期刊
INFORMATION FUSION
卷 14, 期 4, 页码 410-422出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.inffus.2012.08.002
关键词
Adaptive robust combination rule; Discount on evidence; Group decision making; Information fusion; Theory of evidence
资金
- Hong Kong Research Grants Council [CUHK446907]
- Geographical Modeling and Geocomputation Program under the Focused Investment Scheme of The Chinese University of Hong Kong
Dempster-Shafer theory of evidence has been employed as a major method for reasoning with multiple evidence. The Dempster's rule of combination is however incapable of managing highly conflicting evidence coming from different information sources at the normalization step. Extending current rules, we incorporate the ideas of group decision-making into the theory of evidence and propose an integrated approach to automatically identify and discount unreliable evidence. An adaptive robust combination rule that incorporates the information contained in the consistent focal elements is then constructed to combine such evidence. This rule adjusts the weights of the conjunctive and disjunctive rules according to a function of the consistency of focal elements. The theoretical arguments are supported by numerical experiments. Compared to existing combination rules, the proposed approach can obtain a reasonable and reliable decision, as well as the level of uncertainty about it. (c) 2012 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据