Evolutionary Generation of Prototypes for a Learning Vector Quantization Classifier
An evolutionary computation based algorithm for data classification is presented. The proposed algorithm refers to the learning vector quantization paradigm and is able to evolve sets of points in the feature space in order to find the class prototypes. The more remarkable feature of the devised approach is its ability to discover the right number of prototypes needed to perform the classification task without requiring any a priori knowledge on the properties of the data analyzed. The effectiveness of the approach has been tested on satellite images and the obtained results have been compared with those obtained by using other classifiers.
KeywordsFeature Space Recognition Rate Crossover Operator Near Neighbor Reference Vector
Unable to display preview. Download preview PDF.
- 3.Lanzi, P.L., Stolzmann, W., Wilson, S.W. (eds.): IWLCS 1999. LNCS (LNAI), vol. 1813. Springer, Heidelberg (2000)Google Scholar
- 5.Greene, D.P., Smith, S.F.: Competition-based induction of decision models from examples. Machine Learning, 229–257 (1993)Google Scholar
- 6.Janikow, C.Z.: A knowledge-intensive genetic algorithm for supervised learning. Machine Learning, 189–228 (1993)Google Scholar
- 7.De Jong, K.A., Spears, W.M., Gordon, D.F.C., Janikow, Z.: Using genetic algorithms for concept learning. Machine Learning, 161–188 (1993)Google Scholar
- 9.Rauss, P.J., Daida, J.M., Chaudhary, S.A.: Classification of spectral image using genetic programming. In: Genetic and Evolutionary Computation Conference (GECCO), pp. 726–733 (2000)Google Scholar
- 13.Karayiannis, N.B.: Learning vector quantization: A review. International Journal of Smart Engineering System Design 1, 33–58 (1997)Google Scholar
- 15.Blickle, T., Thiele, L.: A comparison of selection schemes used in genetic algorithms. Technical Report 11, Swiss Federal Institute of Technology (ETH), Gloriastrasse 35, 8092 Zurich, Switzerland (1995)Google Scholar