A variable weight approach for evidential reasoning  

作　　者：CHANG Lei-lei LI Meng-jun JIANG Jiang 

机构地区：[1]College of Information System and Management, National University of Defense Technology, Changsha 410073, China [2]Department of Mathematics, University of Birmingham, Birmingham B 152TT, UK

出　　处：《Journal of Central South University》2013年第8期2202-2211,共10页中南大学学报（英文版）

基　　金：Foundation item： Projects（70901074, 71001104, 71201168） supported by the National Natural Science Foundation of China

摘　　要：A variable weight approach was proposed to handle the probability deficiency problem in the evidential reasoning （ER） approach. The probability deficiency problem indicated that the inadequate information in the assessment result should be less than that in the input. However, it was proved that under certain circumstances, the ER approach could not solve the probability deficiency problem. The variable weight approach was based on two assumptions： 1） the greater weight should be given to the rule with more adequate information; 2） the greater weight should be given to the rules with less disparate information. Assessment results of two notional case studies show that 1） the probability deficiency problem is solved using the proposed variable weight approach, and 2） the information with less inadequacy and more disparity is provided for the decision makers to help reach a consensus.一条可变重量途径被建议在证据的推理处理概率缺乏问题(嗯) 途径。概率缺乏问题显示在评价结果的不适当的信息应该是不到那在输入。然而，在某些情形下面， ER 途径不能解决概率缺乏问题，这被证明。可变重量途径基于二个假设：1 ) 更大的重量应该与更足够的信息被给规则；2 ) 更大的重量应该与不太迥异的信息被给规则。二概念的案例研究的评价结果显示出那 1 ) 概率缺乏问题用建议可变重量途径被解决，并且 2 ) 有更少的机能不全的信息和更多的不同被提供决定制造者帮助到达一致。

关 键 词：probability deficiency evidential reasoning （ER） inadequate information variable weight CONSENSUS 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] Q959.8[自动化与计算机技术—控制科学与工程]