A Comparison of the Classification Capabilities of the 1-Dimensional Kohonen Neural Network with Two Pratitioning and Three Hierarchical Cluster Analysis Algorithms
Authors:
- 1 March 1998
- journal article
- research article
- Published by Cambridge University Press (CUP) in Psychometrika
- Vol. 63 (1) , 5-22
Abstract
Neural Network models are commonly used for cluster analysis in engineering, computational neuroscience, and the biological sciences, although they are rarely used in the social sciences. In this study we compare the classification capabilities of the 1-dimensional Kohonen neural network with two partitioning (Hartigan and Späth k-means) and three hierarchical (Ward's, complete linkage, and average linkage) cluster methods in 2,580 data sets with known cluster structure. Overall, the performance of the Kohonen networks was similar to, or better than, the performance of the other methods.Keywords
This publication has 32 references indexed in Scilit:
- A connectionist approach for clustering with applications in image analysisIEEE Transactions on Systems, Man, and Cybernetics, 1994
- Generalized clustering networks and Kohonen's self-organizing schemeIEEE Transactions on Neural Networks, 1993
- An unsupervised learning technique for artificial neural networksNeural Networks, 1990
- Self-organizing feature maps and the travelling salesman problemNeural Networks, 1988
- Comparing partitionsJournal of Classification, 1985
- Algorithm AS 136: A K-Means Clustering AlgorithmJournal of the Royal Statistical Society Series C: Applied Statistics, 1979
- Computer Generation of Random Variables Using the Ratio of Uniform DeviatesACM Transactions on Mathematical Software, 1977
- Objective Criteria for the Evaluation of Clustering MethodsJournal of the American Statistical Association, 1971
- Hierarchical Grouping to Optimize an Objective FunctionJournal of the American Statistical Association, 1963
- A Coefficient of Agreement for Nominal ScalesEducational and Psychological Measurement, 1960