Hostname: page-component-7b7567c8c-wxvwp Total loading time: 0 Render date: 2026-11-14T09:23:50.271Z Has data issue: false hasContentIssue false
  • English
  • Français

A Variable-Selection Heuristic for K-means Clustering

Published online by Cambridge University Press:  01 January 2025

Michael J. Brusco and
J. Dennis Cradit
Michael J. Brusco*
Affiliation:
Florida State University
J. Dennis Cradit
Affiliation:
Florida State University
*
Requests for reprints should be sent to Michael J. Brnsco, Marketing Depaxtment, College of Business, Florida State University, Tallahassee, FL 32306-1110, E-Mail: mbrusco@cob.fsu.edu
Get accessRights & Permissions [Opens in a new window]

Abstract

One of the most vexing problems in cluster analysis is the selection and/or weighting of variables in order to include those that truly define cluster structure, while eliminating those that might mask such structure. This paper presents a variable-selection heuristic for nonhierarchical (K-means) cluster analysis based on the adjusted Rand index for measuring cluster recovery. The heuristic was subjected to Monte Carlo testing across more than 2200 datasets with known cluster structure. The results indicate the heuristic is extremely effective at eliminating masking variables. A cluster analysis of real-world financial services data revealed that using the variable-selection heuristic prior to the K-means algorithm resulted in greater cluster stability.

Keywords

cluster analysisK-means partitioningvariable selectionheuristics

Information

Type
Articles
Information
Psychometrika , Volume 66 , Issue 2 , June 2001, pp. 249 - 270
Copyright
Copyright © 2001 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

Footnotes

We gratefully acknowledge the constructive comments of three anonymous reviewers, the Associate Editor, and Editor, which led to considerable improvements in this article. We hole diat our variable-selection heuristic evolved during the review process. This evolution was attributable to a variety of factors including: (a) the publication of the HINoV procedure (Carmone et al., 1999), (b) a thoughtful comment from an anonymous reviewer regarding correlated masking variables, and (c) a helpful suggestion from the Associate Editor concerning multiple true cluster structures in a single dataset.

