A Clustering Algorithm Using the Ordered Weight Sum of Self-Organizing Feature Maps
Clustering is to group similar objects into clusters. Until now there are a lot of approaches using Self-Organizing Feature Maps(SOFMs). But theyhave problems with a small output-layer nodes and initial weight. This paper suggests one-dimensional output-layer nodes in SOFMs. The number of output-layer nodes is more than those of clusters intended to find and the order of output-layer nodes is ascending in the sum of the output-layer node’s weight. We can find input data in SOFMs output node and classify input data in output nodes using the Euclidean Distance. The suggested algorithm was tested on well-known IRIS data and machine-part incidence matrix. The results of this computational study demonstrate the superiority of the suggested algorithm.
KeywordsCluster Algorithm Output Node Link Weight Learn Vector Quantization Suggested Algorithm
Unable to display preview. Download preview PDF.
- 1.Anderson, E.: The IRIS’s of the Gaspe Peninsula. Bull. Amer. IRIS Soc. 59, 2–5 (1939)Google Scholar
- 5.Everitt, B.S., Laudau, S., Leese, M.: Cluster Analysis. Edward Arnold, London (2001)Google Scholar
- 8.Kohonen, T.: Self-Organizing Maps. Springer, Berlin (1997)Google Scholar
- 9.Kusiak, A.: Computational Intelligence in Design and Manufacturing. John Wiley & Sons, New York (2000)Google Scholar
- 13.Tasao, E.C.-K., Bezdek, J.C., Pal, N.N.: Fuzzy Kohonen Clustering Networks. Pattern Recognition 27(5), 754–757 (1994)Google Scholar