Article ID: | iaor1999368 |
Country: | Netherlands |
Volume: | 93 |
Issue: | 2 |
Start Page Number: | 402 |
End Page Number: | 417 |
Publication Date: | Sep 1996 |
Journal: | European Journal of Operational Research |
Authors: | Chen Shaw K., Mangiameli Paul, West David |
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 depart 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.