可能性理论研究  

Brief analysis of the possibility theory

作　　者：严巍[1] 赵川[1] 

机构地区：[1]成都理工大学信息科学与技术学院,成都610059

出　　处：《数字通信》2014年第5期24-27,共4页Digital Communications and Networks

摘　　要：研究不确定性的传统方法是使用概率论和统计,但现实世界中所遇到的不确定性有多种类型,传统方法并不能全部适用。传统方法仅适于处理发生条件的不确定导致的不确定性,即已知某一个确定的事件,由于其发生条件的微小扰动,导致事件发生与否的不确定。另外一种类型的不确定性则是事件本身就存在着不确定因素,此时传统方法不再有效,而应该使用可能性理论来处理。讨论使用可能性理论处理自然语言型软数据的具体规则和方法,比较可能性理论与概率论方法的差异,并介绍可能性理论涉及的最新热点问题。The conventional research method of uncertainty is using probability theory and statistical approach, however, there are different kinds of uncertainty in the real world, and the conventional methods can not deal with them. The conventional method merely applies to the uncertainty that is generated by the uncertain conditions. That is, given a certain event, the slight deviation of conditions leads to random occurrence. Another kind of uncertainty is the event has fuzzy factors itself and conventional way is invalid here, and possibility theory is necessary in this situation. This paper states the specific rules and approaches of using possibility theory to deal with soft data which is represented by natural language, compares possibility theory with probability theory and introduces the latest information about possibility theory.

关 键 词：可能性理论 模糊集 软数据 软数据推理 BISC 

分 类 号：O159[理学—数学]