摘要
在证据理论中,证据组合方法是一个非常重要的研究工具,然而当面对高度冲突的证据时,用传统的方法直接进行正则化处理将会导致与直觉相悖的结论。较好的解决方法是在对证据进行组合前先进行预处理,即对证据进行加权平均。在考虑了证据理论与Vague集的对应关系后,将Vague集之间的相似度量方法运用到证据组合方法中,通过组合结果比较可以看出,选用的Vague集相似度量方法与改进的Dempster-Shafer证据组合方法有相近的目标识别与收敛效果,这表明可以将Vague集理论的一些有用结论运用于证据理论中,以更好地处理和研究不确定性知识。
Evidence combination rule plays an important role in Dempster-Shafer（DS） evidence theory. However, the Dempster combination result could not identify the actual conditions when faced with seriously conflicting evidence. A better solution is to pretreat the evidence before combination, namely evidence weighted average. After studying the relationship between DS evidence theory＂ and Vague sets, the similarity measurement between different Vague sets is applied to evidence combination. By comparing the combining results,we find that the similarity measurement between Vague sets and ihe modified DS evidence combination has similar effect in object recognition and convergence, indicating that some useful conclusions of Vague set theory can be applied to the evidence theory, which will be helpfid in handling and studying the uncertainty of knowledge.
出处
《济南大学学报（自然科学版）》
CAS
北大核心
2014年第5期366-370,共5页
Journal of University of Jinan(Science and Technology)
基金
国家自然科学基金(61203341)
济南大学教学研究项目(JZC1137)
