Neural Network Clustering Based on Distances Between Objects
We present an algorithm of clustering of many-dimensional objects, where only the distances between objects are used. Centers of classes are found with the aid of neuron-like procedure with lateral inhibition. The result of clustering does not depend on starting conditions. Our algorithm makes it possible to give an idea about classes that really exist in the empirical data. The results of computer simulations are presented.
KeywordsLateral Inhibition Optimal Partition Input Point Close Class Noisy Point
Unable to display preview. Download preview PDF.
- 2.Ayvazyan, S., Bukhstaber, V., Enyukov, I., Meshalkin, L.: Applied statistics. In: Clustering and Reduction of Dimension. Finansy i Satitistika, Moscow (1989) (in Russian)Google Scholar
- 3.Zagoruyko, N.: Applied Methods of Data and Knowledge Analysis. Institute of Mathematics Press, Novosibirsk (1999) (in Russian)Google Scholar
- 5.Giada, L., Marsili, M.: Algorithms of maximal likelihood data clustering with applications. Cond-mat/0204202Google Scholar
- 6.Mantegna, R.N.: Hierarchical structure in financial markets. European Phys. Journal B 11, 193–197 (1999)Google Scholar