An improved conflicting-evidence combination method based on the redistribution of the basic probability assignment

• Published in: Applied intelligence (Dordrecht, Netherlands) Vol. 52; no. 4; pp. 4674 - 4700
• Main Authors: Yan, Zezheng, Zhao, Hanping, Mei, Xiaowen
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.03.2022; Springer Nature B.V
• Subjects: Artificial Intelligence; Computer Science; Data integration; Entropy (Information theory); Fault diagnosis; Information sources; Machines; Manufacturing; Mechanical Engineering; Methods; Processes; Dempster–Shafer theory; Conflicting evidence; Basic probability assignment; Evidence fusion
• ISSN: 0924-669X; 1573-7497
• DOI: 10.1007/s10489-021-02404-4

Dempster–Shafer evidence theory is widely applied in various fields related to information fusion. However, the results are counterintuitive when highly conflicting evidence is fused with Dempster’s rule of combination. Many improved combination methods have been developed to address conflicting evidence. Nevertheless, all of these approaches have inherent flaws. To solve the existing counterintuitive problem more effectively and less conservatively, an improved combination method for conflicting evidence based on the redistribution of the basic probability assignment is proposed. First, the conflict intensity and the unreliability of the evidence are calculated based on the consistency degree, conflict degree and similarity coefficient among the evidence. Second, the redistribution equation of the basic probability assignment is constructed based on the unreliability and conflict intensity, which realizes the redistribution of the basic probability assignment. Third, to avoid excessive redistribution of the basic probability assignment, the precision degree of the evidence obtained by information entropy is used as the correction factor to modify the basic probability assignment for the second time. Finally, Dempster’s rule of combination is used to fuse the modified basic probability assignment. Several different types of examples and actual data sets are given to illustrate the effectiveness and potential of the proposed method. Furthermore, the comparative analysis reveals the proposed method to be better at obtaining the right results than other related methods.