A comparison of SOM neural network and hierarchical clustering methods. (English) Zbl 0912.90209

Summary: Cluster analysis, the determination of natural subgroups in a data set, is an important statistical methodology that is used in many contexts. A major problem with hierarchical clustering methods used today is the tendency for classification errors to occur when the empirical data departs from the ideal conditions of compact isolated clusters. Many empirical data sets have structural imperfections that confound the identification of clusters. We use a self organizing map (SOM) neural network clustering methodology and demonstrate that it is superior to the hierarchical clustering methods. The performance of the neural network and seven hierarchical clustering methods is tested on 252 data sets with various levels of imperfections that include data dispersion, outliers, irrelevant variables, and nonuniform cluster densities. The superior accuracy and robustness of the neural network can improve the effectiveness of decisions and research based on clustering messy empirical data.


90B99 Operations research and management science
62H30 Classification and discrimination; cluster analysis (statistical aspects)
68T05 Learning and adaptive systems in artificial intelligence


Full Text: DOI


[1] Anderberg, M. R., Cluster Analysis for Applications (1973), Academic Press Inc: Academic Press Inc New York · Zbl 0299.62029
[2] Berry, W.; Bozarth, C.; Hill, T.; Klompmaker, J., Factory focus: segmenting markets from an operations perspective, Journal of Operations Management, 10, 3, 363-387 (1991)
[3] Chandra, C.; Shahrukh, A.; Arora, S., Clustering effectiveness of permutation generation heuristics for machinepart matrix clustering, Journal of Manufacturing Systems, 12, 5, 388-407 (1993)
[4] Giuliano, G., Subcenters in the Los Angeles region, Regional Science and Urban Economics, 21, 163-182 (1991)
[5] Helsen, K.; Green, P. A., Computational study of replicated clustering with an application to market segmentation, Decision Science, 22, 5, 1124-1141 (1991)
[6] Ibis, D., Cluster analysis targets prospects, Credit World, 80, 2, 38-41 (1991)
[7] Kamrani, A. K.; Parsaei, H. R.; Chaudry, M., A survey of design methods for manufacturing cells, Computers and Industrial Engineering, 25, 487-490 (1993)
[8] Kangas, J. A.; Kohonen, T.; Laaksonen, J. T., Variants of self organizing feature maps, IEEE Transactions on Neural Networks, 1, 1, 93-99 (1990)
[9] Kohonen, T., Adaptive, associative, and self organizing functions in neural computing, Applied Optics, 26, 23, 4910-4918 (1987)
[10] Kohonen, T., Self Organization and Associative Memory (1988), Springer-Verlag: Springer-Verlag New York · Zbl 0659.68100
[11] Mathieu, R., A methodology for large scale r&d planning based on cluster analysis, IEEE Transactions on Engineering Management, 40, 3, 283-291 (1993)
[12] Milligan, G. W., An examination of the effect of six types of error perturbation on fifteen clustering algorithms, Psychometrika, 43, 5, 325-342 (1980)
[13] Milligan, G. W., An algorithm for generating artificial test clusters, Psychometrika, 50, 1, 123-127 (1985)
[14] Mulvey, J. M.; Crowder, H. P., Cluster analysis: an application of Lagrangian relaxation, Management Science, 25, 4, 329-340 (1979) · Zbl 0415.90085
[15] Neural Computing (1993), NeuralWare: NeuralWare Pittsburgh, PA
[16] Pal, N. R.; Bezdek, J. C.; Tsao, E. C.-K., Generalized clustering networks and Kohonen’s self organizing scheme, IEEE Transactions on Neural Networks, 4, 4, 549-557 (1993)
[17] Rao, M. R., Cluster analysis and mathematical programming, Journal of the American Statistical Association, 66, 622-626 (1971) · Zbl 0238.90042
[18] Ritter, H.; Martinetz, T.; Schulten, K., Neural computation and self organizing maps (1992), Addison Wesley: Addison Wesley Reading, MA · Zbl 0752.68068
[19] Robles, F., International market entry strategies and performance of united states catalog firms, Journal of Direct Marketing, 8, 1, 59-70 (1994)
[20] (SAS/STAT Users Guide (1988), SAS Institute, Inc: SAS Institute, Inc Cary, N.C), 283-357
[21] Spisak, A., Cluster analysis as a quality management tool, Quality Progress, 25, 12, 33-38 (1992)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.