A New Partition Criterion for Fuzzy Decision Tree Algorithm

Decision trees represent a simple and powerful method of induction from labeled instances. Fuzzy decision tree is the generalization of decision tree in fuzzy environment. The knowledge represented by fuzzy decision tree is more natural to the way of human thinking, but it's preprocess and tree-constructing are much costly. In this paper, we propose a modified fuzzy decision tree model (MFD). Entropy of multi-valued and continuous-valued attributes is both computed with fuzzy theory after fuzzification, while entropy of other attributes is dealt with General Shannon method. Experiment results suggest that the proposed model is more effective and efficient and can leads to comprehensible decision trees.

[1]  John Mingers,et al.  An Empirical Comparison of Pruning Methods for Decision Tree Induction , 1989, Machine Learning.

[2]  Ronald R. Yager,et al.  Measures of Entropy and Fuzziness Related to Aggregation Operators , 1995, Inf. Sci..

[3]  James C. Bezdek,et al.  Measuring fuzzy uncertainty , 1994, IEEE Trans. Fuzzy Syst..

[4]  Jennifer Widom,et al.  Clustering association rules , 1997, Proceedings 13th International Conference on Data Engineering.

[5]  Johannes Fürnkranz,et al.  Pruning Algorithms for Rule Learning , 1997, Machine Learning.

[6]  H. Trussell,et al.  Constructing membership functions using statistical data , 1986 .

[7]  Ronald R. Yager,et al.  Entropy and Specificity in a Mathematical Theory of Evidence , 2008, Classic Works of the Dempster-Shafer Theory of Belief Functions.

[8]  M. Shaw,et al.  Induction of fuzzy decision trees , 1995 .

[9]  B. Bouchon-Meunier,et al.  Learning Fuzzy Decision Rules , 1999 .

[10]  L. Zadeh.  Fuzzy sets as a basis for a theory of possibility , 1999 .

[11]  Krzysztof J. Cios,et al.  Continuous ID3 algorithm with fuzzy entropy measures , 1992, [1992 Proceedings] IEEE International Conference on Fuzzy Systems.