«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

期数:

2013年05期

页码:

58-62

栏目:

智能、算法、系统工程

出版日期:

1900-01-01

文章信息/Info

Title:

A New Improved D-S Algorithm Based on Weight of Evidence

文章编号:

1673-629X（2013）05-0058-05

作者:

董增寿¹; 邓丽君¹; 曾建潮²

[1]太原科技大学 电子信息工程学院;[2]太原科技大学 机械工程学院

Author(s):

DONG Zeng-shou; DENG Li-jun; ZENG Jian-chao

关键词:

D-S证据理论; 组合规则; 证据距离函数; 冲突证据

Keywords:

D-S evidence theory; combination rule; evidence distance function; conflict evidence

文献标志码:

A

摘要:

D-S(Dempster-Shafer)证据理论是一种有效的不确定性推理方法,但在组合高冲突证据时, D-S证据理论得到的结果却往往有悖常理.为了解决冲突证据的合成问题,考虑到不同的证据在合成过程中的重要程度不同,提出了一种新的基于证据权重的D-S改进算法.该方法首先引入一个度量证据体间相似度的证据距离函数,建立相应的证据距离矩阵,求出系统中各证据到证据集的平均欧式距离,然后通过信任函数来获得描述各证据重要程度的权重系数并对证据源进行修正,最后利用D-S组合规则对修正后的证据进行合成.通过算例的分析以及与其它改进算法的比较,验证了新方法的有效性和优越性

Abstract:

Dempster-Shafer evidence theory is an effective method for dealing with uncertainty problems,but the results obtained are counterintuitive when the evidences highly conflict with each other. According to the importance of the evidences,a new improved D-S algorithm based on the weight of evidence is presented to solve this problem. Firstly,a distance function between the bodies of evidences is introduced,the average Euclidean distance between evidence and evidence subset is obtained after establishing the evidence distance matrix,and then modify the evidence through the belief function to get weight coefficients which describe the importance degree of evi-dence,finally these modified evidences are combined together according to the D-S rule. Through the numerical study and compared with the other improved methods verify the validity and superiority of the new method

常用功能

更新日期/Last Update: 1900-01-01