基于群体决策和选择性融合的证据组合方法

Robust Evidence Combination Based on Group Decision Making and Selective Fusion

多数学者认为,利用修改证据体的方法来解决证据冲突问题较为合理,然而现有的融合方法均采用组合性融合方法,当焦元数目较多时运算量过大.针对此问题,提出一种基于群体决策和多准则选择性融合的证据组合方法,首先利用不同的群体决策方法对证据体进行预处理,以达到消除证据冲突的目的,然后采用选择性融合方法代替传统的组合性融合方法,选取证据可信度、证据信息散度和证据冲突度量三个评价指标并基于排序融合实现多个准则的综合利用,最终从多个待组合证据中选取一组最优的证据作为最终的融合结果.实验结果表明本文所提方法是合理有效的.

Most researchers hold the viewpoint that revising mass function based methods are reasonable to deal with the problem of conflicting evidence combination. Actually, they did not effectively reduce conflict among evidences by revision. And the existing fu- sion approachs are all using the clustering fusion method, the amount of calculation will explosion when there more focal elements. So in order to solve these intractable problems, a new evidence fusion approach is proposed based on group decision making and multi-cri- teria. First of all, different group decision making methods are used for preprocessing the conflict evidence in order to alleviate the con- flict between evidences. Second, the selective fusion approach is used for instead the traditional clustering fusion approach, and three e- valuation criteria are used which are credibility of evidence, evidence information divergence and evidence conflict measure to realize comprehensive utilization of multiple criteria based on rank level fusion. At last, a optimal evidence is selected to set as the final fusion result, the experimental results show that this proposed method is reasonable and effective.