Interval type-2 fuzzy logic in hybrid neural pattern recognition systems

Abstract

We describe in this paper an overview of new methods that we have been working on for building intelligent systems for pattern recognition using type-2 fuzzy logic and soft computing techniques. Soft Computing (SC) consists of several computing paradigms, including type-1 fuzzy logic, neural networks, and genetic algorithms, which can be used to create powerful hybrid intelligent systems. In this paper, we are reviewing the use of a higher order fuzzy logic, which is called type-2 fuzzy logic. Combining type-2 fuzzy logic with traditional SC techniques, we are able to build powerful hybrid intelligent systems that can use the advantages that each technique offers in solving pattern recognition problems. © 2013 Springer-Verlag Berlin Heidelberg.