Induction of fuzzy decision trees based on fuzzy rough set techniques

Efficient criteria to select fuzzy expanded attributes are important for generation fuzzy decision trees (FDTs). Given a fuzzy information system (FIS), fuzzy expanded attributes play a crucial role in fuzzy decision making. Besides, different fuzzy expanded attributes have different influences on decision making, and some of them may be more important than the others. This paper makes an attempt to improve fuzzy decision tree by extending FDT to the fuzzy rough set theory. One of the main contributions of this paper is a new criterion to select the expanded attributes by using accuracy measure of fuzzy expanded attributes with respect to fuzzy decision attributes.