Defuzzification methods in intuitionistic fuzzy inference systems of Takagi-Sugeno type: The case of corporate bankruptcy prediction

Recently, an intuitionistic fuzzy inference system (IFIS) of Takagi-Sugeno type has been proposed. Previous results have shown that by adding non-membership functions, the average errors may be significantly decreased compared with FISs. In this paper, we design defuzzification methods for this class of systems. The methods are based on weighted average and weighted sum of the consequents of rules in IFIS. The empirical comparison of the methods is carried out on a dataset for corporate bankruptcy prediction.