How Many Clusters? An Investigation of Five Procedures for Detecting Nested Cluster Structure
The paper addresses the problem of identifying relevant values for the number of clusters present in a data set. The problem has usually been tackled by searching for a best partition using so-called stopping rules. It is argued that it can be of interest to detect cluster structure at several different levels, and five stopping rules that performed well in a previous investigation are modified for this purpose. The rules are assessed by their performance in the analysis of simulated data sets which contain nested cluster structure.
KeywordsCluster Structure Single Link Local Rule Cluster Criterion Global Rule
Unable to display preview. Download preview PDF.
- Beale, E. NI. L. (1969): Euclidean cluster analysis. Bulletin of the International Statistical Institute, 43(2), 92–94.Google Scholar
- Cooper, M. C. and Milligan, G. W. (1988): The effect of measurement error on determining the number of clusters in cluster analysis. In Data, Expert Knowledge and Decisions, Gaul. W. and Schader, M. (eds.), 319–328, Springer-Verlag, Berlin.Google Scholar
- Gordon, A. D. (1996): Cluster validation. Paper presented at IFCS-96 Conference, Kobe, 27–30 March, 1996.Google Scholar
- Jain, A. K. and Dubes, R. C. (1988): Algorithms for Clustering Data. Prentice-Hall, Englewood Cliffs, NJ.Google Scholar
- Milligan, G. W. and Cooper, M. C. (1985): An examination of procedures for determining the number of clusters in a data set. Psychometrika, 50, 159–179.Google Scholar