Intuitionistic fuzzy information – Applications to pattern recognition

This paper addresses the issue of information-theoretic discrimination measures for intuitionistic fuzzy sets (IFSs). Although many measures of distance, similarity, dissimilarity, and correlation between IFSs have been proposed, there is no reference regarding information-driven measures used for comparison between sets. In this work we introduce the concepts of discrimination information and cross-entropy in the intuitionistic fuzzy setting and we derive an extension of the De Luca–Termini nonprobabilistic entropy for IFSs. Based on this entropy, we reveal an intuitive and mathematical connection between the notions of entropy for fuzzy sets (FSs) and IFSs in terms of fuzziness and intuitionism. Finally, we demonstrate the efficiency of the proposed discrimination information measure for pattern recognition, medical diagnosis, and image segmentation.