References

Anderberg, M.R. (1973). Cluster analysis for applications. New York, NY: Academic Press.Google Scholar
Arabie, P., Hubert, L.J. (1994). Cluster analysis in marketing research. In Bagozzi, R.P. (Eds.), Advanced methods in marketing research (pp. 160–189). Oxford, England: Blackwell.Google Scholar
Arabie, P., Hubert, L.J. (1996). An overview of combinatorial data analysis. In Arabie, P., Hubert, L.J., De Soete, G. (Eds.), Clustering and classification (pp. 5–63). River Edge, NJ: World Scientific Publishing.10.1142/9789812832153_0002CrossRefGoogle Scholar
Art, D., Gnanadesikan, R., Kettenring, J.R. (1982). Data-based metrics for cluster analysis. Utilitas Mathematica, Series A, 21, 75–99.Google Scholar
Balakrishnan, P.V., Cooper, M.C., Jacob, V.S., Lewis, P.A. (1994). A study of the classification capabilities of neural networks using unsupervised learning: A comparison with K-means clustering. Psychometrika, 59, 509–525.10.1007/BF02294390CrossRefGoogle Scholar
Balasubramanian, S., Gupta, S., Kamakura, W., Wedel, M. (1998). Modelling large data sets in marketing. Statistica Neerlandica, 52, 303–323.10.1111/1467-9574.00086CrossRefGoogle Scholar
Berry, M.J.A., Linoff, G. (1997). Data mining techniques: For marketing, sales, and customer support. New York, NY: John Wiley & Sons.Google Scholar
Blattberg, R., Glazer, R., Little, J. (1994). The marketing information revolution. Boston, MA: Harvard Business School Press.Google Scholar
Box, G.E.P., Muller, M.E. (1958). A note on the generation of random normal deviates. Annals of Mathematical Statistics, 29, 610–611.CrossRefGoogle Scholar
Breckenridge, J.N. (1989). Replicating cluster analysis: Method, consistency, and validity. Multivariate Behavioral Research, 24, 147–161.10.1207/s15327906mbr2402_1CrossRefGoogle Scholar
Carmone, F.J., Kara, A., Maxwell, S. (1999). HINoV: A new model to improve market segmentation by identifying noisy variables. Journal of Marketing Research, 36, 501–509.10.1177/002224379903600408CrossRefGoogle Scholar
Chaturvedi, A., Carroll, J.D., Green, P.E., Rotondo, J.A. (1997). A feature-based approach to market segmentation via overlapping K-centroids clustering. Journal of Marketing Research, 34, 370–377.CrossRefGoogle Scholar
Cheng, R., Milligan, G.W. (1996). K-means clustering methods with influence detection. Educational and Psychological Measurement, 56, 833–838.CrossRefGoogle Scholar
Cormack, R.M. (1971). A review of classification (with Discussion). Journal of the Royal Statistical Society, Series A, 134, 321–367.10.2307/2344237CrossRefGoogle Scholar
DeSarbo, W.S., Carroll, J.D., Clark, L.A., Green, P.E. (1984). Synthesized clustering: A method for amalgamating alternative clustering bases with different weighting of variables. Psychometrika, 49, 57–78.CrossRefGoogle Scholar
DeSarbo, W.S., Manrai, A.K., Manrai, L.A. (1993). Non-spatial tree models for the assessment of competitive market structure: An integrated review of the marketing and psychometric literature. In Eliashberg, J., Lilien, G. (Eds.), Handbook in operations research and management science: Marketing (pp. 193–257). New York, NY: Elsevier.Google Scholar
De Soete, G. (1986). Optimal variable weighting for ultrametric and additive tree clustering. Quality and Quantity, 20, 169–180.10.1007/BF00227423CrossRefGoogle Scholar
De Soete, G. (1988). OVWTRE: A program for optimal variable weighting for ultrametric and additive tree fitting. Journal of Classification, 5, 101–104.CrossRefGoogle Scholar
De Soete, G., DeSarbo, W.S., Carroll, J.D. (1985). Optimal variable weighting for hierarchical clustering: An alternating least-squares algorithm. Journal of Classification, 2, 173–192.CrossRefGoogle Scholar
Fowlkes, E.B., Gnanadesikan, R., Kettenring, J.R. (1987). Variable selection in clustering and other contexts. In Mallows, C.L. (Eds.), Design, data, and analysis (pp. 13–34). New York, NY: John Wiley & Sons.Google Scholar
Fowlkes, E.B., Gnanadesikan, R., Kettenring, J.R. (1988). Variable selection in clustering. Journal of Classification, 5, 205–228.10.1007/BF01897164CrossRefGoogle Scholar
Fowlkes, E.B., Mallows, C.L. (1983). A method for comparing two hierarchical clusterings (with comments and rejoinder). Journal of the American Statistical Association, 78, 553–584.10.1080/01621459.1983.10478008CrossRefGoogle Scholar
