Hu Chunyan,Hu Liangping,Cluster analysis of disordered samples: based on the methods of the second set of distances and the centroid method[J].SICHUAN MENTAL HEALTH,2024,37(S1):78-83
Cluster analysis of disordered samples: based on the methods of the second set of distances and the centroid method
DOI:10.11886/scjsws20240310002
English keywords:Distance  Dissimilarity  Non metric type  Hierarchical cluster analysis  Dendrogram
Fund projects:
Author NameAffiliationPostcode
Hu Chunyan Graduate School Academy of Military Sciences PLA China Beijing 100850 China 100850
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 
100029
Hits:
Download times:
English abstract:
      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 similarity, dissimilarity, metric and non metric type. The calculation methods involved the second set of distance methods (12 in total) for measuring the distance between two samples and the centroid method for measuring the distance between two clusters of samples. The data in the two examples were "survey data on the expenditure of farmers in 16 regions of China in 1982" and "measurement values of 13 sports events on two quantitative indicators". Using SAS software, cluster analysis of disordered samples was performed on the data from two instances, and the SAS output results were explained.
View Full Text   View/Add Comment  Download reader
Close