计算机应用 ›› 2018, Vol. 38 ›› Issue (10): 2794-2800.DOI: 10.11772/j.issn.1001-9081.2018030532

• 人工智能 • 上一篇    下一篇

降低相似度碰撞的证据融合方法

王剑1, 张志勇2, 乔阔远2   

  1. 1. 郑州大学 信息工程学院, 郑州 450001;
    2. 河南科技大学 信息工程学院, 河南 洛阳 471003
  • 收稿日期:2018-03-15 修回日期:2018-05-05 出版日期:2018-10-10 发布日期:2018-10-13
  • 通讯作者: 王剑
  • 作者简介:王剑(1978-),女,河南洛阳人,副教授,博士,CCF会员,主要研究方向:多媒体社交网络、信息安全、可信计算和使用控制;张志勇(1975-),男,河南新乡人,教授,博士,IEEE高级会员,主要研究方向:多媒体社交网络、数字版权管理、可信计算和使用控制;乔阔远(1991-),男,河南濮阳人,硕士研究生,主要研究方向:信任度计算、数据挖掘、信息安全、多源信息融合。
  • 基金资助:

    国家自然科学基金资助项目(61772174,61370220);河南省自然科学基金资助项目(162300410094)。

Evidence combination rule with similarity collision reduced

WANG Jian1, ZHANG Zhiyong2, QIAO Kuoyuan2   

  1. 1. School of Information Engineering, Zhengzhou University, Zhengzhou Henan 450001, China;
    2. School of Information Engineering, Henan University of Science and Technology, Luoyang Henan 471003, China
  • Received:2018-03-15 Revised:2018-05-05 Online:2018-10-10 Published:2018-10-13
  • Supported by:

    This work is partially supported by the National Natural Science Foundation of China (61772174, 61370220), the Natural Science Foundation of Henan Province (162300410094).

摘要:

针对相似度碰撞引发证据融合结果错误的问题,提出一种新的证据融合方法。首先,提取证据的焦元序列特征并将其转化为排序矩阵以弥补证据相似度易碰撞的不足;其次,联合证据的排序矩阵和信息熵完成对证据权重的确定;最后,生成归一化证据(MAE)并使用Dempster融合公式将MAE融合n-1次获得最终的结果。基于在线的鸢尾花数据集对证据平均融合方法、余弦相似度证据融合方法、证据距离融合方法和证据信誉度融合方法进行了花类型识别准确性的F-Score对比,上述四种方法的F-Score分别为0.84、0.88、0.88和0.88,而所提方法的F-Score为0.91。实验结果表明,所提方法的决策准确率更高,融合结果更加可靠,能为证据决策提供了有效的解决方案。

关键词: D-S证据理论, 证据融合方法, 相似度碰撞, 基本信任分配, 冲突管理

Abstract:

Aiming at the problem of decision error caused by similarity collision in evidence theory, a new combination rule for evidence theory was proposed. Firstly, the features of focal-element sequence in evidence were extracted and converted into a sort matrix to reduce similarity collision. Secondly, the weight of each evidence was determined based on sort matrix and information entropy. Finally, the Modified Average Evidence (MAE) was generated based on the evidence set and evidence weight, and the combination result was obtained by combing MAE for n-1 times by using Dempster combination rule. The experimental results on the online dataset Iris show that the F-Score of average-based combination rule, similarity-based combination rule, evidence distance-based combination rule, evidence-credit based combination rule and the proposed method are 0.84, 0.88, 0.88, 0.88 and 0.91. Experimental results show that the proposed method has higher accuracy of decision making and more reliable combination results, which can provide an efficient solution for decision-making based on evidence theory.

Key words: D-S evidence theory, combination rule method, similarity collision, Basic Probability Assignment (BPA), conflict management

中图分类号: