New Similarity Measures on Intuitionistic Fuzzy Sets

An intuitionistic fuzzy set (IFS), an extended version of a fuzzy set, is more applicable for representing vague information since it provides a way to control a degree of hesitation. One of most prominent research directions on the theory of IFSs is measurement of similarity between IFSs. Even several similarity measures have been proposed, some unreasonable cases can be made by such measures. In this paper, we present two similarity measures: one is a set-theoretic measure, the other combines the former measure and the concept of similarity measures on a concept lattice. In our experimental study, we focus our attention on two basic directions of performance evaluation: one is how much the proposed measures are reasonable and the other is how much accuracy the measures produce when they are applied to pattern recognition. The experimental results show that the both measures are reasonable and achieve satisfactory performances on pattern recognition. Mathematics Subject Classification: 03E72, 68T10