Toward a development of general type-2 fuzzy classifiers applied in diagnosis problems through embedded type-1 fuzzy classifiers

Nowadays, with the emergence of computer-aided systems, diagnosis problems are one of the most important application areas of artificial intelligence. The present paper is focused on a specific kind of computer-aided diagnosis system based on General Type-2 Fuzzy Logic. The main goal is the generation of General Type-2 Fuzzy Classifiers that can handle the data uncertainty. The concept of embedded Type-1 Fuzzy membership functions has been proposed to be used in the design of General Type-2 Fuzzy Classifiers. A methodology for generating the embedded Type-1 fuzzy membership functions is introduced, and the subsequent approach for developing the Footprint of Uncertainty of the General Type-2 Fuzzy Classifier is presented. On the other hand, the proposed approach performance is evaluated by the experimentation with different diagnosis benchmark problems. In addition, a statistical comparison with respect to another existing approach of General Type-2 Fuzzy classifiers is presented.