Complex Belief KL Divergence in Complex Evidence Theory
Junjie Huang, Fuyuan Xiao · 2022
A generalized Dempster-Shafer evidence theory, known as complex evidence theory can explain regularly fluctuating data and offer a better framework for uncertain information representing and processing. In this study, we propose a probabilistic transformation method to transform the complex basic belief assignments of complex evidence theory into probability distributions. Additionally, we propose a generalized Kullback-Leibler divergence to measure the discrepancy of two complex basic belief assignments. Besides, we prove the properties of the proposed divergence and validate its availability by numerical examples.