Improved intuitionistic fuzzy cross-entropy and its application to pattern recognitions

In this paper, considered the information carried by the membership, the non-membership and the intuitionistic index in intuitionistic fuzzy sets (IFSs) as probability spaces, a modified cross-entropy measure between two IFSs is proposed to overcome the drawback of the intuitionistic fuzzy cross-entropy measure proposed by Vlachos and Sergiadis and the notion of symmetric discrimination information measure for IFSs is introduced. Finally, numerical examples are presented to demonstrate the efficiency of the proposed discrimination information measure for pattern recognition and medical diagnosis.