Gordon, A. D. Identifying genuine clusters in a classification. (English) Zbl 0900.62311 Comput. Stat. Data Anal. 18, No. 5, 561-581 (1994). Summary: The paper addresses the problem of assessing the validity of clusters produced by a clustering procedure. Several null models for data are described. Previous research is reviewed, it being shown how much of it can be formulated in terms of properties of sets of within-cluster and between-cluster pairwise dissimilarities. A Monte Carlo test for assessing the value of a U-statistic based on these sets of pairwise dissimilarities is described and illustrated on four data sets. The final section includes further discussion of ways of specifying relevant null models. Cited in 2 Documents MSC: 62H30 Classification and discrimination; cluster analysis (statistical aspects) Keywords:Cluster validation; Monte Carlo test; Poisson model; Random dissimilarity matrix model; Unimodal model; U-statistic PDFBibTeX XMLCite \textit{A. D. Gordon}, Comput. Stat. Data Anal. 18, No. 5, 561--581 (1994; Zbl 0900.62311) Full Text: DOI