点击次数：

论文类型：期刊论文

发表时间：2010-08-01

发表刊物：INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL

收录刊物：SCIE、EI、Scopus

卷号：6

期号：8

页面范围：3737-3749

ISSN号：1349-4198

关键字：Dempster-Shafer evidence theory; Least point's principle; Decision making

摘要：On the basis of attributes of basic element in combination with those of non-basic element, a solution to a decision model based on least point's principle is substantively improved, which makes the model further perfect in theory. By introducing relative plausibility function, expected support and decision, matrix, respectively, we,systematically solve some key problems in the model, such as exhaustive search in the power set of a frame of discernment, selection strategy of choosing a certain basic element from candidates to remove it, and formal estimation for control parameters. As a summary of studies, we modify and redefine the general algorithm of the least point's decision model in order to make it a practical decision making method which can be widely used in the fields of expert system, intelligent decision making, and pattern recognition, etc. Finally, two real system examples are employed to examine the validity of the decision results by the improved approach, experimentally, which also helps us to have an insight into the essential nature of the decision model based on least point's principle.