Friedman, H.P., Rubin, J. (1967). On some invariant criteria for grouping data. Journal of the American Statistical Association, 62, 1159–1178.10.1080/01621459.1967.10500923CrossRefGoogle Scholar
Gnanadesikan, R., Kettenring, J.R., Tsao, S.L. (1995). Weighting and selection of variables for cluster analysis. Journal of Classification, 12, 113–136.10.1007/BF01202271CrossRefGoogle Scholar
Green, P.E., Carmone, F.J., Kim, J. (1990). A preliminary study of optimal variable weighting in K-means clustering. Journal of Classification, 7, 271–285.10.1007/BF01908720CrossRefGoogle Scholar
Helsen, K., Green, P.E. (1991). A computational study of replicated clustering with an application to market segmentation. Decision Sciences, 22, 1124–1141.CrossRefGoogle Scholar
Hubert, L., Arabie, P. (1985). Comparing partitions. Journal of Classification, 2, 193–218.CrossRefGoogle Scholar
Knuth, D.E. (1997). The art of computing: Vol. 1. Fundamental algorithms. Reading, MA: Addison-Wesley.Google Scholar
Krieger, A., Green, P.E. (1999). A generalized rand-index method for consensus clustering of separate partitions of the same data base. Journal of Classification, 16, 63–89.CrossRefGoogle Scholar
Kruskal, J.B. (1972). Linear transformations of multivariate data to reveal clustering. In Shepard, R.N., Romney, A.K., Nerlove, S.B. (Eds.), Multidimensional scaling: Theory and applications in the behavioral sciences (pp. 181–191). New York, NY: Seminar Press.Google Scholar
MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, I, 231–297.Google Scholar
McIntyre, R.M., Blashfield, R.K. (1980). A nearest-centroid technique for evaluating the minimum variance clustering procedure. Multivariate Behavioral Research, 15, 225–238.CrossRefGoogle Scholar
Milligan, G.W. (1980). An examination of six types of the effect of six types of error perturbation on fifteen clustering algorithms. Psychometrika, 45, 325–342.10.1007/BF02293907CrossRefGoogle Scholar
Milligan, G.W. (1985). An algorithm for generating artificial test clusters. Psychometrika, 50, 123–127.CrossRefGoogle Scholar
Milligan, G.W. (1989). A validation study of a variable-weighting algorithm for cluster analysis. Journal of Classification, 6, 53–71.CrossRefGoogle Scholar
Milligan, G.W. (1996). Clustering validation: Results and implications for applied analyses. In Arabie, P., Hubert, L.J., De Soete, G. (Eds.), Clustering and classification (pp. 341–375). River Edge, NJ: World Scientific Publishing.CrossRefGoogle Scholar
Milligan, G.W., Cooper, M.C. (1986). A study of the comparability of external criteria for hierarchical cluster analysis. Multivariate Behavioral Research, 21, 441–458.10.1207/s15327906mbr2104_5CrossRefGoogle ScholarPubMed
Milligan, G.W., Cooper, M.C. (1988). A study of the standardization of variables in cluster analysis. Journal of Classification, 5, 181–204.10.1007/BF01897163CrossRefGoogle Scholar
Milligan, G.W., Soon, S.C., Sokol, L.M. (1983). The effect of cluster size, dimensionality, and the number of clusters on the recovery of true cluster structure. IEEE Transactions on Pattern Analysis and Machine Intelligence, 5, 40–47.10.1109/TPAMI.1983.4767342CrossRefGoogle ScholarPubMed
Morey, L.C., Blashfield, R.K., Skinner, H.A. (1983). A comparison of cluster analysis techniques within a sequential validation framework. Multivariate Behavioral Research, 18, 309–329.CrossRefGoogle ScholarPubMed
Rand, W.M. (1971). Objective criteria for evaluating clustering methods. Journal of the American Statistical Association, 66, 846–850.10.1080/01621459.1971.10482356CrossRefGoogle Scholar
Rohlf, F.J. (1970). Adaptive hierarchical clustering schemes. Systematic Zoology, 19, 58–82.CrossRefGoogle Scholar
Salstone, R., Stange, K. (1996). A computer program to calculate Hubert and Arabie's adjusted Rand index. Journal of Classification, 13, 169–172.CrossRefGoogle Scholar
Ward, J.H. (1963). Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association, 58, 236–244.10.1080/01621459.1963.10500845CrossRefGoogle Scholar
Waller, N.G., Kaiser, H.A., Illian, J.B., Manry, M. (1998). A comparison of the classification capabilities of the 1-dimensional Kohonen neural network with two partitioning and three hierarchical cluster analysis algorithms. Psychometrika, 63, 5–22.CrossRefGoogle Scholar
Wedel, M., Kamakura, W.A. (1997). Market segmentation: Conceptual and methodological foundations. Boston, MA: Kluwer Academic Publishers.Google Scholar