Complex interval number‐based uncertainty modeling method with its application in decision fusion

Abstract

Complex evidence theory, a generalization of Dempster–Shafer evidence theory, is an effective uncertainty reasoning for decision fusion in complex-valued domain. In particular, the generation of complex basic belief assignment (CBBA) is a key issue for uncertainty modeling in complex evidence theory. In this paper, we first construct complex interval number (CIN) model. In this context, we propose a novel CBBA generation method to model uncertainty in the framework of complex planes. Furthermore, we propose a novel decision-making algorithm on the basis of the CIN-based CBBA generation method. Through an application in pattern recognition on several real-world data sets, the efficiency of the proposed decision-making algorithm is verified.