Review of Fuzzy Rule Based Classification systems

Abstract

Fuzzy Rule Based Classification systems (FRBCs) have received significant attention among the researchers due to the good behaviour in the real time databases. An important issue in the design of fuzzy rule-based classification system is the optimized generation of fuzzy if-then rules and the membership functions. The inductive learning of fuzzy rule classifier suffers in rule generation and rule optimization when the search space or variables becomes high. This creates the new idea of making the fuzzy system with precise rules leading to less scalability and improved accuracy. Accordingly, different approaches have been presented in the literature for optimal finding of fuzzy rules using optimization algorithms. Among the different techniques available in the literature, choosing the type, number of membership functions and defining parameters of membership function are still challenging tasks. In this paper, the optimization algorithms for optimal design of membership function and optimal rule generation are reviewed.