A software program is described that performs K-means clustering with several enhancements. The software system includes the influence detection methodology developed by Cheng and Milligan. As such, the applied user can identify those data points that affect the results of a given cluster analysis. The user interface is interactive and allows for multiple analyses in the same session. Finally, the software can handle data set sizes much larger than those provided for by traditional hierarchical clustering methods.
Get full access to this article
View all access options for this article.
References
1.
Anderberg, M. R. (1973). Cluster analysis for applications. New York: Academic Press.
2.
Cheng, R. , & Milligan, G. W. (1995). Hierarchical clustering with influence detection. Educational and Psychological Measurement, 55, 237-244.
3.
Cheng, R. , & Milligan, G. W. (in press-a). Measuring the influence of individual data points in a cluster analysis. Journal of Classification.
4.
Cheng, R. , & Milligan, G. W. (in press-b). Identifying influence regions in hierarchical clustering. Multivariate Behavioral Research.
5.
Forgy, E. W. (1965). Cluster analysis of multivariate data: Efficiency versus interpretability of classifications. Paper presented at the meeting of the Biometric Society (WNAR), Riverside, CA.
6.
Hubert, L. J. , & Arabie, P. (1985). Comparing partitions. Journal of Classification, 2, 193-218.
7.
Jancey, R. C. (1966). Multidimensional group analysis. Australian Journal of Botany, 14, 127-130.
8.
Milligan, G. W. (1980). An examination of the effect of six types of error perturbation on fifteen clustering algorithms. Psychometrika, 45, 325-342.
9.
Milligan, G. W. , & Sokol, L. (1980). A two-stage clustering algorithm with robustness recovery characteristics. Educational and Psychological Measurement, 40, 755-759.
10.
Punj, G. , & Stewart, D. W. (1983). Cluster analysis in marketing research: Review and suggestions for application. Journal of Marketing Research, 20, 134-148.
