A Hybrid Decision Tree Model Based on Credibility Theory
Decision-Tree (DT) is a widely-used approach to retrieve new interesting knowledge. Fuzzy decision trees (FDT) can handle symbolic domains flexibly, but its preprocess and tree-constructing are much costly. In this paper, we propose a hybrid decision tree (HDT) model by introducing credibility entropy into FDT. Entropy of multi-valued and continuous-valued attributes are both calculated with credibility theory, while entropy of other attributes is dealt with general Shannon method. HDT can decrease the cost of preprocess and tree-constructing significantly. We apply the model into geology field to find out the factors which cause landslide. Experiment results show that the proposed model is more effective and efficient than fuzzy decision tree and C4.5.
KeywordsClassification Credibility Theory Fuzzy Decision Tree Fuzzy Entropy
Unable to display preview. Download preview PDF.
- 1.Quinlan, Induction on decision trees.Machine Learning 1(1), 81-106 (1986)Google Scholar
- 2.Bouchon-Meunier, B., Marsala, C., Ramdani, M.: Learning from imperfect data. In: Dubois, D., Prade, H., Yager, R.R. (eds.) Fuzzy Information Engineering: a Guided Tour of Applications, pp. 139–148. John Wiley and Sons, Chichester (1997)Google Scholar
- 3.Bouchon-Meunicr, B., Marsala, C.: Learning fuzzy decision rules. In: Dubois, D., Bczdek, J., Prade, H. (eds.) Fuzzy Sets in Approximate Reasoning and Information Systems. Handbook of Fuzzy Sets, vol. 3, Kluwer Academic Publishers, Dordrecht (1999)Google Scholar
- 9.Li, X., Liu, B.: New independence definition of fuzzy random variable and random fuzzy variable. World Journal of Modelling and Simulation (2006)Google Scholar
- 19.Li, P., Liu, B.: Entropy of credibility distributions for fuzzy variables, Technical Report (2005)Google Scholar
- 22.Kohonen, T.: Self-Organization and Association Memory. Springer, Berlin (1988)Google Scholar
- 25.Breiman, L., et al.: Classification and Regression Frees. Wadsworth International Group, Behnont (1984)Google Scholar