Trust Revision for Conflicting Sources

被引:0
作者
Josang, Audun [1 ]
Ivanovska, Magdalena [1 ]
Muller, Tim [2 ]
机构
[1] Univ Oslo, N-0316 Oslo, Norway
[2] Nanyang Technol Univ, Singapore, Singapore
来源
2015 18TH INTERNATIONAL CONFERENCE ON INFORMATION FUSION (FUSION) | 2015年
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Sensors and information sources can produce conflicting evidence for various reasons, including errors and deception. When conflicting evidence is received, some sources produce more accurate evidence than others with regard to the ground truth. The reliability of sources can be expressed by assigning a level of trust to each source. In this situation, multiple fusion strategies can be applied: one strategy is to directly fuse the evidence based on a priori trust in each source, another strategy is to first revise a priori trust assignments as a function of the degree of conflict, before the evidence is fused. This paper focuses on the latter approach, and describes a method for applying trust revision in case of highly conflicting evidence. The trust revision method is expressed in the formalism of subjective logic.
引用
收藏
页码:550 / 557
页数:8
相关论文
共 11 条
[1]  
[Anonymous], P 4 INT C TRUST MAN
[2]  
[Anonymous], 1994, Formalizing trust as a computational concept
[3]  
Ding L., 2004, P 13 INT WORLD WID W
[4]  
Fullam K., 2005, P 8 INT WORKSH TRUST
[5]  
Gambetta D., 1990, Trust: Making and Breaking Coopeative Relations, P213
[6]  
Josang A., 2004, P 2 INT C TRUST MAN
[7]  
Josang A., 2013, P 16 INT C INF FUS F
[8]  
Josang A., 2006, CRPIT, V48
[9]  
Josang A., 2012, P 6 IFIP INT C TRUST
[10]  
Josang A., 2012, 15 INT C INFORM FUSI, P1225