Acquisition and Combination of Evidence Based on Rough Set Theory

摘要：

In the evidence theory, the most popular rule of combination, called Dempster’s rule of combination, has several interesting mathematical properties. Evidences used in Dempster’s rule of combination are given by experts and distinct information sources have the same weight. However, experts’ knowledge related with problem domain is limited, subjective and sometimes difficult to obtain. This may result in the belief function obtained from experts unreliable. The averaging assignment of weight may also get counterintuitive result when evidences conflict. This paper proposes a new approach to acquire evidences from large decision tables using the rough set theory and clustering. And then an investigation of how to compute the weights and support degrees of evidences are undertaken. Finally, an improved rule of combination is proposed. Experimental results compared with id3 and J48 show that the proposed methods can be used effectively in decision making.

关键词： Evidence theory Rough set theory Evidence acquisition Evidence combination

作者: Yezheng LIU Yuanchun JIANG Xiao LIU

作者单位: School of Management,Hefei University of Technology,Hefei 230009,China

会议类型: 国际会议

会议名称: 工业工程与系统管理2007年国际会议(International Conference on Industrial Engineering and Systems Management)(IESM 2007)

会议地点: 北京

会议语种:英文

在线出版日期: 2007-05-30（万方平台首次上网日期，不代表论文的发表时间）