A new fuzzy reasoning method based on the use of the Choquet integral

Abstract

In this work we use the Choquet integral as an aggregation function and we apply it in the fuzzy reasoning method of fuzzy rule-based classification systems. We study the behaviour of several fuzzy measures and we propose a genetic learning method of an appropriate fuzzy measure to model the interaction among the set of rules of each class. In the experimental study we show that the new proposal allows the performance of the fuzzy reasoning method of the winning rule to be outperformed when dealing with classification problems in which several fuzzy rules are fired to classify each example.