期刊
IEEE TRANSACTIONS ON FUZZY SYSTEMS
卷 29, 期 7, 页码 2018-2031出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TFUZZ.2020.2991296
关键词
Uncertainty; Entropy; Measurement uncertainty; Fuses; Atmospheric measurements; Particle measurements; Decision making; Complex information quality; complex-valued distribution (CvD); credibility measure; decision-making; entropy; information fusion; quality measure; uncertainty
资金
- Research Project of Education and Teaching Reform in Southwest University [2019JY053]
- Fundamental Research Funds for the Central Universities [XDJK2019C085]
- Chongqing Overseas Scholars Innovation Program [cx2018077]
This article introduces a generalized intelligent quality-based approach for fusing multisource information, with the aim of maintaining high quality fused results by considering credible information sources. The method defines vector representation and compatibility criteria for complex-valued distributions, devises an information quality measure using Gini entropy, and proposes uniform and weighted fusion methods for complex-valued distributions. These methods aim to achieve the highest quality fused results by taking into account the credibility of information sources.
In this article, we propose a generalized intelligent quality-based approach for fusing multisource information. The goal of the proposed approach is to fuse the multicomplex-valued information while maintaining a high quality of the fused result by considering the usage of credible information sources. First, a vector representation of complex-valued distribution is defined, as well as the definitions of compatibility and conflict degrees between complex-valued distributions. Based on that, the information quality measure of complex-valued distribution is devised by leveraging the concept of Gini entropy. After that, we study some special cases of the information quality measure in maximally certain and uncertain complex-valued distributions. Additionally, a uniform fusion method for complex-valued distributions is proposed on the basis of the complex-valued information quality as an initial feasible basis of decision-making. Taking into account a credibility measure in terms of the subsets of information sources, a weighted fusion method is then presented for complex-valued distributions. Particularly, the weighted fusion method can achieve the highest quality of the fused result from the associated aggregations of information that are modeled in complex-valued distributions. Finally, some examples are illustrated to demonstrate the feasibility and effectiveness of the proposed methods.
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