基于改进Jaccard系数的证据间相似性度量方法  

Similarity Measurement Between Evidences Based on Improved Jaccard Coefficient

作　　者：董仕 马怀祥 Dong Shi;Ma Huaixiang(School of Mechanical Engineering,Shijiazhuang Tiedao University,Shijiazhuang 050043,China)

机构地区：[1]石家庄铁道大学机械工程学院,河北石家庄050043

出　　处：《石家庄铁道大学学报（自然科学版）》2021年第2期66-71,共6页Journal of Shijiazhuang Tiedao University(Natural Science Edition)

基　　金：国家自然科学基金(11872254)。

摘　　要：合理准确地描述证据之间的相似性,是证据有效合成的前提。针对现有方法在度量证据之间相似性时的不足,提出了一种度量证据间相似性的新方法。首先将Jaccard系数矩阵分块归一化并引入余弦相似度模型,根据证据间相似度对证据源加权平均,最后利用Dempster组合规则进行组合。该方法突出单元素焦元在计算证据间相似度时的重要度,在对含有多元素焦元的证据合成时可靠性更高。仿真算例验证了该方法的有效性。Reasonable and accurate description of the similarity between evidences is the premise of effective evidence synthesis.To overcome the shortcomings of existing methods in measuring similarity between evidences,a new method for measuring similarity between evidences was proposed.Firstly,the Jaccard coefficient matrix was divided into blocks and normalized,then cosine similarity model was introduced,and then the evidence sources were weighted and averaged according to the similarity between evidences.Finally,Dempster combination rule was used to combine.This method highlights the importance of single element focus element in calculating the similarity between evidences,and is more reliable in the fusion of evidences with multi-element focus element.The effectiveness of the method has been verified by a simulation example.

关 键 词：证据合成 相似性度量 Jaccard系数 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]