Fuzzy classifcation of imbalanced data sets for medical diagnosis

In this paper we have proposed a new method for medical diagnosis which is a hybridization of fuzzy logic and Ant Colony Optimization (ACO). At first, we utilize an oversampling method to balance the input datasets. Then, a set of fuzzy rules are discovered by using of an ACO algorithm. These fuzzy rules are made up our classifier. In next stage, testing samples are classified by an averaging based fuzzy engine. Our results indicate that the proposed method is efficient as a decision support tool for medical diagnosis.