Conflict Measurement Method Based on Improved Evidence Distance
School of Mechanical Engineering, Inner Mongolia University of Technology, Hohhot 010051, China
| Abstract: | In facing the challenge of extracting difficult data from complex rolling bearings, the uncertainty of data has become a significant issue. Among various approaches, evidence theory stands out as a crucial method. However, in practical applications, traditional evidence theory exhibits certain limitations, especially in its core component—the conflict management mechanism—when dealing with highly conflicting evidence. Accurately quantifying and managing conflicts between pieces of evidence becomes a key issue in enhancing fusion effectiveness during the process of multi-source information fusion. To address this issue, this study introduces an innovative evidence conflict measurement method aimed at improving the capability of traditional evidence theory to handle highly conflicting evidence. This method starts from the perspective of "distance" and conducts a thorough analysis of the properties required by such a distance measurement. By proving these properties, the theoretical rationality and applicability of the proposed conflict measurement method are ensured. Further, through a series of theoretical proofs and numerical calculations, this study demonstrates the accuracy of the method in dealing with completely conflicting and completely non-conflicting evidence. Simultaneously, the method can effectively measure evidence with varying degrees of conflict, showing higher flexibility and broad applicability. This is achieved by applying the new method to classic cases of conflicting evidence and comparing it with existing methods. The comparison not only verifies the effectiveness of the new method on both theoretical and experimental levels but also reveals its significant advantages in handling highly conflicting evidence. The innovation of this research is primarily reflected in the following aspects: Initially, a novel conflict measurement framework is introduced, predicated on an exhaustive analysis and understanding of the "distance" between evidences. Subsequently, the rationality and efficacy of this measurement approach are corroborated through rigorous theoretical validation. Finally, comparative analyses with conventional methodologies underscore the effectiveness of this new approach in addressing highly conflicting evidences in practical applications. |
| Keywords: | Evidence Theory; Evidence Distance; Conflict Measurement |
| DOI: | 10.57237/j.mse.2024.01.005 |
View:
Downloads:
Received: 2 February 2024
Accepted: 15 March 2024
Published Online: 18 March 2024
References
| [1] | Dempster A P. Upper and lower probabilities induced by a multiple valued mapping [J]. The Annals of Mathematical Statistics, 1967, 38(2): 325-339. |
| [2] | Shafer G. A mathematical theory of evidence [M]. Princeton: Princeton University Press, 1976. |
| [3] | Zadeh L A. A simple view of the Dempster-Shafer theory of evidence and its implication for the rule of combination [J]. AI Magazine, 1986, 7(2): 85-90. |
| [4] | Smets P, Kennes R. The transferable belief model [J]. Artificial Intelligence, 1994, 66(2): 191-234. |
| [5] | Liu W R. Analyzing the degree of conflict among belief functions [J]. Artificial Intelligence, 2006, 170(11): 909-924. |
| [6] | Jousselme A L, Grenier D, Bossee E. A new distance between two bodies of evidence [J]. Information Fusion. 2012(2): 91-101. |
| [7] | 严志军, 陶洋. 基于证据距离和不确定度的冲突数据融合算法 [J]. 计算机工程与应用, 2018, 54(10): 54-58+104. |
| [8] | 彭颖, 胡增辉, 沈怀荣. 一种修正证据距离 [J]. 电子与信息学报, 2013, 35(7): 1624-1629. |
| [9] | 邓勇, 王栋, 李齐等. 一种新的证据冲突分析方法 [J]. 控制理论与应用, 2011, 28(6): 839-844. |
| [10] | 罗贺, 尹艳平, 胡笑旋等. 基于可信因子的证据源修正方法 [J]. 系统工程与电子技术, 2015, 37(06): 1459-1464. |
| [11] | 宋亚飞, 王晓丹, 雷蕾等. 基于相关系数的证据冲突度量方法 [J]. 通信学报, 2014, 35(5): 95-100. |
| [12] | 蒋雯, 彭进业. 一种新的证据冲突表示方法 [J]. 系统工程与电子技术, 2010, 32(3): 562-565. |
| [13] | 孙贵东, 关欣, 衣晓等. 冲突证据的相关系数度量方法 [J]. 通信学报, 2018, 39(12): 30-39. |
| [14] | 杨建平. 证据理论及其复杂系统可靠性分析方法与应用研究 [D]. 电子科技大学, 2012. |
Join the Editorial Board
We invite active, qualified and high profile scientists and researchers to join as Editorial Board Members.
Join UsBecome a Reviewer
Scholars with a strong interest in reviewing are invited to join the reviewer panel to ensure the quality of the research to be published.
Join Us