Quine空间与二相不确定度量  

Quine-Space and Measure of Binary Phases Uncertainty

作　　者：洪龙[1,2] 

机构地区：[1]南京邮电大学计算机学院,南京210003 [2]软件开发环境国家重点实验室,北京100191

出　　处：《计算机学报》2013年第4期809-817,共9页Chinese Journal of Computers

基　　金：国家自然科学基金(61170322);软件开发环境国家重点实验室开放课题(SKLSDE-2011KF-04);国家"八六三"高技术研究发展计划项目基金(2009AA043303)资助~~

摘　　要：Pearl指出:不确定是智能系统所面临的核心问题.存在形式是不确定的基本问题,而至今未见不确定的存在形式的报道.文中试图建立一种有关不确定的一般理论框架和度量方法.在区分不确定和不确定性之后,采用向量描述不确定的独立属性,构造不确定的存在形式——Quine空间;在建立属性标准度后,提出单相不确定和二相不确定的度量模型.作为度量模型应用的示例,以中介真值程度讨论模糊性度量,以现代概率理论讨论随机性度量,以模糊随机性讨论二相不确定度量.结果表明:文中所提出的方法具有自然、普适的特征,且具有计算机可以处理的定量形式.文中试图使各种不确定性的度量值归一化,使得它们在程度方面具有可比性.Pearl indicated that uncertainty was a core problem faced by intelligent systems.And an existing modality is the basis of uncertainty.However,to the best of author＇s knowledge,there has not been any report on the research of the existing modality of uncertainty.This paper aims at founding a general theory framework and the corresponding measure method for processing uncertainty phenomenon.Two concepts are distinguished： uncertainty and uncertainty-ness;the unattached attribute of uncertainty is described with vector;Quine-space,one existing modality of uncertainty,is constructed.The attribute standard pointer is established,and two types of the measuring models of uncertainty-ness are proposed： unitary phase and binary phase.As the given demonstrations of measuring models,the measure of fuzziness and the measure of randomness are discussed with medium truth degree and modern probability respectively;fuzzy-randomness is too discussed.The results show that the method proposed in this paper possesses features of natural-ness and universality and quantitative form processed by computers.We attempt to make the measure-value of various uncertainty attributes normalization in order that they may compare with each other in uncertain degree.

关 键 词：不确定 存在形式 框架 模型 度量 应用 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]