Self-adaptive fuzzification in fuzzy decision tree induction
Xiaodong Dai, L. Q. Gao, Chun-Ru Dong · 2010
One of the most important issues in fuzzy decision tree learning is the fuzzification of input data. This paper proposes a self-adaptive data fuzzification algorithm based on the self-organizing map (SOM) technology, which can automatically determine the number and coordinates of centers in triangular membership functions. Then the membership degree of each sample to all fuzzy subsets can be calculated. Finally, a fuzzy decision tree can be learned from the fuzzified training samples by any selected fuzzy decision tree heuristic algorithm. Experimental results on UCI data set iris show that the new approach outperform the traditional fuzzification methods.