance,and the SAS output results were explained.

关 键 词：不对称名义变量 无序样品 Dice系数 谱系聚类分析 树形图 

分 类 号：R195.1[医药卫生—卫生统计学]