A proposal of fuzzy ID3 with ability of learning for AND/OR operators

摘要

The fuzzy ID3 is a powerful method to acquire fuzzy decision trees. However, the fuzzy ID3 has a couple of problems, i.e., a problem of a lack of representation and an adjusting problem. In our fuzzy ID3, AND/OR operators are formulated using t-norm and t-conorm connectives with parameters and each parameter is adjusted using golden section method. By adjusting parameters, the proposed fuzzy ID3 gives more accurate fuzzy rules for representing the data sets. If t-conorm connective is selected as AND/OR operator, the decision tree has a sub-decision tree.