标题:PGrey rough influence diagrams method and its medical application
作者:Hu, Haiqing ;Liu, Bingqiang ;Shen, Tao
作者机构:[Hu, Haiqing ;Liu, Bingqiang ] School of Mathematics, Shandong University, Jinan, Shandong, China;[Hu, Haiqing ;Shen, Tao ] School of Electrical Engin 更多
会议名称:12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
会议日期:13 August 2016 through 15 August 2016
来源:2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
出版年:2016
页码:977-983
DOI:10.1109/FSKD.2016.7603311
关键词:approximation; decision rules; grey rough sets; influence diagrams; uncertain information system
摘要:Influence diagrams is getting extensively applied as rule acquisition in economics, business, and military fields, in traditional influence diagrams, computational models of dependence relations are established by probability distribution, which has good performance in deterministic information system. However, information systems(IS) in real world usually are uncertain or approximate, such as IS composed by interval valued data, rules from large scaled uncertain data is also imprecise, the suitability of probability distribution model is questioned, how to represent approximate knowledge is a focus issue of traditional influence diagram model. Grey sets and rough sets are two mathematical tools in dealing with uncertain IS, consequently, this paper proposes a new alternative computational model grey rough influence diagrams to solve that problem, it is a combination of grey rough sets and influence diagrams. In the proposed scheme, the random relationship between the nodes and approximate rules are denoted by grey rough sets from uncertain IS. This paper provides a general influence diagrams model for uncertain IS, finally, a decision analysis example in medical fields is illustrated. © 2016 IEEE.
收录类别:EI;SCOPUS
资源类型:会议论文;期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-84997634584&doi=10.1109%2fFSKD.2016.7603311&partnerID=40&md5=5522e2bf1f8ab78c69010aa3da592cd3
TOP