Title
Using Z-number to measure the reliability of new information fusion method and its application in pattern recognition
Abstract
Information fusion has traditionally been a concern. In the fusion process, how to effectively take care of the ambiguity and uncertainty of data is a fascinating problem. Dempster-Shafer evidence theory shows powerful functions in dealing with uncertainty information, and Z-number can comprehensively model the ambiguity and reliability of information. Inspired by this, this paper proposed a new information fusion method based on Dempster-Shafer theory and K-means clustering and it established the reliability evaluation criterion based on Z-number. Comparison and discussion verify the rationality of the proposed method, which also illustrates the method has better robustness and sensitivity than existing methods, some critical issues in DST, e.g., conflict management, evidence stuck, are well investigated and overcome by the proposed method. Number examples and the application further shows the application potential of the proposed method in a data-driven intelligent system. (C) 2021 Elsevier B.V. All rights reserved.
Year
DOI
Venue
2021
10.1016/j.asoc.2021.107658
APPLIED SOFT COMPUTING
Keywords
DocType
Volume
Information fusion, Dempster-Shafer evidence theory, Z-number, K-means clustering, Target Recognition
Journal
111
ISSN
Citations 
PageRank 
1568-4946
1
0.35
References 
Authors
0
5
Name
Order
Citations
PageRank
Ye Tian110.35
Xiangjun Mi263.44
Huizi Cui341.39
Pengdan Zhang411.03
Bingyi Kang511.70