Comparison of Various Feature Selection Methods in Application to Prototype Best Rules
Prototype based rules is an interesting tool for data analysis. However most of prototype selection methods like CFCM+LVQ algorithm do not have embedded feature selection methods and require feature selection as initial preprocessing step. The problem that appears is which of the feature selection methods should be used with CFCM+LVQ prototype selection method, and what advantages or disadvantages of certain solutions can be pointed out. The analysis of the above problems is based on empirical data analysis.
KeywordsFeature Selection Feature Subset Feature Selection Method Ranking Method Feature Selection Algorithm
Unable to display preview. Download preview PDF.
- 5.Shanonn, C., Weaver, W.: The Mathematical Theory of Communication. University of Illinois Press (1946)Google Scholar
- 6.Setiono, R., Liu, H.: Improving backpropagation learning with feature selection. The International Journal of Artificial Intelligence, Neural Networks, and Complex Problem-Solving Technologies 6, 129–139 (1996)Google Scholar
- 8.Chi, J.: Entropy based feature evaluation and selection technique. In: Proc. of 4-th Australian Conf. on Neural Networks (ACNN 1993) (1993)Google Scholar
- 9.Yu, L., Liu, H.: Feature selection for high-dimensional data: A fast correlation-based filter solution. In: Proceedings of The Twentieth International Conference on Machine Learning (2003)Google Scholar
- 10.Duch, W., Biesiada, J.: Feature selection for high-dimensional data: A kolmogorov-smirnov correlation-based filter solution. In: Advances in Soft Computing, pp. 95–104. Springer, Heidelberg (2005)Google Scholar