Research on Fusion of Dependent Evidence Based on Kendall Correlation Coefficient
Zhou Jie, Xiaoyan Su, Hong Qian · Proceedings of the 4th International Conference on Computer Science and Application Engineering · 2020
The Dempster-Shafer evidence theory (D-S evidence theory) has been applied in many fields for its superiority in dealing with uncertainty information and ignorance. However, independent evidence cannot be well dealt with by the classical D-S theory, which restricts its applications. This paper has proposed a method based on Kendall correlation coefficient for dealing with dependent evidence. Firstly, the Kendall correlation coefficient is used to calculate the degree of dependence between evidence; Secondly, the discount coefficient is the inverse of the degree of dependence; Finally, the total fusion result combines D-S theory combination rules and discount coefficients. Give an example to explain the proposed method in this article, and compare the recognition results with other methods, so as to illustrate the superiority of proposed method.