An improved evidence combination approach based on credibility of evidence

Yaozhong Yu, Chen Yang, Hong Jiang, Qingdong Li, Jie Xiang

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

In this paper, an improved evidence combination method is proposed to address the paradox problem of Dempster-Shafer theory when dealing with highly conflict evidences. Considering that the counter-intuitive combination result of Dempster's theory is related to both the unreliable evidence and the unreasonable distribution of the conflict evidences, the method combines the evidence preprocessing and conflict reassignment closely. The credibility of each evidence is firstly obtained based on Jousselme's distance. And then the original evidences are preprocessed by using the credibility degree as the discounting factor. The preprocessed evidences are lastly combined by the modified combination rule which distributes the conflict to each combining proposition, also taking the credibility of each evidence into account. The numerical examples demonstrate that the modified method can solve the paradox problem dealing with evidence conflict and has high convergent speed.

源语言英语
主期刊名Proceedings of the 36th Chinese Control Conference, CCC 2017
编辑Tao Liu, Qianchuan Zhao
出版商IEEE Computer Society
5518-5522
页数5
ISBN(电子版)9789881563934
DOI
出版状态已出版 - 7 9月 2017
已对外发布
活动36th Chinese Control Conference, CCC 2017 - Dalian, 中国
期限: 26 7月 201728 7月 2017

出版系列

姓名Chinese Control Conference, CCC
ISSN(印刷版)1934-1768
ISSN(电子版)2161-2927

会议

会议36th Chinese Control Conference, CCC 2017
国家/地区中国
Dalian
时期26/07/1728/07/17

指纹

探究 'An improved evidence combination approach based on credibility of evidence' 的科研主题。它们共同构成独一无二的指纹。

引用此

Yu, Y., Yang, C., Jiang, H., Li, Q., & Xiang, J. (2017). An improved evidence combination approach based on credibility of evidence. 在 T. Liu, & Q. Zhao (编辑), Proceedings of the 36th Chinese Control Conference, CCC 2017 (页码 5518-5522). 文章 8028232 (Chinese Control Conference, CCC). IEEE Computer Society. https://doi.org/10.23919/ChiCC.2017.8028232