An enhanced approach to rule base simplification of first-order takagi-sugeno fuzzy inference systems

Fuzzy rule base simplification is used to reduce the complexity of fuzzy models identified from the data. In this paper, an enhanced approach is proposed for simplifying the rule base of fuzzy inference systems when all the membership functions for a variable are highly similar to one another. In this case it is possible to remove a variable from the rule antecedent, but keep it in the rule consequent. Experimental results show that simpler rules can be obtained while barely sacrificing accuracy.