Pv. Balakrishnan et al., A STUDY OF THE CLASSIFICATION CAPABILITIES OF NEURAL NETWORKS USING UNSUPERVISED LEARNING - A COMPARISON WITH K-MEANS CLUSTERING, Psychometrika, 59(4), 1994, pp. 509-525
Citations number
39
Categorie Soggetti
Social Sciences, Mathematical Methods","Psychologym Experimental","Mathematical, Methods, Social Sciences
Several neural networks have been proposed in the general literature f
or pattern recognition and clustering, but little empirical comparison
with traditional methods has been done. The results reported here com
pare neural networks using Kohonen learning with a traditional cluster
ing method (K-means) in an experimental design using simulated data wi
th known cluster solutions. Two types of neural networks were examined
, both of which used unsupervised learning to perform the clustering.
One used Kohonen learning with a conscience and the other used Kohonen
learning without a conscience mechanism. The performance of these net
s was examined with respect to changes in the number of attributes, th
e number of clusters, and the amount of error in the data. Generally,
the K-means procedure had fewer points misclassified while the classif
ication accuracy of neural networks worsened as the number of clusters
in the data increased from two to five.