基于不确定理论的CBRB规则激活权重计算方法  

A new calculation method of rule active weight for conjunction belief rule-base based on uncertainty theory

作　　者：方炜杰 尹泽锋 巩晓婷[2] 蓝以信[2] 傅仰耿[1] FANG Weijie;YIN Zefeng;GONG Xiaoting;LAN Yixin;FU Yanggeng(College of Mathematics and Computer Science,Fuzhou University,Fuzhou,Fujian 350108,China;Decision Sciences Institute,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]福州大学数学与计算机科学学院,福建福州350108 [2]福州大学决策科学研究所,福建福州350108

出　　处：《福州大学学报（自然科学版）》2020年第3期283-288,共6页Journal of Fuzhou University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61773123);福建省自然科学基金资助项目(2019J01647)。

摘　　要：合取型置信规则库系统的推理方法以概率论为基础,通过乘法实现合取关系,然而这种计算方法并不总能准确反映规则的激活程度.鉴于此,引入不确定理论,改进前件属性合取关系的实现方式,以前件属性中最小的个体匹配度作为核心,并保留信息利用的完整性,提出一种新的规则激活权重计算方法.在此基础上,利用扩展置信规则库系统分别在回归和分类问题上对该方法的推理效果进行验证.实验结果表明:新方法有效提高了系统的推理能力,并能稳定地适用于多个数据集.The reasoning method of conjunction belief rule-based system bases on probability theory, so the conjunction relation is expressed by multiplication. However, such a method may not always exactly reflect the activation degree of rules. Hence, referring to the methodology of uncertainty inference, improving the expression of conjunction relation between antecedent attributes, a new calculation method of rule activation weight is proposed in this paper. The new method regards the minimum individual matching degree of antecedent attributes in each rule as the key and keeps the integrity of information utilization. Using extended belief rule-based system, the reasoning ability of the new method is verified on both regression and classification problems. The experiments illustrate that the new method generally improves the reasoning ability of system, and can apply to several datasets stably.

关 键 词：置信规则库 证据推理 规则激活权重 Dempster/Shafer证据理论 不确定理论 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]