Flexible Neural Tree for Pattern Recognition
This paper presents a novel induction model named Flexible Neural Tree (FNT) for pattern recognition. FNT uses decision tree to do basic analysis and neural network to do subsequent quantitative analysis. The Pure Information Gain I(X i ;ϑ), which is defined as test selection measure for FNT to construct decision tree, can be used to handle continuous attributes directly. When the information embodied by neural network node can show new attribute relations, FNT extracts symbolic rules from neural network to increase the performance of decision process. Experimental studies on a set of natural domains show that FNT has clear advantages with respect to the generalization ability.
KeywordsNeural Network Decision Tree Continuous Attribute Hide Unit Current Node
Unable to display preview. Download preview PDF.
- 1.Huan, L., Rudy, S.: Feature Transformation and Multivariate Decision Tree Induction. In: Arikawa, S., Motoda, H. (eds.) DS 1998. LNCS (LNAI), vol. 1532, pp. 279–291. Springer, Heidelberg (1998)Google Scholar
- 4.Atlas, L., Cole, R., Uthusamy, M., Lippman, A.: A Performance Comparison of Trained Multi-layer Perceptions and Trained Classification Trees. In: Zhong, S., Malla, S. (eds.) Proceedings of the IEEE International Conference on Computer Vision, Osaka, Japan, vol. 78, pp. 1614–1619 (1990)Google Scholar
- 5.Setiono, R., Huan, L.: Symbolic Representation of Neural Networks. Int. J. Computer 29(5), 71–77 (1996)Google Scholar
- 7.Dougherty, J.: Supervised and Unsupervied Discretization of Coninuous Features. In: Armand, P., Stuart, J.R. (eds.) Proceedings of the 12th International Conference on Machine Learning, Tahoe City, California, USA, pp. 194–201 (1995)Google Scholar